Full text of "Designing information : new roles for librarians"

ILLINOIS LIBRARY

AT URBANA-CHAMPAIGN

STACKS

CENTRAL CIRCULATION BOOKSTACKS
The person charging this material is re-
sponsible for its return to the library from
which it was borrowed on or before the
Latest Date stamped below.

Theft, mutilation, and underlining el book* are reason*
for disciplinary action and may result In dismissal from
the University.
TO RENEW CALL TELEPHONE CENTER, 333-84OO

UNIVERSITY OF ILLINOIS LIBRARY AT UUANA-CHAMPAIGN

APR 1 5 1994
MAY 1 3 199i

Trrn REO'D

FEB 03 1997
MAR 2 199?
MAR 2 1997

When renewing by phone, write new due date below
previous due date. L162

Designing Information:

New Roles for Librarians

apers presented at the 1992 Clinic on Library Applications
of Data Processing, April 5-7, 1992

Sponsored by

Graduate School of Library and Information Science
University of Illinois at Urbana-Champaign

Clinic on Library Applications
of Data Processing: 1992

Designing Information:
New Roles for Librarians

Edited by
LINDA C. SMITH

and
PRUDENCE W. DALRYMPLE

Graduate School of Library and Information Science
University of Illinois at Urbana-Champaign

1993 by The Board of Trustees of the University of Illinois
ISBN 0-87845-088-2 ISSN 0069-4789

Printed in the United States of America
on acid-free paper

CONTENTS

Introduction 1

Linda C. Smith and Prudence W. Dalrymple

Embedding the Library into Scientific and Scholarly

Communication through Knowledge Management 5

Richard E. Lucier

Building Electronic Bridges between Scholars and

Information: New Roles for Librarians 19

Carolyn M. Gray

The Gateway to Information: Development,

Implementation, and Evaluation 34

Virginia Tiefel

Design and Development of a Library Information

Workstation 48

Timothy W. Cole, Leslie Troutman,
William H. Mischo, and Winnie Chan

Somebody Knockin': The Public Library at the

Electronic Door 77

Jean Armour Polly

Electronic Information in School Libraries 96

David V. Loertscher

Principles and Strategies for Designing Effective

Computer-Mediated Instruction 1 16

Ruth V. Small

The Development of Computer-Based Training in a

Systematic Staff Training Program 131

Joe C. Rader

Designing for the Computer Screen 147

Ronnie Peters

Contents (Cont.)

Hypertext . . . Information at Your Fingertips 164

Peter Scott

Delivering a Variety of Information in a Networked

Environment 178

Katharina Klemperer

Free-Net in Cleveland and Case Western Reserve University

Library: Linking Community and University 187

Arlene Moore Sievers

New Technology, New Tools, New Librarians:

Shaping the Future 204

M. E. L. Jacob

Contributors 212

Index. .217

Introduction

The twenty-ninth annual Clinic on Library Applications of Data
Processing was held April 5-7, 1992, at the University of Illinois at
Urbana-Champaign. The clinic theme, "Designing Information: New
Roles for Librarians," reflects the availability of increasingly
sophisticated hardware and software that provide librarians with new
tools for designing information. Tools include software for electronic
publishing, database development, and interface design, as well as
hardware and software for hypermedia/multimedia. These developments
present an opportunity for librarians in all types of libraries to assume
new roles and to collaborate with others to produce new products and
services.

INFORMATION DESIGN

Simon (1981, p. 129) has observed that "design ... is the core of
all professional training; it is the principal mark that distinguishes
the professions from the sciences." Increasingly there is recognition
that librarians in the future will have opportunities to serve not only
as "information navigators," but also as architects or designers of
information products (Borah, 1992). Orna (1992, p. 305) suggests that
to be successful as designers, librarians need to

Know about users and what they do

Understand the nature of the information they need

Have understanding and skills in

1. Conceptually organizing information

2. Visually organizing it

LINDA C. SMITH b PRUDENCE DALRYMPLE

The work of Edward R. Tufte (1983, 1990), keynote speaker for
the clinic, provides a rich source of ideas and principles for information
design. His The Visual Display of Quantitative Information and
Envisioning Information offer numerous illustrations of effective means
of communicating complex information. Attention to visual design
"with care given to color, typography, layout, icons, graphics and
coherency" can contribute to the quality and usability of information
delivered through computer screens as well as on paper (Tufte, 1992,
p. 15). Ronnie Peters's paper on "Designing for the Computer Screen,"
included in this volume, offers additional guidance for the task of
organizing a large amount of information in the small area offered
by current computer screens.

NEW ROLES FOR LIBRARIANS

As M. E. L. Jacob observes in her paper summarizing the clinic,
the authors represented in this volume are among the leaders, pioneers,
and early adapters of new technology. Their descriptions of projects
in which they have been involved provide insights into roles that
librarians can fill.

Richard E. Lucier and Carolyn M. Gray explore roles for librarians
in knowledge management. Lucier proposes a new role for librarians
in collaboration with scholars as creators and maintainers of scholarly
and research databases and presents the Genome Data Base at Johns
Hopkins University as a working prototype. Gray describes the Gesher
Project, a joint effort of Digital Equipment Corporation's Cambridge
Research Laboratory and the Brandeis University Libraries to
understand the changing nature of scholarly research and to develop
computer-based tools to assist in these activities.

Two specific design projects to enhance library users' access to
information are described by Virginia Tiefel of Ohio State University
and by a group of librarians from the University of Illinois at Urbana-
Champaign (Timothy W. Cole, Leslie Troutman, William H. Mischo,
and Winnie Chan). Ohio State University's Gateway to Information
provides guidance and instruction for students on how to proceed
through an information search that integrates the use of print and
computerized information. The Illinois Library Information
Workstation project gives integrated and largely transparent access from
a single terminal to a wide range of library resources. Its user-friendly
interface facilitates patron searching of bibliographic databases, with
flexibility to allow terminal-specific customization of the interface to
accommodate localized patron needs and library resources. Both projects
seek to provide "one-stop shopping" for the user and to address problems

INTRODUCTION

that users have in selecting information resources and formulating
questions.

In the area of instructional design, Ruth V. Small reviews principles
and strategies for designing effective computer-mediated instruction,
recognizing that librarians are increasingly asked to design or adapt
instructional programs. Joe C. Rader describes the development of
computer-based materials for staff training at the University of Tennessee
Libraries. Rader's case study explains each step in the development
process, including choice of librarians to serve on the development team,
topic selection, selection of hardware and software (HyperCard),
development of instructional materials, evaluation, implementation,
and replication at another site.

Jean Armour Polly and David V. Loertscher address applications
in diverse environments. Polly demonstrates that the Internet has a
number of resources of potential value to public library patrons. While
eventually users may be able to access such material from home, Polly
sees a place for librarians as long as the Internet remains difficult to
use. To encourage librarians to get connected to the Internet, she
identifies resources for getting onto the Internet and learning more about
it through user guides. Loertscher describes the various ways in which
school librarians have applied technology, identifying certain trends
such as the use of microcomputers in managing school libraries and
the possibilities for involving students in online searching, creating
local databases, data gathering and analysis, and creation of multimedia
productions. Those students who gain experience with storage, retrieval,
and production of text, sound, and pictures will come to expect access
to such technology in public and academic libraries as well.

Librarians have an important role to play in making information
available via networks. Peter Scott explains the use of hypertext tools
in the development of HYTELNET, a tool providing instructions for
and access to information resources available on the Internet. The
challenge is to organize the information necessary to access these diverse
sites in as simple and straightforward a fashion as possible, and Scott
demonstrates that hypertext is well suited for this purpose. Katharina
Klemperer describes the different categories of information resources
that libraries handle (indexes, structured full text, full text, numeric,
and multimedia) and the different needs of each with regard to access
and delivery. The challenge is to develop the tools that will accomplish
this. Arlene Moore Sievers provides an overview of the Free-Net concept
and its operation in Cleveland through Case Western Reserve University.
Sievers notes that major libraries in Cleveland have been actively
involved in Free-Net from the beginning and that public libraries have
been active in Free-Nets in other cities. Librarians can both contribute
to the development of information resources made available through

LINDA C. SMITH b PRUDENCE DALRYMPLE

Free-Nets and use such databases as community resource files to enhance
their services to library users.

OTHER COMPONENTS OF THE CLINIC

In addition to the papers assembled in this volume, the clinic
included a keynote speech by Edward R. Tufte on "Envisioning
Information" and an illustrated talk by Richard Greenfield entitled
"Tying It All Together: Designing Graphical User Interfaces to Integrate
and Evaluate Information Resources." The clinic began with three
preconference workshops covering desktop publishing (presented by
Nan Goggin and Kathleen Chmelewski), database design (presented
by Carol Tenopir and Gerald W. Lundeen), and expert systems (presented
by Lloyd A. Davidson, Judy E. Myers, and Craig A. Robertson), made
possible with support provided by the Council on Library Resources.
A poster session gave several clinic participants an opportunity to make
presentations on topics related to the clinic theme. Presenters included
James E. Agenbroad on "Browsing Classification Data: Feasible?
Useful?," Mark Crook and Craig Henderson on "OCLC's Batch
Services," James S. Foster, Javed Mostafa, and Beatriz Calixto on "CAI
Packages for Microcomputer Competency," Gregory B. Newby on
"WAIS: A New Model for Information Retrieval," and Eric Rumsey
on "Use of HyperCard to Teach Medline CD-ROM." The editors
gratefully acknowledge the contributions of all these individuals to the
success of the clinic.

LINDA C. SMITH
PRUDENCE W. DALRYMPLE

Editors

REFERENCES

Borah, E. G. (1992). Beyond navigation: Librarians as architects of information tools.

Research Strategies, 70(3), 138-142.
Orna, L. (1992). Information design and information services: What information

professionals should know about design, in order to deliver value-added information

products. Aslib Proceedings, 44(9), 305-308.

Simon, H. A. (1981). The sciences of the artificial (2nd ed.). Cambridge, MA: MIT Press.
Tufte, E. R. (1983). The visual display of quantitative information. Cheshire, CT: Graphics

Press.

Tufte, E. R. (1990). Envisioning information. Cheshire, CT: Graphics Press.
Tufte, E. R. (1992). The user interface: The point of competition. Bulletin of the American

Society for Information Science, 18(5), 15-17.

RICHARD E. LUCIER

University Librarian and Assistant Vice Chancellor
for Academic Information Management
University of California, San Francisco

Embedding the Library into Scientific

and Scholarly Communication
through Knowledge Management

ABSTRACT

Knowledge management is a new role for academic research libraries
that has the potential to integrate the library into scholarly and scientific
communication in a significant way. Work in knowledge management
is advancing in both the sciences and humanities. The Genome Data
Base at the Johns Hopkins University is currently the most advanced
knowledge management prototype. As part of its new Center for
Knowledge Management, the University of California, San Francisco
is undertaking several initiatives to create a campuswide knowledge
management environment.

INTRODUCTION

The University of California, San Francisco (UCSF) is one of the
nine campuses of the University of California (UC) system. With schools
in medicine, pharmacy, nursing, and dentistry, and graduate programs
in the behavioral and social sciences, UCSF is unique within UC in
that it is the only campus devoted to research, education, and service
in the health sciences.

1992 Richard E. Lucier

RICHARD E. LUCIER

In September 1990, as the result of a decade of planning, UCSF
opened a new library building of great beauty and utility that is a
visual representation of the importance of the library to the UCSF faculty
and student community (Cooper, 1991). The critical challenge in the
current decade is to articulate and realize a programmatic vision that
will (a) embed the library into the scientific and clinical research,
educational curricula, and professional practice programs of this diverse
and distributed campus; (b) position the library as a campus focal point
for knowledge-based applications of information technology; and (c)
establish the library's leadership in the development of knowledge bases
and online tools for the health sciences.

OUR VISION:
A KNOWLEDGE MANAGEMENT ENVIRONMENT

Historically, the function of the research library has been storage
and retrieval. This will remain at the core of the library's responsibilities.
More recently, the library has extended its role to include information
transfer, or the delivery of information over high-speed communications
networks. Responsibilities and activities in this area are increasing
rapidly, driven by users' needs and the growing availability and
reliability of the Internet or the National Research and Education
Network (NREN). A new, more experimental and challenging role for
the library is that of knowledge management, the insinuation of the
library at the beginning of the scientific and scholarly communication
process for the purpose of building and maintaining specialized
knowledge bases in unique collaborations with scientists and scholars.

Our vision for the UCSF Library, and its innovative new Center
for Knowledge Management, embraces all three functions: storage and
retrieval, information transfer, and knowledge management. Figure 1
graphically represents this vision, which we call a Knowledge
Management Environment.

This Knowledge Management Environment is an integration of
knowledge sources, access and delivery systems, education and training
programs, and personalized services with the following components:

online bibliographic databases of the library's physical collection;

the "full text" of the published literature online, including images;

high-quality, interactive knowledge bases critical to the daily work
of scholars and scientists;

online tools for the peer review of data and collaborative knowledge
base management;

high-speed communications for the conduct of scientific and scholarly
work from the local to international levels; and

KNOWLEDGE MANAGEMENT

an integrated access tool, or wide area information server, to retrieve
information from local and remote bibliographic databases, "full-
text" information sources, and specialized knowledge bases.

Figure 1. The Knowledge Management Environment

SCIENTIFIC AND SCHOLARLY COMMUNICATION

The need for a Knowledge Management Environment emerges from
problems inherent in the current scientific and scholarly communication
process. Figure 2 depicts the information transfer cycle as we know
it today. Scientists and scholars discover new knowledge and commun-
icate it through both writing and teaching. Publishers disseminate that
information through a variety of primary and secondary information
products. In their traditional storage and retrieval role, research libraries
build collections and make available to users the world's published
literature. Since the 1970s, network access (information transfer) to this
stored knowledge through online catalogs and indexes, along with a
new emphasis on service and education, has assumed major importance.
However, the roles and functions of scholars, publishers, and librarians
have remained fundamentally the same.

By the late 1980s, the limitations of this prevailing model for
scientific communication were becoming apparent. The length of the
hard-copy publishing process makes it increasingly difficult for scholars

RICHARD E. LUCIER

Scientists and Scholars

Discovery and Communication

Libraries/Vendors

Storage and Retrieval

Publishers/Libraries

Indexing and Cataloging

Figure 2. The information transfer cycle

and scientists to communicate their findings in a timely fashion. With
the rising cost of publishing and a limited resource base, libraries and
universities can no longer afford to support comprehensive collections.
The financial crisis facing libraries is not short-term; rather, it is
structural in the current environment. Most importantly, it is clear that
the presentation of knowledge in static form, whether in print or as
part of the emerging electronic library, is grossly inadequate. Scientists
and scholars, often on their own and with inadequate support, are
augmenting this passive presentation of knowledge with a growing
number of interactive, discipline-based knowledge bases that are
developed, maintained, and shared across networks. Knowledge
management has emerged from this situation as a creative response
to managing the world's knowledge base.

THE LIBRARY AS KNOWLEDGE MANAGER

Knowledge management represents a new model for scientific and
scholarly communication in which faculty and research librarians share
the responsibility for the collection, structuring, representation,
dissemination, and use of knowledge using electronic information
technologies. Encompassing the entire information life cycle, from
creation of new knowledge to its dissemination and use, knowledge
management is a collaborative enterprise, where scholars, scientists, and
research librarians work together to develop and maintain knowledge
bases and derivative information products. Knowledge bases are
developed and maintained through knowledge management processes,
which ensure content integrity and usefulness. A variety of products
and services can be derived from the knowledge base. The collaborative
nature of knowledge management, embodied in techniques of shared

KNOWLEDGE MANAGEMENT 9

development of functional specifications, rapid prototyping, and user
acceptance testing, fosters an interdependency among all involved. The
ongoing management of the knowledge base also requires funding and
administration strategies that crosscut traditional departmental,
disciplinary, and institutional boundaries.

Knowledge management consists of four primary components:

1. Collaboration: the shared responsibility for the development and
management of knowledge bases, products, and services. Effective
collaboration requires a balanced relationship among peers,
recognizing the unique value of each person's contributions to the
success of shared work. A multidisciplinary team of collaborators
includes discipline-based scholars and scientists, librarians, computer
scientists, and software engineers.

2. Knowledge base: a collection of scholarly knowledge structured for
computational storage and representation. A knowledge base may
contain all or some part of the intellectual core of a scholarly
discipline. The contents of the knowledge base are chosen and
validated by consensus at some level within the scholarly community
that develops, uses, and maintains it.

3. Knowledge management processes: those activities of collaborators
related to the creation, structuring, representation, dissemination,
and use of scholarly knowledge. They result in knowledge bases,
patterns of collaboration and communication that ensure the integrity
and continuing usefulness of those knowledge bases, and knowledge
products.

4. Knowledge products and services are the output derived from the
knowledge base: books, articles, computer-based educational ma-
terials, database subsets, and typesetting tapes are examples of know-
ledge products. Knowledge products are market driven, developed
in response to the immediate information needs of scholars, scientists,
educators, students, and other information seekers. Product services
are the customer support activities associated with each knowledge
product. Examples include production of typesetting tapes or camera-
ready copy for hard-copy publication or education and training
programs to provide skills and abilities needed for full and appro-
priate use of knowledge products derived from the knowledge base.

What is remarkably different about the knowledge management role
is that it insinuates the library at the beginning of the information transfer
cycle rather than at the end and focuses on information capture rather
than access and use (Figure 3).

The long-term implications for building and maintaining a portion
of the library's collection in this manner are enormous. Knowledge
management transforms the various roles in the scientific communication

RICHARD E. LUC1ER

process and potentially places ownership and control back in the hands
of the scholarly community. It also has enormous potential for closing
the gap between research faculty and their students and integrating
the library into research and education programs in a significant way.

Scientist* and Scholar*

Discovery and Communication

Figure 3. New roles in a Knowledge Management Environment

THE GENOME DATA BASE:
A KNOWLEDGE MANAGEMENT PROTOTYPE

To date, the most advanced knowledge management prototype is
the Genome Data Base (GDB), developed at the Laboratory for Applied
Research in Academic Information, William H. Welch Medical Library,
the Johns Hopkins University. (The following section is an adaptation
of sections from Lucier [1990].) GDB is a working prototype, which
serves the international scientific community on a daily basis. The most
technologically advanced systems possible are not our primary goal in
knowledge management; instead we are more concerned with designing
systems that work and that people use in their everyday environments.

GDB is a gene-mapping database that serves as a repository for
data collected by scientists engaged in the international human genome
effort. GDB integrates several types of data including descriptions and
map locations of human genes and other markers, descriptions of DNA
probes used to characterize the markers and polymorphisms, contacts
for obtaining probes, and more than 25,000 linked bibliographic citations.

To see GDB as an example of knowledge management, it is essential
to have an understanding of the sociology of the human genetics
community, namely the Human Gene Mapping (HGM) Workshops.
The First International Human Gene Mapping Workshop, held in 1973,

KNOWLEDGE MANAGEMENT 11

was instituted to develop and maintain a consensus human gene map.
Since that time, similar workshops have been held either annually or
biennially. These workshops are one of the community's primary data
filters. The HGM workshops are organized by committee, one for each
chromosome as well as several specialized committees, e.g., nomencla-
ture, DNA, mitochondrial, and comparative (mouse). These committees
collect, review, analyze, and synthesize all the mapping data from the
published literature to produce the consensus human gene map.

Two aspects of HGM work have specially driven the need and design
for GDB: (a) the growing volume and complexity of data and (b) the
interactive character of the peer review work of HGM committees. The
amount of information that committees must process has increased
proportionately with the greatly heightened scientific activity in this
area. It is estimated that the information doubles every two years. In
1973, 75 people attended HGM 1, and 25 genes were mapped. At HGM
10 in 1989, 700 scientists were in attendance, and 1,630 genes were
mapped. Until now, during the four- or five-day workshop, members
would collect and input information concerning their particular
chromosome. With the larger volume of data, this has become a nearly
impossible task, even with the introduction of computers to the
workshops beginning in 1983. Making use of the Internet and public
data networks, GDB provides the committees with an online,
continuous, interactive system into which information can be added
and verified at the committee members' convenience throughout the
year. The committees will continue to meet annually, but the workshops
can now focus on science and the analysis and significance of data
rather than on data entry.

Although the various chromosome committees do their work mostly
independent of each other, there is considerable interaction among the
chromosome, nomenclature, and DNA committees. Certain data
elements are shared; these elements, e.g., gene symbol, cannot become
part of the database until they have been validated by the appropriate
members of various committees, in consultation with each other. An
"online peer review process" has been integrated into GDB editorial
interfaces, again making significant use of national and international
networks and a completely modularized design.

At present, GDB draws primarily on the HGM workshops and
the literature for the major portion of its data. Already, it is beginning
to include unpublished and unvalidated data submitted directly by users
for consideration and subjected to quality control by both GDB staff
and a special group of scientific editors. Figure 4 illustrates this data
flow and highly dynamic form of scientific communication possible
in a networked environment. It also represents a true electronic journal
in a knowledge management environment.

RICHARD E. LUCIER

DM* SOURQS

HGMW
Literature
Other Databases
Scientific Community

Invalidated Data

Validated Data

GDB User Community

Figure 4. GDB data flow in a networked environment

GDB is designed so that it is possible to develop other information
products that the user community demands in order to accomplish its
work. In addition to the various interfaces provided for HGM committees
and GDB Editors, a more generalized online searchable version of GDB
is available to the scientific community. The HGM Reports, published
by Karger in a special issue of the journal Cell Genetics and Cytogenetics,
are produced from GDB data structures.

IMPLEMENTATION OF A KNOWLEDGE MANAGEMENT
ENVIRONMENT AT UCSF

Although the library will continue to build a high-quality paper-
based collection in the health sciences, excellent service in a distributed
environment as well as educational programs will assume a far higher
priority than in the past. We anticipate a rapidly increasing emphasis
on information transfer and knowledge management over the next 10
years, and we will focus our technology-based efforts on these roles.

Figure 5 depicts the primary areas in which we plan to develop
or adopt technological innovations over the next three to five years,
as we implement the first phase of our Knowledge Management
Environment.

Driven by the needs of our customers, the continuously changing
external environment, and new advances in technology, we are fashion-
ing a dynamic, multidisciplinary organization with three programmatic
divisions.

KNOWLEDGE MANAGEMENT

CONTENT

ACCESS

EDUCATION

SERVICE

STORAGE

Bibliographic db's

RETRIEVAL

Online Indices

'Full Text'

INFORMATION
TRANSFER

Delivery of 'Full Text'
information
on-demand

Wide Area
Information Server
(GALEN) for inte-
grated access to
local/remote db's

Training
programs
which facilitate
access and use

Information
services to
distributed faculty
and student
workplaces

KNOWLEDGE
MANAGEMENT

Discipline-specific
knowledge bases of
high currency,
value, and integrity

Interactive tools for
collaborative data
maintenance in a

Information
Retrieval Software
for 'Full Text'
Electronic Library

Online tools for
information
access and analysis

Practice-based
health sciences
Informatics
Curriculum

Information
products to support
curriculum

Direct collaboration
between faculty
and librarians in
the development/
maintenance of
knowledge bases
and products

networked

environment

Figure 5. Development areas for the first phase of the Knowledge
Management Environment

1. Information Resources and Services is responsible for storage and
retrieval and information transfer functions including the following:

collection management and processing;

public information services to our distributed customer base;

document and information delivery;

special collections including the History of the Health Sciences,
the Oriental Collection, and University Archives; and

bibliographic instruction component of a broad educational
program.

2. The Center for Knowledge Management, created by the Division of
Academic Affairs and the library, is responsible for information
transfer and knowledge management functions. In collaboration with
faculty, the center's staff performs the following functions:

develops new information products and services, e.g., knowledge
bases and online tools for the health sciences;

pursues applied research projects related to UCSF informatics
problems;

advises graduate students in computer science and the health
sciences who are using the center as their laboratory;

consults with faculty, staff, and students in the development of
private databases, etc.; and

14 RICHARD E. LUCIER

supports the state-of-the-art systems and infrastructure that
underpin the development, maintenance, and use of knowledge
resources and information services.

3. The Interactive Learning Laboratory has primary responsibility for
our educational and instructional programs including the following:

development of a health sciences informatics curriculum;

integration of educational technology resources into the curriculum
of the various schools and professional training programs;

instructional computing and the development of multimedia
software for education; and

educational and external publications.

A fourth division, Finance, Planning, and Administration, supports
our storage and retrieval, information transfer, and knowledge
management functions through the efficient and effective management
of our financial and human resources and facilities. This division is
also responsible for development. In order to implement the UCSF
Knowledge Management Environment, it is critical for the library to
implement long-term financial planning for the effective use of state
funds as well as broaden its financial resource base beyond state-
appropriated funds. Important sources of support include grants,
contracts, business-university agreements, gifts, and information
consulting and brokering activities. An endowment for the Center for
Knowledge Management has been established as an important priority
in an upcoming campus capital campaign.

As we move towards realizing our Knowledge Management
Environment vision, it has also been necessary for us to examine and
refashion the library's organizational culture as well. Several principles
guide us in this challenging and long-term task that has been greatly
aided by the opportunity to recruit several new professional staff from
the library, computing, and biomedical science professions:

high value placed on technological innovations that solve practical
and recognized problems;

continuous involvement of faculty, staff, and students in the
University of California tradition of shared governance;

an informed, knowledgeable, and service-oriented staff a critical
factor;

technology a tool, not an end;

strong management essential for program development and the
effective use of human and financial resources;

processes and tasks organized around outcomes;

pragmatism and principle as a dual basis for decision making;

outcomes as the principal evaluation measure; and

KNOWLEDGE MANAGEMENT 15

entrepreneurial responsiveness to environmental changes, opportun-
ities, and emerging information technologies a key to success.

Several new projects have already begun. In collaboration with our
Human Gene Mapping Center, we have successfully sought funding
to build and maintain a chromosome 4 database, which will be our
first efforts at collecting and making available source data. Discussions
are continuing with (a) Springer- Verlag for an experiment with several
important online journals, (b) clinical researchers for an AIDS
knowledge base, and (c) medical educators for the creation of a
comprehensive database that would support undergraduate medical
education.

CLR STUDY OF KNOWLEDGE MANAGEMENT

In 1987, the Council on Library Resources (CLR) awarded a grant
to Richard E. Lucier and Nina W. Matheson to address the changing
roles of research libraries, the scholarly community, and university
publishers in scientific and scholarly communication through
examination of the knowledge management model as implemented in
Lucier's work at the Laboratory for Applied Research in Academic
Information, the William H. Welch Medical Library, the Johns Hopkins
University. (This following section is an adaptation of sections from
Lucier & Matheson [1992].) The CLR grant had three major objectives:

1. Documentation of the knowledge management model. The collection,
examination, and synthesis of statements, definitions, and
descriptions of the knowledge management model and its components
have been major documentation activities of the project. Briefing
materials for the Symposium on Knowledge Management drew
heavily upon these files. A monograph on the knowledge manage-
ment model, coauthored by the principal investigators, will be
published by the Johns Hopkins University Press in late 1992.

2. Diffusion of the knowledge management model to academic settings
outside medicine. Initial diffusion of the concept of the model
occurred through presentations made by the principal investigators
to high-level staff at numerous academic institutions that seemed
to possess the requisite human, technical, and financial resources
to implement the model. Follow-up calls, interviews, and site visits
monitored the possibility of actual implementation in these settings.
Presentations were also made at several national meetings over the
life of the grant.

3. Sponsorship of a national meeting on knowledge management. Early
on, the principal investigators formed a special executive committee

16 RICHARD E. LUCIER

to oversee this component of the project. This group decided on
the strategy of a special invitational symposium as most appropriate
to a full discussion of the key issues raised by the knowledge
management model.

From October 27-29, 1991, 63 invited guests gathered at the Coolfont
Conference Center in Berkeley Springs, West Virginia, for the Invita-
tional Symposium on Knowledge Management, a policy-level forum
for examination of knowledge management. Included among these
experts were scholars, university administrators, academic librarians
from major public and private universities, association directors,
independent consultants, and others whose work and interests have led
to innovations in scholarly and scientific information management.
Major private and public funding agencies such as the National Science
Foundation, the National Endowment for the Humanities, and the
Andrew W. Mellon Foundation were also represented. During the three-
day gathering, participants' time was divided between working group
meetings and plenary sessions.

Symposium registrants directed their attention and activities toward
four desired outcomes:

1. Shared understanding of the knowledge management model,
including the economic and political advantages and disadvantages
of different approaches and social and other noneconomic barriers
to wider implementation of knowledge management.

2. Clarification of implications for scientific/scholarly communication,
comparing the current situation to communication in knowledge
management environments, and suggesting solutions for problem
areas.

3. Scenario development outside human genetics, applying knowledge
management to other scholarly information problems; examining
existing knowledge management projects, especially in the human-
ities; and identifying the advantages, disadvantages, opportunities,
and barriers to knowledge management within particular disciplines.

4. Recommending implementation strategies for knowledge manage-
ment, providing a rationale, time frame, level of intensity, projected
resource requirements, technological initiatives, and, where possible,
priority audiences.

At the symposium's concluding session, the leaders of these groups
presented recommendations for future actions in each of the five areas.
Strong consensus emerged in support of wider implementation of know-
ledge management. In the words of Donald N. Langenberg, registrants
should take action to colonize carefully selected distant locations in

KNOWLEDGE MANAGEMENT 17

intellectual space with the practice of knowledge management. Groups
also pinpointed a set of complementary actions to develop the conceptual
and curricular infrastructure for knowledge management environments.

Knowledge management is a transformational activity. Working
examples of knowledge management serve as proof of concept for the
approach. They also help to highlight the areas where immediate work
is needed if an infrastructure to nurture new implementations is to
emerge in the next three to five years. Next steps involve actions with
national and international impact; individual initiatives must be
supplanted by broader based, mainstream action targeted to reduce
barriers and leverage opportunities.

Themes running through the plenary and working group
discussions and recommendations highlight three action items that
require immediate attention:

1. Financial strategies. The future diffusion and success of knowledge
management rests in large part on the development of reasonable
and creative financial strategies and on an economic model that
considers the needs of all important players. In particular, the model
must consider that the current state of research institutions and
information producers, in an era of limited resources and constrained
public agencies, requires the ability to mesh pricing, costing, and
allocation strategies among various organizations and groups both
internal and external to the research enterprise.

2. Intellectual property. As a next logical step in moving towards a
knowledge management environment, it is increasingly important
to convert existing published works to electronic form for online
access and management. The symposium's work group on intellectual
property proposed pulling a group together to describe the climate
needed for knowledge management, particularly the elements of
collaborative ownership. Such a group would identify current copy-
right status for each class of information (e.g., source data, consensus
data, the published literature, and bibliographic records), project what
is needed, and recommend actions to be taken over the next five
years.

3. Technology strategy. A functional architecture that will serve as a
reference model is needed for knowledge management. This structural
definition can serve as a rationale for institutional infrastructure
planning and technological development. The architecture should
take into account the available technologies but must also offer a
plan for incorporating future developments. Though there will
ultimately be several architectures for knowledge management, a
general one is needed to begin with that defines how to deal with
communications, content organization, standards, and other related
issues.

18 RICHARD E. LUCIER

CONCLUSION

It is clear that the knowledge management concept is a vital, effective
approach to scientific communication in networked environments.
Working implementations of knowledge management exist, and numer-
ous projects in the sciences and humanities can be identified where
the knowledge management approach will provide identifiable benefits
to disciplines and institutions. Wider implementation of knowledge
management approaches requires that the focus of action and attention
be redirected to issues beyond those that arise from individual university-
or discipline-specific projects. Enthusiasm exists for initiating new
knowledge management experiments in a number of disciplines, but
it is not likely that any coordinated effort can emerge until additional
work is done to reduce technological, legal, and financial barriers. The
involvement of new participants, including people who bring legal
and economic expertise and who share an interest in and commitment
to shaping new roles and processes in scholarly and scientific
communication, is critical.

REFERENCES

Cooper, R. S. (1991). A library for the fifteenth through the twenty-first centuries. Bulletin

of the Medical Library Association, 79(2), 147-158.
Lucier, R. E. ( 1990). Knowledge management: Refining roles in scientific communication.

EDUCOM Review, 25(3), 21-27.
Lucier, R. E., & Matheson, N. W. (1992). Invitational symposium on knowledge

management: Overview and recommendations. Baltimore, MD: Welch Medical

Library.

CAROLYN M. GRAY

Associate Director

Brandeis University Libraries

Waltham, Massachusetts

Building Electronic Bridges between Scholars
and Information: New Roles for Librarians

ABSTRACT

Through a description of information science, communications, and
knowledge utilization information models, this paper provides an
introduction to the conceptual framework for the use of information
in knowledge work activities and outlines one approach to studying
knowledge work. The Gesher Project a design effort undertaken by
Brandeis librarians and Digital Equipment Corporation software
engineers is presented, with details of a group study of the Brandeis
Radio Astronomy Group (BRAG). A prototype information manage-
ment system developed by Digital Equipment Corporation researchers
is also described.

INTRODUCTION

The traditional response of librarians to the study of information
needs has been to study what users are doing in the library studying
how they are using information that is available. Maurice Line has
suggested that we should instead hypothesize about need based on the
nature of the activities in which individuals are involved (Line, Brittain,
& Cranmer, 1971). Whether one is involved in designing a new bib-
liographic tool, designing a new approach to library instruction, or
designing a new library system, it is important to understand what
the needs of end-users are in relation to a specific information activity.

20 CAROLYN M. GRAY

Libraries have been developing in an evolutionary process in
relation to information products and information services. Each new
product has been built upon the models of the past. Edward Tufte (1990)
has suggested that we must envision information and information
activities in a different manner in order to develop products and services
that are truly revolutionary. If one assumes that a major library
constituency is the scholars and researchers in the user community, then
one can begin to think about their "needs." Instead of building better
tools based upon historical precedent, librarians can begin to think
beyond the confines of the past by beginning to examine the knowledge
work of scholars.

This paper provides an introduction to the conceptual framework
for the use of information in knowledge work activities, outlines one
approach to studying knowledge work, and presents an overview of
a design effort undertaken by Brandeis librarians and Digital Equipment
Corporation software engineers.

INFORMATION AND KNOWLEDGE WORK

Consider three potential outcomes that are desirable and possible
by examining "knowledge work":

Through the process of examining scholarly activity, librarians may
be able to assist in the development of scholarly support software
that is not just an electronic analogue of existing electronic- or print-
based library reference works but a medium for a kind of scholarly
support activity that is genuinely new and three-dimensional.

By examining how scholars use information, librarians can develop
simulations of complex information activities. It is possible to sim-
ulate for the undergraduate the complex situation of a scholar analyz-
ing a literary text. Librarians can develop truly innovative, interesting,
and educational library instruction programs.

By examining how scholars use information, we can begin to develop
collections and connections to collections that truly reflect scholars'
needs and support their scholarly work in an enabling fashion.

Elsewhere, the author (Gray, 1992) has suggested that information
is a dynamic process with distinct phases forming a life cycle that can
be defined, isolated, and examined. In this criterion, information is
said to be a dynamic process, to be diverse and cumulative in effect,
and to lead to informed action (Kochen, 1970; Taylor, 1980, 1986).
Following are three types of models that contribute to understanding
the dynamic nature of information. It may be helpful to view information
as having a distinct life cycle that begins with creation, involves

B UILDING ELEC TR ONIC BRIDGES 21

dissemination, collection by a potential user, analysis, subsequent use,
and storage. Various iterations of these phases in the information life
cycle are present in the models described. To study the "information
life cycle," one must combine elements of each of the models into an
iterative model that includes data collection, analysis, action, and
feedback loops.

To understand the concept of "information" as a dynamic process,
various linear models that depict information on a continuum are
reviewed. The Kochen (1970) model depicted in Figure 1 shows a pro-
gression from information to wisdom with two intermediate transfor-
mations along the way.

Information- > transformation into knowledge-- >
assimilation of knowledge into understanding- >
fusion of understanding into wisdom

Figure 1. Kochen information model

Another version of the Kochen model can be seen in Taylor (1980)
as he outlines four steps for refining data and transforming it into
information for decision making. The first step is the organization of
data, in which he includes collection, sorting, grouping, classifying,
formatting, presenting, and displaying. The second step is synthesizing,
which is a systematic approach to selecting, analyzing, interpreting,
adopting, and compressing. The third step is judgment, which is a
more critical act of selecting and evaluating against established
parameters. The final step in the model is the decision process. In the
decision process, useful knowledge is assessed and decisions are made
based upon the goals of the organization or decision maker. Choosing
among alternatives, compromising, bargaining, and consultation with
experts are all elements of this final process.

The Taylor model is often depicted as a pyramid rather than as
a linear model, with the raw data forming the base of the pyramid
and the decision process forming the apex. In a later work, Taylor ( 1986)
retains the steps but depicts the model as an hierarchical spectrum from
data to action.

The Taylor model shown in Figure 2 illustrates the unrefined "data"
at the base of the hierarchy, followed by "information," the first level
of refinement or organization of data. "Informing knowledge" is

22 CAROLYN M. GRAY

organized and synthesized to create in the recipient some greater
understanding. "Productive knowledge" has attached to it some form
of critical or evaluative element. The apex represents the action of the
decision maker.

Action

1W* i Wil

Productive Knowledge

i |-

Informing Knowledge

11 I : :vK

Information

Dkta

Figure 2. Taylor information model

Variations on these models can be found in the management
information systems (MIS) literature. Boulton and Saladin (1983) and
Hodge, Fleck, and Honess ( 1984) depict a flow from raw data to a decision
point in their information system continuum illustrated in Figure 3.
The "data processing" step is roughly analogous to the "informing
knowledge," and "data output" corresponds to "productive knowledge"
in the Taylor model. In the MIS school of research, "information
utilization" has as an underlying assumption that some refining process
has been undertaken to turn raw data into useful information.

One may contrast these models to a model used in communications
theory as depicted in Davis and Olson's (1985) general model of a
communication system. The above models place more emphasis upon
the use, and Davis and Olson study the process. The communications
school approach is depicted in Figure 4. The communications model
begins with a message or information source, a transmission device,
a channel or conduit through which the message travels, and a receiver

BUILDING ELECTRONIC BRIDGES 23

Raw data or Input -->
Data Processing >
Data Output- >
Information-- > Decision

Figure 3. MIS information model

source- > transmitter- >
channel > receiver/decoder- >
destination

Figure 4. Communications model

that relays the message to its destination. In this model, there is concern
that the message remain intact from source to destination. The ideal
is for the "destination" to understand the message as it was intended
by the "source." Noise and distortion often arise in the channel. Thus,
the communication approach is concerned with maintaining the
integrity of the message, and the information-processing approach
focuses upon transformation.

These variations reflect the difference in approaches between the
information scientist and the communications theorists. There is yet
a third approach that arises out of the knowledge utilization literature.
Havelock (1972, 1976) develops what he calls a knowledge flow system.
The unique nature of this system is that it is not a strict linear model
but has a series of "feedback loops."

Figure 5 depicts a strict linear model, but in fact there are a series
of feedback loops with information from applied research feeding back
to basic research, e.g., engineers feeding information back to basic re-
search scientists, information from consumers being fed back to prac-
titioners or retailers, or practitioners feeding information to the applied
researcher to create understanding of what is or is not working. The
field of knowledge utilization is primarily concerned with studying the
flow of research to practitioners. Some of the earliest studies in knowledge
utilization were done in the agricultural field, which studies the

24 CAROLYN M. GRAY

utilization of advances in applied agricultural techniques literally at
the "grass roots" level. More recently, we see the use of knowledge
utilization studies in the human services field to assess the use of
innovations (both techniques and technology).

Basic research- >
Applied Research -->
Practitioners/Producers/
Manufacturers/Retailers- >
Consumers/Clients/Citizens

Figure 5. Havelock knowledge system

The information science, communications, and knowledge
utilization information models contribute to understanding the dynamic
nature of information. This conceptual framework is a part of the
foundation needed for the study of knowledge work activity. Knowledge
work may entail using the scientific method of research, or it may involve
activities such as literary analysis that cannot be framed by the scientific
method. Knowledge work assumes reliance upon information-intensive
sources for "work" to be accomplished. It may be helpful to examine
one attempt to frame the concept of knowledge work by analyzing the
activities that may be involved when one engages in knowledge work.
Davis and Olson (1985) identified seven major categories of knowledge
work activity:

Diagnosis and problem finding
Planning and decision making
Monitoring and control
Organizing and scheduling
Authoring and presentation
Communication
System development

Davis and Olson's divisions of knowledge work are complemented
by Mackenzie Owen and van Halm's ( 1989) description of the information
cycle that includes the following:

Production

processing of data

BUILDING ELECTRONIC BRIDGES 25

text processing
communication

Distribution

editing (link between production and distribution)
quality control, e.g., peer review
marketing
physical production

Acquisition (booksellers and libraries)

selection, physical acquisition, and storage of materials
cataloging and indexing
provision of documents by selling or lending
instruction and advice to the end-user

Use

orientation

problem formulation

searching and selecting relevant information

acquisition of selected information

processing of information

establishing relationships between various items of acquired

information
production of new information

The Gesher Project team began their research with this broad
conceptual understanding of knowledge work. The project is designed
to build a bridge between the information cycle and scholars to support
their knowledge work.

GESHER PROJECT

What follows is a description of a joint project between Brandeis
University Libraries and Digital Equipment Corporation's Cambridge
Research Laboratory (CRL) that seeks to understand the changing nature
of scholarly research and to develop computer-based tools to assist
scholars in their research activities. This joint project is entitled the
"Gesher Project." Gesher is a Hebrew word meaning bridge, which
is intended to symbolize a link between the scholar and the scholarly
information most traditionally found in libraries.

The Gesher Project had its birth when computer scientists at
Digital's CRL and librarians at Brandeis began to discuss the possibility
of building a personal information management system for scholars
that would use bibliographic data from our online catalog. As discussions
evolved, we tried to imagine what scholars might want to do and how
a system might be designed to meet their scholarly information needs.
A project was designed with the following broad goals:

26 CAROLYN M. GRAY

1. to study the information-seeking process within a research university
setting as a paradigm of how people locate and utilize information
in the course of their work, and

2. to assist DEC/CRL staff in evaluating a personal information
management system to be developed by CRL project members with
participation by Brandeis faculty, doctoral candidates, and senior
research librarians.

As the project team began work, a set of assumptions were developed
that have helped to guide the research. These assumptions include the
following:

1. Scholarly research is changing.

2. Understanding the scholarly research process can help librarians
design services to address the changing needs of researchers. (See
Belkinetal., 1990.)

3. Scholars must participate in the design of any new and improved
system that aims to help manage their scholarly information.

4. Ethnographic field research techniques are useful in understanding
the scholarly process.

5. The role of academic librarians in relation to scholarly research
support is changing or needs to change.

6. Skills for librarianship are changing.

Grounded in the conceptual foundations of knowledge work, with
the broad project goals in mind, and these diverse assumptions, the
project team began its research.

Participative Design

Librarians in the project have concentrated on the goal of
understanding the scholarly information management process. The
research team from CRL and Brandeis decided to use a participative
design technique in the system development. To understand our initial
work, it is helpful to have a general knowledge of the tenets of
participative design. Participative design is a technique pioneered by
Mumford and MacDonald in the 1980s in their expert system design
efforts. Participative systems design means giving responsibility for all,
or part, of the design of a new system to the group who will use it.

Participative design is a concept that is best applied in a single
organization where workers are engaged in a common pursuit. Scholarly
research, of course, varies by discipline. But as Mumford and MacDonald
(1989) point out, "Participation is a means to an end and not an end
in itself. It is there to assist the creation of good systems that work
efficiently, increase human effectiveness and contribute to a stimulating
and satisfying work environment" (p. 27).

BUILDING ELECTRONIC BRIDGES 27

The participative design technique developed by Mumford and
MacDonald is called "The ETHICS Method." (ETHICS stands for
Effective Technical and Human implementation of Computer-based
Systems.) Initial examination made it apparent that the ETHICS method
was not appropriate to adopt wholesale for the Gesher Project, but
rather it was decided to use the elements of the method that seemed
to match project needs. The five steps in participative design follow:

Step 1. Diagnosis of Need
Step 2. Discrepancy Analysis
Step 3. Agreeing on Objectives
Step 4. Designing the System
Step 5. Implementing the System

The initial research efforts concentrated on step 1, which involves
the following:

Describing the existing scholarly research systems being used by
scholars at Brandeis. In this descriptive process, it is important to
look at day-to-day tasks in the scholarly process of collecting
information, storage, and retrieval.

It is important to assess the efficiency needs of the scholar by asking
such questions as the following: What are the problems impeding
progress in the scholarly process? Slowing it down? Causing errors?

It is important to assess the effectiveness needs by describing the
key tasks and establishing which contribute to scholarly goals. One
can do this by asking two questions: Are the tasks being performed
effectively? Are there tasks that are not being performed that should
be introduced? In scholarly work, effectiveness may be related to the
coordination of activities of other scholars in the same discipline
locally or at other institutions.

It is important to understand future change: How is scholarship
changing in general, and how will this impact the individual scholar
or institution?

It is important to understand the needs of the scholar on several
different levels:

Knowledge needs. How, ideally, would each scholar or group like

their existing skills and knowledge to be used? What opportunities

for developing them further would be advantageous? How well

are needs being met?

Psychological needs. What are scholars' needs for responsibility,

status, esteem, security, and advancement, and how do they define

these needs?

Support and control needs. What kind of support services would

enable them to carry out their work responsibilities more

efficiently?

28 CAROLYN M. GRAY

Task needs. What kinds of task structures and role responsibilities
do different kinds of scholars find motivating, interesting, and
challenging? What opportunities exist for self management, for
developing new methods and services? This area is very relevant
to the teaching role of faculty and an area where new working
relationships can be developed among teaching faculty, computer
professionals, and librarians.

Ethical needs. How do scholars wish to be treated by the library?
Do policies on communication, consultation, and participation
meet their expectations?

An examination of the ETHICS method helped to clarify project
staff understanding of what kinds of questions to ask about the scholar's
work, but this method did not help in knowing "how to ask" to ensure
that the eventual design solves the right problem or set of problems.
After an examination of alternative methods of surveying or interviewing
faculty, it was decided to use ethnographic field research techniques
as a way to interview scholars.

To carry out the research, a two-stage process of interviews was
designed. The first stage of interviews involves an in-depth small group
study in a specific discipline. The second stage involves using the
findings from initial interviews to design a broader survey in other
disciplines.

Ethnographic field research techniques were adopted for this study
for four primary reasons. First, the method helps the researcher to
understand behaviors according to how they are embedded in social
and historical fabric of everyday life. The focus is on the relationships
between the parts. The design of any component has implications for
the rest of the system. Scholarly research is thus placed in the context
of the discipline and the day-to-day work life of the scholar.

Second, because the method is descriptive, the researcher withholds
judgment about the behavior described or observed. The researcher can
look for how seemingly inefficient behavior is embedded in the social
system. By describing first, the researcher does not jump to offer technical
solutions.

Third, ethnographic research helps researchers understand other
people's behavior from their point of view. The researcher must focus
on how they categorize their activities and functions and not on artifacts
of the environment. The interviewer must not impose her view of the
world on the situation being observed.

And fourth, researchers learn about others by encountering their
situation firsthand. Researchers look at everyday, naturally occurring
talk and action. An important part of a work group's interaction may
occur around the coffeepot or watercooler. By making naturalistic

B UILD1NG ELEC TR ON 1C BRIDGES 29

observations, researchers can record and understand the use of
technology within the environment of its use.

In studying the ethnographic approach, the project team learned
that it is not a method to generate good questions in a mechanical
way; that is, researchers cannot go into every interview with a script
to be followed for three reasons. First, good questions emerge from
an understanding of the group. Second, good questions emerge in the
course of the interview. And third, results emerge from the interaction
between interviewer and interviewee.

The Small Group Study

The following case study was prepared by Sue Woodson-Marks who
has training and experience in ethnographic research. The assignment
to the research team was to describe the research habits of a single
research group in terms of their use of information.

The group studied was the Brandeis Radio Astronomy Group
(BRAG) that includes the following:

two faculty members;

one advanced graduate student;

two post-docs, one who is still working at Brandeis and one who
has already moved to another area university but returns to Brandeis
to attend "Astrolunch";

several lower-level graduate students who are in the process of deciding
whether they want to join the group; and

one honors undergraduate student.

For Gesher Project purposes, the faculty and advanced graduate
students are considered the core members of the group the ones
concentrated on in the interviews. They are all working on various aspects
of a single issue: measuring the linear polarization of astronomical
objects. In his dissertation work, the senior scientist devised a means
of measuring this aspect of astronomical objects that has not been
previously recorded. Their work now involves developing the technology
for taking these measurements and using the data they gather to better
understand the structure and function of these radio sources.

The approach used was to conduct in-depth interviews of the core
group with an eye towards understanding how BRAG works as a group,
what work they do, and how they do it. Project team members also
observed at two "Astrolunches," the lunch time forum for reviewing
current literature in the field and presenting work done for the group.
This report is based upon five interviews in all, lasting from one to
three hours, which have been conducted by a team including the
ethnographer and a librarian with the software engineer participating
in one interview.

