Skip to main content

Full text of "A History of the Personal Computer"

See other formats

Personal Computing in the 1960's 2/1 





1964 1965 





Small Computer Syiltmi 

I < 

>| MIT UNC prototype 

i 1 

| DEC LINC-8 ! 

w-e | dec PDP-a/L 

| DEC PDP-8/S \ 


| DEC PDP-5 | DEC P 
Aug j Apr 



i Apr 

| IBM 1 





| HP 21 IB Controller \ 
| HP9100A ! 

Tims sharing 

: 1 i 1 


j May j 

j|< >| llMTCTSS I | iMIT Project MAC 


Software j 

F0 1 PTH 

|< >| Dartmouth BASIC 

I APL May 

1 Logo 

<< >| \C0B0L |<- 

Dec | 

| Spare Warx tjs.-n* 
| Algol \ May j 



! | Sketchpad 


! D«.c 

| Mo 

derr, invented 

| ARPA IPTO established j 

: SRI Augmented Heveen;!: Center 

| IBM System/360 
| Apr j 

f/lflO funrfe 

ur Computer Society (ACS) 


| Canon/Texas Inst. 

Pocketronic Calculator 

FIGURE 2.1: 

A graphical history of personal computing in the 1960's. 

2/2 Part 1 The Historical Background 



'MBfaH 1 

■fflHirii" 1 ! | 

T\ r '.'li!i 

■ jni 

IS- 1 

Figure 2.2: Wesley A. Clark and the MIT LINC computer. A 
number of people consider LINC as being the first 
personal computer . 

Photograph is reprinted with permission of MIT Lincoln 
Laboratory, Lexington, Massachusetts . 

Chapter 2 Personal Computing in the 1960's 

Most mainframe computers used a sequential batch 
process type of operation for computing tasks at the end 
of the 1950's. This mode of operation resulted in a slow 
and cumbersome interaction with the user. 

The 1960's was a decade that saw many improvements 
to personalize the computer user interface. The 
government funded research in organizations such as the 
Massachusetts Institute of Technology (MIT) and the 
Information Processing Techniques Office (IPTO) by the 
Advanced Research Projects Agency (ARPA) of the 
Department of Defense. This resulted in innovations of 
significant importance for personal computing. 
Educational institutions such as MIT, Dartmouth College 
and Stanford Research Institute (SRI) created items such 
as the first personal computer, the "mouse," computer 
networks, BASIC programming language and time sharing. 

2.1 ... Time sharing 

A time sharing computer system is one that 
interacts with many simultaneous users through a number 
of remote consoles. An interleaving in time of two or 
more jobs on one processor gives what appears to each 
user, as the dedicated use of the computer. The first 
paper published describing time sharing was by 
Christopher Strachey at the Paris International 
Conference on Information Processing in June 1959. 

MIT developed and tested the concepts of time 
sharing between December 1958 and early 1959 on an IBM 
704 computer. Principals in this early development were 
John S. McCarthy, an early advocate of time sharing and 
Herbert M. Teager. 

A Study Group investigated MIT's computational 
requirements for research and teaching in 1960 and made 
recommendations in early 1961. This resulted in the 


2/4 Part 1 The Historical Background 

development of a time sharing system for an IBM 709 
computer by Fernando J. Corbato and his staff that was 
first demonstrated in November 1961. This group 
developed the system further on an IBM 7090 computer and 
it became known as the Compatible Time Sharing System 
(CTSS) . MIT presented a paper describing CTSS at the San 
Francisco Spring Joint Computer Conference in May 1962. 

Jack B. Dennis wrote a proposal for a time sharing 
system on the MIT TX-0 computer in 1959. Then in 1961, 
MIT received a Digital Eguipment Corporation (DEC) PDP-1 
minicomputer. Under the direction of Dennis, a graduate 
student John E. Yates developed a time sharing system 
for the DEC PDP-1 that became operational in the spring 
of 1963. 