30 CAROLYN M. GRAY

Although the BRAG members are each individually involved in
a number of different activities (e.g., teaching, taking classes, serving
on university committees, etc.), this project has concerned itself
primarily with the work they do as researchers in astrophysics. The
information-related tasks involved in this research include designing
research projects, gathering and organizing data, producing and refining
tools for analyzing the data, analyzing the data, and disseminating the
results of their work.

Designing New Observations and Writing Proposals for
Grant Money and for Time on the Radio Telescopes

Like most other astronomers, BRAG members collect their data
using groups of radio telescopes owned by other institutions. They must
not only petition for funds to support their research, they must compete
with other radio astronomers around the world for time on these
telescopes. Although one particular grant may cover more than one
year, proposals for research funds and time on the telescopes are generally
written each year. These proposals are generally written by the faculty,
although graduate students may be writing their own grant proposals
as well. In either case, the writer must not only have a clear notion
of the work being proposed, he must also know what has been done
in the field recently and how the proposed work fits into ongoing,
already funded, research.

A weekly lunch seminar, Astrolunch, serves, in part, the function
of keeping BRAG members up to date on current literature. Members
of the group are assigned individual responsibility for reporting on
particular journals in this seminar. When new issues of a journal arrive,
the responsible member reviews the abstracts and table of contents of
the new issues and reports on any articles that would be of interest
to members of the group. Faculty members also use this venue to report
to students on current funding issues. Although this may seem a rather
labor-intensive means of reviewing the literature, project staff cannot
recommend a streamlining of this process through computerization
because it seems clear that the Astrolunch serves an important teaching
function. The field of issues covered in this seminar is much greater
than "the most recent developments in astrophysics." Here students
learn about the values that guide research, the appropriate ways of
evaluating other researchers' work, the nature of collegiality, and the
history of the field including important personal information about
the researchers that would not easily be available otherwise.

Collecting and Organizing Data on Radio Sources

Though this is central to the activities of BRAG, it does not serve
the purposes of this paper to describe this process in detail. Suffice

BUILDING ELECTRONIC BRIDGES 31

it to say that there are two types of data that are collected and organized
the group's own data and data from other radio astronomy groups.
The control over the BRAG members' data is managed in-house and
seems to work quite well.

More problematic is the retrieval, recording, and organization of
information gathered from other researchers. It is the understanding
of the needs in this area where the Gesher Project may be able to provide
some assistance. The need for information generated by others is a
relatively common bibliographic problem finding a work in which
the desired data is reported. Access to NASA's intergalactic database
may prove to be the best bibliographic solution.

The other half of the problem is capturing the needed information
in a useful form. The data may be in the form of a spreadsheet with
many different observations of a number of different objects, and no
one is interested in keying pages and pages of this data into their own
computers. Project staff have recommended that rather than investing
time and money in solving this technical problem, BRAG would be
better off using a service that scans documents for a fee.

Writing Software for the Analysis of Data Collected by the Group

Although BRAG members use a number of programs from other
institutions to manage their software, a substantial portion of the group's
work seems to involve writing and updating computer programs
developed by the group for reducing their data and representing it
graphically. This constant "tinkering" can cause significant difficulties;
a proliferation of versions develops, and one loses track of which version
is the appropriate one to use. The software engineer on the Gesher
Project has proposed a software management program solution to this
problem.

Disseminating the Results of the Group's Work through Published
Writings, Attendance at Meetings, Correspondence with Other Radio
Astronomers, and Public Lectures

Though the Gesher Project may have design components that aid
in the development of formal papers, such as creating bibliographies,
this was not explored very extensively in the initial interviews.

Conclusions from the Small Group Study

The approach of asking how the whole process of doing research
in astrophysics works allowed project staff to place information needs
in context. Without spending time at the Astrolunch seminar just
relying on interviews there would only have been evidence like one
scientist's complaint that even having the journals delivered to his

32 CAROLYN M. GRAY

mailbox is inconvenient he wants someone to read the journals and
tell him which articles to read. Project staff would not have been able
to see the broader role that Astrolunch plays in the process of keeping
abreast of the literature.

Spending the time up front, trying to learn the basics of their
research project, and doing multiple interviews within a single research
group are strategies that paid off in the end. The issues of loading
masses of data into spreadsheets and managing software updates both
came out of the last interview with the senior researcher. Both had
been mentioned in passing in earlier interviews, but their importance
was only evident in the last interview because one could see the research
in a larger context.

This was not an in-depth ethnographic study. Project staff were
only able to conduct a few interviews and attend a couple of seminars.
If there were more time, it would be good to attend more seminars
besides Astrolunch and some graduate level classes. Even with this
minimal work, the Gesher Project team was able to come up with several
interesting ideas about how the BRAG team members do their research
and what some of their information needs are.

SYSTEMS DESIGN IMPLICATIONS

Digital Equipment Corporation researchers are developing a
prototype information management system. The prototype has been
developed using a medical model and was first previewed publicly at
DEC World in Boston in April-May 1992. The medical model includes
menuing capability, live links to in-room patient monitors, graphic
images such as radiology reports, patient records, physician records,
databases resident at various locations, security levels, and links out
to external sources such as MEDLINE. The system is running on a
DecStation 5000 as a server with DecStation 2000s as primary
workstations. Primary requirements of the system design include high-
resolution graphics terminals with windowing capabilities.

The library system, yet to be programmed, includes the following
design elements:

a menu of possible activities that can be customized for a scholar
or group of scholars;

network links to bibliographic databases, catalogs, indexing and
abstracting services;

links to local custom programs and files such as BRAG's own data
analysis system;

personal databases created with pointers to external files;

bibliography-formatting software;

BUILDING ELECTRONIC BRIDGES 33

PC/fax to receive scanned images; and

high-resolution graphics capability and links to image files.

CONCLUSION

The work presented here is preliminary and reflects research-in-
progress at Brandeis University. The work offers one model of how
librarians can be involved in the design of new computer products for
library users. The primary contribution at this time is methodological.
Using qualitative research techniques, involving users in the design
of systems, and librarian participation in research and design reflect
new roles for librarians.

REFERENCES

Belkin, N. J.; Chang, S.-J.; Downs, T.; Saracevic, T.; fe Zhao, S. (1990). Taking account

of user tasks, goals and behavior for the design of online public access catalogs.

In D. Henderson (Ed.), Information in the year 2000: From research to applications

(Proceedings of the 53rd ASIS Annual Meeting) (Vol. 27, pp. 69-79). Medford, NJ:

Learned Information.
Boulton, W. R., & Saladin, B. A. (1983). Let's make production/operations management

top priority for strategic planning in the 1980s. Managerial Planning, 32(1), 15.
Davis, G. B., fc Olson, M. H. (1985). Management information systems: Conceptual

foundations, structure, and development (2nd ed.). New York: McGraw-Hill.
Gray, C. M. (1992). Information for management planning and decision making: Toward

a comprehensive model. Unpublished doctoral dissertation, Brandeis University,

Waltham, MA.
Havelock, R. G. (1972). Bibliography on knowledge utilization and dissemination. Ann

Arbor, MI: Center for Research on Utilization of Scientific Knowledge, Institute for

Social Research, University of Michigan.
Havelock, R. G. (1976). Planning for Innovation through dissemination and utilization

of knowledge. Ann Arbor, MI: Center for Research on Utilization of Scientific

Knowledge, Institute for Social Research, University of Michigan.
Hodge, B.; Fleck, R. A., Jr.; & Honess, C. B. (1984). Management information systems.

Reston, VA: Reston Publishing.
Kochen, M. ( 1970). Stability in the growth of knowledge. In T. Saracevic (Ed. ), Introduction

to information science (pp. 44-55). New York: R. R. Bowker.
Line, M. B.; Brittain, J. M.; & Cranmer, F. A. (1971). Information requirements of

researchers in the social sciences. (Investigation into information requirements of

the social sciences: Research report no. 1). Bath, England: Bath University of

Technology, University Library.
Mackenzie Owen, J. S., & van Halm, J. (1989). Innovation in the information chain:

The effects of technological development on the provision of scientific and technical

information. London: Routledge.
Mumford, E., & Mar Donald, W. B. (1989). XSEL's progress: The continuing journey

of an expert system. New York: Wiley.
Taylor, R. S. (1980). Value added aspects of the information process [Summary]. In A.

R. Benenfeld 8c E. J. Kazlauskas (Eds.), Communicating information (Proceedings

of the 43rd ASIS Annual Meeting) (Vol. 17, p. 344). White Plains, NY: Knowledge

Industry Publications.

Taylor, R. S. (1986). Value-added processes in information systems. Norwood, NJ: Ablex.
Tufte, E. R. (1990). Envisioning information. Cheshire, CT: Graphics Press.

VIRGINIA TIEFEL

Director, Library User Education

Ohio State University Library

Columbus, Ohio

The Gateway to Information:
Development, Implementation, and Evaluation

ABSTRACT

The Ohio State University Library has developed The Gateway to
Information, which is a computer-assisted program for undergraduate
students. The program guides students in identifying, locating,
evaluating, and selecting information independently. The Gateway has
been in development for six years, funded by four grants, and has
undergone continuous evaluation. No help screens or handouts are
needed to use the system.

INTRODUCTION

The Gateway to Information was designed to help undergraduate
and eventually graduate students identify, locate, evaluate, and select
the most useful information for their needs. Running on Apple
Macintosh workstations, The Gateway provides guidance and
instruction for students on how to proceed through an information
search that integrates the use of print and computerized information.
The microcomputer program serves as an online "bridge" to other
computer-based systems, enabling the user to apply major elements of
a search strategy process by gaining ready access to the text of relevant
CD-ROM-based encyclopedia articles and journal indexes as well as
print sources. Each time users begin to search the catalog, the

THE GATEWAY TO INFORMATION 35

microcomputer program offers a comprehensive search strategy option
to lead the user through information sources beyond the catalog. Goals
of the project are to teach students the following skills:

find, evaluate, and select materials that meet their needs regardless
of format;

access and integrate the content of online catalogs and CD-ROM
databases easily (even as novice researchers); and

apply information-seeking and critical-thinking skills with a high
degree of independence.

The Gateway has been continuously evaluated by users, and
revisions have been made based on the results of the evaluations.
Available on nine workstations since spring 1991, access to The Gateway
will steadily increase as the library replaces 50 to 100 public catalog
terminals with workstations that provide Gateway access. The Gateway's
technology is basic and adaptable so the project is transferable to other
libraries both conceptually and technically.

THE NEED

Effective problem solving in a complex society requires educated
citizens who possess the ability to identify, acquire, and evaluate available
information on a particular topic, question, or set of problems. With
vast increases in the quantity of information available, most people
are simply not capable of coping with this phenomenon, especially
because of the increasing necessity for them to use computerized data-
bases to gain access to much of this information.

College students, in particular, need instruction not only in the
use of individual databases but, much more importantly, in a
comprehensive approach to finding and integrating information
whether in print sources or in online sources. Most instruction that
has taken place at the postsecondary level, however, has focused on
teaching use of individual sources, including databases, with little if
any guidance provided in how to integrate and weigh the usefulness
of information obtained from a variety of online and print sources.
Moreover, the proliferation of information has intensified the need for
students to be able to evaluate information: the challenge often lies
not in students' being able to find enough information but in their
being able to evaluate and select the most useful for meeting their specific
needs. Thus, two instructional needs in "information-seeking" skills
must be satisfied: to teach students how to find needed information,
using whatever formats are most efficient, then to evaluate that
information to select what is most appropriate to the task at hand.

36 VIRGINIA TIEFEL

Many studies have shown that most undergraduate students never
learn how to use libraries or other information sources effectively.
Without instruction specific to information seeking, it has been found
that most students will scan the library catalog to identify a few books
on a topic, check out the titles that may be available, and attempt to
complete the assignment. This is obviously a wholly inadequate
approach to information seeking in today's society.

The Gateway to Information was conceived six years ago in the
Office of Library User Education at the Ohio State University (OSU)
Library in response to the burgeoning demands of the library user
education program. The user education program has been in place since
1978, and as OSU Library Director Bill Studer observed, the program
had become the victim of its own success. Meeting the staffing demands
of the program was becoming increasingly difficult, and given the
library's budget, there would be no additions to the staff. The program
was reaching annually over 30,000 students with some form of course-
related instruction, and another 5,000 students were taught in clinics
and workshops. Although that is a large number, it was an inadequate
one considering the 53,000 students on campus. Furthermore, it was
recognized that to become information literate, students need multiple
sessions of instruction. An additional point of concern was the realization
that students were beginning to use remote access to the online catalog;
this practice resulted in students' reversion to total dependence on the
catalog a dependence librarians had been trying to break by instructing
students about the variety of library resources beyond the catalog.

DEVELOPMENT OF THE GATEWAY

Most of the library instruction at OSU has focused on the search
strategy concept that is a step-by-step process moving from general to
specific information through evaluation and selecting the best
information for the need (see Figure 1). In pursuing how to continue
the expansion of the instruction program without more staff, it was
decided to try putting the search strategy on a computer that would
be connected to the online catalog and to CD-ROMs. A grant was sought
from the Fund for the Improvement of Postsecondary Education (FIPSE),
a granting agency that funds innovative but largely embryonic projects.
The proposal came close to acceptance in 1986, and the following year
a revised proposal received funding. The project has subsequently
received two grants from the Higher Education Act II-D, College Library
Technology and Cooperation Grants Program, and a grant from the
William Randolph Hearst Foundation for a total of half a million
dollars. These four grants were critical to The Gateway's development.

THE GATEWAY TO INFORMATION

Search Strategy: An Efficient Research

SELECT A TOPIC

' r

I ENCYCLOPEDIAS

(General or Special for an overview

\ ^/- nir.-nn^/ARIFF! ^

(for unknown or

) ] \ODSCUTB terms/words^

BOOKS

/ N

bv subject bv author or title

PERIODICAL ARTICLES

Newspapers Magazines Jouma

fin

acquired
before
1972

(Ci
I Cat

S acquired after 1972 C \ t^C ^\

S N

Periodical and Newspaper

Sti_ J LOO

Indexes
print
CD-ROM

I ' online )

y "[computerized!
I catalog J

ud Y**^

atog I

' r
OTHER SOURCES

- >
' r

Biographical
Indexes

\ [ U.S.Gov. | j Statistical
I Documents I I Sources

Book Essay and
I I Reviews I I General Literature

LCS provides only information about location and availability of books and journals: it does not list individual journal articles.
Library of Congmt Subject Headings

Figure 1. Search strategy concept taught at Ohio State University

38 VIRGINIA TIEFEL

The University Library provided a full-time equivalent (FTE)
position for directing the project, two FTE professional positions, and
considerable staff time. The library assigned a professional librarian
to the project full time for four months and provided some equipment.
The university's Instructional Development and Evaluation unit
provided a 10% equivalent FTE evaluation and computer technology
expertise for the first two years, with the assistance of a graduate teaching
associate. University Systems, the support and provider for the library's
online catalog, loaned $30,000 worth of equipment and provided one-
third of a programming position. The Academic Computing Center
also provided staffing and equipment support.

The library has provided student programming time, fees for lines
to the library's online catalog, software and equipment, and valuable
space in the library for The Gateway team. When the project was begun,
the programmer/analyst-senior and the systems programmer par-
ticipated in the evaluation of needed computer equipment. Based on
their findings, equipment and software were selected and purchased
using funds provided by the University Library and the related university
computer center. This equipment included microcomputer work-
stations, a local area network (LAN), and a connection to the university's
Amdahl mainframe computer, which runs the online catalog system.

Macintosh HyperCard 2.0 was used for prototyping The Gateway
narrative because it offered the easiest method for creating the narrative
and making the necessary revisions. In the beginning, programming
activities centered on developing the microcomputer "front-end" for
the University Library's mainframe catalog system. This was com-
plicated by the need for the microcomputer to process special characters
(e.g., diacritical marks) that are needed for the several foreign languages
supported by the online catalog system. The development of this
capability, however, had other benefits. It permitted a more flexible
user interface that could place all or part of the catalog information
anywhere on a microcomputer screen and make possible the combining
of catalog data with that from other information sources. Like most
online catalog systems, the OSU online catalog was developed for
mainframe display terminals that have a fixed display format and access
to only one information source. Therefore, this new capability offered
a major improvement over existing library information systems and
could be adapted by many institutions that had the same limitations
in their catalogs.

Programming was started with these underlying structures because
they were necessary for implementing the overall project design to
bring together information from different sources utilizing various
learning and access strategies. Work began on a single user workstation
connected to the library's mainframe-based catalog system and

THE GATEWAY TO INFORMATION 39

conversion to the LAN environment where users on several workstations
could share a single link to the library's online catalog system.

Incorporated into the project's design was the ability to update
both information sources and the narrative/instruction. These features
were needed to keep pace with the always changing environment within
information systems and information itself. It also enhanced the project's
transferability by permitting other institutions to tailor the system to
their particular needs. The data communications connection to in-
formation sources was intended to be transportable to other institutions
with little or no modification: there are only a limited number of ways
to connect microcomputers for data transfer, and most of them will
have been included in the design of The Gateway.

The Gateway software runs on Apple Macintosh Ilex computers
that are connected to the campus computer network through which
the library user may access available information services. Currently,
The Gateway workstation user may access the University Library's online
catalog and 12 CD-ROM databases that are housed in CD-ROM towers
and mounted on a LAN. The Gateway software, which includes
HyperCard 2.0, MAC/TCP, and MitemView, is installed on each Gate-
way workstation. The Gateway workstation was designed to function
as the catalog workstation with the intent that every public terminal
for the OSU Library's online catalog would, in time, be a Gateway
workstation. The OSU Library also intends to make The Gateway avail-
able for remote users of the online catalog. The content of the narrative,
instruction in The Gateway software, and the system design have been
developed to migrate easily to other library environments.

The design group decided to begin writing the narrative with the
journal section, and when that became operable, attention was turned
to the development of the first step of the search strategy encyclopedias.
The intent was to design a common interface to the databases so users
would see the same screen design regardless of the database they were
using. The Academic American Encyclopedia was added to The Gateway,
which also now offered the journal section.

The first step in the search strategy is finding background
information on the topic, for which an encyclopedia is usually the best
source. Using a dictionary for definitions of unknown words and terms
is the next step, with searches of journals usually a third step, depending
on the topic. The additional steps of the search strategy were added
until all were operable.

When the technology to link the CD-ROM versions of the en-
cyclopedia, dictionary, and indexes was perfected, The Gateway in-
corporated that format into its instruction. The Gateway continued
to add indexes in electronic format as they became available. The
decision, however, on what to instruct the students to use was not decided

40 VIRGINIA TIEFEL

by what was available in electronic format but by what was deemed
the best choice to meet the students' needs. The Gateway is designed
to make the best use of technology but not to be driven by it. Links
from the microcomputers to the catalog and CD-ROMs were completed
near the end of the second year.

IMPLEMENTATION OF THE GATEWAY

As of May 11, 1989, a Macintosh workstation with The Gateway
prototype was set up in the main library for library personnel to use
and evaluate. For security reasons, the workstation was located in the
administrative wing, and librarians and staff were encouraged to
schedule a time or simply to stop by to use it. Evaluation forms were
available at the terminal.

Based on the evaluation data, the journal section was revised, and
a number of subjects and indexes were added to this section. This
development was supervised by Nancy O'Hanlon, head of the Reference
Department at the Undergraduate Library, who was on temporary
assignment to the project. She brought to the project considerable
knowledge of how undergraduates seek information and what is needed.
Her appointment ran from March through June 1989, and she did an
excellent job of pulling the narrative and the staff ideas together, adding
consistency to the screen design, and expanding the journal search and
other sections of the narrative. Testing and evaluation involving 10
randomly selected students were conducted by the Center for Teaching
Excellence evaluation personnel in May. Based on the evaluations, The
Gateway was revised and the narrative was expanded. Considerable time
was spent on how to determine and analyze user needs. As a help in
accomplishing that goal, as well as to provide baseline data, a user
study was conducted in the 1988 fall quarter.

The highest priority of the project's programmers was to program
and implement the instructional design ideas of the library staff and
others. Thus far, programming had been done using a high-level
language. Programming using an authoring system that allowed easier
and faster development was preferable, but a graphics-based authoring
system with the required communications features had not been found.
The search for tools that would allow easier implementation and
modification of design ideas continued.

At the end of two and one-half years, the narrative had been
developed from the original journal search into five major areas of
information: Facilities, Explain, Strategies, Sources, and LCS (online
catalog). The Gateway provided a subject approach to encyclopedias
and journal indexes, recommending which materials students should

THE GATEWAY TO INFORMATION 41

use depending on their information needs. The Gateway also answered
simple questions about the library system, e.g., library locations, floor
plans, and services.

By early 1990, connectivity between The Gateway, CD-ROM
materials, and LCS had been achieved. This meant that from a single
Gateway workstation, the user could access the Academic American
Encyclopedia on CD-ROM, the catalog, periodical indexes, and the
needed guidance to make decisions about what to use and where the
information was. A prototype LAN was set up with five workstations
sharing a line to the library's online catalog system and databases on
CD-ROM. Project goals were seamless access to the CD-ROM database
from within The Gateway program. This was a real breakthrough for
the project in both instruction and technology.

The Gateway has met the needs of many students and programs.
One specific program can be cited as an example of improved library
instruction with The Gateway. The Young Scholars Program was imple-
mented by the university in 1988 and is designed to prepare minority
students for college. Beginning with a class of pre-seventh graders the
first year, the program brings to campus for two weeks 400 students
at each grade level. They are taught subject matter and certain skills
including information-seeking skills. The first year of the library's
involvement with the program was 1990, and the library instruction
was ineffective. In 1991, a special limited edition of The Gateway was
prepared for the class of tenth graders to use in completing an assignment
on Martin Luther King. Students did their work in a computer lab,
and evaluations showed that use of The Gateway was very successful.
Students liked it and appeared to learn from it. Without The Gateway,
the library would have had a difficult time creating and implementing
a meaningful library assignment that was also popular with students.

EVALUATION OF THE GATEWAY

Evaluation results provide evidence of how The Gateway has
benefited students and improved library services. In 1988-1989, copies
of proposed narrative sections were periodically distributed to library
and faculty and staff who had expressed an interest in critiquing them.
About 30 copies were distributed, and a wide variety of responses were
received. These provided some of the material Nancy O'Hanlon used
in the development of the search strategy narrative.

The project has been continuously evaluated, primarily through
written evaluation forms left at the workstations. The first evaluation
study, which is a summative evaluation, was done in fall 1988 when
baseline data were collected on how students found information in the

42 VIRGINIA TIEFEL

library using traditional methods. This will be compared with how
students use The Gateway, and the results of both methods will be
examined. This comparative analysis has not yet been done. The other
evaluations have been formative ones, and the results have been used
to revise and expand the narrative. In mid-1989, the first public work-
station was set up, and library staff evaluated The Gateway using forms
that asked about screen design, logic of the narrative, and the content.
In addition, 11 students were intensively interviewed using The Gateway
in fall 1989.

When the first public workstation became available in the main
library in January 1990, evaluation forms were placed next to the
terminal. These forms were similar to the staff evaluation forms. In
mid-1990, two freshman classes of 41 students were required to use The
Gateway for an assignment. Evaluations were very positive. The eval-
uation form was changed considerably in mid-July of 1990 and has
remained much the same since. After the CD-ROM access became
available on The Gateway in mid-February 1991, satisfaction and usage
both increased.

The Gateway was also evaluated by special classes a graduate class
and two industrial design classes. Their comments were not summative
but formative, i.e., how to improve The Gateway. The Center for Teach-
ing Excellence provided an industrial design expert to evaluate The
Gateway in January 1990. In spring 1990, a library science class evaluated
The Gateway.

The Gateway was designed for use by lower level undergraduates
with the intention of increasing its complexity and sophistication in
materials to meet the needs of advanced undergraduate and graduate
students and ultimately faculty. Evaluations showed that upper level
students, faculty, and staff used The Gateway and were successful in
their searches. Of 1,190 evaluation forms turned in voluntarily at the
workstations from July 16, 1990, to July 31, 1991, the breakdown by
academic level of user was as follows: freshman, 106 (9%); sophomore,
127 (11%); junior, 170 (14%); senior, 226 (19%); graduate students, 306
(26%); faculty and staff, 77 (6%); other, 102 (9%); no answer, 72 (6%).
In summary, 629 or 53% of the users were undergraduates, 306 (26%)
were graduate students, and a total of 935 students made up 79% of
The Gateway usage during that period.

Nine Gateway to Information workstations help students to identify
their information needs and locate, evaluate, and select the information.
The Gateway's success rate in accomplishing this is documented in the
results of the project's evaluations. Results of 1,656 evaluation forms
dated from July 16, 1990, to January 31, 1992, indicated that 78% were
"completely" or "mostly" successful in their searches: 89% rated the
screens "very" or "mostly" clear. Ease of use of The Gateway was rated

THE GATEWAY TO INFORMATION 43

"very" or "mostly" easy by 84%. From 964 of those evaluations, 83%
said they would use The Gateway again.

Sample topics searched included reflexology, women in politics,
waste water pollution, medieval period dress and costumes, gum
chewing/bubble gum, reunification of Germany, and social welfare.
Comments were varied but mostly very positive. Some examples included
the following: "Easy as pie." "This thing takes your hand and leads
you right down the path." "This was incredible! What a time saver."
"I could see exactly what I was doing, and I knew my status all the
time." "Every thing you could want is at your fingertips!" "I'm addicted:
great visual format." "Really easy to use please get more of these."

Design issues were settled by evaluation results when possible, and
the impact of evaluation can be seen in Figure 2. The improvement
in the evaluation results can be directly attributed to the revisions that
were made based on the evaluations. Figure 3 shows two screens in
the Gateway's early development. These were opening screens on The
Gateway before it was made available to the public when it was still
being evaluated by library staff only. The first screen showed type of
material books or journals and was too limiting in its options. The
second screen attempted to anticipate the user's needs and was also
too limiting: users were unable to identify with the options. Neither
approach worked well. Opening screens of The Gateway now offer a
research strategy diagram that works well (Figure 4): the screen provides
the users with several options and allows them to better control their
searching.

The evaluation studies revealed some basic tenets. One was that
most students will not read more than two lines: they prefer to skim
text. Another was that students usually select the first or second choice,
especially when using the system for the first time. As they become
accustomed to using the system, this tendency diminishes. Most users
did not understand the meaning of icons or how to use them. This
lack of knowledge extends to arrows, but they do understand boxes.

THE FUTURE OF THE GATEWAY

The Gateway will continue to undergo expansion in its narrative
and number of available databases and workstations. Immediate plans
include continued revision of the narrative based on evaluations and
the addition of special subject sections. The first one of these sections
is on communication and is being tested by students. A business section

VIRGINIA TIEFEL

The Gateway to Information:
Comparison Statistics

r7/] Original Evaluations: (715)
\A July 16, 990 -March 31. 1991

100%_

July 16, 1990 -May 20, 1991

Most Recent Evaluations: (256)

April 1, 1991 - May 20. 1991

90%_

89%

82%

80%_

80%

, Aa 75% 76%

74%

70%_

-
-

66%

60%_
50%_
40%_
30%_
20%_

/
t

'
/

,
/
/

/
/
/

In 1 '.

Comp etely Sea

rched Very easy Very c ear Very easy

ormosUy LCS ormosUy ormosUy ormosUy

successful easy to use clear screens easy to use

with search Search Strategy The Gateway

Map

Figure 2. Evaluation results of revisions made on The Gateway

will soon be ready, and by fall 1992, a section of women's studies will
be available. These additions will be tested by users and revised until
they are as user friendly as the existing Gateway narrative. Commercial
databases will be added to The Gateway as money becomes available.

Replacement of public online catalog terminals began in spring
1992, and 20 new workstations with Gateway capability will be available
by the end of May. The library will continue this replacement activity
until all 106 public workstations have Gateway capability. All of these
activities narrative revision and expansion and addition of databases
will be ongoing. The project will never be finished: it is a forever project.
It was envisioned that way, and its development has substantiated that
vision.

Plans are already underway to expand the subject list of 100 topics
to one based on the Library of Congress Subject Headings classification.
The Gateway now recommends specific materials for each subject, and
the enhanced list of subjects with recommended sources will expand
The Gateway's ability to guide users to the best information. Some
have suggested that The Gateway ultimately be programmed to respond

THE GATEWAY TO INFORMATION

Subject - You have a topic and you want to know what
has been written about it

Select one of the following sources:

f Magazines/Journals *

If you're not sure where to begin your search for information,
start by analyzing your information need.

Which of the following best represents the ultimate use
of the information you seek?

A paper

A bibliography

A speech

Figure 3. Two screens used in early versions of The Gateway

VIRGINIA TIEFEL

Books/Periodicals
(Online Catalog System - LCS)

Figure 4. Opening screen used in later versions of The Gateway

to the user's selected subject with not only which materials to use but
pages, subject headings used, etc. Some say the cost of doing this would
be prohibitive, and from a librarian's viewpoint, it doesn't teach
information skills: it's the vending machine approach. However, it is
a concept worth exploring.

In The Gateway's immediate future is the development of a UNIX-
based system. This would provide remote access and make The Gateway
compatible with any type of computer. The narrative would probably
not be as appealing as the Macintosh version, but the use of windows
would permit the use of some graphics. It is hoped to have this version
ready next year, but obtaining money to buy the equipment and do
the programming will determine the timetable.

Formative evaluation will continue to provide the basis on which
The Gateway narrative is revised and expanded. It would be valuable
to do a summative evaluation to determine what impact The Gateway
has on students' information seeking. Do they find more or less material
using The Gateway than with traditional searching? Is the information
found more or less appropriate for their needs? How does The Gateway
affect students' attitudes?

THE GATEWAY TO INFORMATION 47

In terms of physical expansion, The Gateway will be available on
59 terminals in the library system by summer 1992. This is more than
half of the number of public terminals in the library system. The
remaining 47 terminals will be replaced within the next year making
The Gateway available in all OSU libraries. When the UNIX version
is finished, The Gateway will be available across campus in dormitories
and offices and off-campus for OSU users who have access to computers.
This will be a very popular move; it is one our students have consistently
asked for from The Gateway's inception.

The OSU Library plans to share The Gateway with other academic
institutions and school and public libraries. The complication is the
Library's inability to support such sharing, having neither the necessary
staff nor the resources. There have been ongoing talks with several
companies about marketing The Gateway. And although there is some
interest in a collaborative effort, there is nothing definite to date. Many
other academic libraries have expressed an interest in acquiring The
Gateway for their institutions. The leaders of a statewide project to
link all Ohio primary and secondary schools electronically are interested
in incorporating The Gateway into their project. Public libraries have
expressed a desire to collaborate on a Gateway version for public libraries.
Envisioned is an information system that teaches and guides students
from primary through secondary and postsecondary institutions to the
public libraries on how to find, evaluate, and select information. The
system based on the search strategy concept will make students
information literate. In fact, students will learn the search strategy so
well that they will be able, ultimately, to apply the concept in libraries
without Gateway terminals.

The Gateway to Information is already a success with users, and
its potential for development and expansion is virtually limitless. User
satisfaction and usage are very encouraging, and The Gateway has
demonstrated that it can change how libraries are used. Although no
other institution has the right to use The Gateway, there is promise
of and an interest in transporting it to other institutions. OSU is
committed to sharing The Gateway and to encouraging its adoption
by as many other institutions as possible.

Evan Farber, the preeminent expert on bibliographic instruction
in the world and one of The Gateway consultants, summarized the
project this way:

I was so pleased with the progress you all have made with Gateway. As
I told you I said to the LOEX group, I felt proud to be associated with
the project. It's very impressive, and I think academic librarians are going
to feel indebted to you for many, many years. To be sure, others will build
on it, improve it, but the credit for developing the first really effective
computer-assisted bibliographic instruction program will belong to you.
Congratulations and thanks so much for permitting me to take part in
it. (E. Farber, personal communication, May 28, 1991)

TIMOTHY W. COLE

Assistant Engineering Librarian
University of Illinois at Urbana-Champaign

LESLIE TROUTMAN

Music Library User Services Coordinator
University of Illinois at Urbana-Champaign

WILLIAM H. MISCHO

Engineering Librarian
University of Illinois at Urbana-Champaign

WINNIE CHAN

Automated Records Maintenance Coordinator
University of Illinois at Urbana-Champaign

Design and Development of a
Library Information Workstation

ABSTRACT

This paper describes the design and continuing development of the
University of Illinois at Urbana-Champaign Library Information
Workstation, which provides the primary, in-library patron access to
ILLINET Online Plus, the Library's extended online public access cat-
alog system. The automated library information resources and infor-
mation processing environment are briefly described as they have
influenced the design and development of the Library Information
Workstation. The Library Information Workstation philosophy and
approach are discussed in the context of relevant information access
issues and patron needs and requirements. Features of the current Library
Information Workstation implementation are then described using
illustrations focused particularly on integrated access to local (resident
on individual workstations) information files and an integrated end-
user interface for bibliographic database searching. Ongoing
development plans also are discussed briefly.

LIBRARY INFORMATION WORKSTATION 49

INTRODUCTION

This paper discusses general principles and precedents relating to
library information workstation design and development, focusing
specifically on interface and workstation development at the University
of Illinois at Urbana-Champaign (UIUC) Library. The paper is divided
into five sections: (a) a brief description of the ILLINET Online Plus
(IO+) extended online catalog, (b) an overview of the UIUC Library
Information Workstation (LIW) philosophy and approach, (c) a
discussion of user searching behaviors and needs and the interface design
considerations that have driven the UIUC LIW design and development,
(d) an examination of the local file access capability integrated into
the UIUC LIW, and (e) a detailed look at the integrated UIUC LIW
interface developed to facilitate end-user searching of bibliographic
databases accessed via the local BRS/SEARCH implementation.

Concurrent with the Library Information Workstation de-
velopments described in this paper, and to an extent the driving force
behind it, has been the emergence of the Extended Online Public Access
Catalog (E-OPAC). These developments in turn have been possible
because of improvements in telecommunications technologies, computer
hardware and software, and advances in the accessibility and breadth
of bibliographic databases. Together these developments are allowing
libraries to provide enhanced access to local and remote bibliographic
resources. This is being done principally through the model of the
E-OPAC (Hildreth, 1989, 1991; Potter, 1989).

E-OPACs typically provide value-added access to resources beyond
the conventional OPAC such as campus or community information
resources, locally created bibliographic files, locally mounted and remote
periodical index databases, online bibliographic database vendors and
utilities, and the OPACs of other major Association of Research Libraries
(ARL) or regional libraries. One of the primary roles of the E-OPAC
is to serve as a node in a campuswide information system (CWIS).

Supporting the E-OPAC have been a number of important emerging
information technologies such as powerful yet affordable microcom-
puter workstations; optical disk and enhanced magnetic storage media;
graphical and imaging technologies and standards; local, campus,
regional, and national telecommunication networks; and sophisticated
information retrieval search engines (e.g., BRS/SEARCH). These
technologies supply system designers and developers with the tools
needed to provide enhanced access within the E-OPAC environment.

The UIUC Library has incorporated certain of these technologies
into the IO+ E-OPAC (Mischo, Sandore, Clark, fc Gorman, 1990). The

50 TIMOTHY W. COLE ET AL.

development and implementation of IO+ has been integrally connected
to the evolving UIUC LIW, a multifeatured microcomputer workstation
developed in the Library to serve as a public terminal for IO+ (Mischo
& Cole, 1992). The UIUC LIW is the primary means of accessing the
IO+ information resources and plays a key role in integrating the
multiple technologies that comprise IO+.

Specifically, the LIW provides interface, gateway, and database
management software to facilitate direct patron access to IO+ resources
and services. The major objectives of the UIUC LIW are (a) the integrated
and largely transparent access from a single terminal to a wide range
of library and information access resources, (b) the inclusion of user-
friendly, expert-system interfaces that facilitate patron searching of
bibliographic databases and lessen end-user searching difficulties, (c)
the built-in flexibility to allow terminal-specific customization of LIW
menus and interfaces to accommodate localized patron needs and library
resources, and (d) the utilization of emerging multimedia and image
transmission technologies to enhance end-user interfaces and to provide
more rapid and more complete patron access to information.

The later sections of this paper illustrate some of this functionality
with specific examples, focusing on two of the information resources
available in the current UIUC LIW implementation: the capability of
storing, searching, and integrating local, customized databases stored
on the workstation hard disk and a microcomputer-based interface for
the locally mounted BRS/SEARCH implementation IBIS (Illinois
Bibliographic Information Services). These two particular LIW features
serve as illustrations of the enhanced information access provided jointly
by the LIW and the E-OPAC.

Finally, the paper concludes with a brief discussion of current
planning and development work in progress to further extend the scope
and functionality of the UIUC LIW.

ILLINET ONLINE PLUS

The IO+ extended OPAC provides access to a variety of local and
remote information resources via two different campus networks. These
resources include the following:

1. The statewide online catalog ILLINET Online (IO) with holdings
data from over 800 libraries in the state of Illinois, totalling over
9 million records. Access to IO is through both microcomputer and
mainframe interfaces.

LIBRARY INFORMATION WORKSTATION 51

2. A locally mounted mainframe-based BRS/SEARCH retrieval system
(IBIS) and attendant periodical index databases, presently comprised
of Current Contents (seven sections, updated weekly), eight Wilson
databases (Readers' Guide Abstracts, Social Sciences Index,
Humanities Index, Business Periodicals Index, Applied Science and
Technology Index, Biological and Agricultural Index, General Science
Index, and a combined file), and the ERIC databases. The IBIS system
is presently comprised of 5.1 million periodical citations.

3. Locally generated and maintained data files stored on the
microcomputer workstation hard disk and searched using a locally
developed sequential search database management system. These files
can be customized by departmental library site and include databases
such as hot topic bibliographies, faculty interest profiles, staff
directories, new book lists, and reserve lists.

4. A gateway to the CARL (Colorado Alliance of Research Libraries)
UnCover periodical database system.

Additional information resources are currently being tested and
integrated into the LIW at selected UIUC campus library sites. These
resources include the following:

1. Gateway access to database and telecommunications resources on the
campus fiber-optic network (UIUCNet), which include the Oxford
English Dictionary (OED), weather information, current news, class
listings, and campus telephone directory.

2. Gateway access to Internet resources, including the OPACs of selected
Committee on Institutional Cooperation (CIC) and ARL research
libraries and consortia, such as Northwestern University, Indiana
University, and the MELVYL California statewide union catalog, and
access via a local Gopher client to many other Internet resources.

3. Access to databases stored on CD-ROM networks (including full-
text files such as the UMI periodical article data) using the CD vendor
search engines, run from the interface using shell software.

4. Multimedia and graphical files (in .PCX format) such as building
maps, floor plans, and mixed graphics, sound, and text (including
hypermedia files). Software has been developed to retrieve and display
scanned images with voice-over being provided by a programmable
speech synthesizer.

5. The capability of invoking specific commercial microcomputer
software application packages such as expert system and database
management packages from the interface and returning to the
interface menu level using shell software.

In addition, several important access and linking mechanisms are
in place within the Information Workstation and IO+ for providing

52 TIMOTHY W. COLE ET AL.

maximum access to available information resources. These mechanisms
include the following:

1. UIUCNet file transfer capabilities used to transmit BRS/SEARCH
search results (bibliographic citations and abstracts) to users'
electronic mailboxes.

2. The on-the-fly linking and display of call number and limited
holdings information from a displayed BRS/SEARCH search citation.

3. Software-controlled gateway paths within the workstation including
automatic logon to local and remote resources and hidden password
entry for applications requiring passwords.

The above information resources and linking technologies are made
available to library users and staff through the LIW software, presently
deployed in 39 UIUC departmental libraries on some 110 public
terminals. The software is also being tested at selected institutions in
the 40-member ILCSO (Illinois Library Computer System Organization)
network consortium. The workstation software is also being tested in
networking environments utilizing TCP/IP telecommunications
protocols.

LIBRARY INFORMATION WORKSTATION

The UIUC LIW is presently implemented on a range of IBM PS/2
platforms from Model 30 286 machines to PS/2 Model 70s. The LIW
employs interface, gateway, and database management software to
enhance user access to local and remote information resources available
in IO+. The LIW is the center of a client-server user access model
for IO+ that features a distributed retrieval network with databases
on local and remote file servers and the interface and gateway functions
residing on the microcomputer workstation. Our implementation of
this model is illustrated in Figure 1.

The information resources accessed by the LIW may reside on local
or remote mainframes, on CD-ROM files in stand-alone or networked
environments, or as files stored on the microcomputer hard disk and
accessed via a microcomputer database management application. One
of the advantages of the microcomputer workstation approach is that
the main search menu presented to the user can be customized to suit
the needs of specific departmental libraries. A sample main menu from
the UIUC Music Library is shown in Figure 2.

The LIW project has focused on the development and testing of
microcomputer software and hardware technologies to (a) enhance the
user-computer interface, (b) provide expert-system searching techniques

LIBRARY INFORMATION WORKSTATION

Library Information
Workstation

(IBM PS/2 55SX, 70)

Microcomputer Applications

Software, Expert Systems,

DBMS

UIUCNET
Campus Fiber-Optic
Network
OED

-Campus Directory
-Class Timetable
-Weather Info.
-Campus Bulletin

Board
-BITNET
Reference &

Referral Service

National Bibliographic

Databases

(OCLC, EPIC)

Library Information Workstation: UIUC Library

Figure 1. Library Information Workstation: Implementation of a client-
server user access model for IO+

and guided assistance in user searching, (c) utilize multimedia
technologies in providing assistance with user instruction and point-
of -con tact help, and (d) provide extended access to information resources
on the IO+ statewide network, the campus network, and the Internet.

This approach facilitates a "one-stop-shopping" approach to a
broad array of information resources. The LIW is designed to use
multimedia techniques in providing access to bibliographic, numerical,
graphical, and full-text resources. The long-range goal of the LIW is
to merge the three types of workstation technology: bibliographic
database and gateway services, multimedia and imaging technologies,
and scholarly user needs such as data analysis, scientific computing,
and word processing.

TIMOTHY W. COLE ET AL.

ILLINET ONLINE PLUS

Select Database to be Searched

MAIN MENU SEARCH OPTIONS:
USE ARROW KEYS/HOUSE TO
SELECT, then Press ENTER

Online Catalog

* References to Articles
in Journals I Magazines

Music Library Indexes and
Lists

UIUCNET Services

Information about the
School of Music

EXPLANATION: References to Articles . . .
Illinois Bibliographic Information Service
(IBIS). Search for references to articles
in journals, agazines, and conferences.

CHOICES:

Wilson Indexes ~ 1,000 Acadesic,
Business I Popular Journals, 1983+

Current Contents Databases Last
2 years, 7,000 Acadeaic Journals

ERIC Databases Education
Journals, Reports, Papers

CARL UNCOVER -- 2 years
of Popular a Acadesic Journals

EXIT THIS HENU

ees University of Illinois - version c

Figure 2. Main menu from the UIUC Music Library

Several academic and special libraries have pursued the development
of microcomputer scholar's workstations or similar multifunctional
microcomputer-based desktop systems (Arms, 1990). These institutions
include Brown University, the University of Southern California, Ohio
State University (Tiefel, 1991), and Carnegie- Mel Ion University.

USER SEARCHING BEHAVIORS AND NEEDS

In May 1990, the UIUC Library Online Catalog Advisory Committee
formed an Interface Design Subcommittee with the charge to design
and implement LIW interfaces, beginning with the interface to the
local BRS/SEARCH implementation IBIS (Norlin et al., 1992a, 1992b).
To provide context for planning the design of interface features for
the LIW, the literature on user needs and searching behaviors was
examined.

LIBRARY INFORMATION WORKSTATION 55

Numerous studies of both online catalog and end-user bibliographic
searching services have been reported and reviewed (Mischo & Lee, 1987;
Drabenstott, 1991; Hildreth, 1989). As Borgman (1986) has noted, the
users of all online bibliographic retrieval systems exhibit similar
behavior patterns and have the same types of mechanical and conceptual
difficulties. Studies of direct end-user search services and online catalog
use show the following:

1. Users are enthusiastic about performing searches on easy to use,
quickly learned, inexpensive search systems.

2. Search strategy formulation and the use of Boolean operators pose
difficulties for users.

3. Users have difficulty with the simpler interfaces provided in the after-
hours services and CD-ROM systems.

4. Several recent online catalog and CD-ROM transaction log analyses
and repeat search studies show high user failure rates for both subject
searches and known-item searches (Peters, 1989; Hunter, 1991; Charles
& Clark, 1990; Puttapithakporn, 1990; Schultz & Salomon, 1990).

5. High levels of reported user satisfaction with search results may not
reflect true success rates (Ankeny, 1991; Nielsen, 1986).

6. End-user search services can demand a significant investment of
library staff time in training and assistance.

7. End-users resist formal training sessions and the use of printed
instructions, preferring computer-assisted instruction (CAI) and direct
one-to-one instruction from library staff or peers.

8. The typical user searches relatively infrequently; even the frequent
users search infrequently enough so as to require retraining or
refamiliarization with the system.

In addition, the online catalog use studies have revealed several
facts important to designers of E-OPACs containing periodical index
databases:

1. Most catalog users want materials on a topic.

2. Subject searching is the predominant mode of searching; it accounts
for more than one-half of all searches.

3. Catalog users report the most problems with subject searching.

4. One-third to one-half of searches result in no items retrieved.

5. Conversely, a large percentage of subject searches provide a partial
match with controlled vocabulary terms and produce a large number
of retrieved citations.

6. User-entered subject search terms match the Library of Congress
Subject Headings controlled vocabulary only 20% to 40% of the time.

56 TIMOTHY W. COLE ET AL.

7. Systems with keyword searching appear to receive more subject
searching.

8. Catalog users place the highest priority for improvements on various
subject search enhancements.

9. Users approach online catalogs expecting to find access to a broader
field of materials, including periodicals, than are covered by the
traditional card catalog.

The results of the end-user and online catalog use studies have
important ramifications for the design of all bibliographic retrieval
systems, including OPACs, online periodical index databases, and optical
disk search systems.

On the one hand, the overwhelming acceptance of E-OPACs by
users and the high degree of user satisfaction with such systems can
be interpreted as a mandate for enhanced subject access (Hildreth, 1987;
Mathews, 1991). Historically, the card catalogs of the late 19th century
provided access to periodical articles via 3X5 cards supplied by vendors
or prepared in-house, so the renewed interest in shaping the modern
online catalog into an "analytic" catalog capable of providing the same
function is not surprising. Locally mounted periodical index databases
provide users with access to the periodical literature from the same
terminals used to search the online catalog. They serve to complement
periodical index databases made available in stand-alone and networked
CD-ROM workstations. Local access to the periodical literature has
become a common feature of the E-OPAC (Seiden, 1991; Locally loaded
databases, 1989).

On the other hand, the use studies also show that the objective
quality and success of end-user searches often are not high. The interface
plays a particularly critical role in the searching of bibliographic
retrieval systems that employ sophisticated information retrieval
techniques and contain records with subject-rich fields.

Yet, while it has become clear to library system designers that better
interfaces and "front-end" technologies can greatly enhance end-user
searching of today's large bibliographic databases, examination of the
information science and computer science literature reveals that there
are no prescriptive models that can be followed to arrive at an optimum
interface design (Grudin, 1989; Sutcliffe & McDermott, 1991; Yee, 1991).
There are no complete human-computer interaction theories (Fischer,
1989), and stable and complete guidelines for interface design are felt
to be several decades away (Shneiderman, 1987, p. 417), although a few
key interface design principles have been identified and accepted (Gould
fc Lewis, 1985; Wilson & Rosenberg, 1988, p. 865). The LIW end-user
searching interface described below, therefore, was developed from first

LIBRARY INFORMATION WORKSTATION 57

principles and in response to the specific considerations described above
rather than according to any existing prescription. It continues to be
refined and developed based on experience and observation.

A CUSTOMIZED IMPLEMENTATION EXAMPLE

As an illustrative example of the UIUC LIW as currently
implemented, Figure 2 shows the opening LIW menu as defined for
the UIUC Music Library installation of the system. Menu pick number
1, "Online Catalog," provides access to the statewide online catalog
(IO). At the Music Library, as at most sites on the UIUC campus, access
to IO is provided via user-friendly microcomputer interface software
developed by UIUC Prof. C.-C. Cheng (1985). Elsewhere in the state,
most patron access to IO is via the more recently developed mainframe
interface.

Menu pick number 2, "References to Articles in Journals &
Magazines," provides access to the statewide BRS/SEARCH imple-
mentation for searching bibliographic databases (IBIS). Note that the
database availability indicated in Figure 2 is specific to UIUC. Exact
database availability varies slightly on other campuses in the statewide
ILSCO consortium. The LIW interface to IBIS is discussed in detail
in a later section.

Menu pick number 4, "UIUCNet Services," provides access to
database and telecommunications resources on the UIUC campus fiber-
optic network. In this particular installation, one may access the
electronic version of the OED, the campus phone and e-mail address
directory, and preselected OPACs from other institutions. Figure 3 shows
a selection of UIUCNet resources specific to the Music Library
implementation of the LIW.

Data files generated and maintained by each library (menu picks
number 3 and 5) are stored on the microcomputer hard disk and searched
using a locally developed, sequential search database management
system integrated into the LIW. The search software, written in the
Microsoft BASIC Professional Development System language, was
authored by UIUC faculty members William Mischo, Timothy Cole,
and David Stern. A primary goal of developing the search software
in-house was to facilitate the interchange between IO+ applications.

This sequential search application is intended for ASCII files up
to a few megabytes. Since data files are unindexed and standard ASCII
in format, they can be created in a variety of ways. Data files can be
created by downloading from sources such as IO or IBIS; files may also
be created with standard word processors and saved as ASCII text. The

TIMOTHY W. COLE ET AL.

ILLINET ONLINE PLUS ISO*}

12:05

Select Database to be Searched

KAIN HENU SEARCH OPTIONS:
USE ARROW KEYS/HOUSE TO
SELECT, then Press ENTER

LIBRARY CATALOG Choices:

Indiana University
Northwestern University
Ohio State University
U. of California (HELVYL)
U. of Illinois at Chicago

EXIT THIS HENU

EXPLANATION: Other Library Catalogs
Search for iteas at other universities
using their online catalogs.

CHOICES:

Oxford English Dictionary
Caipus Telephone Directory
Other Library Catalogs

EXIT THIS MENU

opyright (C) i?9i, 199E Board of Trustees University of Illinois - version 3,30
Figure 3. UIUCNet resources specific to the Music Library

files can be customized for a very specific user population. Files created
for the Music Library include the Journal List, Current Acquisitions
List, Cumulative Acquisitions List, Compact Disc List, Video List,
Dissertation List, and Resource Guide (see Figure 4). A help screen
supplements the description given for each of these files in the IO+
menu "Explanation Box" (see upper right-hand corner of Figure 4).

In some cases, files are created to complement access to materials
in the online catalog. For example, in IO it is not possible to limit
one's search to only CDs, yet many of the Music Library's clientele
request CDs specifically. For several years, a separate dBase III database
was maintained. When that was no longer feasible, the sound recording
portion of the electronically prepared monthly acquisition list was
appended to a printout of the dBase III file. These files have now been
combined and may be searched using this application. Because files
are unindexed, they are updated easily; new material is simply appended.

LIBRARY INFORMATION WORKSTATION

ILLINET ONLINE PLUS (10+)

12:06

Select Database to be Searched

HA IN KENU SEARCH OPTIONS;
USE ARROW KEYS/ HOUSE TO
SELECT, then Press ENTER

Music Library Indexes and
Lists

EXPLANATION: Dissertation List

Covers School of Husic theses I dissertations

Coverage: 1936 to the present.

CHOICES:
Journal List

Current Acquisitions List
Cumulative Acquisitions List
COipact Disc List
Video List
Dissertation List
Resource Guide

EXIT THIS HEMU

opyright (C) 1991, 1992 Board of Trustees University of Illinois - version 3.30
Figure 4. Files and help screen created for the Music Library

Files are searched sequentially from beginning to end, byte by byte.
The inherent power and speed of the IBM PS/2 machine the IBM
PS/2 Model 30 286 is the recommended minimum platform for LIW
implementation combined with recent improvements in Microsoft
BASIC permits a file to be searched very quickly, even in this sequential
search manner; file indexing is unnecessary. Since the files are not
indexed, the user need not worry about searching specific fields or using
controlled vocabulary. The string search algorithm used permits both
right- and left-hand truncation.

For flexibility, the integrated LIW sequential search software does
accommodate record delineation. By creating discrete records within
the file, the use of Boolean logic becomes possible. A set of reversed
brackets is used to delineate the end of a record. End-of-record markers
may be introduced into a file by means of a word processing macro.

60 TIMOTHY W. COLE ET AL.

UNIVERSITY OF ILLINOIS HUSIC LIBRARY VIDEORECORDIM6 LIST
OPERAS

Beethoven, Ludaig van, 1770-1827,

Fidelio [videorecording] / Beethoven ; libretto, Josef Sonneleithrier, N'e*
York : Video Arts International, ci935, Cast! Elisabeth Soderstrou, Anton de
Ridder, Robert Ailasn, The London Fhilharsonic Orchestra and Slyndsbourne
Chorus.
VIDRECH1500B33F5E1985

VIDRECH1500B33F521985 BEETHOVEN, LUDHIB VAN, 1770-1827. FIDELIQ* NEW YORK
NOLC 3320793 1985 1 ADDED: 87092? NEN6

01 001 W HUQ RCALL 9H 1030/930 10* DC

02 001 SAVE 921212 UC

Press ENTER to resuae search, or type an LCS comsand,

Figure 5. Local file search results and their circulation status

In addition, files created by downloading from IO using another locally
created library staff application, Illinois Search Aid, can be automatically
supplied with reverse brackets between bibliographic records.

Available options include searching a single term, searching for
two or more terms within one record (the logical operator "and"), and
searching for any of two or more terms in a record (the logical "or").

Mentioned above was the goal of facilitating the interchange of
information between applications. One example is a local data file that
contains catalog call number information. The LIW software permits
a dynamic link between local file search results and current IO circulation
status and holding information (see Figure 5).

SEARCHING FOR JOURNAL ARTICLES ON THE LIW

Selection of the "References to Articles in Journals & Magazines"
LIW main menu pick starts a microcomputer-mediated session on IBIS,
the local implementation of the BRS/SEARCH information retrieval
system. Figure 6 shows the opening IBIS interface menu screen for an
ERIC database search. The three-windowed approach of the LIW main
menu screens is preserved providing a sense of integration and
continuity. In addition to this basic three-windowed menu approach,

LIBRARY INFORMATION WORKSTATION

the IBIS interface also uses pop-up dialog boxes and pop-up and bar
menus. Wherever possible, the interface uses menus to solicit user inputs.
Where dialog boxes are required, illustrations and detailed prompt texts
are provided.

The extensive use of menus minimizes the need for IBIS users to
know explicit command syntax or specific database features or
nomenclature. Instead, menus tailored to each IBIS database are
provided. Using the menu shown in Figure 6, even users unfamiliar
with ERIC are immediately made aware of powerful ERIC search features
such as educational level terms and age descriptor codes. The menu
approach allows patrons to use these features without having to learn
and memorize specific field or search codes. Similar special feature search
menus are provided for other IBIS databases. Consistent across these
search menus, selections are listed from broad (e.g., keyword searches;

IBIS (Illinois Bibliographic Intonation Syitea)
SEARCHING: Coaplete ERIC LAST UPDATE; 12/01/9E Choosing Type of Search

CHOOSE TYPE OF SEARCH:
USE ARROU KEYS & ENTER
or HIGHLIGHTED LETTER
Keywords or phrases
Title and subject Words
Subject Headings only
Title words only
Author
Journal title search

* Educational Level Teras *
Age Level Descriptors
Publication Types

Diagras of Boolean logic
EXIT this Database

EXPLANATION; Educational Level Tens
Search for article references by assigned
educational level tens, Can be used to
narrow search results.

CHOICES;

Early Childhood Education
Preschool Education
ELesentary Secondary
Elementary Education
Secondary Education
POstsecondary Education
Higher Education
Two Year Colleges
EXIT THIS HENU

Funded in part by BECKNAN INST for ADV SCI I TECH
Figure 6. Opening IBIS interface menu for an ERIC search

62 TIMOTHY W. COLE ET AL.

title and descriptor searches) to narrow (e.g., corporate author searches;
publication type searches).

The IBIS microcomputer interface uses an "expert systems
approach," incorporating a great deal of experienced searcher expertise
behind the scenes. All entered search terms and phrases are parsed and
checked by the microcomputer interface software before being
formulated into the proper syntax and sent off to the mainframe BRS/
SEARCH implementation for processing. Appropriate operator and
search field codes are added. Search phrases are checked for database-
specific stopwords, system-reserved words (e.g., Boolean operators;
adjacency operators) that might lead to ambiguous results and obvious
redundancy or incompatibilities with previously entered terms. The end-
user is warned or asked to reenter the search argument according to
the nature of the specific problem.

To further facilitate end-user searching, the microcomputer IBIS
interface uses a search strategy formulation technique centered around
the software creation and combination of user-entered search terms and
concept groups. This approach is patterned after the "concept building
block" approach to online searching, one of three classical techniques
for performing effective searching (Pfaffenberger, 1990, pp. 106-107).
The concept group approach was demonstrated by Marcus (1981) in
an experimental system and has been adapted in several commercial
systems, including the DIALOG CONNECTION systems (Large, 1990,
pp. 30-32) and BRS/AFTER DARK (Guidelines, 1989), and several
academic end-user systems (Pollitt, 1990; SearchMate, 1990). In addition
to being a logical method for building and modifying search strategy,
the concept building block approach also facilitates bibliographic
instruction, both in group settings and one-to-one, and eliminates the
need for user mastery of the various Boolean search operators.

Figure 7 shows the help screen that describes and illustrates this
building block search approach as implemented in the LIW IBIS
interface. This screen comes up automatically when the user begins
his or her search. A more elaborate, extensive description of this search
process can be requested by the patron from this summary help screen.
Experience with this approach in various forms at UIUC has shown
that it indeed helps address and reduce many of the end-user searching
difficulties described above (Mischo & Moore, 1989).

Of course, not all cases can be covered in manageable menus, and
there may be unanticipated occasions where parsing of a search string
may not be desirable. To accommodate this case and to allow for a
librarian to help a patron without having to exit the IO+ interface
code, two forms of BRS native mode command "pass-thru" are allowed

LIBRARY INFORMATION WORKSTATION

IBIS (Illinois Bibliographic Information Systea)
SEARCHING: Coiplete ERIC LAST UPDATE: 12/01/92 Choosing Type of Search

CHOOSE TYPE
USE ARROW KE
or HIGHLISHT
Keywords or

-EXAMPLE TOPIC SEARCH-

Construct a search topic for exasple:
PREDICTIONS OF EARTHQUAKES IN THE MIDWEST
and divide it into its separate subject
ideas or CONCEPTS (related tens).

CONCEPT 1 CONCEPT I CONCEPT 3
Earthquakes Midwest Predictions
SeisBic Activity New Hadrid Forecasts
Fault Lines Illinois
St. Louis

Enter the search ters in separate CONCEPT
groups. An 'AND' Search will retrieve
references with at least ONE TERR fro*
each CONCEPT group or coluen.

* Press ENTER to continue Search
I Help Screens on Concept searching

ases (Keywords)
is.

sate sentence.

in part by 8ECKHAN IN5T for ADV SCI 1 TECH
Figure 7. IBIS help screen

by the microcomputer interface program. For a single search term or
process that will result in the generation of a single search set, program
parsing can be disabled. For more extensive native mode activities or
to review what has been done so far in native mode, the interface can
be "turned off" completely, allowing for a direct native mode session
between the terminal and the mainframe.

Finally, allowance has been made in the software for some
workstation-specific interface tuning. The IBIS microcomputer interface
has the built-in capability to trap for search inputs specific to a particular
subject or library clientele and make automatic substitutions before
forwarding the search argument to the mainframe BRS/SEARCH
implementation (the user is notified on the search term entry screen).
Trap/substitution lists used by the interface in performing this function
can be both database and workstation specific. (An example of this
substitution operation is described below.) Additionally, available
interface output options can take advantage of specific, local workstation

TIMOTHY W. COLE ET AL.

printing and downloading capabilities. Lastly, the help screens that
describe and illustrate the search process can be modified easily for
a specific workstation or cluster of workstations.

Figures 8-14 show a sample microcomputer-mediated IBIS search
for information about interfaces to online catalogs done in the ERIC
database. In addition to performing a keyword search, the hypothetical
searcher is also aware of an author writing in the subject area and
so wants to add in any works by that author whether picked up in
the keyword search or not. Finally, having a fairly large retrieval set,
the searcher decides to limit the final search set to conference paper
or speech citations.

Figure 8 shows an initial keyword concept term entry dialog box.
Note the illustration included in the prompting for term entry. After
a user has searched for the first term of a concept, he is given the
opportunity to add additional related terms to the concept. Choosing

IBIS (Illinois Bibliographic Inforaation Systea)
SEARCHIN8: Complete ERIC LAST UPDATE: 1E/G1/92 Choosing Type of Search

CHOOSE TYPE OF SEARCH:
USE ARROW KEYS & ENTER
or HIBHLI6HTED LETTER
K

EXPLANATION: Keywords or phrases

Retrieve articles by Words or Phrases (Keywords)

in titles, subjects, and abstracts.

tence.

Entering first ter* in CONCEPT i

Enter a single ord (e.g. Earthquakes)

or a phrase (e.g. Seissic Activity in Illinois).
First teri in Concept 1:1+ ENTER Key)
online catalogs

..SEARCHING

Ver. 3.36 Funded in part by BECXHAN INST for ADV SCI & TECH

Figure 8. Initial keyword concept term entry dialog box

LIBRARY INFORMATION WORKSTATION

to do so will bring up the slightly modified search term entry box
shown in Figure 9. The user is reminded of the first search term in
the concept in the prompt for all later related terms in a concept.

After the user has entered all the terms in a given concept, he
is asked if he wants to narrow his search with an additional concept
(i.e., do a Boolean AND search), broaden his search with an additional
concept (i.e., do a Boolean OR search), or take one of several other
actions (see Figure 10). In this example, the user chooses to narrow
with an additional keyword search term. (Different types of searches
can be combined within the concept building block approach, as
illustrated by concepts 3 and 4 described below.) Figure 11 is the dialog
entry box for entering the first term of concept 2. Note the automatic
substitution for the patron's entry of "interface" as a search term. As
mentioned above, this substitution was made using a list of terms
particular to the workstation on which the search is being done.

SEARCHING,' Cosplete ERIC LAST UPDATE: 12/01/92

Current Concept 1: online catalogs;
Last Tens Entered: online catalogs

RESULT: 958

Entering SYNONYMS or RELATED TERHS for CONCEPT 1 teras.

Other topic ideas should be put in separate Concepts.

Enter ONE TERM (Word or Phrase) AT ft TIME {+ ENTER Key)

F3 to FINISH this CONCEPT, cosbine results, 60 to NEXT CONCEPT

F4 to Print or Display Results

Another Alternate ters for online catalogs :
opacs

..SEARCHING

Tunned in part by BECKKAN INST for ADV SCI 1 TECH
Figure 9. Modified search term entry box

TIMOTHY W. COLE ET AL.

SEARCHING: Cssplete ERIC LAST UPDATE: 12/01/92

Last Concept: online catalogs; opacs;
Coabine-j Result: online catalogs; opacs;

Display/Send Coabined Result
Display/Send Any Search Result

Lisit Cosbined Result to 1985 Present
* NARROW Results with another Concept, Author, t

Organization ..(AND Search)
BROADEN Results with another Concept, Author,

Organization ..!OR Search)
Review Search History

Begin new Search in this Database
EXIT this Database

RESULT: 959
RESULT: 959

.W....TT ......... _ ,

i===~i=i= Concept
Concept 1 AND"Concep"t 2

Ver. 3.30 - 10-1-92

Funded in part by KCKflAN 1NST for ADV SCI & TECH
Figure 10. Search options menu

After adding a second concept, our hypothetical patron next decides
to broaden his search by adding in all citations attributed to a particular
author. Figure 12 shows an author entry dialog box with entry template.
This template removes any ambiguity about order or form of personal
name entry. The template is consistent across databases; database-specific
syntax and field nomenclature are taken care of behind the scenes by
the interface software.

Finally, the patron uses the main IBIS microcomputer interface
search menu to narrow his search set to "Speeches, Conference Papers."
In response to the menu selection, the interface generates and sends
the appropriate publication type search command and displays an
already filled in search term dialog box to the user (Figure 13). The
resulting retrieval set is 11 documents. An interface-generated summary
of the entire search is shown in Figure 14. Contrast this with the summary
generated by the BRS/SEARCH native-mode "..d all" command shown

LIBRARY INFORMATION WORKSTATION 67

SEARCHING: Coaplete ERIC LAST UPDATE: 12/01/ Narrowing Search with Concept 2
CONCEPT 1 online catalogs; opacs; RESULT: 959

Entering Ter<s):
Combined Results online catalogs; opacs; RESULT; 959

NARROWING SEARCH RESULTS EXPLANATION: Keywords or phrases

Entering first ten in CONCEPT I
K.

Type first Ters (word or phrase) in CONCEPT 2
interface

Itsnce.

The tera '(interfaceSI or front adj end$l or gateway*!)
has been substituted for interface.
..SEARCHING

no=

Ver. 3.30 - 10-1-92 Funded in part by BECKHAN INST for ADV SCI I TECH
Figure 11. Dialog entry box for entering the first term of concept 2

in Figure 15. Note the work done behind the scenes by the interface
software.

Appropriate adjacency and Boolean operators have been inserted
by the interface. The automatic substitution of "(INTERFACE!! OR
FRONT ADJ END$1 OR GATEWAY$1)" for the original user input
of "interface" has been made (search set 4). The author name has been
hyphenated, a preferred form constructed to improve recall, and the
name has been searched in both the author field and the abstract field
(which in ERIC is usually the only place where authors of individual
conference papers are indexed). Finally, the appropriate ERIC-specific
publication type code for "Speeches, Conference Papers" has been sent
by the program in response to the patron's menu selection. As much
as feasible, the burden for knowing proper syntax and database-specific
index practices has been shifted from the end-user to the interface
software.

TIMOTHY W. COLE ET AL.

SEARCHES: Complete ERIC LAST UPDATE: IE/01/ Broadening Search with Concept 3
CONCEPT 2 interface; RESULT: S3**

Entering Tersfs):
Combined Result: CONCEPT i AND CONCEPT 2 RESULT: 99

BROADENING SEARCH RESULTS! EXPLANATION: Personal Naie

Entering AUTHOR Search
AUTHOR'S LAST HAHE__iischo

AUTHOR'S FIRST NAHE/INITIAL__wiiIias
(If unknown, press EHTER~KEY)

AUTHOR'S MIDDLE INITIAL__h

(If unknown, press ENTER KEY)

** USE THE PgUp Key or UP Arrow to CHANGE ENTRIES *

..SEARCHING

naed in

'er. 3.30 - 10-1-92 Funded in part by BECKHAN INST for ADV SCI I TECH
Figure 12. Author entry dialog box with entry template

Having facilitated the patron's search, the interface also simplifies
the display/output of search results. Again the intent is to have the
user make his or her output selections from a menu and have the interface
software interpret those selections and generate an appropriate BRS/
SEARCH native mode command. Figure 16 shows the output options
available to patrons using the IBIS microcomputer interface including
e-mail and downloading of citations to diskette. After selecting the
output mode, the user then selects the desired output format as shown
in Figure 17. A typical citation printout (from an IBIS search of ISI's
Current Contents database) is shown in Figure 18.

Note the full-paragraph labelling and the formatting of the citation
printout. This is done on the mainframe by the BRS/SEARCH print-
time formatting process. The added line at the end of the citation
showing UIUC library call number and three-letter library location

LIBRARY INFORMATION WORKSTATION

SEARCHINS: Ccaplete ERIC LAST UPDATE: IE/01/ Narrowing AUTHOR Search
AUTHOR S iischo, illia h RESULT: 5

Entering Tersis):
Combined Result: (CONCEPT 1 AND CONCEPT 2} OR AUTH 1 RESULT: 103

NARROWING AUTHOR SEARCH I EXPLANATION: Speeches, Conference Papers

Entering first ter in CONCEPT 3

Type first Ten (word or phrase) in CONCEPT 3
Speeches, Conference Papers

..SEARCHING

Statistical Data
Audiovisual/Non-Print Ntls
EXIT THIS MENU

Ver. 3.30 - 10-1-92 Funded in part by BECKHAN INST for ADV SC! & TEC
Figure 13. Search term dialog box

code is added by the microcomputer interface software. The interface
recognizes the citation source field as it displays or prints the citation
and then performs a search of a local database file stored on the
workstation hard disk to find the call number of the item (using the
same algorithm developed for local database searching and described
previously). The fact that this look-up file resides on the workstation
easily accommodates the campus-to-campus variations in information
about journal availability, call numbers, and location information.

CONCLUSION: FUTURE DEVELOPMENTS

The software functions of the Library Information Workstation
have evolved over the course of the project as the various IO+ resources

70 TIMOTHY W. COLE ET AL.

SET NUMBER SEARCH TERN RESULT

01 t-nline catalogs (Keyword} 95?

02 opacs (Keyword) 31

03 CONCEPT 1 online catalogs; opacs; 959
0* CONCEPT 2 interface; 23^

05 CONCEPT 1 AND CONCEPT 2 99

06 AUTHOR 1 tischo, williai h 5
0? (CONCEPT 1 AND CONCEPT 2} OR AUTH 1 103

08 CONCEPT 3 Speeches, Conference Papers ; 68969

09 ((CONCEPT i AND CONCEPT 2) OR AUTH 1) AND CONCEPT 11

RETURN to Previous Kenu
* DISPLAY/SEND Search Results (Set)

Figure 14. Interface-generated search summary

have become available. IO+ and the UIUC campuswide information
system will continue to expand and add new resources, and the UIUC
LIW will add new features and continue to refine its present
functionality in response.

Specific work is underway to expand the UIUC LIW to provide
enhanced access to a number of new information resources including
many outside the UIUC library system itself. In December 1993, the
UIUC will open the new Grainger Engineering Library Information
Center. The Grainger Center will feature state-of-the-art computing
and networking facilities. All LIWs within the center will be able
to access campus and Internet network resources. Additionally, several
high-capacity, in-house network servers will provide a number of
supplemental information access resources and capabilities.

In-house services currently under design and development for the
new Grainger Center include networked CD-ROM databases; a

LIBRARY INFORMATION WORKSTATION 71

ERIC SCREEN 1 OF 2

ERIC LAST UPDATE: 12/01/92

1 ONLINE WITH CATALOGS

RESULT 958

2 OPACS

RESULT 81

3 1 OR 2

RESULT 959

* (INTERFACE$i OR FRONT ADJ END$1 OR GATEWAYS!)
RESULT 23W

5 3 AND *

RESULT 99

6 KISCHO-WILLIA$-H$,AU. OR MISCHC-W-H$.AU. OR WILLIAM ADJ H ADJ MISCHD OR
WILLIAM ADJ MISCHC

RESULT 5

7 5 OR 6

RESULT 103

3 150. PT.

RESULT 68969

9 ? AND 8

ERIC SCREEN 2 OF 2
RESULT 11

END OF DISPLAY mt

BRS SEARCH MODE - ENTER 2UERY

Figure 15. BRS/SEARCH search summary

((CONCEPT 1 AND CONCEPT 2) OR AUTH 1) 11 Refs

SELECT DISPLAY/PRINT/E-MAIL OPTION (+ ENTER)

* Display References on Screen
DOWNLOAD References to 3,5 in. Diskettes
Print to High-Speed Printer (NOT YET AVAILABLE)
Send Result to Electronic Mailbox
Print on Attached Printer
EXIT THIS MENU

Figure 16. Output options using the IBIS interface

TIMOTHY W. COLE ET AL.

((CONCEPT i AND CONCEPT 25 OR AUTH I) 11 Ref5

SELECT DISPLAY/PRINT/E-HAIL OPTION (+ ENTER)
Display References on Screen

FORMATS!

Author/Title List
Title, Source I Major Descriptors
Author, Title, Source & Abstract
* Full Forsat, including Abstract
KENU

Figure 17. Output formats using the IBIS interface

Grainger Center BRS/SEARCH implementation to support access to
databases of interest to Grainger patrons but not of sufficient interest
to justify loading on the statewide BRS/SEARCH implementation IBIS;
Internet access to online catalogs at other CIC schools; access from all
workstations to campus network resources such as the OED, local area
weather forecasts and alerts, and campus phone/e-mail directory; state-
of-the-art scanning, computer-aided-design, and computer-aided-
instruction hardware and software; and coordinated full-text article
retrieval and delivery via fax or image transmissions over network
connections. Figure 19 provides a preliminary indication of the kind
of services that will be available from the Grainger Center LIWs.

The LIW interfaces as well will evolve to incorporate enhanced
search techniques such as Best Match and Partial Match algorithms,
term query expansion, and knowledge-based query expansion. We are
interested in exploring the "find another like this one" approach to

LIBRARY INFORMATION WORKSTATION 73

CONCEPT 1 AND CONCEPT 2 22 Refs SCREEN i OF 2

Record i of 22

Accession nuaber: 92^3.

Author: HUSAIN-S. OBRIEN-A.
Author affiliation: LOU6HBOROUBH UKIV TECHNOL,DEPT INFORNAT i LIB STUDIES,

LOU6HBOROU8H LEli 3TU, LEICS, ENGLAND.

Article title; RECENT TRENDS IN SUBJECT ACCESS TO OPACS - AN EVALUATION.
Source (journal): INTERNATIONAL -CLASSIFICATION. 1992, V19, N3, P1W-K5.
Cited references: 0043.
Genuine Article I: Ji395.

Language: ENGLISH (EN).
Publication type: ARTICLE 1U).
Subject keywords: HACH1NE-READABLE-LCSH. ONLINE-CATALOGS.

INFORMATION-RETRIEVAL, ELEMENTS .
Subject category: LIBRARY-i-INFORHATION-SCIENCES (in BEKA).

Abstract: Research conducted in the early 80's has shown that subject
acces= is still one of the lost dostinant approaches in OPACs. However, while
sose of the subject searches result ir, no recall, others often retrieve so such
Call No.: Q.025.U612 LSX

Next Prev NUaber *dark* Change PRint/Send TOC SOrt EXit
Hark Reference for Print ing /Down load ing iNOT E-HAIL)

Figure 18. Citation printout from an IBIS search

expanding search results. Interfaces will increasingly make use of state-
of-the-art multimedia and graphical techniques to accommodate and
facilitate more effective user information-seeking behaviors.

The UIUC Library IO+ extended OPAC and campuswide
information services provide users with access to a myriad of local and
remote information resources. The Library is committed to making these
resources available and facilitating their use through the continued
evolution of the Library Information Workstation.

TIMOTHY W. COLE ET AL.

1 111 NET ONLINE PLUS (10*)

Select Database to be Searched

MAIN MENU SEARCH OPTIONS:
USE ARROW KEYS/ROUSE TO
SELECT, then Press ENTER

Online Catalog

References to Articles
in Journals I Magazines

Inforaation I Help files
* CoasurucatioTiS
Materials Processing
Word Processors
Exit Options

EXPLANATION: Coaaunications
Connect to Building LAN, UIUCNet,
or Internet network resources, or
or open a aodea connection.

CHOICES:

Oxford English Dictionary
Catpus Telephone Directory
Heather Infortation
Gopher Information Systems
Other Library Catalogs
Dasher Connection
TCP/IP Connection
Modes Connection

EXIT THIS MENU

opyright 1C) S?91, 199E Board of Trustees University of Illinois - version 3,30
Figure 19. Future services available from the Grainger Center LIWs

REFERENCES

Ankeny, M. L. (1991). Evaluating end-user services: Success or satisfaction? Journal of

Academic Librarianship, 16(6), 352-356.
Arms, C. (Ed.). (1990). Campus strategies for libraries and electronic information.

Rockport, MA: Digital Press (EDUCOM Strategies Series on Information Technology).
Borgman, C. L. (1986). Why are online catalogs hard to use? Lessons learned from

information-retrieval studies. Journal of the American Society for Information Science,

37(6), 387-400.
Charles, S. K., & Clark, K. E. (1990). Enhancing CD-ROM searches with online updates:

An examination of end-user needs, strategies, and problems. College b Research

Libraries, 5/(4), 321-328.
Cheng, C.-C. (1985). Microcomputer-based user interface. [Special section: In depth

The online catalogue of the University of Illinois at Urbana-Champaign]. Information

Technology and Libraries, 4(4), 346-351.
Drabenstott, K. M. (1991). Online catalog user needs and behavior. In N. Van Pulis

(Ed.), Think tank on the present and future of the online catalog: Proceedings (pp.

59-83). Chicago: Reference and Adult Services Division, American Library Association.

LIBRARY INFORMATION WORKSTATION 75

Fischer, G. (1989). Human-computer interaction software: Lessons learned, challenges

ahead. IEEE Software, 6(1), 44-52.
Gould, J. D., & Lewis, C. ( 1985). Designing for usability: Key principles and what designers

think. Communications of the ACM, 25(3), 300-311.
Grudin, J. (1989). The case against user interface consistency. Communications of the

ACM, 32(10), 1164-1173.

Guidelines: Search strategy development. (1989). BRS Bulletin, 13(3).
Hildreth, C. R. (1987). Beyond Boolean: Designing the next generation of online catalogs.

Library Trends, 35(4), 647-667.
Hildreth, C. R. (1989). General introduction; OPAC research: Laying the groundwork

for future OPAC design. In C. R. Hildreth (Ed.),, The online catalogue: Developments

and directions (pp. 1-24). London: Library Association.
Hildreth, C. R. (1991). Advancing toward the E-3 OPAC: The imperative and the path.

In N. Van Pulis (Ed.), Think tank on the present and future of the online catalog:

Proceedings (pp. 17-38). Chicago: Reference and Adult Services Division, American

Library Association.
Hunter, R. N. (1991). Successes and failures of patrons searching the online catalog at

a large academic library: A transaction log analysis. RQ, 30(3), 395-402.
Large, J. A. (1990). Software developments. In P. T. Bysouth (Ed.), End-user searching:

The effective gateway to published information (pp. 19-43). London: Aslib.
Locally loaded databases in online library systems [Special issue]. Information Technology

and Libraries, 8(2), 99-185.
Marcus, R. S. (1981). An automated expert assistant for information retrieval. In L. F.

Lunin, M. Henderson, 8c H. Wooster (Eds.), The information community: An alliance

for progress (Proceedings of the 44th ASIS Annual Meeting) (Vol. 18, pp. 270-273).

White Plains, NY: Knowledge Industry Publications.
Mathews, J. R. (1991). The online catalog: Time to move beyond the boundary of a

catalog! In N. Van Pulis (Ed.), Think tank on the present and future of the online

catalog: Proceedings (pp. 5-16). Chicago: Reference and Adult Services Division,

American Library Association.
Mischo, W. H., & Cole, T. W. (1992). The Illinois extended OPAC: Library information

workstation design and development. In M. Ra (Ed.), Advances in online public

access catalogs (pp. 38-57). New York: Meckler.
Mischo, W. H., & Lee, J. (1987). End-user searching of bibliographic databases. In M.

E. Williams (Ed.), Annual review of information science and technology (Vol. 22,

pp. 227-263). New York: Elsevier.
Mischo, W. H., & Moore, A. F. (1989). Enhanced access to periodical literature within

an online catalogue environment. In C. R. Hildreth (Ed.), The online catalogue:

Developments and directions (pp. 107-126). London: Library Association.
Mischo, W. H.; Sandore, B.; Clark, S. E.; & Gorman, M. (1990). University of Illinois

at Urbana-Champaign. In C. Arms (Ed.), Campus strategies for libraries and electronic

information (pp. 117-141). Rockport, MA: Digital Press (EDUCOM Strategies Series

on Information Technology).
Nielsen, B. (1986). What they say they do and what they do: Assessing online catalog

use instruction through transaction monitoring. Information Technology and

Libraries, 5(1), 28-34.
Norlin, D. A.; Cardman, E. R.; Davis, E. B.; Dossett, R.; Henigman, B.; Mischo, W.

H.; & Troutman, L. (1992a). Interface design and development: The human factor.

Library Hi-Tech, 70(3), 7-24.
Norlin, D. A.; Cardman, E. R.; Davis, E. B.; Dossett, R.; Henigman, B.; Mischo, W.

H.; & Troutman, L. (1992b). Dynamics of interface design and development [Abstract].

In T. Kirk (Ed.), Academic libraries: Achieving excellence in higher education:

Proceedings of the Sixth National Conference of the Association of College and

Research Libraries (pp. 374-376). Chicago: Association of College and Research

Libraries.

76 TIMOTHY W. COLE ET AL.

Peters, T. A. (1989). When smart people fail: An analysis of the transaction log of an
online public access catalog. Journal of Academic Librarianship, 75(5), 267-273.

Pfaffenberger, B. (1990). Democratizing information: Online databases and the rise of
end-user searching. Boston: G. K. Hall.

Pollitt, A. S. (1990). Intelligent interfaces to text retrieval systems. In P. Gillman (Ed.),
Text retrieval: The state of the art (Proceedings of the Institute of Information Scientists
text retrieval conferences, "The User's Perspective" [1988] and "Text Management"
[1989]) (pp. 192-208). London: Taylor Graham.

Potter, W. G. (1989). Expanding the online catalog. Information Technology and Libraries,
8(2), 99-104.

Puttapithakporn, S. (1990). Interface design and user problems and errors: A case study
of novice searchers. RQ, 30(2), 195-204.

Schultz, K., & Salomon, K. (1990). End users respond to CD-ROM: What users really
think. Library Journal, 7/5(2), 56-58.

SearchMate strategy development. (1990). In SearchMate search manual. Los Angeles,
CA: University of Southern California Health Sciences Library.

Seiden P. (Ed.). (1991). Survey of libraries providing locally mounted databases. Chicago:
Reference and Adult Services Division, American Library Association.

Shneiderman, B. (1987). Designing the user interface: Strategies for effective human-
computer interaction. Reading, MA: Addison-Wesley.

Sutcliffe, A. G., & McDermott, M. (1991). Integrating methods of human-computer
interface design with structured systems development. International Journal of Man-
Machine Studies, 34(b), 631-632.

Tiefel, V. (1991). The Gateway to information: A system redefines how libraries are used.
American Libraries, 22(9), 858-860.

Wilson, J., 8c Rosenberg, D. (1988). Rapid prototyping for user interface design. In M.
Helander (Ed.), Handbook of human-computer interaction (pp. 859-875). New York:
Elsevier.

Yee, M. M. (1991). System design and cataloging meet the user: User interfaces to online
public access catalogs. Journal of the American Society for Information Science, 42
(2), 78-98.

JEAN ARMOUR POLLY*

Assistant Director, Public Services

Head, Microcomputer Services

Liverpool Public Library

Liverpool, New York

Somebody Knock in':
The Public Library at the Electronic Door

ABSTRACT

A suburban public library with a long history of technological innovation
chronicles its adventures during its first year of Internet connectivity,
including staff use of electronic mail, TELNET, and File Transfer
Protocol (FTP). Future plans include public use from the computer
lab located in the library. A resource section includes information on
how to get on the Internet and how to learn more about it through
user guides.

COMPUTER USE AT THE LIVERPOOL PUBLIC LIBRARY

The Liverpool Public Library, in central New York State, has a
long history of computer use for both staff and patrons. In October
1991, we celebrated ten years of public computing. What had begun
in 1981 with a 48K Apple 11+ has grown into a multimedia
lab/playground with seven computers and related peripherals. Over
1,400 hours per month are reserved on two Macintosh LC's, a Macintosh
SE, an Apple HGS, an Apple He, or an IBM compatible. Clients tinker
with their rsums and learn how to use application software like word

*The author is now the Manager of Network Development and User Training at
NYSERNet, Inc.

78 JEAN ARMO UR POLLY

processors, databases, and spreadsheets. Home-schooled students use the
computers for drill and practice. Churches produce newsletters; clubs
produce mailing labels.

The seventh system is called the Emerging Technologies
Workstation. It is a color Macintosh Ilci with attached videodisk and
CD-ROM players as well as an audio mixer and headphones. This
equipment was part of an Apple Library of Tomorrow grant, which
the library received in 1990. Currently, we are beta-testing the Library
of Congress's American Memory Project interactive videodisks, but more
on that later.

About a year ago, our mid-level regional network, NYSERNet, Inc.,
approached us with a deal that we could not refuse. We accepted a
grant from them and became the nation's smallest public library with
Internet connectivity. NYSERNet, Inc., is an unusual nonprofit
corporation "whose mission is to advance science, education, and
research through the interchange of information via computer
networks." Affiliates include over 40 academic sites, libraries, nonprofit
organizations, and research and government facilities. Their goal is New
York Statewide connectivity for everyone. NYSERNet is aggressively
bringing new users to the Internet. Recently, a number of sites have
come online due to their New Connections grant program. Besides our
library, others include the American Museum of Natural History, the
Russell Sage Foundation, the New York Public Library, and various
K-12 sites.

Under our grant, NYSERNet subsidizes network costs for a trial
period of varying lengths. The site must provide its own phone line,
9600-baud modem, and computer (MS-DOS compatible, Macintosh, or
Sun system). At the end of the grant period, we can elect to keep the
software and pay for our connection, which at this time is about $200
per month.

Sometime during 1992, we hope to offer Internet access to lab clients.
It is unclear at this time what form this access will take. Will users
have "accounts" and private mailboxes so that they can exchange
electronic mail (e-mail) and subscribe to listservs and newsgroups? Or
will the front-end be just a screen or two of a "Top Ten" list of interesting
places to telnet to?

We also have developed a proposal to provide a "Free-Net-type"
dial-in system that would allow up to 64 concurrent sessions via a
standard, dedicated Internet connection. The Cleveland Free-Net,
described later in this paper, is a wonderful pilot, but it operates under
the UNIX operating system. We think that most small libraries will
not have a UNIX box nor a UNIX wizard on-site, so we propose to
develop this project under the Macintosh operating system. Not that
we think most public libraries are able to afford a standard connection

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 79

and a support staff like ours, either, but we hope to pave the way and
show what can be done with these resources.

We think that ALL librarians need access to e-mail right away.
They can get this connectivity with a minimum of equipment: a
computer, phone line, communications software, and low-speed modem.
Average costs are about $20 per month for e-mail only.

LOCAL ELECTRONIC MAIL

Our site uses QuickMail (CE Software) as its in-house mailer. Most
full-time staff members have e-mail addresses on our system. They can
log in to these mailboxes from any computer on our local area network
(LAN). Mail is stored on a server located in the Computer Department.
When someone logs in, the server is checked for mail addressed to that
person. New mail may be read, stored, forwarded, printed, or deleted.
Users can compose new mail and easily send it to individuals, or
"groups" of individuals may be set up under one address, like "all
Department Heads."

Mail can be carbon copied to others as well as blind carbon copied
meaning the main recipient does not know you have also copied others.
Mail can even be "unsent" if it has not been read by the addressee.
This function is useful if you change your mind about sending that
nasty note!

E-mail has many benefits, although there are drawbacks, too. Some
of the good things:

Saves paper, although not if you print out every message. Most users
do not do this, instead storing messages electronically on their
computers.

Improves communication among those on different work schedules
and in different work groups. It is time efficient to send an electronic
message to many people rather than making one memo, using a
photocopier to make many copies, then physically distributing all
the copies.

Avoids face-to-face confrontation. Sometimes this is also more time
efficient since meeting with someone to communicate one message
may lead to more conversation about other issues, which time
management may not allow.

One of the negative things about e-mail is that it allows users to
avoid face-to-face confrontation. Everyone likes to distance themselves
from conflict, and e-mail is just one more way to do it. Our system
even lets you write a memo and time it to be sent after you have left
the building and are safely 10 miles away!

80 JEAN ARMOUR POLLY

Reprimands via e-mail are never a good idea, and many of us have
a reminder note taped to our monitors that says "bad news should be
delivered only in person."

USE OF THE QUICKMAIL-TO-INTERNET GATEWAY

We have had e-mail throughout our building for many years, and
everyone is used to the easily understood interface. Once connected
to the Internet, we feared that its users would need to learn another
mailer. Fortunately, we use a combination of products that allows our
Internet mail to appear in our QuickMail mailboxes alongside our in-
house library mail.

QuickMail Administrator runs all traffic through the various
mailcenters we have set up. QM Admin takes care of where the mail
is, who it is addressed to, who has read what, what has been deleted,
etc.

One mailcenter is called Internet. Anything sent to this mailcenter
is handled by a third-party product called UMCP QuickMail. This
software "gateway" has been configured to grab the 9600-baud modem
and a standard phone line every 30 minutes and call out to the local
PSI (Performance Systems International) POP (point of presence).

From there it hops to a UUCP (UNIX-to-UNIX Copy Program)
mail server located Out There, exchanges mail from and to addresses
at lpl.org (our Internet domain name), and logs off.

UMCP QuickMail then releases the line and resets the modem.
It goes through all mail received and re-sorts it into the local mailboxes
of our users. QuickMail Remote lets our users call in from their home
systems and send and receive mail just as they would do at their desks
on-site. Our only problem with this is speed it is much faster at work
than at home, and only one person at a time can log on from a remote
location.

Readers used to academic computing centers and many incoming
lines and modem ports will scoff at the "paper clip and baling twine"
method of getting Internet mail to our public library clients but on
the other hand, it's cheap and it works.

TELNET AND FTP

QuickMail will not help us if we want to connect to a remote
computer and use its resources using TELNET or if we want to acquire
files or software using anonymous File Transfer Protocol (FTP). For
that we use Intercon's TCP Connect II. Its graphical user interface

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 81

is a big step toward making Internet use much less arcane. This program
allows us to call the local POP and initiate a SLIP (Serial Line Internet
Protocol) connection to the fast network backbone.

Most academic and research sites have a dedicated connection to
the Internet and a static numeric address. Our SLIP connection gives
us a dynamic address it changes depending on what modem we have
hit on the way in. Once we have been given our address for that particular
session, we become what amounts to any other "host" computer on
the Internet but only for as long as we stay connected. Once we log
off, we are truly gone, known only to the domain name servers Out
There in dataspace.

TCP Connect II's TELNET and FTP tools are easy to use and
support the Macintosh interface. But we do not think they are easy
enough for the public to use. Getting the assigned dynamic IP (Internet
protocol) number slid into the right place is something we do not feel
clients in our lab should have to do. We want a macro to log on, get
the IP, put it where it needs to go, and then ask what the client wants
to do. We have not been able to do this using this product. Many other
telecommunications programs allow creation of scripts like this, but
none of them works using a SLIP connection so far. Or if they do
support SLIP, they must run under Ethernet at your site and we are
running only PhoneNet/LocalTalk at this point. Remember, I said this
was a cheap solution! Nothing is easy on the frontier of telecom-
munications. . . .

WHAT WE HAVE BEEN DOING WITH OUR
INTERNET CONNECTION

In the past year, we have "subscribed" many employees to various
discussion groups. Most of these are job related, such as PACS-L (Public-
Access Computer Systems) or LIBADMIN (Library Administration),
although we do have some "fun" subscriptions such as DOROTHY-
L (for mystery readers) and BIRD. EAST, which is the Audubon hotline
for announcements of unusual bird sightings. Recently, after receiving
one of these hotline reports, we all piled into the van to inspect a
rare arctic visitor: a hawk owl who had settled on a tree about 30 miles
from the library.

We are disseminating some of this Internet wealth in print form
to our library patrons. Our Mystery Readers Club in particular enjoys
the information they get from DOROTHY-L. Audubon sightings have
been distributed to the local Audubon Club, birdseed boutiques, and
other interested birders.

82 JEAN ARMOUR POLLY

It is estimated that there are now over 12,000 moderated
lists/discussion groups on the Internet. The figure does not count the
many unmoderated lists available. Lists exist on everything from
Arthurian legend to origami. It is true that a good percentage of the
communication on these lists amounts to so much "line noise" that
is, unnecessary use of bandwidth, personal flames (tirades), and other
distractions. However, MUCH of the information exchanged is solid
and useful to our lives as professionals as well as helping us achieve
personal growth.

No more are we an isolated, small public library hardly looking
beyond our own service area borders. Now we are discussing quality
circles, coping with shrinking budgets, and personnel issues with our
counterparts all over the world.

We compare notes on computer hardware and software and are
able to get personal recommendations of products from real users rather
than relying on vendor hype and reviews, if we can find them. We
are expressing our views on public library on-ramps to the National
Research and Education Network (NREN) and policies that will affect
our patrons' use of new technologies. Telecommunicating through the
Internet has helped us level the playing field for discussions with
government officials, network providers, policy makers, and other users.
Now that we have got Internet connectivity, we do not remember life
without it.

NYSERNET AND THE NEW YORK STATE LIBRARY

Of course now we want all the other libraries in New York State
to have what we have. There is hope. In 1989, the New York Statewide
Automation Committee released a report on the telecommunications
future of New York State's 7,000 libraries. It proposed the idea of the
"Electronic Doorway" through which even the smallest and most remote
libraries could access the resources of other libraries around the state.

The New York State Library is the largest state library in the nation,
with over 5.5 million items in its collection. In 1992, it will join
NYSERNet and make its online public access catalog (OPAC) available
to Internet researchers.

The State Library and NYSERNet will collaborate on a joint
initiative to begin implementation of the Electronic Doorway concept.
They will investigate possibilities for staging a replacement of the New
York State interlibrary loan system (NYSILL) and deployment of a
statewide e-mail system among libraries and library systems.

Although there are some small New York libraries currently without
even telephones, let alone computers, we support this initiative. Herbert

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 83

S. White says that the poorer and more remote a library is, the more
it needs technology.

Recently, wireless communications technology has taken off, and
we see demonstrations of the following:

wireless LocalTalk networks;

the wireless equivalent of Tl network speed over a line-of-sight
distance of three miles;

products such as the Mobidem modem, which provides two-way packet
radio Internet connectivity;

products such as the HP 95-LX palmtop computer and the Motorola
Newstream pager (dial 1-800-Skyword with your modem, address a
friend's personal ID number, and send a message directly to the
computer in his backpack; it uses satellites and magic to deliver your
instantaneous thoughts for about $40 a month).

All this means is that we will not continue to be dependent on wired,
land-based telecommunications infrastructures. This is of particular
interest to rural libraries.

MULTIMEDIA ON THE NET?

The Library of Congress's American Memory Project is currently
in beta-test at 37 libraries around the United States. Our library is one
of the handful of public libraries using it. It is available for use in
our public computer lab.

The project brings primary source material "out of the archives
and into the streets." In its latest incarnation, it includes three videodisks
of material including the following:

25,000 postcards of turn-of-the-century American landmarks from the
Detroit Publishing Company;

Nation's Forum audio archives of political speeches and portraits
of the speakers;

films of the building of New York City, boat tours of the harbors,
films of the 1901 Pan-American Exposition in Buffalo, New York;

18th century broadsides;

political cartoons.

Eventually, the Library of Congress hopes to make resource material
like this available over the NREN. We think their graphic interface
is particularly easy to use.

Interesting uses of telecommunications technology were dem-
onstrated at the March National NET'92 Conference in Washington,
DC, including packet video. Conference goers saw and spoke to a

84 JEAN ARMO UR POLLY

researcher in North Carolina, who appeared on a color computer screen.
The quality was excellent although it works best for "talking heads"
teleconferencing and not for full-motion. The video was coming over
the Internet, not being broadcast from a satellite.

WHAT USERS NEED

As cheap connectivity and hardware become more ubiquitous in
homes and as video dial tones and ISDN (Integrated Services Digital
Network) proliferate, will the public library survive? Many copyright-
free electronic texts can be downloaded from Project Gutenberg now
(get to it via Gophers); some journals now exist in electronic-only
editions; publishing books on demand, tailored to the reader's interests,
is now a reality. Brian Kahin's new book Building Information
Infrastructure, published under the Primis imprint by McGraw-Hill,
is one of these. Kahin writes, "the chapters can be printed individually
or in combination with any other material in the Primis database to
create custom textbooks ... we wanted a Protean publication whose
many parts could be assembled and reassembled to fit the interests of
many readers" (pp. 3-4).

Users can even telnet to the Stanford University Bookstore (via
MELVYL, telnet 31.1.0.11 or telnet melvyl) and electronically peruse the
wares. Rumor has it that students at least will soon be able to purchase
their books online using a credit card. Although the Rest of Us cannot
order books via Internet, we can call the bookstore during normal business
hours and place an order (order desk phone: (800) 533-2670, fax: (415)
322-1936).

If library users can do all this from home, will public librarians
still have jobs? The answer is yes as long as the Internet remains difficult
to use.

Although interfaces are becoming easier, many people prefer to
simply ask a human. In our computer lab, we note that people will
not read a simple one-paragraph "start-up instructions" note posted at
the computer workstation. They much prefer help from another person.

We wonder how automatic teller machines (ATMs) have impacted
bank business. Have they lessened traffic through the line inside? Will
we see less telephone and walk-up reference as people are empowered
to search with Gophers and archies and whatnot from home? Or will
we simply see our emphasis shift from a focus less on a facility-based
collection to more of an individualized client-based "collection" spread
over host computers on many continents? Will we see publicly funded
"911 for Information" network information centers, operated around
the clock, staffed by librarians who know where the Internet goods
are and who's got 'em?

And do we as a profession have the Right Stuff? How do we avoid
being caught in traditional library backwaters so we can emerge as

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 85

librarians surfing the Internet? Training! Training! Training! Get e-
mail as fast as you can even if you have to pay for it personally. Join
some discussion groups even if you just absorb what others are saying.
Get involved somewhere and gopher ill

86 JEAN ARMO UR POLLY

APPENDIX A
WANT TO GET CONNECTED TO THE INTERNET?