Bolt Beranek and Newman Inc., (BB&N) in Cambridge, 
developed another time sharing system for the DEC PDP-1 
in 1962. Two of the principals in the development were 
J.C.R. Licklider and John McCarthy. BB&N presented a 
paper entitled "A Time-Sharing Debugging System for a 
Small Computer" describing these developments at the 
Spring Joint Computer Conference of 1963. 

The Advanced Research Products Agency (ARPA) 
appointed J.C.R. Licklider in 1962 to be in charge of 
the new Information Processing Techniques Office (IPTO). 
Convinced that time sharing would be an important 
computer technology, Licklider selected Robert M. Fano 
at MIT to head the development of a major new system. 
This new time sharing system became known as Project MAC 
(Multiple-Access Computer or Machine-Aided Cognition) 
and an early version was operational by November 1963. 
Initially it could serve 24 users simultaneously. In 
less than a year it was serving 200 users with 100 
teletypewriter terminals. Improvements made to the 
system throughout the 1960 ' s resulted in it becoming an 
important node in the ARPANET. 

In late 1963 the Project MAC Group began a search 
for a more suitable time sharing computer. This resulted 
in the selection of the General Electric GE-645 computer 
in 1964. Shortly after, General Electric and Bell 
Telephone Laboratories joined MIT in the development of 
a new comprehensive time sharing system. This new system 
became known as Multics (Multiplexed Information and 

Personal Computing in the 1960's 2/5 

Computing Service) . It became operational at MIT in 
October 1969, and within two years was serving 500 
users . 

These early developments of time sharing 
technology were important in providing personal 
computing capabilities in a more friendly interactive 
mode at lower cost to an increasing number of users. 
Other institutions such as Dartmouth College started to 
evaluate these new concepts. 

2.2 ... Dartmouth DTSS and BASIC 

Between 1956 and 1962, a small university called 
Dartmouth College in Hanover, New Hampshire, started 
developing simple high-level programming languages on a 
small LGP-30 computer. Principals in these developments 
were John G. Kemeny and Thomas E. Kurtz. Kemeny and 
Kurtz had a conviction that, "...knowledge about 
computers and computing must become an essential part of 
a liberal education." To implement this objective and 
achieve acceptance by the students they had to simplify 
both the computer interface and the programming language 
used by the students. 

Existing languages such as FORTRAN and ALGOL were 
too complex for the majority of students. The early 
languages developed at Dartmouth were: DARSIMCO 
(DARtmouth SIMlified COde) , DART, ALGOL 30, SCALP (Self 
Contained ALgol Processor) and DOPE {Dartmouth 
Oversimplified Programming Experiment) . During this 
period Kurtz became aware of time sharing technology at 
MIT and BB&N. To meet their computer educational 
objectives they decided to develop and implement a new 
time sharing system and a new simple programming 
language for interactive computing. 

In 1963, Dartmouth College selected a new hardware 
system consisting of a General Electric (GE) GE-225 
computer for user programs, a GE Datanet-30 computer for 
communications and scheduling, a disk drive and other 
peripherals. Kurtz supervised undergraduate students in 
the development of the time sharing software, 

2/6 Part 1 The Historical Background 

emphasizing simplicity of use for the novice. This 
software became known as the Dartmouth Time Sharing 
System (DTSS) . At the same time Kemeny developed a 
compiler for the simple high-level programming language 
they had designed, named Beginner's All-purpose Symbolic 
Instruction Code (BASIC) . Features incorporated in the 
BASIC programming language were influenced by the 
knowledge and experience of using ALGOL, FORTRAN and the 
early languages developed at Dartmouth. The College 
received the new computer system equipment in February 
1964 (the GE-225 computer was changed to a GE-235 in the 
summer) . The new time sharing system, BASIC compiler and 
the first test programs operated successfully on May 1, 
1964. Subsequently, the College put the programming 
language in the public domain in order to improve its 
widespread acceptance . 

Between 1964 and 1965, Dartmouth College began an 
association with GE that resulted in a joint effort to 
develop a time sharing system and a BASIC compiler for a 
new GE-635 computer. The College developed the BASIC 
compiler that became known as GE-BASIC. GE became a 
significant contributor in the dissemination of the 
BASIC programming language on larger computers. Other 
companies such as Digital Equipment Corporation (DEC) 
and Hewlett-Packard (HP) developed interpreter 
implementations of BASIC for smaller computers in the 
late 1960's. 