To get connected to the Internet, we recommend talking to your regional
network provider first since they may be able to provide libraries, schools, and
nonprofits low-cost or subsidized connectivity. If you do not know who your
regional is, contact the NSFNet people at 10 Moulton Street, Cambridge, MA
02138; phone: (617) 873-3400. There are many low-cost ways of exchanging global
e-mail, including FidoNet and FrEdMail, which are not described here. A few
other methods of connecting to the net, which include the ability to use TELNET
and FTP, follow.

CERFnet

The California Education and Research Federation (CERFnet) offers DIAL
N'CERF USA. It allows access to the Internet from anywhere in the continental
United States. Users dial a toll-free number to log in to remote machines, transfer
files, and send and receive e-mail. The cost is |20 a month with a $10-per-
hour usage fee. There is an installation charge of $50. For more information,
contact CERFnet, California Education and Research Federation, c/o San Diego
Supercomputer Center, P.O. Box 85608, San Diego, CA 92186-9784; e-mail:
help@cerf.net; phone: (800) 876-CERF or (619) 534-5087.

Performance Systems International (PSI)

PSI offers several varieties of network connectivity, including e-mail-only
accounts, e-mail and TELNET accounts, dial-up host connectivity on demand,
and dedicated connections. Costs are competitive, and performance is reliable.
PSILink, e-mail, and delayed FTP are $19 a month for 2400-baud service or
below, $29 per month for 9600-baud service. CDS (Global Dial-up Service)
includes TELNET and rlogins at $39 a month, 2400-baud, 24-hour access. Host
DCS (Dial-up Connection Service), at about $2,000 per year, includes a full
suite of Internet activities (mail, news, FTP, TELNET). PSI has POPs in over
40 U.S. cities. For more information, contact Performance Systems International,
Inc., 11800 Sunrise Valley Dr., Suite 1100, Reston, VA 22091; phone: (800) 82PSI82
or (703) 620-6651; fax: (703) 620-4586; e-mail: info@psi.com. Entering all-
info@psi.com generates an automatic response containing summaries of various
PSI products.

The World

Software Tool & Die runs a public access UNIX system called The World.
Basic rates are $2 per hour and a $5 monthly account fee. Services offered
by The World include Internet e-mail, USENET news, ClariNet (UPI, AP,
and satellite news services), real-time chat, UNIX Software, archie, the Online
Book Initiative (a publicly accessible repository for freely redistributable
collections of textual information a net-worker's library). The World can also
be accessed over the CompuServe Packet Network. You do not have to be a
CompuServe subscriber to use this network, but you will be billed for its use.

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 87

For more information, contact The WORLD, Software Tool 8c Die, 1330 Beacon
Street, Brookline, MA 02146; phone: (617) 739-0202.

AlterNet . . . "IP Networking for the Rest of Us"

Another twist on low-cost networking is AlterNet. They also lease
equipment like routers and modems and will preconfigure "plug and play"
hardware solutions for you. They will also deal with getting 56k, Tl, or T3
lines run to your site. AlterNet Domestic Service Charges follow (effective 9/
13/91):

Type Monthly Rate Startup Charge

$2,000

$5,000

Tl-LV (Low Volume)

$1,000

$5,000

56k DDS

$1,000

$2,000

Subrate DDS

$ 500

$2,000 (19.2k/9.6k)

Async SLIP/PPP

$ 250

$1,500 (V.32 or V.32bis)

For more information, contact UUNET Technologies, Inc., 3110 Fairview Park
Drive, Suite 570, Falls Church, VA 22042; phone: (800) 4UUNET3, (703) 876-
5050; fax: (703) 876-5059; Internet: alternet-info@uunet.uu.net.

Netcom

If you are in the San Francisco Bay area, check these folks for connectivity.
They provide a complete dial-up and communication service including netnews,
e-mail, Internet access, and general computing services. They anticipate
expansion of their network to southern California (Los Angeles) and Sacramento
by early 1992. Personal accounts are based on a fixed monthly fee without
any hourly or connect charges. There are two different pricing schedules:
personal account (invoiced monthly), $19.50/month or personal account (auto-
billing) $17.50/month. Standard connections and business connections have
varying rates. For more information, (408) 554-UNIX; Internet:
info@netcom.com.

88 JEAN ARMOUR POLLY

APPENDIX B
RESOURCES FOR LEARNING MORE

Computer Systems Policy Project

At the March 1992 National NET'92 Conference in Washington, DC, a
representative of the Computer Systems Policy Project provided a broader vision
of what the NREN could accomplish. An illuminating videotape spotlighted
innovative services and health care for seniors, improved education and lifelong
learning opportunities, advances in industrial design and manufacturing, as
well as broad access to libraries, databases, and e-mail. Copies of the full report,
along with the video, are available for $20 from the Computer Systems Policy
Project, 1735 New York Avenue, NW, Suite 500, Washington, DC 20006; phone:
(202) 628-1700.

Emily Postnews Answers Your Questions on Netiquette

Brad Templeton's (brad@looking.on.ca) piece on how NOT to behave on
the net, Emily Postnews is the foremost authority on proper net behavior, giving
satirical and hilarious advice. To ftp the most recent update, enter pit-
manager.mit.edu. Here is a sample:

Dear Miss Postnews: How long should my signature be? verbose@noisy
A: Dear Verbose: Please try and make your signature as long as you can.
It's much more important than your article, of course, so try to have more
lines of signature than actual text. Try to include a large graphic made
of ASCII characters, plus lots of cute quotes and slogans. People will never
tire of reading these pearls of wisdom again and again, and you will soon
become personally associated with the joy each reader feels at seeing yet
another delightful repeat of your signature. Be sure as well to include a
complete map of USENET with each signature, to show how anybody can
get mail to you from any site in the world. Be sure to include Internet
gateways as well. Also tell people on your own site how to mail to you.
Give independent addresses for Internet, UUCP, and BITNET, even if they
are all the same.

Hitchhiker's Guide to the Internet

As many Network Information Centers are doing, the CERFnet NIC stores
many Internet guides and RFCs (requests for comments), including the famous,
if technical, Hitchhiker's Guide to the Internet. These may be obtained via
anonymous FTP to nic.cerf.net (192.102.249.3). Call the CERFnet Hotline at
(800) 876-CERF for assistance. Ask for their Captain Internet and CERFBoy
comic, too. For more information, contact California Education and Research
Federation, c/o San Diego Supercomputer Center, P.O. Box 85608, San Diego,
CA 92186-9784; e-mail: help@cerf.net; phone: (800) 876-CERF or (619) 534-5087.

Library Resources on the Internet: Strategies for Selection and Use

RASD Occasional Paper no. 12, published in 1992, sells for $18 for members,
$20 for nonmembers. It can be ordered from ALA Order Services, 50 E. Huron,

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 89

Chicago, IL 60611, (800) 545-2433. Electronic versions are available via FTP:
ASCII file from host dla.ucop.edu, directory pub/internet, filename libcat-guide;
host ftp.unt.edu, directory library; filename libcat-guide. WordPerfect 5.1 file
from host hydra.uwo.ca, directory libsof t; filename internet.com.

Mining the Internet

There is a guidebook called Mining the Internet available from the
University of California at Davis. Here is how the Gold Country Mining
Instructions begin:

Jist durn tuckered o' workin' eight t' five for a salary, ain't you? An' you
wanna set out for parts unknown. You're hankerin' for an a'venture. Come'n
then go 'Mining the Internet' with me, father of Clementine (that's my
darlin'), and I'll tell you some old timey tales and introduce you to a new
resource for students, faculty, and staff called wide area networking. . . .
'Taint goin' to hurt you any, and the prospect looks good for a lucky strike.

Mining the Internet and Using the Internet AbB are available from
Computing Services, University of California, Davis, CA 95616-8563; phone:
(916) 752-0233; or electronically by anonymous FTP from ucdavis.edu
(128.120.2.1) directory ucd.netdocs/mining.

New User's Guide to Unique and Interesting Resources on the
Internet 2.0

Available from NYSERNet (New York State Education and Research
Network), it is over 145 pages and lists some 50 sources OPACs, databases,
information resources, and more. May be obtained electronically by anonymous
FTP from nysernet.org directory /pub/resources/guides. The cost is $25 or $18
for NYSERNet members.

Beyond the Walls: Networked Information Kit

Linda Carl of NYSERNet describes this as "an excellent introduction to
the possibilities of 'The World of Networked Information'. This instructional
package, of use in presenting the possibilities and benefits of electronic
networking to groups, consists of a videotape and an instructional notebook
package." The cost is $99 (price includes postage and handling) and $49 for
NYSERNet affiliates. Send a check (made out to NYSERNet) or purchase order.
Send with your name, U.S. mail address, and phone number to NYSERNet
New User's Guide, NYSERNet, Inc., 200 Elwood Davis Rd., Suite 103, Liverpool,
NY 13088-6147; phone: (315) 453-2912; e-mail: info@nysernet.org.

North WestNet User Services Internet Resource Guide

NorthWestNet offers a 300-page guide to the Internet, covering e-mail,
file transfer, remote login, discussion groups, online library catalogs, and
supercomputer access. Copies may be purchased for $20 from NorthWestNet.
NorthWestNet, 15400 SE 30th Place, Suite 202, Bellevue, WA 98007; phone:
(206) 562-3000; fax: (206) 562-4822.

90 JEAN ARMO UR POLLY

NSF Internet Tour HyperCard Stack

This guide includes net history, net maps, and net poetry and lore and
is free. The NSF Service Center also publishes a very complete Internet Resource
Guide ($15). Many items, including the HyperCard Tour to the Internet, are
freely available by anonymous FTP from nnsc.nsf.net. For more information,
contact NSF Network Service Center (NNSC), Bolt Beranek and Newman Inc.,
10 Moulton Street, Cambridge, MA 02138; phone: (617) 873-3400; e-mail:
nnsc@nnsc. nsf . net.

Special Internet Connections

Compiled by Scott Yanoff, this indispensable list of network resources is
available using TELNET and FTP and is updated weekly. It includes a few
OPACs, chat lines, weatherservers, campuswide information systems, and
reference resources. Send e-mail to the list manager (Scott Yanoff) at
yanoff@csd4.csd.uwm.edu or FTP to csd4.csd.uwm.edu, and the filename is
inet-services.

There's Gold in Them Thar Networks! or Searching for Treasure
in all the Wrong Places

Written by Jerry Martin at Ohio State University, this document is available
via Internet message to infoserver@nnsc.nsf.net. Once inside the message area,
give the following commands to retrieve the document:

REQUEST: NSFNET

TOPIC: NETWORK-TREASURES

REQUEST: END

Zen and the Art of the Internet

This guide is the BEST and unfortunately hardest to get unless you are
connected. This will be published in book form sometime in spring/summer
1992. Contact the author, Brendan Kehoe, Sun Network Manager, Widener
University, Chester, PA; e-mail: brendan@cs.widener.edu. Electronic editions
at ftp.uu.net (137.39.1.9) in /inet/doc; ftp.cs.toronto.edu (128.100.3.6) in pub/
zen; ftp.cs.widener.edu (147.31.254.132) in pub/zen as zen-l.O.tar.Z, zen-l.O.dvi,
and zen-l.O.PS; ftp.sura.net (128.167.254.179) in pub/nic as zen-l.O.PS. If you
are limited to UUCP, you can get it anonymously by dialing UUNET at 900-
GOT-SRCS and get the file /inet/doc/FILES.

A Cruise of the Internet

This guide is a new, free Internet HyperCard stack from Merit. Merit is
also a treasure trove of Internet information and resources, including Internet
use statistics. For more information, contact Merit Network, Inc., 2901 Hubbard
Ave., Ann Arbor, MI 48109-2016; phone: (313) 936-3000; e-mail: nsfnet-
info@merit.edu.

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 91

APPENDIX C
INTERNET RESOURCES

Besides our use of discussion lists, we have been exploring resources available
for TELNET and FTP visits, with an eye toward their usefulness to public
library audiences. Here are some of the ones we like.

Cleveland Free-Net

Free-Nets are the brainchild of Tom Grundner, Director, Community
Telecomputing Laboratory, Case Western Reserve University, 303 Wickenden
Building, Cleveland, OH 44106; phone: (216) 368-2733; fax: (216) 368-5436;
Internet: aa001@cleveland.freenet.edu; BITNET: aa001%cleveland.freenet.edu@
cunyvm; and the folks at National Public Telecomputing Network (NPTN),
Box 1987, Cleveland, OH 44106; phone: (216) 368-2733; fax: (216) 368-5436; e-mail:
aa622@cleveland.freenet.edu.

Free-Nets use a city metaphor, complete with schools, hospitals (for people
AND pets), the Cleveland Public Library, the courthouse, and other public
services. Free-Nets also provide weather, news, and gateways to other resources,
including other Free-Nets. To access the Cleveland Free-Net, simply telnet to
freenet-in-a.cwru.edu 129.22.8.82 or 129.22.8.75 or 129.22.8.76 or 129.22.8.44 and
select "visitor" at the login menu.

MELVYL

MELVYL includes the union catalog of monographs and serials held by
the nine University of California (UC) campuses and affiliated libraries. It
represents nearly 11 million holdings at UC, the California State Library, and
the Center for Research Libraries. The MELVYL catalog also provides access
to MEDLINE and Current Contents as well as a gateway to many other systems.
Access to some databases is restricted under a license agreement to the UC
faculty, staff, and students; telnet to melvyl.ucop.edu or any of four Internet
addresses (31.1.0.1, 31.0.0.11, 31.0.0.13, 31.1.0.11). For more information, contact
the University of California MELVYL Catalog, Division of Library Automation,
University of California Office of the President, 300 Lakeside Drive, 8th floor,
Oakland, CA 94612-3550; phone: (415) 987-0555 (MELVYL Catalog Helpline);
e-mail: lynch@postgres.berkeley.edu.

CARL

CARL is a gateway to academic and public library online catalogs, as
well as resources like UnCover and Magazine Index, the Academic American
Encyclopedia, and Internet Resource Guide. Access to some items is limited;
telnet to pac.carl.org or 192.54.81.128. For more information, contact Colorado
Alliance of Research Libraries, 777 Grant, Suite 306, Denver, CO 80203-3580;
phone: (303) 861-5319; e-mail: help@carl.org.

North Carolina's bbs.oit.unc.edu

Read USENET newsfeeds, use LibTel, a scripted TELNET gateway to
access both U.S. and international libraries plus such things as Data Research

92 JEAN ARMO UR POLLY

Associates Library of Congress catalog, the Ham Radio Call Book, the National
Science Foundation, the Weather Server, Webster's dictionary and thesaurus,
and more. For more information, telnet to bbs.oit.unc.edu or 152.2.22.80 to
connect to the bulletin board system.

Services

For information on American University's gateway to many interesting
sites, telnet to wugate.wustl.edu or 128.252.120.1 using the login services.

NYSERView

Travel to the resources described in NYSERNet's New User's Guide to Useful
and Unique Resources on the Internet. For more information, telnet to
nysernet.org or 192.77.173.2 to try these. The login is nysrview; password is
nysrview.

Liberty High

This is a pilot project linking secondary education students with college
campuses. NYSERNet plans to expand this service to every campus in New
York State. To try it out, telnet to nysernet.org or 192.77.173.2 and log in as
librtyhi with the same password.

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 93

APPENDIX D
NAVIGATING THE SEA OF 700,000 HOSTS

Recent estimates of host computers on the Internet have exceeded 700,000.
Many of these hosts offer files, data, graphics, audio files, and more for transfer
to local computers. The notion of sifting through these millions of files looking
for one particular item gives even ALA lifetime members the shakes! Fortunately,
various projects are underway that simplify this search process.

Archie

Peter Deutsch, of McGill's Computing Centre, describes the archie server
concept, which allows users to ask a question once yet search many different
hosts:

The archie service is a collection of resource discovery tools that together
provide an electronic directory service for locating information in an Internet
environment. Originally created to track the contents of anonymous FTP
archive sites, the archie service is now being expanded to include a variety
of other online directories and resource listings.

Currently, archie tracks the contents of over 800 anonymous FTP archive
sites containing some 1,000,000 files throughout the Internet. Collectively,
these files represent well over 50 gigabytes (50,000,000,000 bytes) of
information, with additional information being added daily. Anonymous
FTP archive sites offer software, data, and other information that can
be copied and used without charge by anyone with connection to the
Internet. . . . The archie server automatically updates the listing
information from each site about once a month, ensuring users that the
information they receive is reasonably timely, without imposing an undue
load on the archive sites or network bandwidth. (Deutsch, CERFnet News,
Nov. -Dec. 1991, vol. 3, no. 7)

For more information, contact UNIX Support Group, Computing Centre,
McGill University, Room 200, Burnside Hall, 805 Sherbrooke Street West,
Montreal, Quebec, Canada H3A 2K6; phone: (514) 398-3709; e-mail: peterd@
cc.mcgill.ca. Some archie server sites to telnet to include archie.rutgers.edu or
128.6.18.15 (USA), archie.sura.net or 128.167.254.179 (USA), archie.mcgill.ca or
132.206.2.3 (Canada), archie.funet.fi or 128.214.6.100 (Finland/Mainland
Europe), archie.au or 128.184.1.4 (Australia/New Zealand), archie.doc.ic.ac.uk
or 146.169.3.7 (Great Britain/Ireland).

Gopher

A gopher (or go-fer) is a little furry creature who tunnels through the
ground or someone who fetches necessary items from many locations. Telnet
to consultant.micro.umn.edu or 134.84.133.255, log in as gopher, and enjoy
having your very own gopher. Gopher includes fun and games, humor, libraries
(including reference books such as The New Hacker's Dictionary, Roget's 1911
Thesaurus, and the CIA World FactBook), gateways to other U.S. and foreign
Gophers, news, and gateways to other systems. There is also an archie server.

94 JEAN ARMO UR POLLY

WAIS

Wide Area Information Servers (WAIS pronounced wayz) allows users to
get information from a variety of hosts by means of an electronic "client."
This may be the new world order of automated librarianship. The client searches
various WAIS servers around the globe. The user tells the client how relevant
each hit is, and the client can be sent out on the same quest again and again
to find new documents.

WAIStation is an easy to use Macintosh implementation of a WAIS client.
It can be downloaded from think.com as well as a self-running MediaTracks
demo of WAIStation in action.

WAIS developer Brewster Kahle also moderates a thoughtful WAIS
newsletter and discussion group, often speculating about the future of libraries
and librarians. For more information, contact Brewster Kahle, Project Leader,
Wide Area Information Servers, Thinking Machines Corporation, 1010 El
Camino Real, Menlo Park, CA 94025; phone: (415) 329-9300, ext. 228; e-mail:
brewster@think.com.

WorldWideWeb

Tim Berners-Lee describes his Web this way:

The WWW project merges the techniques of information retrieval and
hypertext to make an easy but powerful global information system. The
WWW world consists of documents, and links. Indexes are special documents
which, rather than being read, may be searched. The result of such a search
is another ("virtual") document containing links to the documents found.
The Web contains documents in many formats. Those documents which
are hypertext (real or virtual), contain links to other documents, or places
within documents. All documents, whether real, virtual, or indexes, look
similar to the reader and are contained within the same addressing scheme.
To follow a link, a reader clicks with a mouse (or types in a number if
he or she has no mouse). To search an index, a reader gives keywords (or
other search criteria). These are the only operations necessary to access the
entire world of data.

To get there, telnet to 128.141.201.74 or info.cern.ch. For more information,
contact Tim Berners-Lee, WorldWideWeb Project, 1211 Geneva 23 Switzerland;
phone: +41(22)767 3755; fax: +41(22)767 7155; e-mail: tbl@cernvax.cern.ch.

PUBLIC LIBRARIES AT THE ELECTRONIC DOOR 95

APPENDIX E
MUST-HAVE VOLUMES FOR THE INTERNET SURFER

Kehoe, Brendan P. (1993). Zen and the art of the Internet: A beginner's guide
(2nd ed.). Englewood Cliffs, NJ: Prentice Hall. The first edition is available
for free from many FTP sites. This version has about 30 pages of new material
and corrects various minor errors in the first edition. Includes the story
of the Coke Machine on the Internet. For much of late 1991 and the first
half of 1992, this was the document of choice for learning about the Internet.
ISBN 0-13-010778-6. Index. $22.00.

Krol, Ed. (1992). The whole Internet user's guide b catalog. Sebastopol, CA:
O'Reilly & Associates. Comprehensive guide to how the network works, the
domain name system, acceptable use, security, and other issues. Chapters
on telnet/remote login, File Transfer Protocol, and electronic mail explain
error messages, special situations, and other arcana. Archie, Gopher,
NetNews, WAIS, WWW, and troubleshooting each enjoy a chapter in this
well- written book. Appendices contain information on how to get connected
in addition to a glossary. ISBN 1-56592-025-2. $24.95.

LaQuey, Tracy, & Ryer, Jeanne C. (1993). The Internet companion: A beginner's
guide to global networking. Reading, MA: Addison- Wesley. Beginning with
a foreword by Vice-President Elect Al Gore, this book provides an often-
humorous explanation of the origins of the Internet, acceptable use, basics
of electronic mail, netiquette, online resources, transferring information,
and finding e-mail addresses. The In the Know guide provides background
on Internet legends (Elvis sightings is one), organizations, security issues,
and how to get connected. Bibliography. Index. ISBN 0-201-62224-6. $10.95.

Tennant, Roy; Ober, John; & Lipow, Anne G. (1993). Crossing the Internet
threshold: An instructional handbook. Berkeley, CA: Library Solutions Press.
A cookbook to run your own Internet training sessions. Real- world examples.
Foreword by Cliff Lynch. Library Solutions Institute and Press; 2137 Oregon
Street, Berkeley, CA 94705; phone (510) 841-2636; fax: (510) 841-2926; ISBN
1-882208-01-3. $45.00.

DAVID V. LOERTSCHER

President

Hi Willow Research and Publishing
Castle Rock, Colorado

Electronic Information in School Libraries

ABSTRACT

Microcomputers have progressed from toys to tools in managing school
libraries. Equipment inventory, circulation, online catalogs, ac-
quisitions, and serials management/check-in have all been affected. In
addition, high technology has presented new possibilities for educating
young people, and school librarians are faced with a role change as
they rise to meet this challenge.

INTRODUCTION

Picture the faculty of the College of Education at the University
of Arkansas in 1979 gathering for a symposium that is to explore the
possibilities of the microcomputer in education. Since almost none of
the faculty has seen a microcomputer, they are anxious to attend and
see what this new technology has in store for education. The presenter
is introduced to this august body an expert from out of town a person
with as much experience with microcomputers as about anyone in the
country. (Remember, it is 1979 and the Apple microcomputer has just
made its appearance on the market.) Our expert is Bryan Burdick from
Kansas City. He is 10 years old. The professors come unglued.

As the era of high technology has dawned upon education in the
past 13 years, three phenomena are evident:

1992 David V. Loertscher

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 97

1. Microcomputers and high technology have progressed from toys to
tools in managing school libraries.

2. New possibilities for educating young people with high technology
have developed.

3. School librarians are faced with yet another role change as they rise
to meet the potential that high technology brings to the education
of young people.

The purpose of this paper is to explore these three areas of interest
from the perspective of the school library as it fits into the larger
perspective of education.

MICROCOMPUTERS AS A TOOL
FOR MANAGING SCHOOL LIBRARIES

Educators in virtually every school in the country have adopted
the microcomputer as an administrative tool for the three basics: word
processing, databases, and spreadsheets. Such things as mass letters to
parents, inventory, scheduling, and budgets are commonplace. School
librarians have been a little slower to adopt these same tools partly
because administrators innovated first, but also because librarians tended
to purchase computers for instructional purposes first and only learned
of management possibilities second. Betty Costa was one of the first
to produce a handbook for the school librarian about using
microcomputers as a management tool, and her book remains a classic
in the field (Costa & Costa, 1991). Standard uses of microcomputers
as tools in school libraries in 1992 include equipment inventory,
circulation, online catalogs, acquisitions, card catalog production,
serials management/check-in, and general management.

Equipment Inventory

Librarians/principals use both homegrown databases and
sophisticated programs to track audiovisual equipment and furniture
owned. Sophisticated programs track original vendors, equipment
description, model and serial numbers, location of the equipment,
maintenance records, lists of replaceable parts (such as lamp numbers
for overhead projectors), and equipment condition. Such programs
provide information for security and insurance purposes including
projected costs of equipment repair and replacement.

Circulation

Librarians used microcomputers early to circulate equipment and
specialized materials such as periodicals. Homegrown overdue systems

98 DAVID V. LOERTSCHER

were also popular in the early 1980s. The earliest commercial programs
for school libraries were those designed to circulate books using a bar
code system. Hundreds of systems were sold to school libraries that
tracked patrons, books in circulation, and overdues. Today, the early
adopters of circulation systems are envious as they see the features of
the fully developed online catalogs for the school library. These people
face automating twice with all the attendant problems of school politics,
new equipment and software costs, and in some cases, the reinputting
of an entire collection.

Online Catalogs

Online catalogs utilizing mainly MS-DOS systems are now common
in school libraries. Popular system names include Follett, Winnebago,
Mandarin-Media Flex, and Dynix. Systems generally use the MARC
format, are individually school based (they do not share data with other
schools or libraries), and contain catalog, circulation, and acquisition
components. Catherine Murphy (1992), who has conducted one of the
few dissertation studies to date about online systems in school libraries,
has recently pulled together some of the most current articles that
describe these systems and ways to select, implement, and evaluate them.
As she was collecting the articles on school library automation, each
article had to be sent back to the original author for a major update,
even when the article had been published the previous year an
indication of the rapid rate of change in this area.

A typical online public access catalog (OPAC) for a school library
provides author, title, and subject searches plus Boolean searching of
all data fields. A few systems are beginning to provide such things as
"interactive search assistance, error-correction features, and additional
information about the contents of materials" (Hooten, 1992, p. 145).
Catalog records for these systems are being purchased from vendors
or downloaded from commercial CD-ROMs.

Two major issues for school OPAC systems are the quality of subject
access and user friendliness for the child. When catalog records are
downloaded into a school library catalog, a number of quality control
problems emerge. Many commercial retrospective conversion companies
have not cleaned up their databases. This means that classification
numbers and subject headings reflect only the editions of Dewey and
Sears/Library of Congress (LC) used at the time of cataloging. Thus,
conflicting classification numbers and a mix of LC or Sears or AC subject
headings are transferred into the OPAC, making it no better than its
card form counterpart. Only if the system has global change capabilities
and if the school librarian has enough expertise to recognize and deal
with consistency does a better product emerge. Sloppy data delivered

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 99

by computer are still sloppy. Boolean and keyword searching helps by
locating common terms that children know, but subject searches with
high recall rates are the exception rather than the rule, and shelf order
is not improved at all. It is not surprising to sit at a school OPAC
and find an author's name spelled several different ways requiring that
the user look in several places to track an author's works. Errors abound
if a combination of local input and downloading from commercial
services has been utilized.

Vendors of OPACs for schools are making plans for higher quality
cataloging data now that basic automation features are refined. For
example, a program could be written to automatically present changed
numbers from one Dewey edition to another, allow the librarian to
accept/reject the change, print out a bibliography of those changes,
and print the spine labels that could be applied to books in a few
minutes. As the librarian approved these global changes, she could
override national suggestions in favor of local preferences. Thus, sports
biographies could be classed with the biographies or with the sport,
and if with the sport, each sport would have a consistent number.

In subject cataloging, troublesome headings such as NEGROES,
BLACKS, or AFRO-AMERICANS could be easily changed to a consistent
term. Modules creating a cross-reference structure for the OPAC could
be created and sold as add-ons so that consistency could be assured.
Consistency in information systems is more important to children than
it is to adults. Adults have a higher developmental level, are more flexible,
and, one hopes, are more tolerant.

Early studies of children using OPACs are quite negative. Children
do not seem to be able to use computer catalogs any better than printed
ones. Edmonds, Moore, and Balcom (1990, 1992) found that Piaget's
concrete operations probably apply to the use of OPACs just as they
apply to other information or learning tasks. Thus, designers of screens
and searching procedures for OPACs need to constantly investigate and
utilize the research dealing with how children see and interpret
information. Much needs to be done with picture subject searches and
automatic transfer from misspelled words to the correct word form or
to a "kid-known" synonym. Voice requests to OPACs need to be designed.
If a child asks the computer for a Dr. Seuss book, the computer might
print out a map showing the shelf location and pictures of some of
Seuss 's book covers and their spine labels.

As results of experiments with young children are forthcoming,
many potential improvements will become apparent. Then again,
perhaps it is too much to expect a child under age 9 or 10 to navigate
a complex information system without the assistance of an older child

100 DAVID V, LOERTSCHER

or an adult. Are commercial companies willing to design information
systems with children in mind? Librarians must demand such features
and be willing to pay for them.

One of the more exciting aspects of quality cataloging for school
libraries is the sharing of in-depth analysis of books. For example, one
high school on Long Island analyzed every one of its play collections
and made the amplified cataloging records available to all libraries
in the area. Online indexes not only to plays but to poems, collective
biographies, and songs are but a few of the possibilities that make small
collections serve patrons to their full potential.

One barrier to school library automation is the lack of a networking
capability that would connect schools and libraries to other schools
and libraries. Although a few states such as Pennsylvania have produced
statewide CD-ROM catalogs of holdings in school libraries, generally
school libraries have yet to become a part of national systems such
as OCLC (there are exceptions).

Acquisitions

Although OPACs may contain acquisition modules, these systems
are not usually sophisticated enough to do any type of collection analysis
and acquisition assistance based on collection segments. This author
has proposed a simple system of collection mapping that chunks the
collection into segments that support the curriculum of the school.
Thus, a school library would have a general collection (providing
breadth) and several to many specialty collections (providing depth)
designed to serve special topics studied by the students. Even if the
designer of the system has not planned for collection segmentation,
the librarian can "cheat" the system. Using any field that has some
form of user control, the librarian can control collection segments
through a coding system.

For example, suppose the librarian wishes to build an in-depth
collection of dinosaur books because the topic is studied by several classes
and grade levels of students each year. The librarian assigns a code
in a controllable data field to the "dinosaur collection," scans all
materials relevant to that collection (both print and audiovisual), and
then maintains this special collection in the OPAC or automated
circulation system. Items deleted from the collection are automatically
deleted from the emphasis collection. Each time the topic is studied,
the librarian prints out the dinosaur collection bibliography, assigns
the dinosaur code to any new materials, scans and deletes any irrelevant
materials, and in a few moments has complete control of this emphasis
collection. Use/circulation statistics can be analyzed easily helping with
decisions such as when to add duplicate titles and when to discard

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 101

outdated or unused titles. This information can also be used as the
basis of collection building in that emphasis area. The author's book
Computerized Collection Development for School Library Media
Centers describes this system of collection building using manual, simple
database managers or OPAC systems (Loertscher & Ho, 1986).

Card Catalog Production

Librarians still using card systems can purchase quite sophisticated
programs that will produce card sets from original input data or will
print out card sets from CD-ROM databases. Catalog cards, spine labels,
circulation cards, and card pockets can all be printed from these
programs.

Serials Management/Check-in

Several commercial programs exist to manage a periodical collection
that would not only be used in check-in but in claiming, weeding,
subscription renewals, and circulation.

General Management

Popular programs such as Apple Works or Microsoft Works are used
by school librarians to write letters, generate reports, keep statistics,
track and prepare budgets, keep address files of suppliers, and manage
a myriad of other files that are better kept on computer than in card
files.

School librarians have been quite slow to become computer literate
enough to make the computer a useful management tool. The same
fear and trepidation that confront a generation of people who grew
up in the manual world prevent librarians from taking advantage of
the computer. At our company, we still get many calls from persons
who cannot use the ubiquitous Print Shop program and the graphic
disks we market. We are amazed at the number of people who don't
know how to turn a computer on or accomplish even the simplest of
tasks electronically.

Both the word processor and the database manager are the school
librarian's best friends. Once these programs are mastered, school
librarians are extremely creative in using these tools to save time and
to manage their centers more effectively. Every year, Libraries Unlimited
publishes a number of products, usually indexes created by these
"hacker" librarians. Examples are indexes to multicultural projects,
science projects, or specialized bibliographies of wordless picture books
or reference books about Indians.

102 DAVID V. LOERTSCHER

Librarians who want to succeed with computers in managing their
libraries must not only become computer literate, they must learn several
major programs in depth. Whether it requires being tutored by a
knowledgeable person, taking computer classes, or teaching oneself,
the time investment in learning about computing pays major dividends.
Librarians must keep current with developments in computing
including hardware and software. A few suggestions include the
following:

Read several popular computing journals each month such as
MacUser or PC World. If you don't understand the articles, at least
read the ads.

Read several library-oriented computing journals such as the ALUG
Newsletter (free from Apple Computer, Inc.), Library Hi Tech, or
Computers in Libraries.

Find a friend who is interested in computers and "talk shop" on
a regular basis.

Share computer stories and concerns with other librarians facing
similar problems. Ideas and solutions to common problems can save
hours and days of frustration.

Attend computer sessions at conferences. If vendors conduct the
session, don't believe all you hear.

Never buy a computer software package or hardware without
comparing features with other programs/machines that will do
similar tasks. Be careful about being one of the first to try out a
new "library package." We would never have good systems for libraries
without those willing to experiment, but one can always expect new
applications to have problems that will take a great deal of time
and effort to fix.

Companies go in and out of business in the computer software and
hardware business. Try to choose the most solid companies that
support their sales and are likely to be in business for a few years.

Discussions about computers with other school, public, and academic
librarians will provide news on developments both local and national.

Don't overlook the expertise of students who are knowledgeable about
computers. They will often provide hundreds of hours of assistance
in exchange for the excitement of building a system that makes a
positive contribution to the library.

Invest in a computer for the home and use it daily. Many software
packages require constant practice if they are to be mastered.

Finally, developing political contacts with administrators will provide
opportunities to update equipment and software as funds become
available.

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 103

Like their counterparts in business, school librarians cannot sit
back and wait for computerization to solve their management problems
like magic. They must continually increase their skill and computer
knowledge if the payoff is to come.

TECHNOLOGY AND EDUCATION

Since the appearance of Edison's motion picture projector in 1910,
there has been a great deal of interest in technological devices that
could be used in education. Wave upon wave of interest has been given
to each new technology, and many have been labeled the saviors of
education. The motion picture projector, the filmstrip projector, the
opaque projector, the 8-mm single concept loop projector, the overhead
projector, radio, television all and more have had their "day in the
sun." The history of this entire movement is best documented in
Saettler's (1990) The Evolution of American Educational Technology.

The major drawback with technology is that none has been
developed specifically for education. Generally, a technology has been
created for a business or scientific purpose, then educators try to figure
out how students can benefit. The cart has always been ahead of the
horse. After a technology is created, a company will try to market that
technology to schools. Modifications to suit students' needs are often
made after some period of adoption in the education market. Such
modification is expensive and time-consuming thus limiting the spread
of the technology.

For example, the motion picture projector was developed as an
entertainment technology first and as an educational one second. School
teachers hated 16-mm projectors because they were never easy to thread,
they were tricky to handle when they malfunctioned, and they required
total environmental control to use in the normal classroom.

In the past few years when videotape, also an entertainment
technology, became cheaper to own than 16 mm, teachers rushed to
adopt a technology they could operate both at school and home, and
a technology that did not require a darkened room. Teachers did not
care that the picture size, color, and clarity were inferior; they wanted
a technology that worked and one they could operate without help.
Today, film companies are having a hard time staying in business, not
because they cannot produce videotape instead of film, but because of
unit pricing. Consumers are so accustomed to buying a videotape in
the store for $30, they think educational videos should be comparatively
priced. Film companies have usually charged |300 to |900 per film

104 DAVID V. LOERTSCHER

print, and films were purchased only by school districts or film libraries.
It takes sales of thousands of prints at $30 to individual schools to
recoup the production costs film companies need to stay in business.

A second problem with technology in education is that producers
generally have made outlandish claims for the contribution of their
particular technology to learning. The 16-mm film was going to replace
teachers. So was educational television. So were teaching machines.
Research projects by the thousands documented the power of these media
to teach. Sometimes research results were positive in favor of the
particular technology; most of the time, research results showed "no
significant difference." Students learned equally well no matter the
medium. It seems logical to assume that a fact can be learned from
a book or a film or a microcomputer screen with equal ease.

What the research did not test was the amount of learning from
the unique characteristics of each medium. For example, since a film
uses motion, it should teach certain things as well as a book and other
things better than a book when a concept is motion oriented. Conversely,
the book should be superior to the film under certain circumstances
for example, for browsing. Today, educational technologists generally
agree that a teaching medium should be selected based on its unique
characteristics as well as its content. Teachers, however, usually choose
a technology based on convenience first, content and characteristics
second.

The research on educational technology does show that new or
different technologies are popular with learners. Students who use a
computer for the first time find it exciting, and they tend to learn more.
Since students' primary objection to schooling is boredom, teachers
and school librarians can exploit this novelty effect if they provide a
wide variety of technological or media experiences rather than con-
centrating on a few. Variety is the spice of teaching and learning.

Because education is such a labor-intensive industry, the costs of
educating a nation of students have risen astronomically in the past
20 years. Theorists have been trying to make the case that technology
could replace some of the teachers and thus lower costs (Molenda, 1992).
Within the past 10 years, a number of schools have been created that
use banks of computers that provide hours upon hours of drills, tutorials,
and simulations. These schools employ lower paid adult aides and a
few teachers who are instructional managers. The computer system in
these schools employs a technique of mastery learning where students
must achieve criterion levels of learning before they can progress onto
the next level of work. Such schools have high costs of investment in
technology, support, and software and are not less expensive to establish
and maintain. For example, WICAT Systems in Provo, Utah, has such
a school. The tuition is comparable to an exclusive private school. They

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 105

do advertise, however, that their students perform much better on
standardized tests than students in traditional schools.

Generally, technology is an add-on cost to schools. That is, the
school will employ enough teachers to teach in the conventional l-to-30
or l-to-25 ratio and then provide technology as tools for the teachers
to use. In this case, technology enhances what a human is able to provide.
Such a plan provides a tremendous challenge to a school district where
90% of the money for education goes to teacher salaries. Technology
must compete for a very small piece of the budget pie. If teachers do
not adopt a technology in such a manner that students are benefiting
from it in a demonstrable way, the technology is considered a frill.

Technology carries with it a tremendous temptation to use it as
a "filler." The teacher may rest and let the machine take up some time.
Then the teacher will attempt to compress the human teaching content
into a smaller time frame. In the past few years, teachers have been
assigned a greater and greater amount to teach in the same or decreased
amount of time. These teachers reassess the time that a technology takes:
Does it really teach as well as I can? Does it cover the content as fast
as I can? Can I get relevant learning materials for the technology the
school owns? Are the materials to be used on a technology of only
peripheral value to the objectives currently being taught?

Many teachers are not comfortable using technology. They may
not be able to operate the equipment. They may fear that when a piece
of equipment fails, it will do so right in the middle of a class, and
they will lose control of the presentation and the students. Many teachers
opt not to use machines. They view technology as either an add-on
responsibility or as irrelevant to the specific task at hand. They laugh
when someone provides a single computer and no software for their
classroom of 30 students and expects something magic to happen.

In an attempt to make technology and materials much easier to
use, schools have been and are being designed around teaching stations.
For example, in Chesterton, Indiana, the new high school building has
been designed around technology. The building has been completely
wired with optical cable. There is a telephone in each classroom. Each
classroom has a computer/technology station complete with at least
two large classroom television monitors. Teachers can command any
number of technologies coming into their rooms at a moment's notice.
They can call up a film at will, connect into a computer network,
call pictures, text, or sound from a CD-ROM to illustrate their lectures,
or use amplified telephone to talk with their class to experts anywhere
in the world. If a malfunction occurs, they can send a message to the
central media center operator on their computer, or they can call the
person on the telephone.

106 DAVID V. LOERTSCHER

This dazzling array of technology is impressive, but it is an add-
on. The community is investing more dollars in each student's
educational experience than in a comparable school using the textbook-
lecture method. Currently, taxpayers are struggling. They are wondering
how much to invest in education when scores on national exams are
declining or at least are not improving significantly. There is no rush
toward technology as "the answer."

THE SCHOOL LIBRARY AND TECHNOLOGY

During the 1960s, when technology first became available to
education on a grand scale, school librarians were resistant to change.
The 1960 Standards for School Libraries advocated the adoption of a
multimedia center concept. At first, many schools hired audiovisual
specialists to handle nonprint media, but with the passage of time,
librarians became more comfortable with a wide range of formats and
generally were put in charge of all media.

When microcomputers came along, many school librarians saw
these machines as learning tools, were early adopters of the technology,
and became computer leaders in their schools. Other librarians ignored
computers, and so others were hired to set up and manage computers
for both computer literacy coursework and computers as instructional
tools. Without the benefit of a national survey (data will be collected
in 1994), this author would guess that a minority of school librarians
are considered to be microcomputer specialists in their buildings. With
the advent of online technology and other high-tech equipment, school
librarians again stand on the threshold of a role choice. Immersion
into the high-tech world of information and media will provide new
frontiers of information use/access for the youth of the nation. School
librarians have a chance to prepare young people for an emerging
information society, but only if they become leaders in the information
technology that has and will appear.

To get a better picture of the potential of the new information
technology as an education tool, let us explore the technology, not by
type of hardware, but by the educational/information functions that
it can provide to young people. Although we may briefly describe a
specific piece of technology and what it does at this moment in time,
the function that it performs is likely to outlast the current model of
whatever gizmo is available.

Access to Bibliographic Information

Currently, bibliographic data are a well-known commodity being
handled by all types of libraries. At first, data such as magazine indexes,

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 107

bibliographies, and other indexes were available principally via
telephone lines; now much is available through CD-ROM or through
locally created databases. Citations of articles or books are only helpful
to a young person if there is easy access to hard copy through local
public or academic libraries. Another problem concerns who pays for
the bibliographies produced from remote online databases. A number
of high schools have budgeted funds for online searching and do not
pass these costs on to students. This policy limits the amount of searching
that is done and requires extensive planning before searches are
conducted (not a bad practice).

Still another problem is the relevance to children and young people
of information on large commercial databases. In the past 10 years,
the availability of databases has proliferated to such an extent that more
and more is available for the younger set. That trend should continue.
Of particular use are information systems containing abstracts of the
books or articles indexed. This format helps students choose a few
relevant articles, and at times the article can be cited in a research paper
when the abstract contains an appropriate fact.

When the school library is connected to a local, regional, or national
network, access to bibliographic data is very useful to young people.
If they know that another school, public, or academic library in the
community has a magazine or book they need, they can usually find
a way to obtain the needed materials for their research. Through the
school OPAC or a separate terminal, the young person can query the
libraries in the area. In Denver, Colorado, any home or school that
has a modem and a computer can access the CARL system. The user
can dial into many libraries on the front range to check bibliographic
citations and do keyword searches of collections throughout Colorado
and beyond. The system is so simple that anyone who can read can
use it.

Another example of networking for young people is Access
Pennsylvania. The collections of hundreds of school libraries in
Pennsylvania have been stored on several CD-ROMs. Young people can
find specific titles or do subject searches and locate materials in their
own school, neighboring schools, regional libraries, or the state as a
whole. Access to distant libraries takes time a commodity that young
people often do not plan for.

School librarians are wise to seek the technology needed to connect
their library to collections other than their own. They should teach
young people how to access other collections and create a simple system
of interlibrary loan. These are the same challenges faced by libraries
of all types and come with the same attendant problems: How can
we provide access? Who will pay? How can we support interlibrary
loan? When shall we own an item or just borrow?

108 DAVID V. LOERTSCHER

The Locally Produced Database

A number of school librarians have discovered database managers
such as Apple Works, Microsoft Works, Claris Works, and others. They
begin to realize that many bibliographies and indexes to local collections
can be created to make certain materials extremely useful. Examples
might include the following:

A song index

A play index

An index to science experiments

A poetry index

A biographical index to collected works

A local newspaper index

A speaker's index

A famous person's address index

Libraries Unlimited has created a number of databases and
bibliographies that can be used as local databases. For example, Mary
Ann Pilger's (1992) Science Experiments Index for Young People is an
index to over 2,000 science experiments books. This index is available
in traditional print form or can be purchased as a database for use
on Apple, Macintosh, or IBM computers. The print form of Pilger's
database has the same drawbacks as other print indexes first you find
the experiment you want, but then you must try to locate the book
in your own or some other library's collection. The database version
has an advantage. The librarian searches which of the 2,000 titles the
library owns, deletes the rest, and prints out the result. The new index
becomes an index to the in-house collection. User success rates jump
to near 100%. The master database can be kept so that titles can be
added or deleted as the collection evolves or as an interlibrary loan
source. Young people and student helpers can be taught how to add
or delete from such an index. In this way, students can understand how
an index is created and how it can be searched to advantage.

If a student can create an index, that student can use many types
of electronic indexes easily and can begin to comprehend what electronic
indexes can and cannot do, how they are built, and how they can be
searched. Although some care must be taken when allowing young
people to save and delete information in a master index, much good
training can accrue. Both library and user benefit.

A second example of a local database is the book and database
by Vandelia VanMeter (1992) entitled World History for Children and
Young Adults. This source contains an annotated bibliography of all
books reviewed in the past 10 years for world history. The printed
version is interesting, but the database is much more valuable from

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 109

a collection-building perspective. Comparisons between what is
currently owned and what is wanted can be done quickly, and purchase
lists can be printed out. Better access to the books is also available
through keyword searching rather than through the traditional subject
index available in the print version. Bibliographies, both of owned
materials and other sources, are easily printed out or included in reference
lists for teachers for use in printed articles or lesson plans. For example,
consider the request, "I need a list of books about the Vietnam War
that have been published in the past 10 years that have been reviewed
positively." Such a request takes moments to prepare either in print
form or in data form that can be pulled into another document.

A third example of a locally owned database is one in which factual
or narrative information is stored for instructional use. Such databases
are available from commercial sources on CD-ROM or as databases that
can be used with common database managers. Examples might be data
about presidents, facts about endangered animals, weather data, address
files, or astronomical data. Such databases are used by students to extract
facts but also to do higher level thinking and for data manipulation.
A few examples and their use include the following:

In a database about the states, which states have the cardinal as their
state bird?

In a database containing weather information, what are the top 10
temperatures recorded in the United States, and where and when
did these temperatures occur?

In two databases, one containing incidents of communicable diseases
and the other weather data, is there any relationship between the
weather and the occurrence of certain diseases?

In a database containing demographic statistics, do certain states have
predominant religions? Are predominant religions regional in nature?
What are the historical reasons for predominant religions being
located in a region?

Students can be expected to use an electronic database in the same
way they would use a reference book. The advantage of the electronic
version is usually both speed and the type of questions that can be
answered/computed in the electronic version. In printed reference works,
the authors and editors must decide how the user is going to approach
the data. In an electronic source, the designers can allow numerous
approaches to the data without allowing the size of the data pool to
increase exponentially.

Young children can learn how to search databases for answers they
need to certain questions, but it is even more educational if young
people can participate in database construction. There are numerous
articles in the literature that describe the process of creating a database

1 10 DAVID V. LOER TSCHER

with children. Ron Martin (1992) in Alaska recently described a database
he had children create about dogs. Each child did research on a different
dog, entered that data into a database Ron had created, and then the
class used the database to help make decisions on what type of dog
each of them might like to own, how dog breeds compare, and what
breeds of dogs are common to certain areas of the country.

Children who learn to construct databases have many advantages.
They learn to research certain facts; they must check those facts to see
that they are as accurate as possible; they must learn how a database
is structured; they must learn how data are entered into a database;
they must learn quality control principles and how to think ahead
as data are entered (do we enter the surname first or the given name?);
and they must learn how to search a database to answer questions of
varying degrees of difficulty. Such classroom activities provide
simultaneous growth in subject competence and in information literacy.
There are few better ways to teach young people the concept of databases.

Databases can be constructed on any microcomputer, even an old
Commodore 64. Happily, database managers simple enough for even
young children are readily available at affordable prices. It does not
require a major investment to reap excellent results.

Children who can create databases can also learn how to chart the
data they enter (budding Edward Tuf tes?). Many programs allow a wide
variety of charts and graphs to be constructed from simple data entered
from a database manager or in its cousin, the spreadsheet. Because
children are constantly tested to see if they can interpret chart and graph
data, they can leap forward in standardized test scores if they know
how to create their own understandable charts and graphs using a
computer.

Data Gathering, Exchange, and Analysis

The American Association for the Advancement of Science is
encouraging teachers to involve young people in more realistic science
experiences. If the school librarian is informed about both data-
gathering opportunities and high-tech possibilities, she can assist the
science faculty in creating marvelous learning experiences. A few sample
scenarios might be the following:

Students gather data on acid rain as a part of National Geographic
Kid Net. They can examine and analyze data from students in all
regions of the country, make comparisons, look for cause and effect,
and pose possible solutions to problems they observe.

Students can gather data from their local streams and rivers as a
part of a major statewide effort sponsored by a university to chart
water availability, usage, and quality. Such cooperation can cultivate

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 111

student/expert relationships, provide career information, and help
students participate in real-world and current scientific problem
solution.

Students might collect data in a cooperative agreement with another
school somewhere in the world. Data can be sent by fax or
electronically mailed back and forth for comparison and analysis.

A school librarian who is comfortable with numerous high-tech
possibilities becomes a valuable team member as data-rich instructional
experiences are designed and carried out. Not only can students gather
data from their local environment, they can enter data from library
reference sources for verification or comparison with the data collected
locally. For example, does the sun rise when the almanac says it should?
Why or why not? Data comparison, analysis, and criticism are the newer
types of information skills students can experience as part of a high-
tech library media center program.

Student Creative Output

In the past 13 years, a whole new world of production has emerged.
Through the word processor, the video camera, CD-ROM, and digital
audio/video, both adults and students can express their ideas and channel
creative energies through a wide variety of media. When students were
limited to transcribing ideas by pencil and paper or by using the
typewriter, only the best students produced any sizable volume of work.
Now, using the new technology, students can produce much more than
they did in the same amount of time. Two factors account for this
phenomenon: a high rate of efficiency and higher motivation through
novelty. A few comments about the various forms of technology that
stimulate student production are illustrative of these factors.

Word Processing

Research shows that students using word processors will write
longer and better reports/papers. Students will revise more because of
the ease of machine revision and the presence of spelling and grammar
checkers, and thesauri allow students to create a higher quality product.
These machine assists actually improve a student's grasp of the language
because spelling, wording, and grammar changes must be approved
by the user; they are not automatic. At first, teachers were hesitant to
accept student work done on word processors because there was some
feeling that, like a calculator in math, students might be getting an
unfair advantage or might be cheating a bit. Most of this fear has been
erased as teachers themselves have become processors of words. There
are many good word processors for children and young adults. Some

112 DAVID V. LOERTSCHER

of these programs can be used just as soon as a child can recognize
letters and how they are used to create words. Many schools require
word processing of all students as a part of their computer literacy
classes. Typing teachers, who were at first very nervous about kids
learning poor keyboarding habits, have resigned themselves to a new
world.

Desktop Publishing

In the past five years, desktop publishing software has become
available for children and young adults. One of the most popular
programs is The Children's Writing and Publishing Center. If students
can handle a very simple word processor, they can automatically handle
a publishing program that allows them to create layouts and add graphics
to their reports. The finished product can be so professional looking
that students become interested in the creation of high-quality printed
products. Young children can produce books, magazine articles,
advertisements, posters, and illustrated reports and term papers. The
school librarian who can envision the potential of desktop publishing
should work with the teacher to plan research or topical studies, have
a plentiful supply of relevant graphics, and provide an open lab for
the creation of products. Students can be required to write more and
better pieces complete with proper citations to print or electronic sources.
Layout requirements can be accomplished through the use of templates
or can be designed by the student depending on the sophistication level.

Desktop Everything Else

In the past three years, a wide variety of new technologies have
emerged that link to a computer to produce not only print but sounds
and images. Perhaps the first move in this direction was the creation
of hypertext. Products like HyperCard for the Macintosh and Linkway
for the IBM allow the user to create databases with buttons. These
programs create something like a box of 3 X 5 cards that can either
be viewed in sequence, or by adding "buttons" (places on the card to
click with a mouse), the user can be linked to any card in the box.
In addition, the user could be linked to both sound sequences and to
pictures.

Even more recent technology allows students to link words to film
clips or video clips or even to other databases by connecting the computer
to a CD-ROM, a VCR, or a tape recorder. Multimedia reports can be
created by young children with a minimum of instruction as long as
the proper equipment, software, and materials are available. It is
interesting to read or hear the technospeak that accompanies the
descriptions of such technology:

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 113

Multimodality. The new media are translatable. Words, images, and sounds
can be placed in one document or can be edited from any other document
to any other document. Information can be presented in text, graphic,
animation, natural voice, music, and special effects on the same platform.
Moreover, multimedia software provide for linking a segment of information,
represented in any mode, to any other chunk of available information.
Hypermedia allows for an omnidirectional search of information, a part
of which may be stored on a CD-ROM or laser disc. The microcomputer
drives the disc player to search and display the information. However,
computers with advanced microchips and internal optical drives provide
an even more powerful platform to designers and producers. Media
computers provide an integrated hardware/software platform for production
and display of instructional information. Digitized audio, video, and text
can be stored and retrieved for the production of new instructional materials
on the same system. Using total digital systems, such as DVI (Digital Video
Interactive), educators are able to represent information in any mode on
demand. Optical scanners and video cameras enable them to transform
information from analog to digital form for editing into multimedia
documents. Using image processing software, visuals can be edited,
enhanced, and transformed products in a matter of seconds. Conducting
such procedures in analog photography remains only in the purview of
professionals in a fully equipped darkroom. Music[al] Instrument Digital
Interface (MIDI) provides the same freedom of expression for music and
sound effects. A keyboard puts composing, interpreting, and performing
power of a full symphony orchestra, a jazz quartet, or a single instrument
at the fingertips of instructional designers and media producers. (Saba, 1992,
p. 129)

The advantages to students in this new multimedia world are quite
simple to understand when a real-world example is given. A report
on the Civil War can contain facts or narrative with accompanying
pictures, stories with accompanying sound effects, or reenactments of
battles complete with animated battle maps and simulated battle sounds.

There is a new term for this manipulation of media re-purposing.
That is, a student can take materials created in one medium for a single
purpose and then edit or merge them into a new product with a new
purpose re-purposing. Although questions of copyright immediately
emerge, the fair-use clause of the copyright law is generally thought
to cover this manipulation as long as the product is created by students
for educational assignments and the resulting product is not sold or
used in public performance.

Teachers and librarians are likely to embrace such products because
these products require the student to spend more time and effort and
require much greater mastery of a topic than was true using either
a pencil or a typewriter. The idea is that students become mini-experts
about a topic as they create a very exciting high-tech product.

An example might be illustrative of the potential. Last year was
the 25th anniversary of Fox-Fire, the oral history project done in the
hill country of Georgia. Using high technology, students can now create,
not only the magazine articles for which the students of Rayburn,
Georgia, are famous, but they can combine commercial or original

114 DAVID V. LOERTSCHER

videotape, commercial or live-recorded sounds, and still photographs
to create their final product. One can imagine the impact of a multimedia
oral history report on killing a hog or attending a mountain country
funeral or even reporting a family reunion country-style.

IMPLICATIONS FOR SCHOOL LIBRARIANS

In the age of high technology, school librarians must be very familiar
with the storage, retrieval, and the production of information, sound,
and pictures. This expanded literacy allows them, with teachers, to be
catalysts in a whole new world of educational possibilities. Many school
librarians are already living in this new world. Others are not even
aware that a new world exists.

Librarians in public and academic libraries can easily recognize
students who are comfortable with this new high-tech world. They will
come into the library, look around, and then ask where all the technology
is located. If the library does not have the technology they like, they
will stomp out complaining about "this rinky-dink place." Students
of this breed will not only be reluctant to use conventional technology,
they may not know how. Being reduced from desktop publishers to
pencil and paper or to the typewriter can be a humiliating and jolting
experience. These students will want instant access to data, full-text
articles, and easy access to fax or quick interlibrary loan. They will
demand current data from the most authoritative sources. They will
want access, not just to words, but to extensive collections of pictorial
and audio material. They will have little patience with librarians who
plead ignorance or poverty as an excuse. Those in academic and public
libraries should be prepared.

In all our enthusiasm for high technology, however, we must be
realistic. Technology must pay its own way if it is to be adopted widely.
We can only expect taxpayers to pay for these expensive tools if

students learn more than they would have by using traditional
teaching and learning techniques;

the technology releases students to be more creative;

students are more information literate and can navigate and use
information technology wisely.

There are lots of reasons not to use technology:

its cost;

the frustration it causes teachers, librarians, and users;

the temptation to just be glitzy;

the increased ability for children to waste time on an expensive
machine.

ELECTRONIC INFORMATION IN SCHOOL LIBRARIES 115

School libraries are on the forefront of the information technology
world. They are laying the groundwork for a whole new generation
of information literate persons. Because they are often alone in a library
media center, without help, they must be superhuman in their breadth
of knowledge, their leadership skills, their technology skills, and their
ability to make a technology pay a true educational benefit. Some of
them complain that there is too much to know. Others dig in and
are constantly refreshed by the excitement and challenge.

If we are truly creating an information society, then we all have
a stake in what goes on in the schools of the nation. The college years
are much too late to begin to build the information skills our society
will demand.

REFERENCES

Costa, B., & Costa, M. (1991). A micro handbook for small libraries and media centers

(3rd ed.). Englewood, CO: Libraries Unlimited.
Edmonds, L.; Moore, P.; 8c Balcom, K. M. (1990). The effectiveness of an online catalog.

School Library Journal, 36(10), 28-32.
Edmonds, L.; Moore, P.; 8c Balcom, K. M. (1992). The effectiveness of an online catalog.

In C. Murphy (Ed.), Automating school library catalogs: A reader (pp. 181-190).

Englewood, CO: Libraries Unlimited.
Hooten, P. A. (1992). Online catalogs: Will they improve children's access? In C. Murphy

(Ed.), Automating school library catalogs: A reader (pp. 144-149). Englewood, CO:

Libraries Unlimited.
Loertscher, D. V., & Ho, M. L. (1986). Computerized collection development for school

library media centers. Castle Rock, CO: Hi Willow Research and Publishing.

(Distributed by Libraries Unlimited, Englewood, CO.)
Martin, R. (1992). Dogs and databases. School Library Media Activities Monthly, 8(5),

39-42.
Molenda, M. (1992). Technology and school restructing: Some clarifying propositions.

In Educational media and technology yearbook (pp. 153-158). Englewood, CO:

Libraries Unlimited.
Murphy, C. (Ed.). (1992). Automating school library catalogs: A reader. Englewood, CO:

Libraries Unlimited.
Pilger, M. A. (1992). Science experiments index for young people, update 91. Englewood,

CO: Libraries Unlimited.
Saba, F. (1992). Digital media: A platform for converging educational technology

'preparadigms'. In Educational media and technology yearbook (pp. 127-133).

Englewood, CO: Libraries Unlimited.
Saettler, P. (1990). The evolution of American educational technology. Englewood, CO:

Libraries Unlimited.
VanMeter, V. (1992). World history for children and young adults: An annotated

bibliographic index. Englewood, CO: Libraries Unlimited.

RUTH V. SMALL

Assistant Professor

School of Information Studies

Syracuse University

Syracuse, New York

Principles and Strategies for Designing
Effective Computer-Mediated Instruction

ABSTRACT

Information professionals are increasingly asked to assist instructional
designers or to be designers of computer-mediated information systems
including the online instruction that facilitates their independent,
skillful use by information consumers. This paper provides some
guidelines for information professionals asked to create effective
computer-mediated instruction. It begins with a discussion of a number
of issues to consider both before and during the design process, describes
a simple yet powerful instructional design model that forms a framework
for making design decisions, and presents a wide range of design
strategies for implementation.

INTRODUCTION

The overwhelming acceptance of computer technologies as
instructional delivery systems requires designing instruction that
considers the unique interactive capabilities of computers as a medium
for providing effective, efficient, and appealing instruction to users.
As more and more information services and resources go "online,"
information professionals are increasingly asked to assist instructional
designers or to be the designers of computer-mediated information
systems, including the online instruction that facilitates their
independent, skillful use by information consumers.

116

COMPUTER-MEDIATED INSTRUCTION

117

Whether designing drill-and-practice programs in which the
objective is the memory of simple skills, tutorials that teach concepts
or rules, or simulations and games that integrate concepts, skills, and
problem-solving activities (Bunderson, 1981), designers find a variety
of creative options through computer-mediated instruction. This paper
is intended to provide some guidelines for information professionals
who are designing or adapting instructional programs or who are serving
in an advisory capacity in that regard. It begins with a discussion of
a number of issues to consider both before and during the design process,
presents a simple yet powerful model that forms a framework for making
design decisions, and concludes with a host of related design strategies
to select based on specific needs and preferences. Figure 1 provides a
graphic overview of the issues and model to be presented in this paper.

Learner Information Task Instruction

Level

Complete
Understandable

Interesting

Attention

Ability

Breadth/depth

Experience

Accurate
Topical

Important

Relevance

Needs

Unbiased
Reliable

Achievable

Confidence

Motives

Useful

Interests

Accessible
Current

Rewarding

Satisfaction

Figure 1. Design considerations for computer- media ted instruction

COMPUTER-MEDIATED INSTRUCTIONAL DESIGN ISSUES

Learners

The first issue to consider is always an information professional's
primary concern the users or, in this case, the learners. Who will receive
the instruction? Who is the target audience? What do I know about
them?

There are a number of important learner characteristics to consider
before beginning the design process. When designing computer-
mediated instruction, these characteristics become even more critical
because the learning process is often an independent activity where
the information professional or instructor serves in a facilitative or
advisory capacity rather than as the source or provider of knowledge.

118 RUTH V. SMALL

The level and ability of learners are important characteristics related
to cognitive knowledge of the subject matter and psychomotor computer
skills. Levels and abilities may range from remedial to gifted, from
beginner to advanced, and from entry level to expert. In most cases,
users of computer- mediated instruction represent a mixture of levels
and abilities requiring a range of options to accommodate them. Other
characteristics such as age and physical limitations may also affect the
design of computer-mediated instruction.

In addition, learners approach the learning experience with a range
of experiences both with the technology and the subject matter. If learners
must possess specific entry behaviors or skills in order to successfully
learn (e.g., English language skills, keyboarding skills, basic reading
skills) and the computer program does not teach them, where and how
will they get them? This is an additional consideration when planning
computer-mediated instruction.

The second group of important learner characteristics relates to
attitudes both toward the subject matter and the technology. Attitudes
encompass individual needs, motives, interests, and preferences. What
do your users need or want to know or do?

There are several ways to determine learner characteristics, including
reviewing the professional literature, consulting academic records,
observing, testing, or interviewing. Although a careful analysis of the
target learner population is recommended before beginning the design
process, it is important to continuously evaluate the design with selected
members of that population to be certain that it continues to be
responsive to their needs and abilities (Marchionini, 1991).

Information

The second issue for consideration is the information itself. How
do you judge the quality of the information (both the subject matter
and the interface) for learning? This is often an overlooked dimension
when designing instruction. If, for example, one designs a motivational,
instructionally sound program in which the information is incomplete
or inaccurate, that instruction becomes at best useless and at worst
dangerous. For instance, if one is designing computer-mediated
instruction that trains nuclear power plant workers to interpret online
information, a change in the function of a key command that is not
reflected in the instructional interface may result in learners becoming
frustrated and not using the system. Furthermore, consider the potential
consequences if the information itself is incorrect or out-of-date.

COMPUTER-MEDIATED INSTRUCTION 119

Taylor's (1986) Value- Added Model identifies a number of
dimensions for evaluating the quality of information. In some cases,
this type of evaluation may require consultation with a subject matter
expert. The information may be evaluated on the following dimensions:

Complete. Does the information provide comprehensive coverage of
the topic for achieving predetermined learning goals?

Understandable. Are the terms, language, and vocabulary used
comprehensible to the intended users? Does it avoid unnecessary
technical jargon and explain necessary complex or interrelated
concepts?

Breadth/ Depth. Is enough information provided for learners to
achieve predetermined learning goals?

Accurate. Is the information error-free so learners can trust the system?

Topical. Is all of the information directly related to the subject matter?
Is there any peripheral or potentially distracting or irrelevant
information that should be eliminated?

Unbiased. Is issue-related information presented in a manner that
includes more than one point of view?

Reliable. Is there a consistent use of terms? Are the directions and
rules consistent throughout the instruction?

Useful. Is all of the information potentially useful to learners?

Accessible. Are important content and instructions available at any
time throughout the program?

Current. Is the information the most recent and up-to-date required
to attain predetermined learning goals?

Including a large amount of information allows the learner access
based on individual needs, interests, or time available (Schaefermeyer,
1990). Winn (1990) suggests using task analysis and field testing
procedures to determine the breadth and depth, completeness, and
accuracy of the information.

Task

The third issue for consideration is the learning task. What are
learners expected to know or do when they are done? Is the program
functionality transparent or self-evident enough that learners can
quickly turn their attention to the learning task rather than
concentrating on how to use the program? For, what the task itself
does not provide, the instruction must.

The first determination about the task is if it has intrinsic interest
to learners. For example, it is probably unnecessary to spend much
time designing strategies that motivate third graders to learn during
a lesson on dinosaurs. On the other hand, it is unlikely that college

120 RUTH V. SMALL

freshmen find instruction on online searching equally intrinsically
stimulating.

Another determination about the learning task is its importance
to learners. If college freshmen must learn online searching techniques
in order to complete a required history assignment, it is likely that
the learning task will be important. If, however, online searching skills
are taught isolated from any required or desired learning goal, the
learning task may not be perceived as relating to learner needs or
interests, and therefore the task does not, on its own, take on a sense
of obvious importance to the learner.

A third task-related concern is whether the task is perceived as
manageable or achievable by all learners. If learners lack the prerequisite
skills or knowledge or if there are external constraints that prevent
learners from successful achievement (e.g., not having enough time to
complete the program), the task may be perceived as unattainable. Low
self-esteem or feeling a lack of personal competence may also interfere
with learning success. Finally, if the task is too complex or abstract
for the target audience to comprehend, the likelihood of learning success
will be greatly diminished unless prerequisite knowledge or skills are
taught first.

The final task characteristic is whether accomplishing the task is,
in and of itself, rewarding. What will learners gain from successful
achievement? Most people need some type of reward for their efforts.
Some rewards take a tangible form, such as grades or the proverbial
"Christmas bonus." Other rewards are more intangible, for example,
praise or when success at one level allows a learner to progress to a
more advanced level. Although the ultimate goal of "learning for
learning's sake" might be desirable, most learners require other types
of rewards. Therefore, if the task is not perceived as intrinsically
interesting, rewarding, and valuable, or if the task is complex and
abstract, then specific instructional design strategies will need to be
systematically included.

THE ARCS MODEL

Once the characteristics of the learner, the information, and the
task have been identified, the design process may be initiated. Although
there are a number of models that provide effective approaches to the
systematic design of instruction, one model that is both powerful and
easy to apply will be presented in this paper. The ARCS (Attention,
Relevance, Confidence, Satisfaction) Model was developed by Dr. John
M. Keller, Professor of Instructional Systems at Florida State University.
The ARCS Model is based on a number of psychological theories, but
its foundation is expectancy-value theory (Keller, 1983, 1987).

COMPUTER-MEDIATED INSTRUCTION 121

Most of the research on expectancy-value theory was conducted
in the workplace to determine how to increase performance on the job.
It is only within the past 15 years, through the work of Keller and
others, that the theories developed for the workplace have been adapted
to applications in education and training.

Therefore, in these latter contexts, expectancy- value theory may
be described in the following manner for learners to put forth the
effort to reach a learning goal, they must (a) value the learning task
and (b) believe that they can successfully accomplish the learning task.
Both must be present. If either or both are absent, the likelihood of
learner effort toward the task is low.

ARCS is a systematic approach to design that addresses both of
these two criteria, as well as the issues previously described. There are
a variety and range of design strategies that relate to each of the four
ARCS principles. Although the ARCS Model may appear to be largely
intuitive, its power lies in its organization and the ability to apply
this intuitive knowledge.

As each of these strategies is described, it may be useful to reflect
on computer-mediated instructional programs used in the past, whether
they were effective or ineffective, and how the strategies presented in
this paper might have improved or enhanced them. They may also
suggest applications to meet current or future design needs.

Attention

The first ARCS principle is to gain the learner's attention and
to sustain it throughout the instruction (Keller, 1987). To gain attention,
consider that variety is "the spice of life." Variety can refer to multimedia
formats selected on the basis of how best to represent the information.
For example, if the information intends to convey motion, a sequence
of events or time-lapse animation may be the appropriate strategy.

Format strategies that may be selected to provide this type of variety
are

text that includes expository information, examples, and practice
items;

graphics (illustrations, figures, diagrams, charts, maps);
sound;
animation;
photographs;
full-motion video.

There are other ways of providing variety. When a concept is
complex or abstract, providing both a textual and visual representation
offers useful redundancy of the concept to the learner because visuals

122 RUTH V. SMALL

convey ideas faster and easier than words, emphasize target information,
and provide an alternative representation to satisfy visual learning styles
(Hazen, 1985). Interspersing information presentation screens with
interactive screens also provides variety (Keller fe Suzuki, 1988). When
information is varied with clear, familiar examples and performance
activities and feedback, learners do not seem to tire of the activity as
quickly (Dick & Carey, 1985).

Variety may also describe a number of attention-focusing techniques
that are relatively easy to activate in computer-mediated instruction.
These include

Flashing (or blinking). This alerts the learner to important
information or that some action is required.

Borders. Borders can be used to set apart important information and
draw the learner's eye to that information. They must be kept separate
from the information they contain (Jones, 1988).

Colors. They not only add aesthetic appeal but also may be used
to facilitate readability or to indicate functionality. Color may also
facilitate subtle discriminations within complex displays
(Shneiderman, 1987). Colors should not be used in a way that
contradicts common expectations (e.g., using red to indicate go
forward and green to indicate exit) (Galitz, 1985).

Shapes. They may be used to quickly indicate similar functional or
navigational command keys or content areas (e.g., a text or graphics
window). Both colors and shapes may be used to delineate areas on
a screen, thereby helping the learner find needed information quickly
and easily.

Highlighting. This may be used to indicate important concepts or
rules.

A number of other attention-focusing devices that draw the learner's
attention to the most important aspects of the information have been
identified (Carson & Curtis, 1991). These include

type size or font and upper- and lowercase that may be used to
emphasize or discriminate information;

mnemonics and other memory devices that may be used to help
organize information for long-term memory;

humor, such as a well-placed cartoon (it must be noted that because
humor is a subjective strategy i.e., what one person finds humorous,
another may not it should be used judiciously);

novel, surprising, or incongruous information that may attract learner
interest and stimulate curiosity.

COMPUTER-MEDIATED INSTRUCTION 123

With all of these attention-focusing devices available to provide
variety and enrich information, the designer may be tempted to infuse
as many different strategies into the instruction as possible. However,
although too few strategies may result in learning boredom, too many
strategies may cause learning anxiety. The optimal number of design
strategies is dependent on all of the issues discussed and, therefore,
must be customized to the specified learning requirements. It is likely,
however, that a design that specifies a menu with 20 flashing items,
all in different colors and type fonts, will cause learning anxiety.
Selection of attention strategies for variety must be purposeful, without
sacrificing stability and structure. Designers should avoid dysfunctional
attention-getting effects as they can be distracting and annoying (Keller
8c Suzuki, 1988). The use of too many attention strategies has been
likened to driving down "The Strip" in Las Vegas for many an
assaulting rather than pleasing experience (Jones, 1988).

Furthermore, learners must have their attention refreshed
throughout the instruction (Marchionini, 1991). That is, the strategies
suggested above may be used to gain the learners' attention, but others
must be implemented in order to sustain their attention thereby
stimulating a deeper level of curiosity and higher level of interest
throughout the instruction (Curtis & Carson, 1991).

An excellent method for sustaining attention is the inclusion of
activities that require learner participation in the learning experience
known as "interactivity" a particularly relevant strategy for computer-
mediated instruction. Interactivity maintains interest and involvement
in the learning process. It may be accomplished through use of
navigational and functional commands or keys in the interface and
strategies such as practice items or embedded questions.

Two useful types of embedded questioning are "overt" and "subtle"
questions. Overt questions typically require a verbal response, either
by the learner typing it or selecting among a choice of responses. Subtle
question ing does not necessarily require an immediate, physical response
from the learner but is intended to promote active thinking by posing
a problem that the learner must mentally ponder, generating alternative
solutions as he proceeds through the program.

A third strategy for maintaining attention relates to making complex
or abstract concepts more concrete. This may be accomplished by using
examples of the concept, personal anecdotes, analogies, or visual
representations.

Relevance

The second ARCS principle is concerned with helping the learner
to understand the relevance of the learning task (Keller, 1987). Anyone

124 RUTH V. SMALL

who has worked with teenage children has probably heard occasional
comments related to subjects such as algebra or earth science, "Why
do I have to learn this? I'm never going to use it!" The importance
of the learning task must be clearly articulated whenever it is not obvious
to the learner.

One relevance strategy is providing familiarity when learning
about or doing something new is similar to something the learner already
knows or does. This may be accomplished through the use of
recognizable examples or anecdotes from the learner's realm of experience
or by relating content to learners' prior experience or maintaining
individual progress records within the program and referring to these
at various points as the learner proceeds through the computer-mediated
instruction (Keller 8c Suzuki, 1988).

Another method of providing familiarity is through the use of
powerful metaphors or analogies to explain complex content making
it easier to relate new ideas to those with which the learner is familiar
(Curtis & Reigeluth, 1984). In a recent project to design multimedia
case studies for learners enrolled in an international executive education
program, the metaphor of a journey was used, and specific navigational
tools strengthened the metaphor. For example, the menu was a "map,"
screens were organized into "regions," and individual screens had
specific "locations" (Curtis & Gluck, 1992).

A second strategy for providing relevance is informing the learner
of the purpose of the learning task at the very beginning of the
instruction and reinforcing its usefulness throughout the instruction
that is, stating what the learner will know or be able to do after
completing the instruction and linking achievement of the task to both
learning goals and real-world applications. Game and simulation
formats are useful for making somewhat obtuse subject matter seem
more relevant.

Another relevance strategy is allowing learners to make choices that
meet their individual needs and interests such as menus and submenus,
varying amounts and sequences of accessible information, flexible entry
and exit options, branching, on-screen notetaking capability, printing
capability, "sound off" options, and full-screen or zooming capability.
Options that include competitive or collaborative teamwork activities
also allow further opportunities for meeting learner needs.

Confidence

The third ARCS principle considers ways to build learners'
confidence levels (Keller, 1987). Instructional strategies that provide

COMPUTER-MEDIATED INSTRUCTION 125

enough learning support for learners to succeed will help to build
learning confidence, competence, self-esteem, the desire to persist at
a learning task, and the motivation to continue learning.

Most learners tend to seek out challenging learning tasks as long
as they believe they are attainable (McClelland, 1965). Most learners
sincerely want to learn, to become competent at the learning task (White,
1959). Learning activities that seem either too easy or too difficult tend
to be avoided because they are not perceived as linked to one's effort.
When a task is perceived as too difficult to learners, it is unlikely that
they will persist because the cost of success is seen as too high. Conversely,
when a task is perceived as too easy to learners, it is unlikely that they
will persist because it is seen as not worth the effort.

Specifying prerequisite skills, attitudes, and knowledge and clearly
presenting the objectives and overall structure of the program at the
beginning of the learning session allow learners to know whether they
are adequately prepared to successfully complete the task (Keller 8c
Suzuki, 1986). Providing feedback that links successful achievement with
personal effort helps prevent perceptions that the results were due to
luck or that the task was easy (Weiner, 1980). Novice computer users
require more "user-friendly" programs and guidance than more expert
users (Keller & Suzuki, 1988).

Confidence-building strategies provide both access to an appropriate
level of difficulty and, when needed, to instructional aides such as
pretesting and placement, accessible online help screens, and control
over type and number of examples and practice items. In addition,
orienting navigational headings and other cues that let learners know
where they are, where they have been, and where they are going, as
well as how long it will take, are desirable (Shneiderman, 1987).

To aid in learning, Carson and Curtis (1991) suggest the use of
alternative, redundant representations of textual information (e.g.,
graphic overviews, diagrams, flowcharts), the use of divergent examples
that proceed from simple to complex, matched nonexamples that are
similar to the example and presented simultaneously to illustrate
common errors, and embedded questions.

Although embedding overt questions within instruction offers a
method for monitoring learning progress, in classroom settings
instructors typically allow their students very little time to respond
to those questions (usually no more than one second). Research indicates
that providing slightly more time (3-5 seconds) results in a greatly
increased likelihood of student responses (Rowe, 1986). Yet instructors
seem to feel uncomfortable with silence and often blurt out an answer
to avoid it. Although the response may be in the instructor's short-
term memory (on the "memory surface") because the instructor has
been thinking about it, it is likely to be in the learner's long-term

126 RUTH V. SMALL

memory, thereby requiring additional time for the learner to retrieve
the information. Computer-mediated instruction is especially "patient"
in providing the appropriate amount of wait-time, based on the learner's
required or preferred pace, which permits the learner to retrieve relevant
information from long-term memory, organize that information, and
respond to the question posed greatly increasing the opportunity for
a successful response.

Some additional useful strategies that help to build learning
confidence are

graphic overviews that provide the learner with a context for
organizing information;

chunking textual information into short, meaningful, and
manageable segments (Keller & Suzuki, 1988) according to the age
level of learners, complexity of material, type of learning taking place,
flexibility of the activity, and learning time requirements (Dick &
Carey, 1985);

scheduled synthesizers that relate ideas within or across lessons and
integrate new material with old and summaries that follow
presentation of significant chunks of information throughout the
instruction, reviewing what has been learned (Reigeluth & Stein,
1983);

divergent examples and content that range from easy to difficult,
known to unknown, and simple to complex;

learning cues or prompts that can "jog" the learner's memory (Cohen,
1983);

menus so that learners do not have to recall or type terms or phrases
in order to access needed information; and

more explanation or guidance for those with little or no previous
knowledge of the topic and a fast track for those who wish to move
quickly through the instruction.

Galitz (1985) recommends a number of confidence-building strategies
including recovery options for retracting or undoing an action, commands
that use familiar, obvious, or common commands, function keys for
frequent actions (e.g., page forward), and labels that provide information
that clearly describes a function key's purpose. Providing supplementary
print materials such as quick reference cards, manuals, job aids, and
keyboard templates allow learners to choose alternative formats that best
match their learning styles and preferences (Cohen, 1983).

Consistency, another confidence-building strategy, may be
implemented by setting consistent standards for achievement that are
fully and adequately described to the learner at the beginning of the
instruction and reinforced by the way the information is presented and

COMPUTER-MEDIATED INSTRUCTION 127

learning is evaluated. For example, instruction that requires the learner
to recall or recognize information would be presented and evaluated
differently than instruction that requires the learner to apply newly
learned information to a new situation (Carson & Curtis, 1991).

Consistency may also be provided through logical organization of
content and use of color, shapes, terms, and key functions, which aid
in screen location. Consistent screen designs that have an orderly, well-
spaced, clutter-free, and clean appearance, and transparent functions;
that use plain, simple English, large enough fonts to read easily, and
clear labeling; and that cohesively group relevant elements on a screen
help to cue appropriate cognitive behavior (Schaefermeyer, 1990;
Hannafin & Hooper, 1989; Hooper & Hannafin, 1988; Jones, 1988; Galitz,
1985). An instructional interface that provides consistent selection
methods, window layout, and positioning of important text and buttons
is also desirable. Consistent location, structure, and terminology should
be preserved, with only occasional variations (Shneiderman, 1987; Keller
& Suzuki, 1988).

Another strategy for building confidence is feedback. Duf field (1991)
defines the three basic steps in human-computer interactions as (a) the
computer receives a response, (b) the computer processes the response,
and (c) the computer provides feedback to the learner. Feedback can
also be used as guidance when the learner uses inaccurate or
inappropriate strategies, immediately following response to a question
or completion of a practice item.

Reinforcing feedback is useful to encourage and support learning
as long as it is used sparingly and is always linked to effort. The most
effective type of feedback is corrective feedback that presents a message
immediately after the learner has made an error. Corrective feedback
not only provides the learner with a simple, concise, and non threatening
error response and directs the learner to the error, but also includes
the steps involved in the correct solution (Wager, Wager, & Duffield,
1989; Cohen, 1983; Tosti, 1978). In this way, the feedback becomes part
of the learning experience.

Satisfaction

The final ARCS principle promotes learning satisfaction (Keller,
1987). One method for doing so is to show or explain the consequences
of successfully achieving the learning task; for example, pointing out
that the learner can now do something she was unable to do previously,
understand something new and difficult, or use something in a new way.

Extrinsic rewards used in computer-mediated instruction such as
motivational feedback linked to effort may take the form of animated
sequences, sound effects, or verbal praise. These are useful as long as

128 RUTH V. SMALL

learners have control over receiving feedback and opting out of it (Keller
& Suzuki, 1988). Although some tasks require extrinsic rewards to
motivate learners, the ultimate goal of instruction is to develop an
intrinsic motivation to learn.

Another way to link learning to its consequences is by providing
opportunities for learners to use their newly learned knowledge or skills
in real or simulated, meaningful applications (Keller & Suzuki, 1988).
Opportunities to immediately apply what is learned to a real problem
such as a class assignment or job responsibility promote learning
satisfaction. As new instructional technologies such as multimedia and
virtual reality develop, learners will be exposed to exciting new ways
of simulating life experiences that are either too difficult, too expensive,
or too dangerous to experience in reality.

Another strategy for promoting learning satisfaction is by providing
an environment of "learner control" in which the learner perceives a
sense of learning empowerment over the learning experience. Computer-
mediated instruction lends itself extremely well to developing this sense
of learner empowerment by allowing, to a greater or lesser degree, control
over the pace of the presentation, the sequence of content, and options
such as returning to a desired screen or module, choosing when and
where to enter and exit the program, reentering where the learner left
off, selecting additional examples or practice items, saving completed
work, printing completed work, or deciding which information to
engage and which to ignore or bypass. However, the amount of learner
control provided must be carefully determined based on the various
design issues presented in this paper. Research indicates that learner
control does not assure greater learning (Steinberg, 1989) and may, in
fact, result in an unintentional lack of access to critical information.
This is an important consideration when making decisions regarding
learner control.

A final strategy for promoting learning satisfaction is through
equity. Equity assures that the goals and evaluation criteria for learning
set and stated at the beginning of the instruction are the same at
completion of the instruction, that the instruction does what it was
purported to do, and that the objectives, content, and test items are
consistent so that learning success is attributable to the effort the learner
has put into the learning activity (Curtis & Carson, 1991).

CONCLUSION

This paper has outlined several learning inputs to consider in the
design of effective computer-mediated instruction. Issues to consider
are learner characteristics, quality of information, and task charac-
teristics. The ARCS Model, in consideration of these issues, proposes

COMPUTER-MEDIATED INSTRUCTION 129

four design principles and a variety of related strategies for the creation
of effective, efficient, and appealing computer-mediated instruction. The
resulting outcomes are improved learning performance and increased
motivation to learn.

REFERENCES

Bunderson, C. V. (1981). Courseware. In H. F. O'Neil, Jr. (Ed.), Computer-based

instruction: A state-of-the-art assessment (pp. 91-125). New York: Academic Press.
Carson, C. H., & Curtis, R. V. ( 1991 ). Applying instructional design theory to bibliographic

instruction: Micro theory. Research Strategies, 9(2), 60-76.
Cohen, V. B. (1983). Criteria for the evaluation of microcomputer courseware. Educational

Technology, 23(1), 9-14.
Curtis, R. V., & Carson, C. H. (1991). The application of motivational design to

bibliographic instruction. Research Strategies, 9(3), 130-138.
Curtis, R. V., & Gluck, M. (1992, March). Multimedia case studies. Paper presented at

Focus on Instruction: Multimedia Conference. Syracuse, NY: Syracuse University.
Curtis, R. V., 8c Reigeluth, C. M. (1984). The use of analogies in written text. Instructional

Science, 13(2), 99-117.
Dick, W., & Carey, L. (1985). The systematic design of instruction (2nd ed.). Glenview,

IL: Scott, Foresman.
Duf field, J. A. (1991). Designing computer software for problem-solving instruction.

Educational Technology Research and Development, 39(1), 50-62.
Galitz, W. O. (1985). Handbook of screen format design. Wellesley Hills, MA: QED

Information Sciences.
Hannafin, M. J., & Hooper, S. (1989). An integrated framework for CBI screen design

and layout. Computers in Human Behavior, 5(3), 155-165.
Hazen, M. (1985). Instructional software design principles. Educational Technology,

25(11), 18-23.
Hooper, S., 8c Hannafin, M. J. (1988). Learning the ROPES of instructional design:

Guidelines for emerging interactive technologies. Educational Technology, 28(7), 14-

18.
Jones, M. K. (1988). Human-computer interaction: A design guide. Englewood Cliffs,

NJ: Educational Technology Publications.
Keller, J. M. (1983). Motivational design of instruction. In C. M. Reigeluth (Ed.),

Instructional-design theories and models: An overview of their current status (pp.

383-434). Hillsdale, NJ: Lawrence Erlbaum Associates.
Keller, J. M. (1987). Development and use of the ARCS model of motivational design.

Journal of Instructional Development, 70(3), 2-10.
Keller, J. M., & Suzuki, K. (1988). Use of the ARCS motivation model in courseware

design. In D. H. Jonassen (Ed.), Instructional designs for microcomputer courseware

(pp. 401-434). Hillsdale, NJ: Lawrence Erlbaum Associates.
Marchionini, G. (1991). Psychological dimensions of user-computer interfaces. ERIC

Digest. (ERIC Document Reproduction Service No. ED 337 203)
McClelland, D. C. (1965). Toward a theory of motive acquisition. American Psychologist,

20(5), 321-333.
Reigeluth, C. M., fc Stein, F. (1983). The elaboration theory of instruction. In C. M.

Reigeluth (Ed.), Instructional-design theories and models: An overview of their current

status (pp. 335-381). Hillsdale, NJ: Lawrence Erlbaum Associates.
Rowe, M. B. (1986). Wait time: Slowing down may be a way of speeding up. Journal

of Teacher Education, 37(1), 43-50.
Schaefermeyer, S. (1990). Standards for instructional computing software design and

development. Educational Technology, 30(6), 9-15.

Shneiderman, B. (1987). Designing the user interface: Strategies for effective human-
computer interaction. Reading, MA: Addison- Wesley.
Steinberg, E. R. (1989). Cognition and learner control: A literature review, 1977-1988.

Journal of Computer-Base d Instruction, 76(4), 117-121.

130 RUTH V. SMALL

Taylor, R. S. (1986). Value-added processes in information systems. Norwood, NJ: Ablex.

Tosti, D. T. (1978). Formative feedback. NSPI Journal, 17(8), 19-21.

Wager, W.; Wager, S.; & Duf field, J. (1989). Computers in teaching: A compleat training

manual for teachers to use in their classrooms. Cambridge, MA: Brookline Books.
Weiner, B. (1980). Human motivation. New York: Holt, Rinehart, and Winston.
White, R. W. ( 1959). Motivation reconsidered: The concept of competence. Psychological

Review, 66(5), 297-333.
Winn, W. (1990). Some implications of cognitive theory for instructional design.

Instructional Science, 19(1), 53-69.

JOE C. RADER

Head, University Archives

University of Tennessee Libraries

Knoxville, Tennessee,

The Development of Computer- Based Training
in a Systematic Staff Training Program

ABSTRACT

In 1990, the Libraries of the University of Tennessee, Knoxville,
successfully developed a seven-unit program of computer-based training
(CBT) for library staff under a Department of Education grant. The
program's development and its implementation mark a first in libraries
for systematic CBT for staff. Named New Horizons in Library Training,
the program has been distributed widely to other libraries. Although
expensive, this pioneering work has promise for further development
and refinement as a staff training tool and for joint utility with computer-
based programs of user instruction.

BACKGROUND

Since 1990, a program of computer-based training (CBT) for library
staff, developed using HyperCard at the University of Tennessee,
Knoxville (UTK), has received wide publicity in the library press and
has been acquired by more than 75 institutions that are known. Others
who are not known may have downloaded the files via File Transfer
Protocol (FTP) over the Internet since the entire program has been
available to the world on the UTK Libraries' VAX (address:
utklib.utk.edu or 128.169.202.177). To date, 211 people have made direct
inquiries about the program by electronic mail, phone, or other means
to the CBT project directors, Pauline S. Bayne and Joe C. Rader. It

131

132 JOE C. RADER

is named New Horizons in Library Training: Computer -Eased Training
for Library Staff, and the following is a discussion of the program,
the process of its development, its reception and implementation, and
some conclusions about the experience.

Since HyperCard became available to the public in 1987, librarians
of all degrees and stations have been exploring its use in the construction
of interactive, computer-based training for library users. The result has
been many programs with widely divergent degrees of sophistication,
thoroughness, and implementation success. Some have been one-
librarian attempts to create something helpful for users of the reference
room. Others have been team efforts with significant institutional
support that were designed to be implemented in a programmatic way.
Project F.O.R.E. (or Focus on Research and Evaluation) from Utah
State University is a good example of this effort (Piette & Smith, 1990).
And there is the multilibrary joint effort called HyperCard Library
Instruction Project (HLIP) (Talan, 1992); that product is to be available
in the fall of 1992. All these projects have the common goal of using
technology to solve a library challenge of human interaction that is
both labor intensive, from an administrative perspective, and highly
repetitive, from the individual staff member's perspective. Computer
software packages like HyperCard (and there are several available now)
have placed into librarians' hands the capability of creating self-paced,
interactive CBT without requiring the expensive services of
programmers, who must write reams of code to create even simple forms
of interactive CBT as had been the case in the past.

THE TENNESSEE CBT PROJECT

In general terms, the project at UTK had as a primary goal the
application of technology to help solve a library challenge: systematic
training of library staff. Despite the widespread enthusiasm for
HyperCard to create library instruction for users, the UTK developers
found no evidence in the literature of the use of CBT in libraries for
staff training (other than a few reports of using some application-specific
programs to teach certain computer skills like DOS or a particular
software like WordPerfect). Yet, among librarians and in the literature,
one encounters much discussion of the training needs and skill
requirements for all levels of library employees.

In reality, the research library of today presents a large, highly
complex environment in which the information needs of users are met
increasingly through technology-mediated processes. The people who
work in this environment must be trained and retrained constantly,
but this is rarely done systematically or, if it is, it is not reported.

COMPUTER-BASED TRAINING 133

Often the training of the new employee is left totally to the supervisor
in the employee's department with little if any coordination from unit
to unit except perhaps for a general welcome and orientation to the
institution given to groups of new employees. Where training does occur,
it is collectively expensive in staff time although staff may be largely
unaware of its cost librarywide. A Carnegie Foundation report has
suggested that industrial corporations spend $40 billion per year in
the training of employees. No one knows what libraries spend.

Into this environment stepped the team of librarians at UTK to
see if they could develop a partial remedy to this management problem.
They proposed CRT with these characteristics:

the program would be accepted by staff and administration;

the program would ensure instruction for all library employees
(student, part-time, full-time) in fundamental, basic library
operations;

the materials would be machine-based (no need for human
intervention after an introduction);

the machines would track trainees' progress;

the materials would be transferable with little effort from institution
to institution.

It was an ambitious proposal and required support beyond that which
a single institution could readily afford. Backing came from the
Department of Education in a $67,000 grant and, later, from Apple
Computer, Inc., in training, encouragement, and additional equipment.
The project also required much support from the University of Tennessee
Libraries as well since this major project drew people away from their
normal positions to work on various aspects of the project, officially
and unofficially, for 15 months. The official "Final Performance Report"
to the Department of Education, available as an ERIC document,
contains details of the activities of the project as well as appendices
of forms, surveys, and other documents used (Bayne & Rader, 1991).

One of the first matters to be decided was who was to be responsible
for developing the instructional units. The grant specified a team of
seven librarians plus two as directors. Team members were solicited
by the codirectors on the basis of instructional experience or interest
in HyperCard applications. To the extent possible, the members were
chosen also to represent a cross section of the libraries. This prevented
the burdening of any one area too heavily by having multiple staff
members away from their normally assigned duties, and it gave a
widespread base of interested parties who, it was hoped, would "infect"
others in their areas with their enthusiasm for the project. Later, team
members were paired with codirectors to create pairs of coauthors, each
of which was responsible for a specific instructional unit. (For those

134 JOE C. RADER

doing arithmetic on the division of labor among the team, one topic
had three persons assigned to it, and all the others had two each.)

Topic Selection

Another thing to be decided early on was which topics out of all
those that might be addressed would be treated in the CBT. Two surveys,
one local and one national, gave guidance. A quick and easy survey
of library supervisors and department heads at the UTK Libraries gave
the developers their initial guidance on subject matter to consider for
inclusion in the program. A more extensive questionnaire was mailed
to all directors or personnel officers of the 1 19 members of the Association
of Research Libraries in late autumn of 1989. An interesting outcome
was that high priority topics from the national survey were among
those that had been ranked high in the local survey also. At the top
were the following:

1. Service Attitudes and Behaviors

2. Orientation to the Academic Library

3. Access to Journal Literature

4. Integrated Online Systems for Libraries

5. Introduction to Reference Work

6. Resource Sharing

7. Acquiring and Processing Library Materials

8. Preservation of Library Materials

9. Introduction to Government Documents

The list was modified somewhat. Number 1 on service attitudes
was omitted because no one could figure out how to develop an effective
unit with the medium that was to be used. Orientation was to be included
but with institution-specific information; it was thought that to be
effective this unit would have to contain local, highly specific
information. It was developed for two reasons: it would be used at UTK,
but it also could serve as a model for other institutions. All other units
were to be designed to present generic information so that they could
be readily used in different libraries. Number 4 on the list, "Integrated
Online Systems for Libraries," became "Computers in Libraries." Six
new units besides a revamped prototype unit introducing the Library
of Congress classification system were the limit called for in the grant;
therefore, the cutoff dropped preservation and government documents
for purposes of the project.

Besides gathering information on topics to be developed, the surveys
also served to plant initial expectations locally and nationally to some

COMPUTER-BASED TRAINING 135

degree for a CBT staff training product "one fine day. " It was a stimulus
to people in libraries to think about the possibility of such, at the
very least.

Preliminary Decisions

At the outset, considerable analysis led to the decision to use
HyperCard and the Macintosh platform. For one thing, there were not
many options in terms of easily accessible hypertext software at the
time, and HyperCard was a relatively simple and highly intuitive tool
in the hands of a Macintosh user of only moderate expertise. A second
consideration was the cost of HyperCard in 1989: free and bundled with
new Macintoshes or $49 if purchased otherwise. And, finally, the
Macintosh, plain and simple, was thought to be the most widespread
kind of computer found in libraries after the IBM and clone PCs.

While waiting for equipment and software, the CBT development
team had much to do. The planning of the content of each unit began.
Each pair of coauthors had to determine exactly what information was
to be conveyed, and, even more difficult, what was the most economical
way to convey that information. To get a group of logocentric people
to use as few words as possible in instruction without sacrificing
important information was a formidable task. This was necessary,
however, for CBT to work effectively; it is not a mechanism that allows
wordy presentation if it is to hold the trainee's interest.

Considerable training was necessary for each team member to be
brought up to speed in the use of HyperCard for development and
in other areas. A two-day intensive HyperCard tutorial taught by Apple
representatives brought the entire team up to the level of doing simple
programming using HyperTalk. A seminar on basic graphic design and
another on the fundamentals of instructional design introduced the
team members to concepts and areas for further individual reading and
exploration. Resource books on HyperCard were made available, and
such standard texts as Gagn, Briggs, and Wager's (1988) Principles of
Instructional Design and Dick and Carey's (1985) The Systematic Design
of Instruction were recommended for further self-education.

Development Activities

When their Macintoshes were available, team members began the
design of "storyboard" stacks that were forerunners of the actual stacks
that would be developed. They created "draft" narration screen-by-
screen. Then reinforcing or presentation ideas for each segment of text
were noted on the screen on which the text was recorded: graphics,
sound resources, special effects, animation, or other production ideas.

136 JOE C. RADER

In this way, an outline of content was being created, and the outline
could mimic the actual features of the ultimate presentation using, for
instance, such devices as loops or other hypertextual connections. This
early conceptualization of the instruction to be presented and the
metaphors and devices of presentation to be used was an important
step on the road to production.

Although team members were topically assigned, the team met as
a group at least once each month for the duration of the project to
review the work done by the pairs of coauthors and to decide matters
of common concern such as the design of screen templates, fonts to
be used, and other questions that affected the consistency of all units
of the series. Team members circulated for comment drafts of texts and
other plans in print and in HyperCard formats so that reciprocal
reviewing could take place as work was being done. This meant that
any creation went through a kind of "pretest" since it was reviewed
by seven other team members before the first review by anyone not
a part of the team.

Evaluation Techniques

Evaluation of the CBT products created was extensive and based
on the Dick and Carey (1985) model for the design of instruction. First,
two or three selected library employees viewed a unit in the presence
of one of the developers of the unit. Discussion between the reviewer
and the developer gave important feedback to the author, who was able
also to observe the interaction of reviewer and material. Authors made
many notes for consideration when revising the instruction. Next what
was termed an "expert review" occurred. All nine team members reviewed
each unit completely and made specific comments and notes in a screen-
by-screen fashion. The coauthor teams then revised their units based
on the information gathered from these evaluation processes.

A second phase of evaluation occurred when 13 library employees
who had not seen the materials reviewed the material. Each examined
all units and made comments on each unit on prepared forms. Again
coauthors made revisions in either instructional content or presentation.
Among the changes made were adjustments to drop the average length
of time to go through a unit from a range of 30 to 50 minutes to 15
to 45 minutes. If trainees are truly learning from the concentrated
instruction presented in CBT, 45 minutes is a long and tiring time
for one sitting.

Late in the year in 1990, the CBT program was ready for one last
"field test." In this phase, library supervisors and new employees went

COMPUTER-BASED TRAINING 137

through the instruction under circumstances that anticipated those in
effect during full future implementation. A total of 49 persons
participated in this evaluation.

After final adjustments and minor changes, the CBT program was
ready for implementation at the UTK Libraries. The main library had
a training room equipped with six Macintoshes, and each branch library
had a Macintosh dedicated to training purposes. Since January 1, 1991,
New Horizons in Library Training has been a part of the training that
all new employees receive. For the first time, perhaps in the history
of the institution, there is a program that ensures that all employees
receive the same basic core of instruction about the libraries, their roles,
and major functions.

Structure of Instruction

Some specific information about the structure of the CBT
instruction might be helpful in understanding both the scope of the
program as it was initiated and how it worked for trainees. The
developers had started with the premise that each trainee would be
issued his or her personal "trainee diskette" that would both control
progress through the program according to a predetermined route and
would be the vehicle by which data on the trainee's progress were
recorded. This information would be used by supervisors to follow a
trainee's progress through the program and would be used for analysis
and evaluation of the program itself. Trainees were to pick up their
diskettes before they started a unit and turn them in again when finished.

Analysis of this procedure soon made it evident that another system
would have to be devised. Keeping up with scores of trainees and their
diskettes would have taken enormous effort. Moreover, the transferring
of files from the individual trainee diskettes to administrative machines,
combining and manipulating the data, and then getting appropriate
information to supervisors in departments throughout the libraries
would have been an even larger personnel administration burden.

At this point, a sound educational principle that had guided the
planning of instruction in the individual units also provided the
inspiration for designing trainee access to the program: namely, "Give
the learner as much choice as possible while learning a set of materials."
Or put another way, "People have a better attitude toward learning
and, therefore, learn more when they can feel that they are in control."
The access solution was simply to provide a menu that offered buttons
for all topical units. Trainees would choose options and have the
responsibility for proceeding through the program as they and their
supervisors determined what was best for them in their particular
situations. Central administrative control, generally resented by

138 JOE C. RADER

employees, was diminished by shifting the responsibility to the
individual trainee and his or her supervisors.

This decision meant that another means of data tracking, gathering,
and consolidation was necessary, however. Scripting in the individual
HyperCard stacks caused certain data to be captured unobtrusively as
a trainee went through the program. At the start of a unit, the trainee
must enter his or her name and the department of employment; after
indicating whether a student employee or full-time employee, the trainee
is not asked for further input of information. But the program
automatically records information on the trainee's performance, for
instance:

the date,

the unit being worked on,
the time work began,

times when the trainee passes certain markers in the unit,
the identifying numbers of questions missed and a score,
any comments the trainee wanted to volunteer when prompted at
the conclusion of the unit,
the time the unit was completed.

The computer writes this information to a text file each time the
trainee uses a "quit" button to exit from the CBT program. Not using
a trainee diskette meant that these data text files would have to remain
on the hard disks of the machines the trainees used until there was
some human intervention to remove them. At UTK, the Library
Personnel Office assumed the responsibility for gathering and massaging
the data. Each Friday, the text file from each training Macintosh is
copied to a diskette and removed from the hard disk, and a "fresh"
empty text file is put on the hard disk to replace the one just removed.
These gathered files are consolidated in the personnel office, extraneous
data (from "false starts" and similar errors) edited out, and reports
generated that are sent to the trainees' supervisors for whatever action
might be appropriate. Supervisors have lists of review and test questions,
for example, so that they can analyze those missed by their trainees
and pursue correction, remediation, or further training. If a book shelver
appears still not to understand the Library of Congress classification
system after going through the CBT module on that topic, the shelver 's
supervisor had best take further action either to train the employee
fully or see that he or she is transferred to another position.

The CBT project directors at UTK did the preliminary analysis
of what data were desired and the programming to capture the pertinent
data and put that data into reports needed for administrative purposes.
These programming requirements had to be integrated with the menu
access decision already made. The directors also had responsibility for

COMPUTER-BASED TRAINING 139

other "front-end" items such as a sequence of "help" screens available
on demand throughout the program.

Experience has shown that the menu approach to providing trainee
access to the CBT program and the consequences of its administration
at UTK have yielded a serendipitous benefit. The cost, formatting,
control, and manipulation of the many diskettes that would have been
necessary with a trainee diskette approach are not necessary. Although
it takes time to gather data and generate the weekly reports, most of
that work would have been necessary anyway, and it is far less time-
consuming to download data from nine machines than it would have
been to download data from potentially scores of trainee diskettes
each week. That both trainee and supervisor, with a sense of control
in their access to and use of CBT, have more of a feeling of "ownership"
is certainly a boon, though New Horizons in Library Training is, in
fact, an activity mandated by central administration.

Supervisors were introduced to the CBT program early. Then,
knowing the contents of the program, they determined the sequence
of units for trainees from their individual departments to follow and
the desired pacing. This information is the basis of a departmental,
paper checklist that is created by the personnel office and maintained
by that office and the trainee or supervisor. Having supervisors who
are responsible for the rest of a new employee's training also responsible
for the CBT portion seemed, in the end, eminently sensible and
appropriate to the developers of the program and the library
administration at UTK.

If for some reason a trainee does not pursue CBT in the time
recommended on his or her checklist, the personnel office sends a
reminder to the department head about the lack of anticipated
progression through the program. At the conclusion of the CBT, a
trainee's supervisor sends the checklist to the personnel office for record-
keeping purposes, and the Dean of Libraries sends the trainee a letter
of congratulations for having successfully completed New Horizons in
Library Training, an important part of library employees' training and
orientation.

Although the emphasis here is on the development of machine-
based training and learning, a few comments on human relations issues
might serve to illustrate the relative success of the CBT program at
UTK. The developers are working on a more detailed piece on this
topic that will be published later if all goes well. The assumption was
that for a new and "foreign" training program to work successfully
in a large organization, it is not enough for an administration to mandate
its use; it might, in that case, very well be viewed as merely some
administrative requirement to be gotten around or negated as much
as possible. Nor is it enough for the program to be innovative and

140 JOE C. RADER

attractive. Rather, to be successful, the program must be understood
by the people who are to use it and be viewed by them as something
that will help them to do their jobs more successfully, easily, or
confidently. In business jargon, a program must be "sold" to the actual
users.

In the prolonged planning, development, and implementation of
CBT at UTK, both the developers and the library administration paid
considerable attention to the notion of "buy in" by rank-and-file
employees and supervisors. The following succinct list of the major
techniques and activities to promote "buy in" illustrates that effort.

1. Even before actual development began, a questionnaire asked selected
staff and supervisors to rank topics thought to be suitable and desired
for development in a CBT program. They understood that their input
would help to shape the content of the proposed program.

2. The project directors negotiated with colleagues about becoming
team members on (among other considerations and characteristics)
the basis of distributed representation throughout the libraries. This
became a "grapevine" means of generating continuing curiosity about
and interest in the project.

3. As units reached a semblance of their intended shape and content,
the development team used every opportunity to stage demonstrations
of the project to different representative groups: administrators group,
department heads council, staff meetings, etc.

4. At the developers' request, staff members of various levels and in
different combinations were asked to evaluate units as they reached
some degree of completion. Evaluative comments were treated with
total seriousness, and the participating staff members recognized that
they were contributing to the project. In due course, such participants
received letters of thanks for their participation.

5. Supervisors, a key group to win over in any library enterprise, were
apprised of plans and progress throughout the project and were
introduced to the various units of instruction early in the final stages
of development. Their familiarity with the project was essential for
any fruitful discussion of implementation planning and acceptance.

6. As both veteran staff and new trainees experienced the program and
then returned to their departments, they became de facto emissaries
of the CBT program. It was helpful that the vast majority had had
a very positive reaction, according to their comments and responses
gathered by machine and paper mechanisms.

7. Finally, the fact that the project enjoyed complete administrative
support from the highest levels from inception of the idea through
implementation must be acknowledged. Announcements in the
Dean's Newsletter and individual memos to affected staff on various

COMPUTER-BASED TRAINING 141

topics throughout the development and implementation of New
Horizons in Library Training served to promote internalization of
a project that otherwise might have been viewed as something largely
external or peripheral to the everyday life of the UTK Libraries.
It may have also been significant that even university administrators
expressed respect and congratulations for the project.

Obviously other libraries who choose to implement the UTK-
developed CBT program could not imitate completely such techniques
and devices for consensus-building for acceptance, but the UTK
techniques should provide some advice in ways to achieve insti-
tutionwide acceptance of programs that otherwise might meet with
resistance. The "top-down" introduction of a new way to train staff
is a particularly sensitive issue since any such program automatically
sets up a situation in which the methods that have been used to train
staff are contrasted with the new way, and individual staff egos,
consequently, are sure to be touched.

THE UNIVERSITY OF KENTUCKY REPLICATION

Despite apparent success and ready acceptance by the staff of the
UTK Libraries, the original directors of the project and the UTK
personnel librarian wanted an opportunity to introduce New Horizons
in Library Training in a research library setting where the staff had
not been "contaminated" by direct exposure to the program or by
publicity about it either in a presentation or from bits and pieces of
information gathered informally. The basic goal was to be able to pretest
a group of trainees to determine whether what appeared to be learning,
as inferred from data on UTK trainees, was indeed occurring because
of the CBT.

Since the libraries of the University of Kentucky and the University
of Tennessee were cooperating on some projects already, the respective
deans approved another project. A team of librarians from both
institutions were to adapt the HyperCard stacks created at UTK to fit
Kentucky's local circumstances and introduce the CBT program there.
In essence, in exchange for assistance in creating appropriate local
information for Kentucky, the staff there would administer a paper
pretest to trainees, gather data on them while they went through the
CBT, and forward that information to the UTK team for analysis.
Coinciden tally, additional work had to be done before starting the
Kentucky project because Claris had released a new version of HyperCard
after the UTK project's development had ended, and Kentucky was
running the latest version (2.0) on their machines where the program

142 JOE C. RADER

was to be mounted. All stacks had to be converted to the new version
and checked for any flaws or discrepancies that resulted from the
conversion.

The Kentucky study has not been concluded at the moment, but
preliminary examination suggests comparable results in terms of trainee
performance. There was apparently some significant difference, however,
in general receptivity to the program. Trainee acceptance at Kentucky
under the circumstances that obtained during the time of the data
gathering was not so positive as that among trainees at UTK.

One readily identifiable problem was the availability of machines
for Kentucky trainees to use. There the CRT program was mounted
on machines in student computer laboratories, placing trainees in the
situation of having to compete with students and others for use of the
machines. There was clear resentment among some library trainees at
the "waste" of their time waiting to get a machine on which to do
some CBT assignment about which they were not too sure in the first
place. That the activity was part of a study being conducted by the
University of Tennessee rankled others. Publication of a formal study
of the Kentucky findings is planned soon.

PRESENT AND FUTURE

Funding for UTK Libraries has not been good in the past two
years. Consequently, morale has been low for everyone: jobs were
eliminated, and services, acquisitions, and even operating hours were
reduced to balance the libraries' budget. In this milieu, no one proposed
further development of CBT units, although the developers had thought
that, once the trail had been blazed, others would follow, quickly, with
the creation of additional units either to enhance the general program
or to supplement it with training units for departmental activities. Only
in the spring of 1992 has this expectation shown signs of being realized,
for development of both kinds of units is now proposed.

One person wants a departmental unit on advanced Library of
Congress classification training designed for stack attendants. The head
of the acquisitions department is developing CBT units to teach more
routine tasks to student workers in her department. And the two directors
of the original project have received encouragement from the
Commission on Preservation and Access to prepare a unit on
preservation issues. If developed, the unit would be added to the menu
of the general program. Or it could be distributed as a stand-alone
unit promoting preservation awareness. Reference librarians have begun

COMPUTER-BASED TRAINING 143

studying sections of the CBT designed for staff with an eye toward
modifying it for user training, either as computer-assisted instruction
in group presentations or library self-instruction devices.

Some successful experimentation has occurred in the UTK Libraries
in the conversion of the CBT HyperCard stacks to run on a DOS platform.
Required to accomplish this are the software products Convert It and
Toolbook and someone competent in both environments, besides a
Macintosh with HyperCard and a DOS/OS machine with lots of
memory. The process does not convert every aspect of the HyperCard
product, and it does not convert without some programming
intervention and other tuning and tweaking. It does work, however,
although some things are lost such as some font varieties and sound
resources; these can be supplied from other sources, just not converted
directly from HyperCard. Because of this experimentation with
converting stacks, current planning for the preservation unit calls for
it to be made available in both formats, thus extending its applicability
to many more sites than one version alone would.

There is additional study at UTK of the opportunities multimedia's
inclusion offers in the realm of staff training. The product QuickTime
has inspired thinking about the mix of real-time video segments in
a general CBT presentation. The prospects are exciting for developers,
but the cost of development rises with the addition of the multimedia
dimension. And the widespread availability of machines capable of
running the product in whatever form it might ultimately take is also
an important question to be answered. All this must be weighed in
the balance of cost-benefit analysis.

CONCLUSIONS

What conclusions can one draw, then, after this long narrative of
design, development, and implementation of a systematic CBT program
for library staff in an academic library?

In the design of CBT, the developers at UTK learned that some
assistance from persons whose expertise was in the practical areas of
art, instructional design, and programming would have been more
efficient than training the team members responsible for content, to
do all work in those areas. They also learned the vital importance of
team effort. Not only was the opinion of many rather than few (or
one) advantageous for the development team in their freewheeling
sessions of brainstorming for ideas, metaphors, or appropriate graphic
representations, but the idea proved beneficial to the development
process throughout. This included extensive use made of reviewers of
various degrees of experience, training, and positions within the

144 JOE C. RADER

libraries. The testing of instruction was found to be absolutely critical
to success; even more so was this true for review and test questions.
The developers found that it is, indeed, an area of expertise itself to
fashion questions that truly test the information intended to be tested
and to do so with clarity and without betraying bias.

Another challenge was for developers to understand the medium
in which they were working, to understand how conveying information
was different in HyperCard from what it was in print or orally. Among
the following concerns the team developed, one can see the
differentiating characteristics of this instructional medium.

Economy of expression became a foremost concern. A HyperCard
presentation is not the place for wordy expression. Related to this is
the necessity for including small amounts of text on a given screen
or at least having it appear to the trainee in small segments. Otherwise,
a hypertext medium begins to look like a print medium that has been
moved to a computer screen.

The quest for potent graphics became obsessive as the importance
of the visual element in holding the trainee's attention was realized.
And sometimes when the "perfect" graphic had been found, successfully
scanned, and imported into HyperCard, it had to be excluded because
of copyright considerations. (Or there was not time enough to obtain
permission for use.)

Although time-consuming to construct, movement on the screen
or outright animation became a powerful tool for punctuating text
or graphic representation.

The design of instruction so that the trainee is in control was an
important concept that was learned. Even when the learning of certain
information is essential and, therefore, required by the instructional
design, it should be presented with as much choice as to sequence as
possible and with as many options as is reasonable. The use of alternate
loops and other devices can assure coverage while allowing for a diversity
of routes through the material. In short, involving the trainee, requiring
some action on his or her part, is vital for successful interactive
instruction.

Trainees found the review or testing after short intervals of
instruction to be helpful and complained during evaluation when it
was not present. They also preferred immediate feedback as to why
their answers to questions were right or, more importantly, wrong; this
in itself can be a powerful teaching and learning device.

To the surprise of developers, trainees liked sound resources even
if they were overly simple or trite. Similarly, trainees appreciated the
humorous or light-hearted style of presentation even though the
fundamental instruction was serious.

The developers heard from reviewers and "field test" trainees or
otherwise learned these points from actual development and evaluation

COMPUTER-BASED TRAINING 145

experience, regardless of whether they had encountered them in their
preparatory studies of instructional design or not.

The distribution of the CRT program at little or no cost as had
been promised in the original proposal to the Department of Education
yielded a surprise or two. Librarians from other institutions sometimes
found that what was intended by author teams to be "generic" instruc-
tion about library procedures was biased by local practice at UTK.
Corollaries to this are that actual details of library practice vary widely
from institution to institution even when the libraries seem comparable,
and these local customs and traditions are important to librarians when
they train employees to work in their institutions. This raises certain
questions: How important are these differences in the delivery of library
and information services to the clienteles of institutions? Or might the
community of libraries or at least those similar in size and mission
move toward standardization of practice in the same way that many
cataloging idiosyncrasies have been lessened over the years since the
advent of shared cataloging? Such standardization might make training
all staff easier (and perhaps using the libraries easier in the long run).
Finding answers to these questions was beyond the scope of the CBT
project at UTK, however, and would, no doubt, raise even further
questions. (There is also the suspicion not a clearly stated opinion
yet that there is little interest in truly generic instruction for training
staff that is not modified to take into account variant local practices.)

Another conclusion the UTK developers reached is that CBT is
an effective technique for staff training (and one that may bring a
consistency to basic training for library staff that does not exist without
CBT), but it is undeniably expensive to develop if developed well.
Furthermore, libraries collectively are not a big enough market
apparently or are not rich enough, or the diversity of practice mentioned
above is too great to warrant the development and marketing of staff
training by private sector vendors. And the expense of development
is often too great for one institution to bear the cost totally for the
creation of an effective program. But two possible solutions to this
dilemma are emerging. One is the development of materials that may
be used commonly to educate the users of libraries about practices and
to train library employees. Distributing the expense of development
between user education and staff training would spread the cost burden.
The other is the group-based project, in which several libraries pool
their resources to create a CBT product useful to all member libraries
for staff training. This would follow the HLIP model, mentioned at
the beginning of this paper. Or, perhaps, a combination of these
approaches might well put CBT development in the realm of the possible
for interested libraries.

An indirect outcome that may occur in the activity of considering
the development of a CBT program for staff is the focusing of attention

146 JOE C. RADER

on the cost of training staff in nonautomated ways. In fact, recognition
of that cost might become a factor in a commitment to an efficient
and effective program of staff training as a mechanism for improving
quality and efficiency of service to users, the ultimate raison d'etre for
libraries after all; that is, a systematic approach.

The CRT project at UTK and its implementation have generally
been deemed a success, and the attention generated nationally and
even internationally has been noteworthy. Clearly there is interest,
and one presumes a need behind the interest, in this area of computer
applications to library problems. But the project at UTK only broke
the ground of a new library management territory. Full exploration
is yet to take place; and as with any good research or project, it may
have pointed a way to move, but it also raised as many questions as
it answered in the process. Perhaps others now will join in pursuing
them.

REFERENCES

Bayne, P. S. ( & Rader, J. C. (1991). Computer-based training for library staff: A

demonstration project using HyperCard. Final performance report. Knoxville:

University of Tennesse Libraries. (ERIC Document Reproduction Service No. ED

333 902)
Dick, W., ic Carey, L. (1985). The systematic design of instruction (2nd ed.). Glenview,

IL: Scott, Foresman.
Gagne, R. M.; Briggs, L. J.; & Wager, W. W. (1988). Principles of instructional design

(3rd ed.). New York: Holt, Rinehart, and Winston.
Piette, M. I., & Smith, N. M., Jr. (1990). Focus on research and evaluation: HyperCard

library instruction. Logan: Utah State University, Merrill Library.
Talan, H. (Project Director). (1992). HyperCard library instruction project. San Francisco,

CA: San Francisco State University, Leonard Library.

RONNIE PETERS

Art Director

Dynamic Diagrams

Providence, Rhode Island

Designing for the Computer Screen

ABSTRACT

Designing for the computer screen poses new challenges for the designer.
While some of the issues are new such as time, motion, and sound,
other aspects such as the readability of typography, the separation and
combination of image and type, and the general issues associated with
projecting the three-dimensional world onto a two-dimensional surface
are part of a complex design tradition. When designing for this new
medium, the designer is faced with the problem of organizing a large
amount of information in a small area and must establish the most
orderly arrangement of information, determine the hierarchic scale of
importance, arrange the easiest accessibility of information, and design
the appearance accordingly.

THE COMPUTER SCREEN AS A DELIVERY PLATFORM

While we still get a great deal of information from looking at the
printed page, more frequently the computer screen is the end delivery
platform for information. Computer users are not just using the
computer screen as the environment on which elements are composed
and created in a preview mode for later production as a printed document.
The computer screen is more often the reading surface on which the
interplay of images, type, and sounds is projected and intended as their
final reading and listening form. Hypertext or hypermedia software
is one of the first kinds of software to exploit the computer screen
as a presentation platform. Hypermedia refers to kinds of linked

147

148 RONNIE PETERS

information and can be various media besides text. Many software
systems include some hypermedia capabilities.

Display terminals are proliferating as information delivery
platforms in many areas of our daily lives. This new reading surface
poses its own set of unique issues and presents new challenges and
opportunities for visual design. While some of the rules are similar,
designing information for this picture surface is different from the
printed page, sign, or billboard. The high resolution and frozen structure
of the printed page are gone, and the coarse resolution of the screen,
the projected rather than reflected light of the printed page, and the
elements of time, motion, and the addition of sound are the qualities
of the computer screen.

The medium of the computer screen and the qualities of that surface
are new, but the issue of projecting a three-dimensional space onto
a two-dimensional surface remains the same. Projection on a two-
dimensional plane allows objects to be rendered that are impossible
to build in three dimensions and are hard to conceive of without
rendering (Figure 1).

What is perceived on the screen of the computer is combinations
of pixels (Figure 2). One pixel represents a point; it is one-dimensional.
Dragging a point forms a line. Dragging a line forms a plane, which
is two-dimensional, and dragging a plane forms a volume, or the illusion
of volume, which is three-dimensional.

RULES OF THE VISUAL WORLD

Designing information on and for the two-dimensional plane of
the computer screen is a new field; however, the computer screen is
a two-dimensional plane, and designing for that space is part of the
graphic design tradition. Within this field, the designer is faced with
using two basic strategies to project information onto the picture plane:
representation and symbol.

Representation is a projection of the world as we literally see it.
The understanding that we have of this image is based on the information
we have stored in our memory. We identify elements in the image with
the elements we remember. For example, we recognize the features of
someone's face in a photograph and construct an image of the person
we associate with the face in our mind.

Using symbol, the designer represents the world in the form of
elements that resemble, imply, or otherwise suggest what we see. Symbol
allows the designer to compress information into a small space and
eliminate unnecessary detail (Figure 3). Five equally sized rings in a
certain configuration become the symbol for the Olympic Games. Each

DESIGNING FOR THE COMPUTER SCREEN

149

Projection on a two-dimensional plane allows objects to be rendered that are impossible
to build in three dimensions and are hard to conceive of in the mind.

Figure 1. Two-dimensional projection

ring and its color represents one of the five continents. With each
compression, the information that passes through our visual channel
is reduced. We expand the information carried by the symbol in our
mind, constructing its relationship to the world in our associative
imagination.

Mixing symbol and representation (Figure 4) can be used by the
designer to great effect. In this image of a Dutch treadmill crane by

150

RONNIE PETERS

Point

Line

Plane

Volume

One-dimensional

Two-dimensional

Breadth -

Three-dimensional

Breadth -

Figure 2. Dimensionality on a two-dimensional picture plane

DESIGNING FOR THE COMPUTER SCREEN

151

Real

Figure 3. Use of symbol

Figure 4. Mixing symbol and representation

Simon Stevin, the representation of the figure, the building, and the
barrel is integrated with the symbolic rendering of the wheel mechanism,

152 RONNIE PETERS

which when drawn in a representational manner would be hidden within
the building.

LIMITED SPACE

The computer screen for ease of navigation and accessibility is the
best place to store information, but it is also the computer's most limited
resource. The amount of storage space required on the computer memory
is rapidly becoming smaller, but the delivery platform, the computer
screen, is increasing in both size and resolution at a much slower pace.
The limited space of the computer screen and the coarseness of the
resolution leave little room for embellishment and decoration.

To make simultaneous information available on the screen requires
a clear and orderly structure. The reader has to be able to differentiate
and retrieve in the easiest possible manner.

MULTIDIMENSIONALITY

The desire to simultaneously represent many aspects of an idea has
been with us for centuries. Various projection methods have been
developed that effectively allow the simultaneous display of views. This
technique of simultaneously projecting views of three dimensions onto
a two-dimensional surface can be used as a device by the designer of
the interactive screen (Figure 5). When compared with the printed page,
the resolution of the computer screen is very coarse. The density of detail
possible on the printed page is many times greater than what is currently
possible on most computer screens. The designer is faced with having
to find alternative methods for projecting images and compressing
information. These two examples demonstrate how visual techniques
can be effectively used to condense information into a small area.

There are a number of projection techniques, each with its own
advantages and drawbacks (Figure 6). Orthogonal or orthographic
projection is when the various views of the object are projected parallel
to the picture plane. The true measures are retained and can be measured.

Oblique projection methods include oblique, axonometric, and
isometric projection. With oblique projection, the frontal plane is drawn
in orthographic projection, and the top or sides are drawn at an angle
of 30 or 45. With an angle of 30, the true measures of the sides are
retained. If an angle of 45 is used, the measures of the sides are halved
in order to retain the correct optical distortion.

An axonometric drawing shows the plane view in orthographic
projection and the side elevations drawn at angles of 30 and 60 or

DESIGNING FOR THE COMPUTER SCREEN

153

Figure 5. Simultaneous projection of views

154 RONNIE PETERS

both at angles of 45. The horizontal measures retain their lengths,
and the verticals are foreshortened.

Isometric projection shows the lateral angles drawn at identical
angles, and the true measures are retained.

Perspective that implies deep space is bound by a horizon line and
one or more vanishing points. The viewer of the perspective image
is outside of the picture frame looking in, his eye being drawn to the
vanishing point. At this point on the plane, all the distorted parallel
lines within the image converge. This image is one point in time; there
is no motion in the still perspective picture.

Flat projections preserve the parallel condition of lines. Flat
projection methods often give the designer more freedom to express
and explain, without being faithful to a horizon line or vanishing point.
They are particularly useful when designing diagrams where elements
of space and time must be projected and cannot be distorted or lost
because of the rules of perspective. Multiple events or processes can
be projected simultaneously and understood by the viewer. These
techniques often allow the designer to describe space and form from
a multitude of angles and positions without being hindered by the
vanishing point or horizon of the perspective image.

Excellent examples of flat projection techniques are found in pre-
Renaissance and Asian art (Figures 7 and 8). Flat projection allowed
the artist to create enormous spaces and project figures and continuous
narrative over space and time within one image. Often a sequence of
events will take place across the picture plane, and the same figures
appear more than once in the same image. British painter and
photographer David Hockney (1988) describes the difference between
the Asian and the Renaissance approach:

The great difference between the Chinese scholar-artist and Renaissance
scholar-artist is this: if the Chinese scholar-artist had a garden ... he would
want to walk in it, so he would make his path so that he'd have a longer
walk. So he walks up the path of his garden and then goes and makes
a picture of that garden, or the experience of walking in it. But the
Renaissance scholar sits in a room and looks out of a window, and then
makes his picture.

He is fixed there with the window picture, and therefore he thinks of
perspective. The Chinese wouldn't because their experience is moving,
flowing, as time is flowing. And so they both start off with very different
locations; one is seated and the other is not. (p. 37)

The arrangement of windows and the layout of the computer screen
are similar to the manner in which the Chinese painting is rendered,
a flat projection. The screen is a flat picture plane on which rectangular
windows are projected. The contents of windows scroll up and down
and left to right, as if continuous in one plane. Windows appear as
though they are lying one on top of the other in plain view. The windows
underneath do not proportionately decrease in size as they would if

DESIGNING FOR THE COMPUTER SCREEN 155

Oblique

Isometric

Figure 6. Projection methods

156

RONNIE PETERS

Figure 7. Example of flat projection in Asian art

the screen were rendered in a perspective projection. The effect of placing
photographic images and perspective projections within the windows
is like mixing the Renaissance perspective image with the flat Chinese
projection.

DESIGN FOR THE COMPUTER SCREEN

Any notion that design for the computer screen is merely decorative
is a misunderstanding. Good computer screen design does not decorate;
it clarifies.

In designing for the computer screen, the designer is for the first
time faced with the problem of organizing such a large amount of
information in such a small area that there is the chance of creating
complete visual chaos. For the best and most persuasive functioning
of the computer, it is necessary to establish the most orderly arrangement
of information, determine the hierarchic scale of importance, arrange
the easiest accessibility of information, and design the appearance
accordingly.

DESIGNING FOR THE COMPUTER SCREEN

157

Figure 8. Example of flat projection in pre- Renaissance art

Rudolf Arnheim ( 1971 ) has described the importance of visual order:

Order is a necessary condition for anything the human mind is to understand.
Arrangements such as the layout of a city or building, a set of tools, a
display of merchandise, the verbal expression of facts or ideas, or a painting
or a piece of music are called orderly when an observer or listener can
grasp their overall structure and the ramification of the structure in some
detail. Order makes it possible to focus on what is alike and what is different,
what belongs together and what is segregated. When nothing superfluous
is included and nothing indispensable is left out, one can understand the
interrelation of the whole to its parts, as well as the hierarchic scale of
importance and power by which some structural features are dominant,
others subordinate, (p. 1)

MULTIPLE WINDOWS

Maximizing the amount of information within the limited area
of the computer screen has led to the use of some interesting display
methods. Defining information by separating it from its surroundings
has led to the development and use of a windowing system. The concept
of multiwindows is not new; wonderful examples of windowing
techniques can be found in Persian miniatures and manuscripts from

158 RONNIE PETERS

the Middle Ages. Windows allow their contents to be separated from
their surrounding while still remaining visible.

The windows create the illusion of a slightly three-dimensional
space on the picture plane of the computer screen (Figure 9a). The
illusion is created by the apparent overlapping of the windows, defined
by their borders. The reality is that they are all projected on the same
two-dimensional picture plane (Figure 9b). By adding frames to the
windows, a layering effect is produced. The windows that are only
partially visible behind the top layer are often distracting to the window
currently being read. While their position in the stack gives reference
to the order in which they were viewed and their existence allows ease
of accessibility, their continued brightness on the screen only serves
as a distraction. Controlling the relative brightness of the windows and
having the possibility of staggering their positions will give a sense
of hierarchy to the stack while retaining accessibility and place emphasis
on the current window (Figure 9c). The illusion of the third dimension
on the screen is more prominent; it helps clarify a structure and gives
order to the documents. Having a photograph or image or even a moving
image within one of the windows can have the effect of creating a
depth hole in the midst of the layer illusion (Figure 9d).

Mixing windows that contain images and areas of high contrast
can create a problem. It is important to clearly define contrast between
foreground and background elements. Understanding how to control
contrast, relative brightness, and color on the screen can be effectively
used to unify, layer, and separate elements (Figure 10). This series of
buttons and icons demonstrates some of the spatial depth and contrast
illusions created by using contrasting shades of gray and black to show
foreground, background, raised, and depressed elements.

THE STRUCTURE OF TYPOGRAPHY
ON THE COMPUTER SCREEN

"Typography is the art of using black to bring out the whiteness."
The high resolution of the printed page allows for a much denser display
of information than the stubborn resolution of the computer screen
(Figure 11). Type for decent quality commercial printing is resolved
between 1,000 and 2,500 lines to the inch, laser printers usually image
at 300 dots to the inch, and most computer screens currently project
at about 72 pixels to the inch.

Type of 8.5 points that might be quite readable on a printed page
at a resolution of 1,270 lines to the inch is not acceptable in a readable
form on the screen of the computer. At less than 12 points, the spaces
between characters become random; some too tight, so that characters

DESIGNING FOR THE COMPUTER SCREEN 159

(figure 9, a)

(figure 9,c)

(figure 9,b)

(figure 9,d)
Figure 9. The computer screen

Foreground

I Foreground J

j r

Figure 10. Spatial depth and contrast

160 RONNIE PETERS

10pt Helvetica

Lbrary Libraiy

14pt Helvetica 14pt Times

Library Library

1270 Lines to the inch, digital typesetting output

Library Library

The bitmapped fonts are shown at about 220% of their actual size. At small point sizes,
the resolution of the computer screen does not allow for the correct character forms or
character spacing to be generated.

Character spacing

very loose Character space Character space

loose Character space Character space

normal Character space Character space

tight Character space Character space

too tight Character space Characterspace

Space that may be saved by using tight character spacing may create readability problems
Words with overlapping characters are hard to read.

Figure 11. Pixel distortion and fonts

DESIGNING FOR THE COMPUTER SCREEN

161

appear to be overlapping, others too loose. The effect is that words
bunch together or break up, and as a result readability is poor. Serif
typefaces, which should be used for large areas of text, become harder
to read than sans serif fonts.

Line spacing or leading should be set proportionately to both the
amount of text that is being displayed and the length of the line of
type. If leading is too tight, it is more difficult for the reader's eye
to find the next line in the paragraph. Leading that is to loose has
the effect of breaking up paragraphs into lines of text and also uses
up valuable screen real estate.

Margins around text within windows are important to help
readability when text windows overlap (Figure 12).

Wide margins for text within
windows allow effective
readability without confusion with

type in thp surrniinriinfi windows

Line lengtWide margins for text withii
be betwet readability without confusio
Leading slwindows. Line lengths for
allow goo 60 characters. Leading

Wide margins for text within win-
dows allow effective readability
without confusion with type in the
surrounding windows. Line lengths
for most text should be between 40
and 60 characters. Leading (line
spacing) should be open enough to
allow good readability.

Figure 12. Margin examples

CONCLUSION

The pixel is the smallest element on the surface of the computer
screen; it is the element that combines to form a line or a plane; it
is the unit that combines to make up a photographic image or a single
letter. It is one point on the plane of the computer screen; it indicates
a position in space; it is static, centralized, and directionless.

Designing for the surface of the computer screen poses new and
interesting challenges for the designer. While some of the issues are
new such as time, motion, and sound, other aspects such as the readability
of typography, the separation and combination of image and type, and
the general issues associated with projecting the three-dimensional world
onto a two-dimensional surface are part of a deep and complex tradition.
When designing for this new medium, it is important to be aware of

162

RONNIE PETERS

what has already been done and what can be used and applied from
the rich history of art and graphic design.

The successful designer of interactive multimedia must understand
how to establish a clear visual language on the computer screen. The
designer must be able to separate information into frames while
separating information within the frames from the information about
the frames. Devices such as the use of an overlaying grid can be effectively
employed while organizing information on the screen (Figure 13). The
grid can prevent random placement and create a good visual sense of
structure while saving significant amounts of screen space.

This is a very basic overview of some of the issues faced by the
designer of the computer screen. As the development of computer
software and hardware continues to become more refined, the designer
will be faced by new and varied issues, but the role of design and the
goal of the designer as the clarifier of information will continue.

9
Biology and
medicine

Sanitation and its
absence
s. I

Diseases

1 Nineteenth Century
Public Health ]

! f

-J Biology and Medicine 1

-
Women and
children

Nineteenth-
Century Public
Health GLP

Tut*

tm \

[Cholera]

1 - 1
[TVphm J

m i

Whold I
ickels "I

rculosis I

| WorkmgClass |

HM^SKrtBrtvMtfl

f Addiction I (

>-"

Nutrition

f=g] ^
Adulteration
& contamtna-
Jlon ol food j

-I' *~* '

'-
Pollcal & Social
Background

-J Add,c,, Q n ' . j-P '

^ TO g |

| NultfltlOn |_JA<)(,.n.r -!< jf. ,,ll, |
LH D,s,as |

i ' i ' i ' ' '

D n

[Nineteenth Century
Public Health j

\ Biology ant

^ Women and Children

H"" " C!J " "

4 Political I, Social Backgroun

-J Sanitation & its AbMHM

._( 333S I
4 ij i ^ 1

HH g |

L| D1 ,.,, |

r^ri f?t^n r^7| r^n i ^^ \

The example above left lacks an overall visual
structure. The position of the boxes, the space
between the boxes and the type within boxes is
randomly spaced. The hierarchical intent of the
contents is lost. The diagram is too close to the
left edge of the window. The diagram above
right is the same information with a more logical
and hierarchical structure applied. Using weight
of line and size and body weight of type can
help create a clear sense of hierarchy. Using an
overlaying grid while composing the structure
helps to define a visual order and save valuable
screen space.

Figure 13. Visual structures for the computer screen

DESIGNING FOR THE COMPUTER SCREEN 163

Two-dimensional design, three-dimensional structures, the surface
of the computer screen, space, and environments do not exist on their
own, separate from one another. They influence each other and extend
into each other's territory, and by doing so, they create the world of
design.

REFERENCES

Arnheim, R. (1971). Entropy and art: An essay on disorder and order. Berkeley, CA:

University of California Press.
Hockney, D. (1988). Hockney on photography: Conversations with Paul Joyce. New York:

Harmony Books.

PETER SCOTT

Small Systems Manager

University of Saskatchewan Libraries

Saskatoon, Saskatchewan, Canada

Hypertext . . . Information at Your Fingertips

ABSTRACT

Hypertext is an alternative to traditional linear text and has been used
successfully to create useful indexes on various types of computers.
HyperRez, from MaxThink, is discussed in detail, as is the creation
of the major Internet index, HYTELNET. Reference is also made to
hypertext utilities currently under development that make use of the
HyperRez software.

HYPERTEXT DEFINED

Simply put, hypertext is the nonlinear representation of text and
graphics. In other words, it is information that is organized in such
a fashion that a user can jump from topic to topic easily and quickly.

Good hypertext allows a user to find, browse, and comprehend
information by indexing ideas and concepts, not merely words. With
just a series of keystrokes, a user can instantly see the relationships
between topics within a given hierarchy. Information is only useful
if it is well organized and accessible. A reader of traditional, linear
text may not be able to comprehend the ideas being propounded by
the author since his ideas may appear unstructured if merely read from
beginning to end. Hypertext, on the other hand, clearly shows those
concepts.

164

HYPERTEXT 165

HYPERTEXT SOFTWARE

Generally speaking, hypertext is best suited for use on computers.
There are many hypertext programs running on many different types
of computer platforms. Most people will have heard of HyperCard,
which runs on Macintosh computers, and many of those people will,
indeed, equate hypertext with HyperCard. However, the Amiga
computer has its own hypertext program, The Thinker; the Atari has
Spectre 128; and the Apple II has Tutor-tech. Hypertext systems are
also available for the IBM personal computer family. These include
such programs as Black Magic, EMPOWERment, HyperPad, Hy-
perSprint, and PC-Browse.

This paper will concentrate its efforts on describing just one
hypertext system, HyperRez, 1 written by Neil Larson of MaxThink in
1988. This particular system has been chosen since it is the one with
which this author is most familiar, and it has been used to run a number
of utilities including HYTELNET, 2 which recently won the Research
and Education Networking Application Award.

HYPERREZ

Here are a few facts about HyperRez. HyperRez was released in
1988. It is an easy to learn and use program, which remains resident
when loaded into the computer's random access memory. It can be invoked
at any time by depressing a hot-key sequence, which a developer may
customize, and can be returned to memory by pressing the Escape key.

HyperRez makes rapid jumps to pure ASCII files and certain types
of graphics files with the keyboard's arrow and page up/page down
keys. The program may be used royalty-free for any utility created by
a developer. To construct a hypertext utility, a developer need only write
files on a word processor that can save files as pure ASCII. This includes
all the major MS-DOS packages such as WordPerfect, Word, and QEdit.
The key to making jumps is to embed angle-bracket links within the
ASCII files. Jumps can be made to the first page of a file, a specified
screen of a file, a specified screen and line, or a specified word within
a file. A developer can create customized help files for other software
packages. For example, it would be possible to embed a link in a database
program, one screen of which could contain an angle-bracket link. Once
that screen is displayed, HyperRez could be invoked and the appropriate
file presented.

For all the reasons stated above, this author chose HyperRez as
the hypertext utility of choice. Many utilities have been designed with
the software, for example, HYDOS, 3 which gives instant access to all

166 PETER SCOTT

known DOS commands; HYPERVAX, 4 which is a browser for getting
help with VAX electronic mail commands; and HYENVOY, 5 which is
a browser for finding help with the Canadian telecommunications
software ENVOY 100. These utilities are freely available to Internet users
and may be downloaded from various sites.

HYDIRECT, a hypertext version of The Directory of Electronic
Journals, Newsletters, and Academic Discussion Lists, published by the
Association of Research Libraries, is in the process of development.
The paper version has entries for 769 scholarly lists, 36 journals, 80
newsletters, and 17 as yet unclassified titles. It is an ideal candidate
for hypertext treatment since new titles are being announced on an
almost daily basis. A user should not have to wait for a new paper
version to be published to keep up with this rapidly changing
information resource.

Perhaps the most popular of this author's utilities is HYTELNET,
a browser that gives a user almost instant access to all known Internet
sites that can be reached with the TELNET utility. For those of you
not familiar with TELNET, it is a program that allows an Internet
user to make remote connections to library catalogs, bulletin boards,
campuswide information systems, Free-Nets, and a host of useful
databases and full-text resources. TELNET can be run on most
mainframes, Macintoshes, and IBM personal computers.

The number of diverse sites currently accessible is enormous. The
number of potential sites is staggering. The challenge is to organize
the information necessary to access the sites in as simple and
straightforward a fashion as possible.

HOW TO CREATE HYPERTEXT

To explain and demonstrate the creation of a hypertext utility that
can be run with HyperRez, let us look at the design of HYTELNET.
Fortunately, HYTELNET is merely a hypertext index there is no
philosophy here, no abstract notions that need to be linked so the
descriptions will be easy to comprehend.

To create a hypertext utility with HyperRez, a developer needs to
set up a new subdirectory that will house the files. Then two major
ASCII files need to be written: START.TXT and HELP.TXT. Figure
1 shows the START.TXT file. These files are essential since HyperRez
needs to load them first in order to operate properly. As soon as the
HyperRez hot-key is invoked, START.TXT will be displayed on the
screen, covering any existing display.

Let's analyze the START.TXT file. Notice the terms surrounded
by angle brackets. These are the initial hypertext jumps. To access one

HYPERTEXT

167

Welcome to HYTELNET version 5.0

What is HYTELNET? <WHATIS>

Telnet-accessible library catalogs <SITES1>

Other telnet-accessible sites <SITES2>

Help files for on-line catalogs <OPOOO>

Internet Glossary <GLOSSARY>

Cataloging systems <SYSOOO>

Understanding Telnet <TELNET>

Key-stroke commands <HELP.TXT>

HYTELNET 5.0 was written by Peter Scott,
U of Saskatchewan Libraries, Saskatoon, Sask, Canada. 1992

Figure 1. The START.TXT file

HYTELNET Program Description.

HYTELNET is designed to assist you in reaching all of the
INTERNET-accessible libraries, Freenets, CWISs,
Library BBSs, & other information sites by Telnet,

HYTELNET is designed specifically for users who access
Telnet via a modem or the ethernet from an
IBM compatible personal computer.

HYTELNET, when loaded, is memory-resident. Once loaded hit
Control + Backspace to activate the program. To
leave the program temporarily hit ESC. To remove
from memory hit ALT-T while in the program.

For information on customizing the program see <CUSTOM>
For accessible Library on-line catalogs see <SITES1>
For other information sites see <SITES2>

For extra information on loading the program and how to
contact the author go to the <READ.ME> file

Figure 2. The WHATIS file

of these files, a user merely moves the cursor with the down arrow

key then hits the right arrow key. The WHATIS file is shown in Figure 2.

Notice that the WHATIS file also contains links. Hitting the link

for accessible library online catalogs brings up the SITES 1 file, as shown

168

PETER SCOTT

in Figure 3. This file is a listing of all the countries that have online
catalogs available.

Library Catalogs

arranged by country

<ATOOO>

Australia

<CNOOO>

Canada

<FIOOO>

Finland

Germany
Hong Kong
Ireland

<ISOOO>

Israel

<MXOOO>

Mexico

<NEOOO>

Netherlands

<NZOOO>

New Zealand

Spain
Sweden

<SZOOO>

Switzerland

United Kingdom
United States

Figure 3. The SITES 1 file

If we select Sweden, we are presented with a list of Swedish sites,
as shown in Figure 4. Arrowing down and selecting SW001, we are
presented with the file for Lund University, as shown in Figure 5.

The SW001 file is a typical site file. It contains all the information
a user needs to connect and log in to this particular site. As an added
bonus, there is also information regarding which cataloging system is
used at the site. In this case, it is VTLS. Before logging into a site,
a user may wish to know which search commands will have to be issued.
Notice that OP017 is a link file. Depressing it brings up the help screen
for the VTLS system (Figure 6).

Let's return for a moment to the START.TXT file. Notice that
there are links to other files, which, in turn, make links to further files.
SITES2 will link a user to a file that lists other types of resources available
on the Internet. OPOOO is a list of help files for online catalogs.
GLOSSARY is a file containing terms relating to various aspects of
the Internet. SYSOOO lists the many different cataloging systems being

HYPERTEXT 169

Sweden

<SW003> Karolinska Institute

<SW001> Lund University

<SW004> Royal Institute of Technology, Stockholm

<SW002> University of Umea

Figure 4. The SWO file

Lund University
To access:

1. Type TELNET LOLITA.LU.SE

2. Select terminal type 10. If you have a terminal
that supports Scandinavian 7-bit ASCII, select 4

3. Enter 100 as the location code

4. Type /LANG 1 to select English

OPAC = VTLS <OP017>
To exit:

1. Type /QUIT.

2. Enter Y to confirm.

Figure 5. The SW001 file

used in libraries and makes deeper links to the sites that use those
systems. TELNET is a file containing a description of that program,
plus various tips and tricks to help make remote sessions successful.
Finally, HELP. TXT describes the actions of the arrow and page keys.
It can also be invoked by depressing the Fl key. All of the files shown
so far are written in pure ASCII and are, therefore, easy to edit with
any word processor. Much of the information gathered to create the
files was found in the Internet sites themselves and captured with a
telecommunications program. Many already existed in other documents
freely available for downloading.

The current version of HYTELNET contains hundreds of small,
discrete ASCII files, so the chore of updating existing files and adding
new ones is not burdensome. Traditional linear indexes, with which
we are all familiar, are extremely tedious to update, especially if they
contain page references that have to be altered each time information
is added or deleted. There is also a tendency with linear indexes not
to update on a regular basis. It is just too much trouble and far too

170 PETER SCOTT

Using VTLS
Advanced User Search System:

Author searches: To search for a particular author, use the A/ searc
command followed by the author's name.

Example: A/Hemingway, Ernest

Title searches: To search for a particular title, use the T/ search
command followed by the title. Omit any leading
articles such as THE, A, AN, LA, ...

Example: T/Sun also rises

Subject searches: To search for a particular subject, use the S/

search command followed by the subject.

Example: S/Metals

Keyword searches: To search for a keyword, use the W/ search

command followed by the subject.

Example: W/ Computers

Boolean Keyword: To search for keywords with boolean operators,

use the B/ search command followed by a
keyword, then an operator, and then the second
keyword .

Example: B/Ocean and Island

B/Dog or Cat
Help: To see help, type /HELP.

Previous Screen: To see previously displayed screens, type PS.
Next Screen: The next screen is usually displayed by typing NS.

Figure 6. The VTLS file

frustrating. Hypertext indexes, on the other hand, are a joy to update.
When changing information in HYTELNET, it is generally only
necessary to alter one or two files. Also, when a particular category
of information begins to grow too large to fit on one or two screens,
it can be split into smaller files and those files renamed. It then becomes
necessary to rename any links to those files. Again, this is not an onerous
task.

The number of accessible Internet sites continues to grow on an
almost daily basis. Keeping current on all the additions has been made
easier by the formation of the LIB HYTELNET 6 mailing list. The
list currently has over 350 members from around the world. When a
new site is discovered, all members of the list are informed. They may,
if they wish, add those new files to their copy of the HYTELNET
program so that it is as up to date as possible. Members freely share
information regarding sites that may have changed their login

HYPERTEXT 171

procedures or that have added new and interesting databases. Some
members have completely rearranged the information in certain files.
For example, one user moved the United States entry to be the first
entry on the SITES 1 file since most of his users were more likely to
want to connect to an American site. Currently, the United States entry
sits last because the file's arrangement is alphabetic. Other members
have decided that they do not wish to have particular files in their
copy because of limited disk space or because they have no interest
in certain information and have, therefore, deleted those files.

A user of HYTELNET may also wish to create links to files that
represent a unique local situation. For instance, it may be felt pertinent
to add an extra help file for one's own online catalog system or TELNET
escape sequence. As long as the new links lead to the correct information,
then all is well.

ADAPTING HYPERREZ TO RUN ON
DIFFERENT PLATFORMS

Hyper Rez, originally a DOS-based program, has recently been
adapted by Earl Fogel of the Computing Services Department at the
University of Saskatchewan to run in a UNIX environment. 7 He wrote
a shell that not only allows for the retention of the original hypertext
links, but also automatically makes a connection to a remote site. This
has allowed users without DOS machines to take full advantage of the
information available in HYTELNET. Internet users are free to
download both the DOS and UNIX versions of HYTELNET. There
is also a version of HYTELNET that runs under Windows, called
CATALIST, 8 designed by Richard Duggan of the University of Delaware.

Perhaps the best example of the UNIX version of HyperRez can
be seen at the University of Western Australia. Deidre Stanton, at Murdoch
University Library, Australia, along with some of her colleagues, is in
the process of designing an Information Resources Access System for
the Australian Academic Research Network (Figure 7).

Note the START.TXT file in Figure 7. Notice the traditional HELP
and WHATIS files. Linking to the file IRASINDEX brings up an index
of available bibliographies, guides, and directories, plus links to the
actual files (Figure 8).

Let's return to the START.TXT file. From this menu, a user can
actually run other programs connected to the UNIX machine. These
programs include archie, the Gopher, NetLib, HYTELNET, and SWAIS.
Archie allows a user to query the anonymous FTP archie server in
Australia, to discover if certain files or programs are available for
downloading. The Gopher 9 is a distributed document delivery service

172

PETER SCOTT

\ Welcome to the Information Resources Access System]

q to QUIT

Key-stroke commands
What is the Information Resources Access System?
Academic Discussion List Directory
Anonymous ftp Database Information
Distributed Document Delivery Service
Information Resources Access System Index
Netlib Library Catalogue Access
Telnet Site Directory
Wide Area Information Servers

Figure 7. IRAS START.TXT

! INFORMATION RESOURCES ACCESS SYSTEM

INDEX !

Key-stroke commands

<HELP>

Bibliographies on Internet Use and Resources

Directories to Network Services

Directories to

Networked Information Resources

Electronic Journals

Guides to Network Services

Guides to

Networked Information Resources

Reports and

Surveys Related to Use of AARNet

Software

Related to Information Resources

Figure 8. IRASINDEX

that allows a user to access various types of data residing on multiple
hosts in a hypertext-like interface. The NetLib program provides a user
interface to library catalogs around the world using the LIBS. COM
software from Sonoma State University. The SWAIS is a collection of
programs that provide access to information distributed over wide area
networks.

HYPERTEXT 173

It is obvious, then, that editing, adding, and deleting information
in a utility based on HyperRez is a fairly straightforward procedure.
In fact, a developer can create many different and useful utilities with
the software.

WHAT LIBRARIANS CAN DO WITH HYPERTEXT

Each year libraries produce thousands of paper documents printed
in various sizes and colors. A typical piece of paper might be a listing
of library-owned information resources of interest to a social studies
student. The library dutifully makes 200 copies on pink paper, whereas
the list for philosophy students is printed on green paper. Shelf space
is made available to house these papers in the hope that the students
might be interested enough to pick them up. Other pieces of paper
are produced for other purposes and scattered around the library system.
How disastrous, though, if there are typographical errors that need
correction. After going to the enormous trouble of compiling the
information, buying the paper, waiting for the photocopier to warm
up, and hand correcting the errors, one might feel that there is a better
way.

Instead, would it not be more sensible to make the information
available on a disk? Not just a disk containing a series of unrelated
files, but, in fact, a hypertext index of all information regarding all
aspects of the library's services. This information could be designed
in such a way as to allow a reader of the utility to gain information
on library hours, special collections, expert staff, library rules and
policies, maps of the stack areas, and so on. The original and fully
updated hypertext files could be sitting on a personal computer in the
reference department. That personal computer itself could be made
available for library users to browse the most up-to-date information
relevant to their needs. The "new" students are computer literate, and
many, indeed, can be seen wandering the halls of academe clutching
their laptops. They are the last people who want to be given a handful
of colored paper.

Librarians could also begin the process of producing customized
hypertext information packages on demand. Not only would this be
a valuable exercise in its own right, but it would also allow the librarians
to gain a renewed credibility in the eyes of their patrons. They would
be seen as disseminators of information rather than its wardens.

There is no reason why librarians should not take an active role
in the design and implementation of a campuswide information system
based on the principles of hypertext indexing. Apart from the usual
opening-hours statement, librarians could also add interesting

174 PETER SCOTT

bibliographies, electronic journals, important conference an-
nouncements, and other useful Internet resource packages. Once the
information has become outdated, it is just a matter of deleting the
appropriate file.

OTHER HYPERTEXT FOUND ON THE INTERNET

When browsing around the Internet, one can find many hypertext
utilities designed by both computer specialists and librarians. One such
is HYCLASS, developed by Clifford Urr, Director of Information and
Library Services, James Martin Associates. HYCLASS is designed to
be run under HyperRez. Its subject matter aids librarians who need
to access Library of Congress classifications assigned to computer- and
software-related materials. A cataloger may instantly access the utility
by depressing the hot-key and browse the files to find the most
appropriate classification number. Once identified, the number is noted,
the program returned to memory, and the number is then entered into
the cataloging software. This process can save the cataloger a great
deal of time, and it is more likely that correct information is found
by this method than by wading through the paper indexes. Of course,
this program could be expanded to include all Library of Congress
classifications.

Librarians in acquisitions departments could also develop and make
use of hypertext indexing. Take for instance the vendor file. This is
usually a computer-produced paper index, sorted alphabetically by
vendor name. This is useful if only a name is being searched, but what
if someone wanted to find a particular vendor specializing in foreign
language material or perhaps a vendor situated in a particular country?
Clearly hypertext is the solution.

Reference was made earlier to the Internet utilities Gopher and
SWAIS. These are both hypertext-like resources that allow a browser
to discover useful information in a hierarchical structure. The Gopher
opens with a main menu, rather like the opening screen of a HyperRez
utility. The user selects a line number to start the process of deep jumps.
One such jump would lead to a submenu containing the names of other
accessible Gophers. Selecting a number would connect the user to that
site almost immediately, even if it were a continent away. The menu
process begins again with the new site. The Gopher allows links not
only to remote sites but also to fully searchable databases, such as
electronic versions of public domain books, newspapers, and
information files concerning many topics. The current version will also
send a retrieved file to the searcher's electronic mail address, wherever
it happens to be.

HYPERTEXT 175

SWAIS, the Simple Wide Area Information Server, also has hypertext
qualities. After logging in, a user is presented with a menu of over
120 databases situated on sites worldwide. Any number of these databases
may be selected by merely hitting the space bar. A keyword or phrase
is entered, and the argument is run against the selected files. The result
is returned in a menu structure, allowing the user to read each individual
file. The SWAIS will also mail selected files to a user's electronic mail
address. Billy Barren, 10 at the University of North Texas, has loaded
the information contained in HYTELNET to a server at his site. A
SWAIS searcher may select that file and issue a keyword argument.
The server will return all the files containing that term. This example
shows how one hypertext utility can interact with another.

These two programs can be run on many different types of computer
platforms, and developers are free to add any type of information that
they feel is useful to the community.

This paper has been stressing the uses to which HyperRez can be
put. MaxThink has also produced a stablemate called Hyplus, 11 which
is based on a similar philosophy of file linking but which contains
some extra features. It is not memory-resident, running instead in a
stand-alone mode. It can search for a term with a glossary utility and
can also run another program simultaneously. It has the ability to dial
a telephone number and allow a user to see a list of all previous jumps
a useful feature if an immediate jump to a particular file is required.

Whichever of the two programs a developer chooses to create
utilities, one thing is for sure. He or she will gain a greater understanding
of the subject matter being organized. This also holds true for the user.
You are invited to test this statement by obtaining a copy of the software
packages described above and by creating your own utilities. (More
information on the software packages is given in the Appendix.)

176

PETER SCOTT

APPENDIX

Hypertext Software and Utilities Available from

wuarchive.wustl.edu
(Directory PD1:<MSDOS.HYPERTEXT>)*

Filename

Type** Length

Date

Description

ARJHLP23.ZIP

327172

920214

HyperText helpfile for ARJ v2.3.

Can be TSR

CPPTOUR.ARC

72187

891208

Hypertext-based tutorial on using

C++

DOSEA5.ZIP

128906

910831

Hypertext program explaining

DOS 5.0.

EBK.ZIP

48545

911005

Topic-oriented hypertext viewing

program

EBKSRC.ZIP

56013

911005

Borland C++ src for EBK.ZIP

hypertext viewer

HS25.ARC

301370

890717

HyperShell hypertext browser,

v2.5

HTEX06.ZIP

246184

920206

Hypertext authoring system,

create/con ver t/edi t

HYCLASS.ZIP

69120

910414

Hypertext LC class & computer-

related material

HYDOS10.ZIP

86977

910615

Hypertext browser for all DOS

commands, vl.O

HYENVOY1.ZIP

64779

910622

Hypertext browser for all

ENVOY 100 commands

HYPER.ARC

171621

880503

Construct your own hypertext

network

HYPERH15.ZIP

63851

900428

Hypercard-like help util. Make

your own stacks

HYPERREZ.ARC

113258

891021

Hypertext reader using plain

ASCII files

HYPLUS.ZIP

155030

910628

Stand-alone Hypertext program

from Maxthink

HYPRVX11.ZIP

75776

901211

Hypertext browser for VAX/VMS

mail help, vl.l

HYTELN50.ZIP

348957

920123

Hypertext browser for finding

TELNET addresses

MAGIC15A.ARC

208998

891126

BlackMagic hypertext reader/

develop, sys. Iof3

MAGIC15B.ARC

217234

891126

BlackMagic hypertext reader/

develop, sys. 2of3

MAGIC15C.ARC

328463

891126

BlackMagic hypertext reader/

develop, sys. 3of3

NUHLP45.ZIP

90642

910628

Norton Utilities 4.5 Hypertext

help browser

SENSES.ZIP

108544

910414

Demonstration of HYPLUS

hypertext compiler

TP55TOUR.ARC

63504

891208

Hypertext-based tutorial on using

OOP in TP5.5

This list was created on Saturday, 28 March 1992 20:16:21 MST.
Some files may have been added or deleted since that date.

See file PD1:<MSDOS.FILEDOCS>AAAREAD.ME for additional information.
Type B is Binary; Type A is ASCII.

HYPERTEXT 177

NOTES

'Larson, Neil. (1989). HyperRez. Berkeley, CA: MaxThink. Available via FTP from
wuarchive.wustl.edu in the mirrors/msdos/hypertext subdirectory as hyperrez.arc.
Change file type to "binary" when fetching.

2 Scott, Peter <scott@sklib.usask.ca>. (1992). HYTELNET. Saskatoon, SK: The
author. Available via FTP from wuarchive.wustl.edu in the mirrors/msdos/hypertext
subdirectory as hyteln50.zip. Change file type to "binary" when fetching.

s Scott, Peter <scott@sklib.usask.ca>. (1991). HYDOS. Saskatoon, SK: The author.
Available via FTP from wuarchive.wustl.edu in the mirrors/msdos/hypertext
subdirectory as hydosl0.zip. Change file type to "binary" when fetching.

'Scott, Peter <scott@sklib.usask.ca>. (1990). HYPERVAX. Saskatoon, SK: The
author. Available via FTP from wuarchive.wustl.edu in the mirrors/msdos/hypertext
subdirectory as hyprvx.zip. Change file type to "binary" when fetching.

5 Scott, Peter <scott@sklib.usask.ca.> (1991). HYENVOY. Saskatoon, SK: The
author. Available via FTP from wuarchive.wustl.edu in the mirrors/msdos/hypertext
subdirectory as hyenvoyl.zip. Change file type to "binary" when fetching.

6 LIB HYTELNET. (1990). To subscribe send an e-mail message to
s< ott@sklib.usask.ca with the subject "Add me to LIB HYTELNET."

7 Fogel, Earl <fogel@skyfox.usask.ca.> (1992). Direct all enquiries
fogel@skyfox.usask.ca.

8 Duggan, Richard <duggan@brahms.udel.edu>. (1990). CATALIST. Newark, DE:
University of Delaware. Available via FTP from ftp.unt.edu in the library subdirectory.

internet Gopher. (1992). Telnet to consultant.micro.umn.edu and log in with
gopher. Emulate vtlOO. Use menus to find "libraries" entry.

10 Barron, Billy <billy@vaxb.acs.unt.edu>. (1992). Telnet to quake.think.com and
log in with wais. Find the entry for the "hytelnet" server, hit the space bar to mark,
select keyword to search, and hit enter. The file(s) containing your keyword will be
retrieved.

"Larson, Neil. (1989). Hyplus. Berkeley, CA: MaxThink. Available via FTP from
wuarchive.wustl.edu in the mirrors/msdos/hypertext subdirectory as hyplus.zip. Change
file type to "binary" when fetching.

KATHARINA KLEMPERER

Director of Library Automation

Dartmouth College
Hanover, New Hampshire

Delivering a Variety of Information
in a Networked Environment

ABSTRACT

The volume and variety of electronic information resources, the increase
in desktop computing power, and the pervasiveness of networks have
combined to make access to information fundamentally different from
that of a decade ago. This paper describes the nature of information
resources that libraries are dealing with now and discusses the different
needs of each with regard to access and delivery.

INTRODUCTION

Information science has undergone a fundamental change during
the past 10 or 15 years. While libraries still provide the same service
that they always have access to information the tools and skills are
entirely different from those that were taught in library schools a decade
ago. The volume and variety of electronic information resources, the
increase in desktop computing power, and the pervasiveness of networks
have combined to challenge the resources of information providers. This
paper will describe the nature of information resources that libraries
are dealing with now and will discuss the different needs of each with
regard to access and delivery.

178

ELECTRONIC INFORMATION DELIVERY 179

The different kinds of electronic information that we have available
for delivery today can be divided into rough categories:

text, including

indexes

structured full text
unstructured full text

numeric

multimedia, including

images

full-motion video
sound

Each of these has different needs in terms of access and delivery.

TEXT

Indexes

The kind of information that libraries have dealt with for years,
and which they have handled with great success, is indexes to larger
bodies of information. Among these we find card catalogs (and their
online cousins), indexes to the journal literature, and catalogs of objects
such as museum artifacts. The characteristics of this kind of textual
information follow:

It divides neatly into "records," all of which include roughly the
same fields.

It is highly structured, that is, each record is composed of distinct
and identifiable fields such as authors and ID numbers.

The records are of similar size.

A whole generation of systems grew up to support online catalogs,
and because other types of indexes are so similar, it was easy to force
them into systems that were designed to handle online catalog records,
usually in MARC format. Each of these indexes has a reasonably small
number of access points that can be indexed and reasonably short fields
that can be displayed and comprehended easily.

Compare the following two examples of data structures from the
Dartmouth College Library Online System. The first is from the online
catalog, the second from the locally mounted MEDLINE database.

180

KATHARINA KLEMPERER

Author: Symposium on Immunology of Milk and the Neonate

(1990 : Miami, Fla.)
Title: Immunology of milk and the neonate / edited by Jiri

Mestecky, Claudia Blair, and Pearay L. Ogra.
Imprint: New York : Plenum Press, c!991.

Series: Advances in experimental medicine and biology ; v. 310.
Location: Dana RJ/216/S945/1990

Author(s): Wilson NW, Self TW, Hamburger RN

Title: Severe cow's milk induced colitis in an exclusively breast-
fed neonate. Case report and clinical review of cow's milk
allergy.

Source: Clinical pediatrics 1990 Feb;29(2):77-80.
NLM ID: 90150935
Location: Health Sciences Serial

Structured Full Text

There is a conceptual difference between index-type databases and
full-text databases. Indexes represent a complete document, whether it
be a book, a phonograph record, or a museum object. A full-text file
is the document; once you have retrieved it you need look no further.
Among full-text formats, we find ourselves on a continuum. At one
end are highly structured data files that are in fact full text but that
can be forced into the traditional online catalog database structure
without too much effort. It may not consist of bibliographic data, but
once new field names have been defined, search and retrieval can proceed
basically as if one were retrieving catalog records. An example of highly
structured full text is a dictionary entry. The following example is from
the American Heritage Electronic Dictionary, as mounted at Dartmouth
College Library:

Word:

Part of Speech:
Inflected Form:
Part of Speech:

Sense:

kin*dle(l)

verb

-died, -dling, -dies.

transitive verb

1. a. To build or fuel (a fire).

1. b. To set fire to; ignite.

2. To cause to glow; light up, as in: The sunset kindled

the skies.

3. a. To inflame; make ardent.

3. b. To arouse; inspire, as in: "No spark had yet
kindled in him an intellectual passion"(George
Eliot).

ELECTRONIC INFORMATION DELIVER? 181

Part of Speech: intransitive verb

Sense: 1. To catch fire; burst into flame.

2. To become bright; glow.

3. a. To become inflamed.
3. b. To be stirred up; rise.

Etymology: Middle English kindelen < Old Norse kynda.
Derivative: kin'dler
Part of Speech: noun

These data, although in fact full text, display the characteristics
of index data proposed above: they divide into records, they are highly
structured, and the records are of similar size. Consequently, it was
a fairly easy task to load the data into the same database manager that
was used for the index-type databases. Notice however the repeating
groups of fields (Part of Speech and Sense). This feature of the data
is not usually found in index-type databases.

When the entire full text is indexed, plenty of new ways to access
information present themselves. For example, in the American Heritage
Electronic Dictionary:

Find all the words of Norwegian derivation in the English language
(sample results: floe, iceberg, fiord, ski, slalom, telemark, lemming,
troll).

Find all the verbs that have to do with "fire" (sample results: anneal,
barbecue, beacon, blaze, burn, crackle, discharge, douse, ignite).

Find all the six-letter words that end in "ism" (sample results: ageism,
cubism, Nazism, nudism, sadism, [Uncle] Tomism).

Unstructured Full Text

At the other end of the text continuum are complete texts of
literature. These texts are minimally structured. Most have sentences,
paragraphs, and chapters, but the information is structured not as a
collection of records but as an ordered string of words.

This changes the methods of access and display significantly. In
record-oriented data, one searches for a known feature (usually a keyword
located in a specific field) in the entire database, and the goal is to
locate all the records containing this feature. Boolean searching means
that two or more features will be found in the same record. Thus, a
search for the author word Hemingway and the title word sun will
retrieve a number of catalog records (mostly Hemingway's The Sun
Also Rises; see Figure 1). Nobody really cares which catalog records
happen to precede or follow these retrieved records; the database records
are basically unrelated to each other except perhaps alphabetically.
Display of retrieved records is straightforward: one provides a short,

182

KATHARINA KLEMPERER

medium, or long display of individual records, giving various depths
of detail to allow users to scan the results and then look more closely
at one or more retrieved records.

Full-text files have a completely different set of needs. Since full
texts consist of a series of ordered words rather than unrelated records,
it can be argued that a literary text is really nothing more than a long
string of characters. Where "records" can be identified, they are of
varying length and tend to consist of multiply occurring "fields," also
of varying length. Fields may also be interleaved and must be displayed
in their original order (e.g., chapter heading, subheading, multiple
paragraphs containing multiple sentences; see Figure 2).

What we are searching for here is not so much records (e.g.,
paragraphs), but matchpoints. If I am searching for the word "rabbit"

Each dot represents a record in the database

Records
containing
"Hemingway" in
an author field

Records

containing "sun"
in a title field

Records containing both

"Hemingway" in an author field

and "sun" in a title field.

Figure 1. Boolean search of a record-oriented database

III

text matchpoints

Ch. 1

t t

Ch. 2 Ch. 3

Ch.4

chapter markers

The line represents the database, which is a string of characters
Figure 2. "Fields" in an unstructured full-text database

ELECTRONIC INFORMATION DELIVERY 183

in the text of Alice in Wonderland, I don't want to find all the paragraphs
or lines that contain the word, I want to locate the occurrences of the
word itself and scan the text preceding and following the matchpoints
I have found. Rather than Boolean combinations (find every "record"
containing rabbit and Alice) full texts are better served by proximity
operators, which locate occurrences within a certain distance of each
other (find every occurrence of rabbit within five words of Alice; find
every occurrence of Alice preceding rabbit by no more than 100
characters). Displays likewise have different requirements; rather than
seeing a list of individual chapters or paragraphs, in which the target
words might appear only after many lines, the user is better served
by a display of matchpoints in context. The user needs to be able to
see all the matchpoints in context at a glance, jump from one matchpoint
to the next, and scroll forward and backward through the text from
any given matchpoint. A further display requirement is to provide the
ability to go to the beginning of the "segment" (e.g., paragraph, chapter,
poem) for each matchpoint. An example will illustrate the initial display
of matchpoints:

sorted by matchpoint:

18 suddenly a White Rabbit with pink eyes ran close by her.

22 of the way to hear the Rabbit say to itself, 'Oh dear! Oh dear!

25 natural); but when the Rabbit actually TOOK A WATCH OUT

28 had never before seen a rabbit with either a waistcoat-pocket, or

95 and the White Rabbit was still in sight, hurrying down

99 to corner, but the Rabbit was no longer to be seen:

07 CHAPTER 1 Down the Rabbit-Hole Alice was beginning to get

30 it pop down a large rabbit-hole under the hedge. In another

35 to get out again. The rabbit-hole went straight on like a tunnel

sorted by appearance in the document:

07 CHAPTER 1 Down the Rabbit-Hole Alice was beginning to get

18 suddenly a White Rabbit with pink eyes ran close by her.

22 of the way to hear the Rabbit say to itself, 'Oh dear! Oh dear!

25 natural); but when the Rabbit actually TOOK A WATCH OUT

28 had never before seen a rabbit with either a waistcoat-pocket, or

30 it pop down a large rabbit-hole under the hedge. In another

35 to get out again. The rabbit-hole went straight on like a tunnel

95 and the White Rabbit was still in sight, hurrying down

99 to corner, but the Rabbit was no longer to be seen: she

The surprise here is that this kind of display has long been with
us, known as a KWIC display (Key Word In Context). Its usefulness
has not been lost.

184 KATHARINA KLEMPERER

The display above gives line numbers for each match; if the text
were not broken into lines, the matches could just as easily be numbered
sequentially. At this point, the user wants to zero in on one match
and perhaps display a certain number of words around it. For example,
displaying 100 words before and after the match at line 25 would result
in the following:

sleepy and stupid), whether the pleasure of making a daisy-
chain would be worth the trouble of getting up and picking
the daisies, when suddenly a White Rabbit with pink eyes
ran close by her.

There was nothing so VERY remarkable in that; nor did Alice
think it so VERY much out of the way to hear the Rabbit
say to itself, 'Oh dear! Oh dear! I shall be late!' (when she
thought it over afterwards, it occurred to her that she ought
to have wondered at this, but at the time it all seemed quite
natural); but when the Rabbit actually TOOK A WATCH
OUT OF ITS WAISTCOAT-POCKET, and looked at it, and
then hurried on, Alice started to her feet, for it flashed across
her mind that she had never before seen a rabbit with either
a waistcoat-pocket, or a watch to take out of it, and burning
with curiosity, she ran across the field after it, and fortunately
was just in time to see it pop down a large rabbit-hole under
the hedge.

In another moment down went Alice after it, never once
considering how in the world she was to get out again.

At this point, the user might want to keep scrolling forward or
might want to go back to the beginning of the chapter in which this
match was found. The whole procedure is more one of navigating around
a text rather than looking at records in a file.

Between the strictly record-oriented indexes and the completely
unstructured full texts, there is a wide variety of full texts that can
be treated either as record oriented or as full-text files, for example,
the Bible, collections of poems, plays, encyclopedias.

NUMERIC DATA

Beyond text files, there are completely different data formats that
are now available electronically, each requiring different methods of
access and display.

A completely different kind of data that has recently received a
lot of attention, thanks to the U.S. government's decision to distribute

ELECTRONIC INFORMATION DELIVERY 185

its census data on CD-ROM, and consequently in great quantity, is
numeric data. Numeric data of course require an entirely different sort
of database management system and user interface. The ideal here is
not simply to provide access to tables as if they were pages out of printed
volume, but to provide access to the raw data that can then be
manipulated statistically. Rather than searching for occurrences of terms
in records or full text, one would like to select a subset of the universe
of data and then perform statistical tabulations on it and produce visually
pleasing displays. For example, using the U.S. census as an example,
select the universe to be all households living in towns with a population
less than 5,000 in the state of Vermont. Then perform statistics on
household annual income: mean, median, standard deviation, frequency
by $10,000 increments. Produce a bar graph to illustrate these results.
Now perform the same operations using the analogous universe in the
states of New York, California, and Mississippi.

A living example of such statistical manipulations is Dartmouth's
SPSS server, which in this simple and somewhat trivial case is showing
frequencies of occurrence of signs of the Zodiac for birthdays of
individuals in a specific population (ICPSR 1991 General Social Survey).
The chart is produced in real time, from variables selected by the user
(Figure 3).

MULTIMEDIA

Of course the newest area of exploration is that of aural and visual
media and the combinations of all media into what is known as
multimedia. The emphasis in audiovisual media has been mostly on
the delivery of the "documents"; no small problem in itself, but access
to the contents is still largely text based.

For example, in a database of musical sound recordings, one would
obviously need to access recordings by name, e.g., Beethoven's Fifth
Symphony conducted by von Karajan with the Berlin Philharmonic
Orchestra. Such indexing is nothing new; the main problem is delivering
the sounds over a network and playing them on a workstation with
reasonable fidelity. One would like to be able to retrieve songs by indexing
the music itself; a user should be able to sing a melody or play it on
a keyboard, or enter a harmonic progression and then retrieve the
citations for the pieces that match it, and then hear the matching music.
This is not an unknown concept; "thematic indexes" that organize
musical themes by their melodic intervals have existed in paper for
a long time. Indexing of images is even more complicated, since the
actual shapes must somehow be encoded into structures that can be
referenced.

186

KATHARINA KLEMPERER

ARIES **************************** 112

TAURUS ******************************* 124

GEMINI ********************************* 131

CANCER ****************************** 120

LEO ****************************** 120

VIRGO ********************************** 137

LIBRA ***************************** H6

SCORPIO ******************************* 122

SAGITTARIUS ********************************* 131

CAPRICORN ********************************* 130

AQUARIUS ********************************** 135

PISCES ****************************** 121

+ + + + + +

40 80 120 160 200
Frequency

Mean

Valid cases

6.577

1499

Missing cases 18

Figure 3. Chart produced from user-selected variables

At this point, the main effort in the multimedia area has been
in the use of hypermedia as a presentation mechanism. Here still, the
emphasis is on the presentation and delivery of media, with access
following a kind of stream-of-consciousness model. Any actual indexing
that is done is still textual.

CONCLUSION

The important point to remember, with the variety of information
that can now be delivered to the desktops of users anywhere in the
world, is that each has different needs regarding access and delivery.
New database engines are needed to provide access to these data resources,
and new delivery mechanisms will display them. The challenge is to
develop the instruments that will accomplish this.

ARLENE MOORE SIEVERS

Head, Information Acquisition Department

University Library

Case Western Reserve University

Cleveland, Ohio

Free-Net in Cleveland and

Case Western Reserve University Library:

Linking Community and University

ABSTRACT

Free-Net Cleveland is a completely free, open-access community computer
system operated and coordinated by Case Western Reserve University
in Cleveland, Ohio. Through Free-Net, users can access a wide range
of electronic services and features including free electronic mail and
the Internet as well as valuable university information sources such
as the university's online catalog, government documents, and links
to faculty members in academic areas. Free-Net also functions as a
communication medium for the university, providing opportunities for
enhanced student/faculty interaction, and it functions as an educational
resource for schoolchildren through its Academy One component. Free-
Net relies heavily on volunteers from the community to operate it and
keep it current. Free-Net Cleveland is part of the National Public
Telecomputing Network (NPTN), which is being created to provide
networked services and links between Free-Nets throughout North
America and the world.

INTRODUCTION

Free-Net is something in which perspective determines what one
knows about it, how one uses it, and most definitely how one perceives

187

188 ARLENE MOORE S1EVERS

its impact. This becomes clear when seeking out diverse perspectives
on something that people in Cleveland, and particularly at Case Western
Reserve University, use in different ways, misunderstand, and yet take
for granted, like any other utility. Cleveland holds the distinction of
being the place where Free-Net began and is its first community site.

The idea of a free community computing network was conceived
of and developed by Dr. T. M. Grundner at Case Western Reserve Uni-
versity in Cleveland. Dr. Grundner has subsequently helped numerous
communities establish their own Free-Nets. He also gave freely of his
time and knowledge in providing the essential background for this paper.
His insights into the philosophy behind the system, how it has
developed, just how one sets up a Free-Net, and his perspective into
the future via the National Public Telecomputing Network (NPTN),
which he is launching, were especially valuable.

Assistance was also provided by one of the few paid employees
who actually runs Free-Net at Case Western Reserve University, and
who knows a great deal about its inner workings, Martha Artzberger,
of Information Network Services. She provided information on campus
use of Free-Net, in academic departments by students and faculty, in
university administration, and its important relationship to the Case
Western Reserve University campuswide computer network,
CWRUNET.

A third major source was George Barnum, University Library's
government documents librarian, who is an experienced sysop (systems
operator) and who has worked on the government documents part of
Free- Net for some years.

Written documentation on Free-Net is virtually nonexistent. For
this reason, it was necessary to rely on information gained through
interviewing Dr. Grundner and others. Written documentation consists
of a description of Free-Net prepared for the National Research and
Education Network (NREN) hearings in Congress and fact sheets. One
of the pitfalls of living in the age of the shift from print to digitized
information is the absence of clear archiving responsibilities and the
lack of a paper trail. One learns this firsthand preparing a paper about
an electronic information source.

WHAT IS FREE-NET?

Free-Net is a completely free, open-access community computer
system operated and coordinated by Case Western Reserve University.
The Free-Net computer allows anyone to call in 24 hours a day and
access a wide range of electronic services and features. These range from
free electronic mail to information in various areas of the "electronic

FREE-NET 189

city." It is largely dependent on volunteers for the development of
"areas," for inputting of information, answering questions, and creating
what Free-Net will be in any community. It draws on a wide range
of subject expertise in many areas as diverse as veterinary science, space
science, a legal service, recipe exchange, and restaurant menus and
reviews. It provides access for the community to the Internet through
the teleport component and to numerous remote information sources
through the library area. It provides numerous forums for users to
communicate, ask questions, and air views (Grundner, 1990, p. 51).

Having been in operation since 1986, it has experienced numerous
growing pains. Probably the greatest complaint anyone has about the
network currently relates to the inadequate number of telephone lines,
which sometimes creates difficulty connecting for those accessing via
modems. For most university users, this problem is virtually nonexistent
because the entire campus is connected to it fiber optically through
the university's campuswide network, CWRUNET.

Perhaps the most important concept behind Free-Net is that of
opening up computing to the widest possible user group. Personal access
to online computer services for those not part of a university, a large
corporation, or the government has been limited to those with the
resources to afford the luxury of a Prodigy or CompuServe service (T.
M. Grundner, personal communication, January 27, 1992). Free-Net
allows access to a multitude of valuable services requiring only a terminal
and modem access. New user registration is even accomplished online.
There is no bureaucracy, forms to fill, or qualifications to meet.

Dr. Grundner sees the development of Free-Net and the Free-Net
movement as similar in nature to the free public library concept that
took hold in the United States in the middle to late 19th century (T.
M. Grundner, personal communication, January 27, 1992). A certain
amount of idealism is behind the concept of this populist computing
network. In a time when access to information appears to be heading
more and more to a privatized system, with perhaps the creation of
new classes of "information wealthy" and "information poor," this
concept is a breath of fresh air to those who still hold fast to the ideals
behind the free public library and believe a democratic society is based
on the principle of equality of access to information.

When asked what his biggest surprise has been in the years since
Free-Net was developed and has been running, Dr. Grundner replied,
"the use of Free-Net by working-class people 'from the neighborhoods'
and their children" (T. M. Grundner, personal communication, January
27, 1992). His interpretation of this phenomenon is that there have
been great numbers of computers bought by families in the hopes of
advancing their children's education. Many of them remain relatively
underused by the families who purchased them. Free-Net, which is

190 A RLENE MOORE SIE VERS

heavily promoted in the schools in the Cleveland area, provides
something interesting and useful to do with a family's personal computer
at no charge (T. M. Grundner, personal communication, January 27,
1992). In the process, computer novices gain confidence and knowledge
using an online computer database and an electronic mail system. The
menus, commands, and functions are designed to be as simple and
transparent as possible, and the organizational structure, developed
around the "electronic city" concept, is an effective and readily
understandable principle.

The community element and the idea of volunteerism are extremely
important to the Free-Net concept. According to Dr. Grundner, there
has been no lack of committed volunteers who create the Free-Net
components. Indeed, the idea came directly from Dr. Grundner's positive
experience with expert volunteers in another project. Free-Net started
from a project Dr. Grundner created when working with the Medical
School of Case Western Reserve University. This was a microcomputer-
based health information system dependent on volunteer effort and the
expertise of the doctors, many of whom were primarily researchers
affiliated with university hospitals. Dr. Grundner was surprised by their
willingness to contribute to the system, to input explanatory information
in the system, and their interest in establishing one-to-one contact with
the people who were asking the questions. The idea of forming a
community-based computer network, based primarily on the efforts of
volunteers but established at a university, came from this experience
(T. M. Grundner, personal communication, January 27, 1992).

CWRUNET, FREE-NET, AND THE UNIVERSITY

Members of the university community are most in contact with
Free-Net because it serves as the university electronic mail system. Many
are only aware of Free-Net as a system rapidly outgrowing its original
capacity. Since the vast number of community users connect via dial-
up modem, and since the system is extremely busy during peak business
hours, it has, at times, been very difficult to get a connection. When
university buildings were fiber optically wired for CWRUNET, of which
Free-Net is a component, this problem ended for most. University users
complain when Free-Net is down for maintenance on Friday mornings
and on the occasional instance when it is not operational, but the truth
is that the electronic mail component has caught on and flourished
on campus where previous efforts in the library and university failed.
This is due to the simplicity of the Free-Net electronic mail component
and the recent almost complete campus access to CWRUNET
connections.

FREE-NET 191

In many ways, for users in the university and University Library,
Free-Net is something of an afterthought. Perhaps explaining that Free-
Net is just one information node of CWRUNET, the campuswide system
network, and that it provides the conduit for electronic mail gives an
indication why. Case Western Reserve holds the distinction of being the
first campus completely fiber optically wired for its system. The system
can handle not only extremely large amounts of data very quickly but
audio, video, and imaging transmissions as well. Cable TV, among
numerous other services such as networked CD-ROMs, shared software,
and data files, are offered to students in their dormitory rooms via
CWRUNET. An exciting electronic imaging and multimedia project,
which Case Western Reserve University is exploring with IBM, is intended
to offer enhanced materials in music and medical science networked
on CWRUNET. Free-Net is nonetheless an important community and
extra-university node of the system, so to speak, and conversely
CWRUNET supplies the Internet connection for Free-Net users.

FREE-NET, STUDENTS, AND FACULTY

These same academic components that connect community users
to university experts function for class members and instructors as a
structured, expansive electronic mail system, a question-and-answer
forum, and a problem-solving medium. Many university academic
departments have Free-Net sections for each class, where not only
students on campus, who have access to CWRUNET, but those living
elsewhere in the community, out of town, or home for vacation can
communicate as well. Expansive course descriptions are given online
and are of great use to students planning their class schedules. It is
interesting that humanities areas such as English and art history, not
only engineering and science departments, use Free-Net options to
enhance student/faculty communication possibilities. Martha
Artzberger confirmed that the Free-Net communication mode is
acknowledged by instructors to be helpful in encouraging shy students
or those with imperfect conversational English to participate in class
discussions on a more comfortable basis and at leisure (M. Artzberger,
personal communication, January 28, 1992).

This paper began with the idea that perspective is significant when
considering Free-Net. Those who work with it in the academic area
see it as an Internet connection, an electronic mail medium, and a way
to provide enhanced access for learning between instructors and students.
It is a convenient campus-access mode for students who live off campus
and want to be part of the learning that goes on between students getting
together, going over assignments, and exchanging views outside of class.

192 ARLENE MOORE SIEVERS

THE UNIVERSITY, FREE-NET, AND THE PUBLIC

The role of the university and the University Library in Free-Net
is important, and it truly is a two-way street. Obviously, the community
has benefited from the university research that created Free-Net and
from the administration's commitment that continues to support its
operation and enhancement. Actually, Cleveland Free-Net, and all Free-
Nets, operate on a minimum of paid labor. The equivalent of only
two and one-half full-time employees are needed for a Free-Net the
size of Cleveland's. Smaller communities operate with only one (T. M.
Grundner, personal communication, January 27, 1992).

The other major university contribution is the expertise of faculty.
There is a Case Western Reserve University component of Free-Net in
the University Circle segment of the electronic city menu that has an
academic section where departmental information, student rosters, office
hours, and course information are posted. Many of the departments
have a general question-and-answer forum not limited to students.

Some academic sections are active, others rarely used, but the
interaction between Free-Net users and faculty in some areas, like the
sciences, is steady and obviously rewarding. These university
components give the public easy access to experts in specialized fields
and at the research university level. Case Western Reserve University,
with internationally recognized programs in chemistry, physics,
engineering, and medicine, is able to offer some of the expertise of
a first-rate faculty to the public at large.

The public aspect of Free-Net brings its own challenges to the
university. With the current debate about freedom of speech and political
correctness on campuses and the heavy use of Free-Net by all age groups
in the community as well as by university undergraduates, the free-
spirited adult exchanges on some of the Free-Net chat groups became
something of a problem. As a community system, Free-Net is like any
public space and is subject to a degree of misuse. The solution has been
the establishment of some obscenity standards for most of Free-Net and
the establishment of "Adult only 18 or older" areas for those who wish
to exercise complete freedom of expression in Free-Net communications
(T. M. Grundner, personal communication, January 27, 1992).

It is interesting that Case Western Reserve University, which no
longer has an education degree program nor a library school, sad to
say, does a great deal for schools and the K-12 age group through Free-
Net. Dr. Grundner, who has a doctorate in education, is especially involved
in this rapidly expanding segment of Free-Net called Academy One.

FREE-NET, CHILDREN, AND SCHOOLS

One of the best uses of Free-Net has been in its K-12 applications
in Cleveland and a number of communities worldwide. To participate,

FREE-NET 193

schools must establish Internet access through a sponsoring university
and must supply their own terminals and modems. All other elements
of participation involved are free. What participating classes and schools
get from Free-Net is unlimited access to the full range of Free-Net
possibilities including personal electronic mail, access to distant library
resources, and curriculum enrichment through specially designed
programs and activities that link schools across the country and the
world (National Public Telecomputing Network, 1991a, p. 1).

The Academy One component sponsors a number of projects that
link schools together and provide interesting learning experiences. For
example, there are simulated space launch projects that school children
participate in through Free-Net. These exercises run simultaneously
on all Free-Net systems and work as organized projects to launch a
space shuttle. School sites are given different roles in each mission,
some being landing sites, tracking stations, alternate landing sites, and
weather stations. Some schools become other shuttles, perform docking
maneuvers, and conduct joint experiments such as monitoring solar
disturbances. Coordination and communication between the shuttle's
mission control and schools are conducted through conferences on
NPTN Free-Net systems. Electronic mail is sent back and forth, hourly
reports are posted, and progress is reported. Schools involved in past
shuttle missions included ones throughout the United States, Finland,
Czechoslovakia, and Russia (T. M. Grundner, personal communication,
January 27, 1992).

Another Academy One Free-Net program is a "virtual worlds"
project that allows students to apply science, mathematics, reading, and
telecomputing skills to solve problems they encounter while conducting
expeditions to other worlds. Each expedition is one school day long
and consists of problems to solve, with each school posting mission
reports on Free-Net about what they encounter. Other schools are given
problems to solve relating to these adventures, such as identifying
creatures they meet. These experiences and exercises appear to be popular
and fun for the children participating and do provide practice for
computer skills as well as an early introduction to networking (National
Public Telecomputing Network, 1991 a, p. 3).

The Academy One component of Free-Net has a lot of projects
in preparation and is limited only by the number of telephone lines
available to it. Ten more will be added if a grant to put the ERIC
database online as a Free-Net resource is successful (T. M. Grundner,
personal communication, January 27, 1992).

FREE-NET AND LIBRARIES

Major libraries, such as the Cleveland Public Library, Case Western
Reserve University Library, the Cleveland Area Metropolitan Library

194 ARLENE MOORE SIEVERS

System (CAMLS), and the Special Libraries Association (SLA) have
been actively involved in Free-Net from the very beginning. The
additional library resources Free-Net provides, such as numerous distant
online catalogs, are popular resources. However, it may be that these
efforts have only scratched the surface of what even more expansive
library/Free-Net involvement could accomplish. Dr. Grundner sees a
great deal of unrealized potential in this area, where there would appear
to be an affinity of missions (T. M. Grundner, personal communication,
January 27, 1992).

At University Library, instituting online reference service, making
the online catalog Euclid available, allowing CARL UnCover to be
accessed on a trial basis, and putting together two innovative government
documents projects have been major contributions to Free-Net.
Certainly, however, these are just a few of the things that could be
done and for which the potential exists. Perhaps self-interest rules
because the electronic mail component has been universally embraced
by the library.

Many of the innovative efforts of the University Library have been
in the direction of CWRUNET, our campuswide information system,
doing such things as putting CD-ROMs on a networked jukebox and
making data files accessible. Much of the University Library's attention
has been diverted by preparations to migrate to a new automated library
system as one of the first-to-be-activated OhioLink sites. University
Library is also in the midst of planning, designing, and building a
new "library of the future," which will be heavily committed to the
new information technologies.

The Cleveland Public Library has been very involved with Free-
Net from the start making their card catalog available online to all
users and allowing registered borrowers, and borrowers in allied systems,
to locate materials, have them checked out, and sent to agencies for
pickup. This is an ideal use of Free-Net because the Cleveland Public
Library, while a very rich resource, is far removed from many people
in the metropolitan Cleveland area who might wish to use it.

In the Medina Ohio Free-Net, the public library also plays a very
central role, as it also does in the Peoria Free-Net. This is the ideal
model and should be a very common one in future Free-Nets in smaller
cities. Where the public library plays a leading, active role in the
community, it will also play an active role in Free-Net. Public libraries,
working with universities, are the ideal combinations to initiate Free-
Netting in their communities.

CAMLS is a very active local consortium of 16 academic libraries,
15 public libraries, 12 medical libraries, 12 corporate libraries, and 3
school systems representing about 344 total sites and approximately
10.5 million volumes. CAMLS has taken the lead in providing a reference

FREE-NET 195

bulletin board manned by librarians from participating libraries,
promising a two-day turnaround time in answering questions. Scanning
through recent questions and responses reveals that questions tend to
be answered the same day. There are the usual interesting reference
questions, queries about libraries' borrowing policies, and other
questions that point to library services Free-Net users would like to
see added. Chief among these appear to be an online encyclopedia and
online magazine sources that could be downloaded. Some of these
materials are on the university's CWRUNET, but are generally limited
to university student use because of site licensing restrictions. A few
sources are on Free-Net in full text, and they are heavily used.

Chief among these in the "library" area of the electronic city is
"The Freedom Shrine." This is a section of full-text freedom documents,
such as the Declaration of Independence, U.S. Constitution, and the
Magna Carta. Famous speeches such as Lincoln's Gettysburg Address
and Martin Luther King's "I Have a Dream ..." speech are in full
text, as are patriotic songs and poems. There are many other documents
accessible, and more are being added. All are searchable by keyword
and are immensely popular with schools and reference librarians. Other
online full-text sources include a few books, primarily limited to those
not covered by copyright. These include religious texts such as the Book
of Mormon and the Koran. An explanation of the Gutenberg Project
and how to participate in the project is also online in this section.

Numerous online catalogs are available directly on Free-Net,
including a Data Research Associates version of the Library of Congress
catalog with title and author search capabilities. Other online catalogs
include almost all the major academic collections in the area, including
that of Ohio State University. These are of real assistance to students
home from college trying to do classwork and papers as well as to
unaffiliated users locating resources on their own. A few major area
resources are not online, and these gaps inevitably limit usability.
University of California's MELVYL, Boston University's online catalog,
and a host of others are reachable through the "library" or the "teleport"
area of the electronic city menu, which provides the Internet connection.
Resources of much more limited use on Free-Net are those such as
Dartmouth's Dante Project, a Renaissance literature database.

Librarians also use Free-Net as a way to communicate with each
other locally. SLA has a large, active, far-flung Cleveland chapter and
uses a Free-Net section for member communication, newsletters, meeting
listings, and a popular jobline. CAMLS, the area consortium, also uses
the medium in this way, as does the Ohio Government Documents
Roundtable (GODORT). The Northeast Ohio Major Academic Research
Libraries (NEOMARL) group, which has a number of sections, also
uses Free-Net to communicate. On the campus of Case Western Reserve

196 ARLENE MOORE SILVERS

University, cooperation among libraries is increasing due to joint
participation in the OhioLink project, and Free-Net has served as a
valuable communication mode in this effort.

When reading through many of the Free- Net listings of the card
files and directories of local service agencies, those with public library
backgrounds may be reminded of the information public libraries have
always endeavored to establish and keep current. These were often called
"community resource files." It was always difficult to determine just
which community sources of information might help patrons, and many
of the agencies and people one would list were difficult to reach or
kept strict business hours. Even successful telephone contact was often
awkward to achieve. Many public libraries were always aware that the
answers to many questions were not in the public library but were
scattered in the community. Looking at Free-Net from a public library
perspective, one cannot help but see that important among its many
benefits is the fact it acts as a gateway to many more resources.

FREE-NET AND GOVERNMENT DOCUMENTS

Case Western Reserve University Library has a number of
connections with Free-Net as an information provider. A specific area
that has great potential for use to the public and the community, and
in which two different approaches have been tried to present
information, is government documents. A six-month pilot project
undertaken by Case Western Reserve University Library and carried
out by George Barnum, government documents librarian, illustrates
the strengths, possibilities, and some shortcomings of Free-Net in
practical applications. The CWRU Government Documents Department
was chosen by the Government Printing Office (GPO) and the General
Accounting Office (GAO) to determine the extent of possible use of
the online Department of Commerce Economic Bulletin Board (EBB)
as a free service. The EBB is generally available for a fee only. The
six months' fee waiver was offered to 100 depository libraries. Case
Western Reserve University Library was one of only a very few libraries
to make the information accessible to users in a digitized form, and
the only one to attempt to present it online to the public at large (G.
D. Barnum, personal communication, January 23, 1992).

In the test he designed, Mr. Barnum, who is a sysop, or systems
operator, in charge of the government documents component on Free-
Net, posted a monthly list of all tables of data that were available from
the EBB. These included such things as monthly retail sales, employment
and earnings, and currency exchange rates from the Federal Reserve.
Each month there would be a listing of around 200 available files.

FREE-NET 197

Free-Netters could browse the list of available tables and listings
and request a file directly or send an electronic mail message to Mr.
Barnum in Free-Net citing the ones they wanted to see and access. On
receipt of such a message, Mr. Barnum would access the Department
of Commerce EBB via modem and download the requested table or
chart to disk. A small amount of editing was required before uploading
the information on Free-Net. It would then be available generally, not
just to whomever had requested it.

It seemed to be a good idea, and it was if considered in terms of
popularity with users and prompt distribution of important government
information. However, it became a tremendously labor-intensive activity,
and there were technical complications such as the 2400-baud rate of
transmission used by the Department of Commerce for downloading,
which made the process very slow and time-consuming. This kind of
thing was compounded by certain technical limitations of Free-Net,
which sometimes necessitated extensive editing of files. Free-Net only
accepted ASCII files, which made uploading of tabular information slow
(G. D. Barnum, personal communication, January 23, 1992).

There were, however, steady requests for tables and statistics, some
from the university community and many others from businesses, local
governments, and other segments of the public. The Cleveland area
is made up of numerous small municipalities, and these were frequent
requesters and users of the data. There were requests from much farther
afield as well, since Free-Net is accessible through Internet. Once users
were aware of the electronic availability of the information, there were
usually repeat requests.

Mr. Barnum soon learned that in dealing with a diverse population
of Free-Net users there were abusers of the system and troublemakers,
even ones smitten with Department of Commerce information. This
is a relatively minor problem that affects nearly all Free-Net sysops
to some extent and occasionally other Free-Net users as well. The system
lets those who are online at any time know who else is using the system
at that time. Occasionally there are rude messages, but this really is
an infrequent problem for most using the system. There are policies
in place now to deal with this sort of disturbance, which really can
be likened to misuse of the airwaves.

Along with the individually requested tables and files, there were
a number that were automatically put onto Free-Net. These included
such popular things as the Gross National Product (GNP), the Consumer
Price Index (CPI), employment statistics, economic indicators, and the
Federal Reserve credit rates. Of course, some of these statistics are
announced publicly immediately on release but usually not in their
complete form, and then only the most general statistics or those that
show significant newsworthy change. The advantage of the online EBB

198 ARLENE MOORE SIEVERS

to Free-Net was that it made all the tables available immediately on
release, which was usually about two months before they would be
received in printed form in one of the usual publication sources.

Since any time lag on time-sensitive statistics makes them
incrementally less valuable in hard-copy publication, the online
immediate access was significant for those who used them for planning
purposes. Case Western Reserve University's Weatherhead School of
Management, business people in the community, and mayors of
suburban communities are examples of information users to whom this
criterion applied.

The test was run from June to December 1990, and was not continued
due to, simply, the inability of the Government Documents Department
to man such a service on a volunteer basis and accomplish what was
required in the amount of time that was needed. Some of the technical
problems encountered could perhaps have been fairly easily solved had
the project continued. Lack of time and staffing was the main deterrent.

The reality is that Free-Net is operated mainly by volunteers. This
means that incorporating its services into those of a university library
must involve stretching limited time and personnel resources. It is a
factor that cannot be ignored. Dr. Grundner sees a more practical role
for public libraries in supporting Free-Net through active involvement.
As already described, the Cleveland Public Library and CAMLS are
actively participating in Free-Net. In other communities, such as Peoria,
the public library has taken a central role as a driving force in Free-
Net. Making terminal access available to the public is a key service
in this effort.

The end of the test involving the Government Documents
Department at Case Western Reserve University, the Department of
Commerce online Economic Bulletin Board, and Free-Net did not mean
a cessation of Free-Net activity for the Government Documents
Department. Their leadership and participation has taken a new turn
and now is committed to a new, perhaps more viable, Free-Net
government documents option.

Free-Net is now providing a means of uniting electronically the
Ohio Government Documents Roundtable (GODORT) members. These
include representatives from the 38 Ohio libraries with government
documents collections. They include the State Library and major
university libraries as well as small public libraries with very limited
collections. GODORT is creating a Free-Net section to aid their own
cooperative efforts, as well as to make information available to the public
more readily. New Free-Net government documents offerings include
an online directory of state government documents collections, the

FREE-NET 199

GODORT newsletter and the GPO depository newsletter available
electronically, and a section for requests, offers, and exchanges of
documents between members.

The State Library also uses Free-Net to publicize a list of available
state documents and their distribution status. It also maintains a current
list of Ohio government officials. A section called "Get government
information" still makes available popular Department of Commerce
Electronic Bulletin Board information such as the CPI, employment
data, and economic indicators. As with all Free-Net areas, there is a
question-and-answer section open to all, which is monitored and
answered by GODORT members. Free-Net Cleveland reaches to such
areas as Oberlin, a GODORT site, and connects with other Ohio Free-
Nets such as the one in Youngstown, Ohio. Other Ohio GODORT
members connect via the Internet. In establishing the GODORT Free-
Net option, much attention is being paid to setting up clear areas of
responsibility for maintaining and updating the information. This is
one of the most important determinants of a successful Free-Net section.
Now Free-Net has even better capabilities to measure use, and these
capabilities will be used to expand or limit the options GODORT makes
available (G. D. Barnum, personal communication, January 23, 1992).

Through the example of the experience of the Case Western Reserve
University Library Government Documents Department in Free-Net,
one can gain an understanding of the ways Free-Net continues to change
and evolve as needs and capabilities of its operators and users do. New
areas of Free-Net activity are proposed and set up continually. Others
lose interest or willing volunteers to operate them.

From a user's standpoint, inactive Free- Net areas are a disap-
pointment. This view is also supported by the university coordinator
for Free-Net who noted that one of the biggest problems in coordinating
Free-Net is in monitoring areas that have gone dormant, contacting
the sysops responsible or finding new ones, and getting the areas active
and current again or deciding whether they should be dropped. Some
sections, such as culinary arts, veterinary medicine, computer groups,
the legal section, and the area run by Cleveland's Lewis NASA Research
Center need no such prodding. They are busy and active all the time,
have lots of enthusiastic contributors and users, and obviously have
good and competent sysops.

FREE-NET, GOVERNMENT, THE LAW, AND POLITICS

One area of intense daily activity on Free-Net is the legal one in
the government area of the electronic city. Many of the elements of
this section are a result of the emphasis in Free-Net on community

200 ARLENE MOORE SIEVERS

participation in government. It is within the framework of a project
called the NPTN Teledemocracy Project. This includes such things as
the Hermes Project, which puts U.S. Supreme Court decisions, dissenting
opinions, and arguments online, as well as other U.S. Federal Court
decisions. It is augmented by an interpretation section to these decisions
and rulings. There is also a question-and-answer legal forum, which
is quite active with participation among the legal community of
Cleveland, business people with legal questions relating to agencies
such as the Occupational Safety and Health Administration (OSHA),
and students and faculty of Case Western Reserve University Law School
asking and answering questions.

Government and politics is an area of substantial Free-Net use with
an even greater potential for growth in the future. In some Free-Nets,
sadly not in Cleveland, discourse between government officials and the
citizenry is an important feature. Only one or two small communities
in the Cleveland area appear to be using Free-Net to provide a one-
to-one link between officials and those whom they represent. In the
few such community links that do exist, the system does perform the
function of getting messages directly to those in charge and of getting
responses. The usual local complaints concerning such matters as
barking dogs, problems with garbage pickup, and needed street repairs
are common messages. There is an incentive to officialdom to reply
since they must know that a larger audience than just the person who
complains will see if the query goes unanswered.

Even taking into consideration those who do not participate, the
government center of Free-Net is one of its most vital elements. OSHA
and the Environmental Protection Agency (EPA) information is
dispersed, interpreted, and commented on in Free-Net. An up-to-date
directory of governmental hotlines, name listings, addresses, and
telephone numbers for elected officials from the president on down
is maintained. The county engineer's office maintains an area issuing
news bulletins and answering questions.

The NPTN Teledemocracy Project consists not only of information
online, but during times of important political campaigns provides all
major candidates with "electronic campaign office space," at least on
the Cleveland Free-Net system. This is intended to allow users to access
information directly from candidates, not only that interpreted by the
media (National Public Telecomputing Network, 1991c, p. 2).

THE FREE-NET MOVEMENT

Dr. Grundner's pioneering efforts are these days directed toward
establishing Free-Nets in other communities and cities and in linking

FREE-NET 201

these through the NPTN, which he created. He sees the NPTN as the
community computing equivalent to National Public Radio or the
Public Broadcasting Service (T. M. Grundner, personal communication,
January 27, 1992).

Linked Free-Nets, which would still be entirely community created
and driven since this is an integral concept of Free-Net, could rely on
some network feeds to fill out their programming, so to speak. Dr.
Grundner calls this cybercasting. Popular examples currently on Free-
Net are the daily USA Today News in an electronic edition and national
weather service forecasts as well as Academy One features. "Go weather"
probably has the most logins next to personal electronic mail on
Cleveland Free-Net. Other such networked offerings would be NASA
space shuttle launch and mission transmissions, which is offered on
Cleveland Free-Net through the NASA Air and Space SIG. Electronic
journals are another possibility as networked utilities of NPTN.

The main role of NPTN currently is in helping communities
establish their own Free-Nets, both in the United States and abroad.
NPTN makes available the technical expertise and software necessary
to operate community systems (National Public Telecomputing
Network, 1991b, pp. 1-2).

Establishing comprehensive electronic mail linkups between Free-
Nets is another objective of NPTN, one that has tremendous capacity
for use by the widest range of people, for example, connecting families
with children in college. NPTN plans to establish a national news
organization to serve the telecomputing public as well as to establish
NPTN international connections with overseas information resources.
Other large cities that would be important NPTN resources are slated
to establish Free-Nets. One of the current sites organizing is Los Angeles
in coordination with UCLA and its library school (T. M. Grundner,
personal communication, January 27, 1992).

A significant international aspect of Free-Net, other than the
Internet connection, is the establishment of direct links to a number
of foreign libraries. Dr. Grundner is keenly interested in establishing
more overseas Free-Nets and in linking them with those in the United
States. During the Gulf War, electronic mail access through Free-Net
was established to enable friends and relatives of troops stationed in
the Middle East to communicate quickly. The most interesting and
visible international connection of Free-Net remains that of the
schoolhouse area of the electronic city, which includes the Academy
One space launches.

ESTABLISHING A FREE-NET

The steps involved in organizing and establishing a community
Free-Net are fairly straightforward. Important questions related to this

202 ARLENE MOORE SIE VERS

issue are how much it costs, who needs to be involved, and what kind
of support NPTN can provide. Dr. Grundner estimates the initial cost
as approximately $10,000, this being the amount one needs to buy the
requisite computer equipment. An energetic, leadership-minded
organizing committee composed of representative members of the
community is essential. Help and advice are available from the start
from NPTN, who provide the software, some organizing information,
and certain network feeds (T. M. Grundner, personal communication,
January 27, 1992).

Of course, an Internet connection, usually through the local college
or university, is essential, as is institutional cooperation in the project.
All Free-Nets operate primarily through volunteers, so it must be
determined if enough dedicated people are available in the community,
with computers, modems, and the willingness to give of their time to
the project. A Free-Net in a medium-sized community might require
only one full-time person to run it. Of course, equipment such as
computers and modems is needed in any public areas to be made available
as public Free-Net sites. Often this equipment exists and is available
in the libraries, schools, and community centers. Free-Net is an obvious
project for communities to put forward for grant aid. In large cities,
the firm commitment of a university would probably be required. In
Cleveland, two and one-half full-time employees from Case Western
Reserve University run Free-Net out of Information Services, which
also administers the university computing center and University
Libraries. Free-Net is an ongoing commitment of the university.

SUMMING UP FREE-NET

This description of Free-Net in Cleveland, from the primary
perspective of an academic librarian, is intended to provide an insight
into what community computing is all about, how it relates to the
university and to libraries of all kinds, and what its potential is. In
a time when it is important that universities demonstrate their relevance
to the states and communities in which they exist, as well as to the
taxpayers who provide funding directly or indirectly, initiating a Free-
Net or becoming an integral part of one that develops is a fairly
inexpensive way to establish this relevance. Currently, universities are
the vital link for community Free-Nets because of the necessity of Internet
access.

When the National Research and Education Network (NREN) was
being debated in Congress, Free-Net documents were presented as an
example of a potential community use of the network. Future Free-
Nets will need to draw on public library resources, but universities

FREE-NET 203

may be bypassed if connections are possible without them. The scenario
of libraries being somehow omitted as important links in providing
information as the information technology revolution forges ahead is
a common one. The public may not relate an information technology
revolution as having to do with public or other kinds of libraries unless
the connection is made apparent. Librarians are often startled by this
misperception since almost all are intensely involved in the shift to
making information available and accessible in digitized form. Reaching
out and being part of collective efforts are essential to staying at the
center of providing access to information in the future. Free-Net is an
avenue for providing this access.

REFERENCES

Grundner, T. M. (1990). Free-Netting: The development of free, public access community
computer systems. In C. A. Parkhurst (Ed.), Library perspectives on NREN: The
national research and education network (pp. 51-52). Chicago: Library and
Information Technology Association.

National Public Telecomputing Network. (1991a). Academy One: A national online
educational community. Project summary. (Available from Dr. T. M. Grundner,
President, NPTN, Case Western Reserve University, 303 Wickenden Building,
Cleveland, OH 44106).

National Public Telecomputing Network. (1991b). Community computing and the
National Public Telecomputing Network. (Available from Dr. T. M. Grundner,
President, NPTN, Case Western Reserve University, 303 Wickenden Building,
Cleveland, OH 44106).

National Public Telecomputing Network. (1991c). NPTN Teledemocracy Project. Project
summary. (Available from Dr. T. M. Grundner, President, NPTN, Case Western Reserve
University, 303 Wickenden Building, Cleveland, OH 44106).

M.E.L. JACOB

M.E.L. Jacob Associates
Columbus, Ohio

New Technology, New Tools, New Librarians:
Shaping the Future

ABSTRACT

Speakers' comments from the Annual Clinic on Library Applications
of Data Processing are summarized. The focus of the clinic was designing
information, and topics discussed include design principles, knowledge
management, applications of technology to information workstations,
graphical interfaces, public library use of the Internet, electronic
information in school libraries, computer-mediated instruction,
computer-based staff training, design techniques, hypertext, information
delivery in a networked environment, and the Cleveland Free-Net.

INTRODUCTION

Most of the authors for the clinic have focused on current state-
of-the-art technology. This is the technology in use in most libraries
and likely to be applied in the near future. While the business of
predicting the future is fraught with minefields and notable failures,
it is likely that the limits of technology will continue to be pushed
forward, and more capability and capacity will be available at ever
lower unit cost. Librarians are both the early adapters of new technology
and the followers. The authors represented in this volume are among
the leaders, pioneers, and early adapters.

204

SHAPING THE FUTURE 205

DESIGNING INFORMATION

Information Technologies

Among the current technologies that affect information services
are workstations, multimedia support, optical and CD-ROM storage
media, and networking. Tools in the form of software and some hardware
are increasingly available, although as Katharina Klemperer noted,
different media require the use of different tools and not all have reached
the same level of refinement and application. Among these software
tools are various hypermedia packages. A number have been described
by Peter Scott. Other tools librarians can and should use in learning
about and adapting technology and tools are the technology itself,
colleagues, continuing education activities, the literature, and of course
the Internet. Themes that have been repeated are information creation,
maintenance, use, and evaluation.

Design Principles

Edward Tufte reminds us that we live in a multidimensional
environment, but our displays are limited to two dimensions. However,
we can use technology and graphic design to achieve apparent
multidimensional presentations. He presented a variety of examples to
illustrate this showing both good and bad ways of displaying
information. His examples, however, required the verbal context he
provided for understanding. The pictures alone were insufficient to
illustrate his points.

Tufte, like Richard Saul Wurman (1989) and Ronnie Peters, believes
that good design and organization can enhance intelligibility of
information. He also agreed with Richard Greenfield that most graphical
user interfaces are examples of poor design. Some supposed advances
he noted are in fact regressions. He also urged using comparisons in
display to enhance analysis and comprehension.

High-resolution displays are essential. Some of these begin to
approach the print medium in resolution, but the displays available
to most users are limited and a poor substitute for the printed page.

Graphical techniques are especially useful when there is an
overwhelming amount of information to be conveyed. These techniques
can focus attention on the primary points while still including some
aspects of the whole. Professional designers can enhance information
displays and should be used. Design by committee is fraught with failure.
There is no substitute for creativity and for the coherence a single good
designer can provide.

206 M.E.L. JACOB

Knowledge Management

Carolyn Gray and Richard Lucier covered related aspects of
knowledge management. Gray described the Gesher Project, a joint
effort of Digital Equipment Corporation and the Brandeis University
Library to study scholarly communication and information use and
to develop a personal information system for scholars. An ethnographer
on the Brandeis Library staff enabled them to apply ethnographic
techniques to their study, providing a better understanding of the context
and a richer view of critical factors. Gray cautioned that what appears
to be inefficient in isolation, such as lunches and coffee klatches, may
be highly effective channels for communicating information and
teaching. She also cautioned that ethnographic techniques are time-
consuming but reveal information that might otherwise be missed.

Gray defined knowledge activity as consisting of seven aspects:
diagnosis and problem finding, planning and decision making,
monitoring and control, organizing and scheduling, authoring and
presentation, communication, and lastly system development. The
information chain involves production, distribution, acquisition, and
use. These two views must be linked. Librarians must build bridges
between scholars and themselves and the information resources. The
world is in a constant state of change. Scholars are changing and so
is scholarly work and research. Librarians must change too.

Richard Lucier provided an overview of the knowledge management
model developed at the Welch Library of Johns Hopkins University.
He proposed a new role for librarians as the creators and maintainers
of scholarly and research databases and illustrated his view with examples
of the support provided by the library to the Human Genome Mapping
project.

Librarians must be active or lose the initiative to others. Budget
problems are not an excuse. The present constraints are part of an overall
systemic change and not just the effect of escalating serial and material
prices. Structural changes are necessary. Librarians must reallocate
existing resources and seek new resources outside traditional bounds.
Changing people is difficult if not impossible.

Too many librarians focus on replacement strategies rather than
innovation or transformation. Knowledge management requires
transforming people, functions, and organizations. Revolution, not
evolution, is needed. Lucier has found it better to separate functions
and set up new units with new people to develop knowledge management
activities. Once such groups become viable, they act as change agents
for the more traditional units such as the library. However, librarians
must find ways to maintain critical traditional services while making
such transitions.

SHAPING THE FUTURE 207

Identifying the critical institutional needs is one way of establishing
priorities and identifying potential new resources. The major barriers
are legal, technical, and financial. Knowledge management will come.
Librarians must decide what role they want to play.

Information Workstations

William Mischo of the University of Illinois and Virginia Tiefel
of Ohio State University (OSU) described applications and innovative
uses of technology at their respective institutions. Mischo discussed the
library information workstation project based on an IBM workstation
providing access to local databases, campus resources, the library catalog,
and external resources including the Internet. Tiefel presented the OSU
Gateway software that helps students to formulate a search query and
execute a search using an encyclopedia, dictionary, CD-ROM indexes,
and the library catalog. The Macintosh-based Gateway encourages use
of both print and electronic resources. Both are focused on providing
seamless, one-stop shopping for the user.

Mischo noted that current information retrieval systems deliver both
too much and too little. They either overwhelm the user with many
documents or fail to retrieve anything. Many users are uncertain which
information resources to use or how to formulate their questions.

Mischo noted that keeping all workstation software updated is a
problem. At present, sneakernet is used; i.e., a staff member individually
loads new software into each workstation. Eventually he hopes to use
the network to distribute such software. Since Illinois has a number
of branch libraries, each can offer local databases and customize the
software interface, particularly help screens, default search values, and
vocabulary. Future changes will incorporate searching multiple
databases, multimedia databases and functions, and more image data
storage and transmission.

Tiefel said that OSU has a continuous evaluation process for The
Gateway. At present, use is limited to units within the main library,
but network access for dormitory and remote use is planned. With such
a large user education program, the tutoring nature is particularly
helpful in providing new students with an easy-to-use access to electronic
resources that requires no prior knowledge of the system or of the
resources used.

Graphical Interfaces

Richard Greenfield, consultant, provided an illustrated tour of
several graphical user interfaces with examples of what not to do as
well as examples that were well done. He urged avoiding most icons,

208 M.E.L. JACOB

pointing out that they were not intuitive and were often confusing.
He also noted that naive users do not care where information comes
from. Like Joe Friday, they just "want the facts, ma'am."

Public Libraries and the Internet

Jean Polly, Liverpool Public Library, continued her paean on the
Internet. While it is not easy to use, it has a wealth of resources that
public libraries can use to better serve their patrons. As more resources
become available only in electronic form, access to the Internet and
its resources becomes increasingly important.

While most university staff have access via stable, permanent
connections, these are too expensive for most public libraries. Instead
they can use dynamic, inexpensive connections to the Internet. While
more difficult to use, they are cheap.

Her list of needs are better interfaces, cheaper interfaces, more and
better training, and more vision regarding Internet use and resources.
Her message to librarians: GET INVOLVED.

Electronic Information in School Libraries

David Loertscher, Hi Willow, discussed the role of the school
librarian and the use of technology in schools. He noted that a number
of the publications, print and diskette, published by Libraries Unlimited
are created by school librarians.

Technology is being used routinely in more schools, and the school
media center is often a major resource in such use. Schools use personal
computers for most administrative tasks such as letters to parents,
scheduling, inventories, and accounts. A large number are also using
computers in the curriculum. He suggested that the best way to teach
students information and computer literacy is to teach them to create
databases in support of projects related to the curriculum.

Computer-Mediated Instruction

Ruth Small, Syracuse University, provided insights into the
principles for designing computer-mediated instructional programs.
Critical factors are the learner, the information, the task, and the
instruction. She recommended use of the ARCS Model: Attention,
Relevance, Confidence, and Satisfaction. Her own examples were clear,
consistent, well ordered, logical, and used repetition.

Computer-Based Staff Training

Joe Rader, University of Tennessee, Knoxville (UTK), described
the project at UTK to create a computer-based training program for

SHAPING THE FUTURE 209

staff. He noted that industry spends some |40 million annually on staff
training. There is no estimate available for what libraries spend or how
much time they devote to staff training.

Lessons learned include (a) staff participation is critical, (b) experts
can make the job easier and the finished products better, (c) technology
can make the task manageable, (d) libraries can adapt tools to their
needs as well as the products created by other libraries, and (e) staff
and supervisors need to feel they have control of the process. He
recommended considering development of packages for both staff and
users or with other libraries to reduce costs. He also asked whether
libraries should consider more standardization of practices and
procedures to enable easy transference of training software.

Design Techniques

Ronnie Peters, graphic designer, reiterated some of Dr. Tufte's
observations and provided copious examples of design techniques.
Among the most interesting of these, however, were his illustrations
from Korean artists showing different approaches to perspectives and
horizons. His illustrations underscored that different cultural
perspectives must also be considered in designing information systems
and graphical interfaces. Context matters.

Peters noted that there is a well-established body of design principles
for print media, and some of these apply equally well to computer
displays. Others must be modified to accommodate the more limited
resolution, size, shape, and color of computer displays. Designers may
use representations of the object itself or symbols standing for the object.
Flags are often used to represent countries. Icons are a formalized symbol.
He noted that icons in a system should have similar shapes and formats.
The eye first perceives the shape and then what it contains.

Type fonts should be used with care. Some fonts are not well suited
to screens. Peters ended by asserting and demonstrating that good design
clarifies and enhances communication, echoing Dr. Tuf te.

Hypertext

Peter Scott spoke as an advocate of hypertext programs. He provided
an overview of available software, focusing mostly on tools created with
Hyper Rez. He said such packages are easy to use, create, maintain, and
adapt for local needs. Paper is a waste, and electronic exchange can
replace it. Scott provides access to information on library and
information services and resources available on the Internet.

Information Delivery in a Networked Environment

Katharina Klemperer provided an overview of Dartmouth services
then focused mainly on three information trends: an increasing volume

210 M.E.L. JACOB

and variety of information, workstations, and networking. She identified
five types of information sources: indexes, structured full text, full text,
numeric, and multimedia. Most information systems handle indexes
periodical indexes, online abstracting and indexing databases, and library
catalogs well. They can also handle many structured full-text files
satisfactorily such as dictionaries, almanacs, and the Bible. They do not
handle unstructured full text such as a large monograph adequately.

Different search engines, display formats, and systems are needed
for unstructured files. Most multimedia files rely on words and codes
for providing access. Work is underway on other access means.

Klemperer echoed Polly on the need for better navigating tools
for locating information resources in networked environments. She
provided examples of some of the categorized displays Dartmouth is
experimenting with.

Cleveland Free-Net

Sievers provided an overview of Cleveland Free-Net: its history, use,
and relationships with Case Western Reserve University. The community
access system is supported by the university, and staff interact with
the community through it. Both gain. Most support comes from
volunteers. It opens access to computing and information resources to
all citizens. Schools and libraries use the resources heavily. Other
communities have established their own Free-Nets.

CONCLUSION

These writers provide multiple answers to what the new roles for
librarians could be. Gray and Lucier suggest that knowledge managers
may be the future: creating bridges to resources, building databases,
and assisting in providing scholarly information systems. Mischo, Tiefel,
Klemperer, Sievers, Scott, and Polly see providing enhanced access to
information and support and guidance to users as major roles. All of
them advocate an active role and involvement in the use of new
technology and resources. Electronic resources are proliferating, and
users need assistance in locating and accessing the information needed
for problem solving, decision making, and even entertainment. The
role librarians play depends on the willingness of the individual librarian
to become involved, participate, learn, and contribute. These leaders
have provided examples of how they have done it and where they expect
to move in the future. The rest of us need to follow their examples

SHAPING THE FUTURE 211

and begin making changes in ourselves and in our environments so
we can remain effective contributors to society and scholarship. The
turtle only moves forward when he sticks his neck out.

REFERENCE

Wurman, R. S. (1989). Information anxiety. New York: Doubleday.

CONTRIBUTORS

WINNIE CHAN is the Automated Records Maintenance Coordinator
and Assistant Professor of Library Administration at the University of
Illinois at Urbana-Champaign (UIUC) Library. She holds a B.S. in
Chemistry from Chinese University of Hong Kong and an M.S. in
Library Science from Louisiana State University. She was previously
a serial cataloger/serials coordinator at the UIUC Library. Her current
responsibilities include coordinating the online catalog production and
batch maintenance activities, involvement in development and design
of interface software for the online catalog, and troubleshooting of
microcomputer software/hardware problems.

TIMOTHY W. COLE received both his B.S. in Aeronautical and
Aerospace Engineering (1978) and his M.S. in Library and Information
Science (1989) from the University of Illinois at Urbana-Champaign
(UIUC). Since 1989, he has worked as Assistant Librarian at UIUC,
with a joint appointment to the Engineering Library and the Beckman
Institute Library. He has been heavily involved with the upgrading
of microcomputer hardware and software throughout the UIUC library
system that has taken place over the past three years. In particular,
he has been one of the principal programmers working on the UIUC's
microcomputer end-user interfaces to bibliographic databases. Prior to
earning his library science degree, Mr. Cole worked as an aerospace
engineer at Martin-Marietta Aerospace and at the Jet Propulsion
Laboratory.

PRUDENCE W. DALRYMPLE is Director of the Office on Accreditation
with the American Library Association (ALA). Formerly a faculty
member of the Graduate School of Library and Information Science
at the University of Illinois at Urbana-Champaign, she received her
Ph.D. in Library and Information Studies from the University of
Wisconsin-Madison. Dr. Dalrymple has published several articles in her
fields of interest, which include information retrieval, particularly
studies of end-user access to electronic information systems, and health
sciences librarianship. Her experience prior to ALA includes more than

212

CONTRIBUTORS 213

10 years as a reference librarian, search analyst, and manager in health
sciences libraries.

CAROLYN M. GRAY is Associate Director at Brandeis University
Libraries. She received a Ph.D. from the Florence Heller School of Social
Policy, Brandeis University, an M.L.S. from the School of Library Science,
University of Oklahoma, and a B.A. in English from the University
of Missouri at St. Louis. Her responsibilities at Brandeis include library
systems, public services, and library development. Dr. Gray has consulted
with a number of different organizations in the area of library technology.
Her fields of interest include information policy and scholarly
communication. Sue Woodson-Marks, Ph.D., an anthropologist on the
library staff, helped with the preparation of the paper.

M. E. L. JACOB is a writer, consultant, and publisher of Entrak. She
teaches workshops in strategic planning and library networking. Ms.
Jacob is active in a number of library and information science
associations and societies and is a frequent speaker at conferences. She
has worked at OCLC and in university, public, and special libraries.

KATHARINA KLEMPERER is Director of Library Automation at
Dartmouth College. She holds a B.A. from Swarthmore College and
an M.A. in Music from San Francisco State University, as well as an
M.L.S. from the University of California at Berkeley. Prior to her
appointment at Dartmouth, she was a member of the MELVYL system
development team at the Division of Library Automation of the
University of California.

DAVID V. LOERTSCHER is President, Hi Willow Research and
Publishing. He has degrees from the University of Utah, the University
of Washington, and a Ph.D. in Library Science from Indiana University.
He has been a library media specialist in elementary and secondary
schools in Nevada and Idaho and taught library media education at
Purdue, the University of Arkansas, and the University of Oklahoma.

RICHARD E. LUCIER is University Librarian and Assistant Vice
Chancellor for Academic Information Management at the University
of California, San Francisco. He holds a B.M. in Music and Philosophy
from the Catholic University of America and an M.L.S. in Library
Science from Rutgers University; he has completed extensive work in
health policy and administration at the University of North Carolina
at Chapel Hill. Previously, Mr. Lucier was the cofounder and director
of the Laboratory for Applied Research in Academic Information at
the Johns Hopkins University. Known for his development of the
Knowledge Management Model, he has special interests in scientific
and scholarly communication, and the development and management

214 CONTRIBUTORS

of scientific databases. Among his publications is a forthcoming book
of knowledge management to be published by the Johns Hopkins
University Press in 1993.

WILLIAM H. MISCHO is Engineering Librarian, Beckman Institute
Librarian, and Professor of Library Administration at the University
of Illinois at Urbana-Champaign. His research interests include
enhanced access to bibliographic resources, interface design and
development, and microcomputer applications in libraries.

RONNIE PETERS is Art Director at Dynamic Diagrams. He was
educated as a graphic designer at the University of Canterbury, New
Zealand, and has a master's degree from Rhode Island School of Design.
He has worked as a designer at numerous positions in New Zealand
and the United States including the Waikato Museum of Art and History,
IRIS (Brown University's Institute for Research in Information and
Scholarship), Studio Works (New York), and Graphic Design
Continuum (Ohio). He won an award at the World Design Expo, Design
Eye Competition in Japan. At Dynamic Diagrams, a design studio
dedicated to the creation of information graphics in both print and
electronic media, his work includes design and production of all aspects
of both print and electronic publications. He also teaches part time
at Rhode Island School of Design. He coauthored a paper, "Design
of Hypermedia Publications: Issues and Solutions," presented as part
of the conference, EP91.

JEAN ARMOUR POLLY is the Manager of Network Development
and User Training at NYSERNet, Inc., the New York State Education
and Research Network. She was formerly the Assistant Director for
Public Services at the Liverpool Public Library in central New York.
She has a B.A. in Medieval Studies and an M.S.L.S., both from Syracuse
University. Author of two books and numerous articles in the library
literature, she is currently a columnist for Library Journal on topics
of technology and computer books. Her interests involve use of the
Internet by the public in library settings as well as human/machine
interface design.

JOE C. RADER is Associate Professor and Head of University Archives,
University of Tennessee, Knoxville, Libraries. Prior to this appointment
in July 1990, he served as Head of Circulation Services. Codirector for
15 months, in 1989-90, of a largely externally funded project to develop
computer-based training for library staff, he has published several
articles and made presentations based on the project and its results.
Pre-librarianship careers, a summer's internship at the Lawrence
Livermore Laboratory in California, and the experience of the Modern
Archives Institute have contributed to the diversity represented in his

CONTRIBUTORS 215

interests. He is a long-time active member of many local and regional
information professional societies and national organizations, including
the American Library Association, the American Association of
University Professors, and the Society of American Archivists.

PETER SCOTT is Manager, Small Systems, at the University of
Saskatchewan Libraries, Canada. He was formerly Order Unit Manager,
Technical Services, at the same institution. He received his B.A. in
Philosophy and Literature at the Middlesex Polytechnic, England, in
1973. He is a frequent speaker at library and networking conferences
and is the author of many hypertext utilities including HYTELNET,
HYPERVAX, HYENVOY, and HYDOS.

ARLENE MOORE SIEVERS is Head of the Information Acquisition
Department at Case Western Reserve University Library. She received
her M.L.S. from Indiana University and worked for a number of years
in public libraries in Indiana. Her work in serials and library automation
began as an account executive with Swets Subscription Service in the
Netherlands. She is active in the United Kingdom Serials Group and
is the North American Correspondent for their Journal Serials.

RUTH V. SMALL is Assistant Professor and Coordinator of the School
Media Program in the School of Information Studies at Syracuse
University. She received her Ph.D. in Instructional Design, Development
and Evaluation at Syracuse University. Her areas of interest focus on
how people access, process, and present multimedia information. She
has designed and evaluated educational and training courses and
materials for business and industry, schools, colleges and universities,
and social services agencies and has published book chapters and articles,
including "Learning Situations and Instructional Models" (with Charles
M. Reigeluth) in Educational Technology: Foundations (edited by
Robert M. Gagn), "The Contributions of Technology to Instruction
and Learning" in the 1990 edition of School Library Media Annual,
and "Information Based Education: An Investigation of the Nature and
Role of Information Attributes in Education" (with Michael B.
Eisenberg) in Information Processing and Management.

LINDA C. SMITH is Associate Professor in the Graduate School of
Library and Information Science at the University of Illinois at Urbana-
Champaign. She joined the faculty in 1977 and teaches in the areas
of reference, science reference, and online information systems. Her
research interests include information retrieval, artificial intelligence,
and science information. She is active in a number of professional
associations including the American Society for Information Science,
the American Association for the Advancement of Science, and the
Association for Library and Information Science Education.

216 CONTRIBUTORS

VIRGINIA TIEFEL has been Director of Library User Education at
the Ohio State University Library since 1978. She is a graduate of the
University of Michigan School of Information and Library Studies and
has published numerous articles on library user education. She is a
recipient of the 1986 Miriam Dudley Bibliographic Instruction Award
and was chosen Outstanding Ohio Academic Librarian in 1984. She
is Project Director of Ohio State's Gateway to Information Project.

LESLIE TROUTMAN is Music-User Services Coordinator and Assistant
Professor at the University of Illinois at Urbana-Champaign. A graduate
of Bowling Green State University, the University of North Carolina
at Chapel Hill, and the University of Illinois at Urbana-Champaign,
her research interests include music materials in the online environment
and the discography of early music. Currently, she is collaborating with
Donald W. Krummel on a book about musical titles.

INDEX

Prepared by Laurel Preece

Academy One (Free-Net): 193

Agenbroad, J. E., 4

AlterNet, 87

American Memory Project, 83

Ankeny, M. L., 55, 74

Archie, 93, 171

ARCS Model, 208; attention principle,
121-123; relevance of learning task,
123-124; confidence of learner, 124-
127; satisfaction of learner, 127-128

Arms, C, 54, 74

Arnheim, R., 157, 163

Artzberger, M., 188, 191

Australian Academic Research
Network: and the Information
Resources Access System, 171

Balcom, K. M., 99, 105

Barnum, G. D., 188, 196, 197, 199

Barren, B., 175, 177

Bayne, P. S., 131, 133, 146

Belkin, N. J., 26, 33

Beyond the Walls: Networked In-
formation Kit, 89

Boolean searching, 62, 65, 67, 181

Borah, E. G., 1, 4

Borgman, C. L., 55, 74

Boulton, W. R., 22, 33

Brandeis Radio Astronomy Group
(BRAG): and the Gesher Project, 29-
32

Brandeis University Libraries: and the
Gesher Project: 2, 25-32, 206

Briggs, L. J., 135, 146

Brittain, J. M., 19, 33

BRS/SEARCH IBIS, 49, 50, 51, 52, 54,
57, 60-69

Building Information Infrastructure,
84

Bunderson, C. V., 117, 129

Calixto, B., 4

Carey, L., 122, 126, 129, 135, 136, 146

CARL (Colorado Alliance of Research
Libraries), 91, 107

Carson, C. H., 122, 123, 125, 127, 128,
129

Case Western Reserve University: and
the Cleveland Free-Net, 3-4, 190-192

CATALIST, 171, 177

Center for Knowledge Management
(University of California, San
Francisco): 6, 12-15

CERFnet (California Education and
Research Federation), 86

Chan, W, 2, 48, 212

Charles, S. K., 55, 74

Cheng, C.-C, 57, 74

Chmelewski, K., 4

Clark, K. E., 55, 74

Clark, S. E., 49, 75

Cleveland Free-Net. See Free-Net

Cohen, V. B., 126, 127, 129

Cole, T. W, 2, 48, 50, 75, 212

Communication model: 22-23

Computer-based training at the
University of Tennessee, Knoxville,
Libraries, 3, 208-209; background of
library, 131-132; description of pro-
ject, 132-134; development activi-
ties, 135-136; evaluation techniques,
136-137; future of, 142-143; project
analysis, 143-146; structure of in-
struction, 137-141; topic selection,
134-135; University of Kentucky re-
plication, 141-142

Computer-mediated instruction, 3,
208; ARCS Model, 120-128; informa-
tion quality, 118-119; learner char-
acteristics, 117-118; task characteris-
tics, 119-120

217

218

INDEX

Computer screen design, 2, 209; char-
acteristics of the computer screen,
147-148; for computer-mediated in-
struction, 122-123; projection tech-
niques, 152-156; representation and
symbol, 148-152; space considera-
tions, 152; typography con-
siderations, 158-161; visual order,
156-157; windows, 157-158

Computer Systems Policy Project, 88

Cooper, R. S., 6, 18

Costa, B., 97, 115

Costa, M., 97, 115

Council on Library Resources: study
of knowledge management, 15-17

Cranmer, F. A., 19, 33

Crook, M., 4

Crossing the Internet Threshold: An
Instructional Handbook, 95

Cruise of the Internet, A, 90

Curtis, R. V., 122, 123, 124, 125, 127,
128, 129. See also Small, R. V.

CWRUNET: and the Cleveland Free-
Net, 189, 190-191

Dalrymple, P. W., 4, 212-213

Databases, local production of, 108-110

Davidson, L. A., 4

Davis, G. B., 22, 24, 33

Designing information products:
skills needed by librarians, 1-2

Desktop publishing: for children, 112

Dick, W., 122, 126, 129, 135, 136, 146

Digital Equipment Corp., Cambridge
Research Laboratory: and the
Gesher Project, 2, 25-32, 206

Directory of Electronic Journals,
Newsletters, and Academic Dis-
cussion Lists, The, 166

Drabenstott, K. M., 55, 74

Duffield, J. A., 127, 129, 130

Duggan, R., 171, 177

Economic Bulletin Board (EBB):
availability through Free-Net,
196-199

Edmonds, L., 99, 105

Education: applications of technology
in, 103-106

Electronic information resources:
indexes, 179-180; multimedia, 185-
186; numeric data, 184-185; struc-
tured full text, 180-181; unstructured
full text, 181-184

Electronic mail: and the Cleveland
Free-net, 190-191; at the Liverpool
Public Library (New York), 79-80

Emerging Technologies Workstation,
78

Emily Postnews, 88

Envisioning Information, 2, 4

ETHICS Method, 27-29

Evolution of American Educational
Technology, The, 103

Expectancy-value theory, 120-121

Extended Online Public Access Cata-
logs (E-OPACs): description of, 49-
50. See also Online Public Access
Catalogs (OPACs)

Farber, E., 47

File Transfer Protocol (FTP), 80-81

Fischer, G., 56, 75

Fleck, R. A., Jr., 22, 33

Fogel, E., 171, 177

Foster, J. S., 4

Free- Net, 3-4, 91; definition of, 187-190;
establishing a Free-Net, 201-202;
government and politics in, 199-
200; government documents avail-
ability, 196-199; K-12 applications,
192-193; role of Case Western
Reserve University, 190-192; role of
libraries, 193-196; role of the
National Public Telecomputing
Network (NPTN), 200-201

Freedom Shrine (Free-Net), 195

Full text: access and delivery of
electronic information, 180-184

Gagne, R. M., 135, 146

Galitz, W. O., 122, 126, 127, 129

Gateway to Information, 2, 207; and
bibliographic needs of students, 35-
36; description of, 34; development
of, 36-40; evaluation of, 41-43; future
of, 43-47; goals of, 35; imple-
mentation of, 40-41

Genome Data Base (GDB): and
knowledge management, 2, 10-12

Gesher Project, 2, 206; BRAG small
group study, 29-32; goals of, 26;
participative design in, 26-29

Gluck, M., 124, 129

Goggin, N., 4

Gopher, 93, 171-172, 174

Gorman, M., 49, 75

INDEX

219

Gould, J. D., 56, 75

Government documents: availability

through Free-Net, 196-199
Grainger Center (University of

Illinois, Urbana-Champaign), 70,

72-74

Gray, C. M., 2, 19, 20, 33, 206, 213
Greenfield, R., 4, 205, 207
Grudin, J., 56, 75
Grundner.T. M., 188, 189, 190, 192, 193,

194, 200, 201, 202, 203

Hannifin, M. J., 127, 129

Havelock, R. G., 23, 24, 33

Havelock knowledge system, 23-24

Hazen, M., 122, 129

Henderson, C., 4

Hildreth, C. R., 49, 55, 56, 75

Hitchhiker's Guide to the Internet, 88

Ho, M. L., 101, 115

Hockney, D., 154, 163

Hodge, B., 22, 33

Honess, C. B., 22, 33

Hooper, S., 127, 129

Hooten, P. A., 98, 105

Human Gene Mapping (HGM)
Workshops, 10-12

Hunter, R. N., 55, 75

HYCLASS, 174

HYDIRECT, 166, 177

HYDOS, 165-166, 177

HYENVOY, 166, 177

HyperCard, 3; and The Gateway to In-
formation, 38; applications to
computer-based training, 131, 132,
133, 135-136, 141-142, 143, 144

Hypermedia, 147-148. See also Multi-
media

HyperRez, 165-166, 171-173, 177, 209

Hypertext, 3, 209; creation of, 166-171;
definition of, 164; HyperRez, 165-
166, 171-173; Internet resources, 174-
177; library applications, 173-174;
software overview, 165. See also
Multimedia

HYPERVAX, 166, 177

Hyplus, 175, 177

HYTELNET, 3, 165, 166-171, 177

IBIS, 49, 50, 51, 52, 54, 57, 60-69
ILLINET Online Plus (IO+), 49-52
Illinois Bibliographic Information
Services. See IBIS

Indexes: access and delivery of elec-
tronic information, 179-180

Information management. See Know-
ledge management

Information models, 20-25, 119. See
also ARCS Model

Information Resources Access System
(University of Western Australia),
171

Internet and public libraries, 3;
computer use at the Liverpool
Public Library, 77-79; connecting to
the network, 86-87; discussion
groups, 81-82; effect on librarians,
84-85; electronic mail, 79-80; File
Transfer Protocol (FTP), 80-81;
learning guides, 88-90, 95; multi-
media applications, 83-84; nav-
igating host computers, 93-94;
QuickMail gateway, 80; resources,
91-92; TELNET, 80-81

Internet Companion: A Beginner's
Guide to Global Networking, The,
95

Internet Resource Guide, 90

Jacob, M. E. L., 2, 204, 213

Johns Hopkins University: and the

Genome Data Base, 2, 10-12
Jones, M. K., 122, 123, 127, 129

Kahin, B., 84

Kehoe, B., 90, 95

Keller, J. M., 120, 121, 122, 123, 124,
125, 126, 127, 128, 129

Klemperer, K., 3, 178, 205, 209, 213

Knowledge management, 206; at Uni-
versity of California, San Francisco,
5-7, 12-15; components of, 9; Digital
Equipment's prototype informa-
tion management system, 32; infor-
mation models, 20-25; knowledge
work activities, 24-25; the Genome
Data Base, 2, 10-12; the Gesher Pro-
ject, 2, 25-32; Invitational Sympos-
ium on Knowledge Management,
16-17; library's role in, 8-10;
scholarly use of information, 19-20

Knowledge Management Environ-
ment (University of California, San
Francisco), 6-7, 10, 12-15

Knowledge work. See Knowledge
management

220

INDEX

Kochen, M., 20, 21, 33
Kochen information model, 21
Krol, E., 95

Laboratory for Applied Research in
Academic Information (Johns
Hopkins University): and the
Genome Data Base, 10-12

Langenberg, D. N., 16

LaQuey, T., 95

Large, J. A., 62, 75

Larson, N., 165, 177

Lee, J., 55, 75

Lewis, C, 56, 75

Liberty High, 92

Library Information Workstation
(LIW) (University of Illinois), 2;
future developments, 69-73;
ILLINET Online Plus (IO+), 49-
52; interface design, 60-69; Music
Library implementation, 57-60;
philosophy and approach, 52-54;
user searching behavior, 54-57

Library of Congress: American Mem-
ory Project, 83

Library Resources on the Internet:
Strategies for Selection and Use, 88-
89

Line, M. B., 19, 33

Lipow, A. G., 95

Liverpool Public Library (New York):
and the Internet, 77-95

LIW. See Library Information Work-
station

Loertscher, D. V., 3, 96, 101, 115, 208,
213

Lucier, R. E., 2, 5, 10, 15, 18, 206, 213-
214

Lundeen, G. W, 4

MacDonald, W. B., 26, 33
Mackenzie Owen, J. S., 24, 33
Management information systems

(MIS) information model: 22-23
Marchionini, G., 118, 123, 129
Marcus, R. S., 62, 75
Martin, R., 110, 115
Matheson, N. W, 15, 18
Mathews, J. R., 56, 75
McClelland, D. C, 125, 129
McDermott, M., 56, 76
MELVYL, 91
Mining the Internet, 89

Mischo, W. H., 2, 48, 49, 50, 55, 62,

75, 207, 214
Molenda, M., 104, 115
Moore, A. F., 62, 75
Moore, P., 99, 105
Mostafa, J., 4
Multimedia: access and delivery of

electronic information, 185-186;

applications for children, 112-114.

See also Hypermedia; Hypertext
Mumford, E., 26, 33
Murphy, C., 98, 115
Myers, J. E., 4

National Public Telecomputing
Network (NPTN): 193, 200-202, 203

Netcom, 87

New Horizons in Library Training:
Computer-Eased Training for
Library Staff. See Computer-based
training at the University of
Tennessee, Knoxville, Libraries

New User's Guide to Unique and In-
teresting Resources on the Internet
2.0, 89

Newby, G. B., 4

Nielsen, B., 55, 75

Norlin, D. A., 54, 75

NorthWestNet User Services Internet
Resource Guide, 89

NPTN Teledemocracy Project, 200

NSF Internet Tour HyperCard Stack,
90

Numeric data: access and delivery of
electronic information, 184-185

NYSERNet, 78, 82-83

NYSERView, 92

Ober, J., 95

O'Hanlon, N., 40, 41

Ohio Government Documents Round-
table (GODORT), 198-199

Ohio State University: and The Gate-
way to Information, 2, 34-47, 207

Olson, M. H., 22, 24, 33

Online public access catalogs (OPACs):
use with children, 98-100. See also
Extended Online Public Access
Catalogs

Orna, L., 1, 4

Participative design, 26-29
Performance Systems International

(PSI), 86

INDEX

221

Peters, R., 2, 147, 205, 209, 214

Peters, T. A., 55, 76

Pfaffenberger, B., 62, 76

Piette, M. I., 132, 146

Pilger, M. A., 108, 115

Pollitt, A. S., 62, 76

Polly, J. A., 3, 77, 208, 214

Potter, W. G., 49, 76

Public libraries and the Internet, 3, 77-

95, 208
Puttapithakporn, S., 55, 76

QuickMail, 80

Rader, J. C, 3, 131, 133, 146, 208, 214-

215

Reigeluth, C. M., 124, 126, 129
Robertson, C. A., 4
Rosenberg, D., 56, 76
Rowe, M. B., 125, 129
Rumsey, E., 4
Ryer, J. C, 95

Saba, R, 113, 115

Saettler, P., 103, 115

Saladin, B. A., 22, 33

Salomon, K., 55, 76

Sandore, B., 49, 75

Schaefermeyer, S., 119, 127, 129

Scholarly communication, 7-10. See
also Knowledge management

School libraries and information
technology, 3, 208; acquisitions,
100-101; card catalog production,
101 ; circulation, 97-98; data gather-
ing and analysis, 110-111; database
production, 108-110; desktop pub-
lishing, 112; educational tech-
nology, 103-106; equipment inven-
tory, 97; implications for librarians,
102, 114-115; library management,
101-102; multimedia applications,
112-114; network access, 106-107;
online catalogs, 98-100; word pro-
cessing, 111-112

Schultz, K., 55, 76

Science Experiments Index for Young
People, 108

Scott, P., 3, 164, 177, 205, 209, 215

Seiden, P., 56, 76

Shneiderman, B., 56, 76, 122, 125, 127,
129

Sievers, A. M., 3, 187, 215

Simon, H. A., 1, 4

Small, R. V., 3, 116, 208, 215. See also
Curtis, R. V.

Smith, L. C, 4, 215

Smith, N. M., Jr., 132, 146

Stanton, D., 171

Stein, R, 126, 129

Steinberg, E. R., 128, 129

Stevin, S., 151

Sutcliffe, A. G., 56, 76

Suzuki, K., 122, 123, 124, 125, 126, 127,
128, 129

SWAIS (Simple Wide Area Informa-
tion Server), 171, 172, 174, 175

Talan, H., 132, 146

Taylor, R. S., 20, 21, 33, 119, 130

Taylor information model, 21-22, 119

TELNET, 80-81, 166

Tennant, R., 95

Tenopir, C., 4

There's Gold in Them Thar Networks!

or Searching for Treasure in all the

Wrong Places, 90
Tiefel, V., 2, 34, 54, 76, 207, 216
Tosti, D. T, 127, 130
Troutman, L., 2, 48, 216
Tufte, E. R., 2, 4, 20, 33, 205

University of California (San
Francisco) Library: knowledge
management at, 5-7, 12-15

University of Illinois (Urbana-
Champaign): and the Illinois
Library Information Workstation
(LIW), 2, 48-76, 207

University of Tennessee, Knoxville,
Libraries: computer-based training
at, 131-146

Urr, C., 174

User interfaces for library work-
stations, 2; development of the
Library Information Workstation
(LIW), 60-69; The Gateway to
Information, 34-47

van Halm, J., 24, 33
VanMeter, V., 108, 115
Visual Display of Quantitative In-
formation, The, 2, 4

Wager, S., 127, 130
Wager, W. W, 127, 130, 135, 146
WAIS (Wide Area Information
Servers), 94

222 INDEX

Weiner, B., 125, 130

White, H. S., 82-83

White, R. W, 125, 130

Whole Internet User's Guide b Cata-
log, The, 95

William H. Welch Medical Library
(Johns Hopkins University): and
the Genome Data Base, 10-12, 206

Wilson, J., 56, 76

Winn, W, 119, 130

Word processing: student use of, 111-
112

Workstations: at the Liverpool Public
Library (New York), 78; Library In-
formation Workstation (LIW), 48-
76; The Gateway to Information,
34-47

World, The, 86-87

World History for Children and Young
Adults, 108

WorldWideWeb (WWW), 94

Wurman, R. S., 205, 211

Yee, M. M., 56, 76

Young Scholars Program (Ohio State
University), 41

Zen and the Art of the Internet, 90,
95

UNIVERSITY OF ILLINOIS-URBANA

30112003381891