The Dartmouth College development of a simple time 
sharing system and the BASIC programming language made a 
significant contribution to personal computing in the 
mid 1960's and following years. Continued improvements 
to the language were made with the release of versions 
two to five between 1965 and 1968. All implementations 
of BASIC at Dartmouth College have been compilers, 
whereas most of the later implementations on small 
personal computers were interpreters. Time sharing 
became widespread and BASIC became a popular programming 
language for personal computing. 

However for intensive computational applications, 
time sharing could not provide the processor resources 
required. Also the convenience and flexibility of having 

Personal Computing in the 1960's 2/7 

one's own dedicated processor led to the development of 
the personal computer. 

2.3 ... The First Personal Computer 

To avoid confusion, one has to define the term 
"personal computer" as being a computer designed for use 
by one person. In the 1960's most computers were large 
mainframes, shared by many users. 

A number of people consider the MIT LINC to be the 
first personal computer. The Massachusetts Institute of 
Technology (MIT) developed LINC to facilitate the use of 
computer technology in biomedical research laboratories. 
LINC is an acronym for Laboratory INstrument Computer. 
Principal designers were Wesley A. Clark and Charles E. 
Molnar . MIT demonstrated a prototype in March 1962 at 
the Lincoln Laboratory and completed sixteen units in 
mid 1963. The scientific users assembled the units to 
improve their understanding of the system that cost 
about $32,000. Initial software was a text editor, an 
assembler and some utilities. 

The LINC system had four console modules, an 
electronics cabinet and a keyboard. The electronics 
cabinet was about the size of a refrigerator. The 
processor logic circuits used transistorized system 
modules from Digital Equipment Corporation (DEC). Memory 
was a magnetic core type with a basic capacity of 1,024 
twelve-bit words, expandable to 2,048 words. 

The four console modules consisted of a control 
console, an oscilloscope module, a tape module with two 
magnetic-tape drives and a terminal module. The 
oscilloscope module could display a 512 by 512 point 

A small number of scientific laboratories used the 
LINC computer in dedicated applications. In 1966, DEC 
released a refined version of the LINC computer that 
they named LINC-8 and sold for $43, 000 . During this 
period of time other companies were extending the low 
end of the market by developing small computer systems. 

2/8 Part 1 The Historical Background 
2.4 ... Small Computer Systems 

Other organizations developed small computer 
systems or minicomputers during the 1960's for 
scientific and commercial users (See Chapter 1.6 for 
earlier small computer systems) . This continued to lower 
the cost of computers and extend the concept of personal 

DEC delivered the first production version of the 
18-bit PDP-1 minicomputer in early 1961. A minimum 
system cost $85,0000. This was followed by the PDP-4, 
then the 12-bit PDP-5 that cost $27,000 in August 1963. 
The 12-bit PDP-8, which became very successful, was 
announced in late 1964 and the first units delivered in 
April 1965 . DEC marketed the PDP-8 as "the world' s 
lowest-priced, fully programmable computer system." The 
computer used new technology such as small scale hybrid 
integrated circuits to reduce the price of the unit to 
$18, 000 . The PDP-8 had 4K words of core memory, 
expandable to 32K words. Then in August 1966, DEC 
introduced the low cost PDP-8/S system with a console 
teletype for $10,000. In 1967, DEC developed the PDP-8/I 
computer that incorporated TTL technology in place of 
the hybrid integrated circuits. This was followed in the 
summer of 1968 by a similar machine with fewer options 
called the PDP-8/L. The PDP-8/L with 4K of memory and a 
teletype sold for $8,500. 

IBM introduced small systems for technical and 
professional users starting with the Model 1620 in 1961. 
This was a batch-oriented system for FORTRAN users. Then 
IBM released the Model 1130 computing system in 1965. 
The 1130 was a single-user system with an integrated 
disk-based operating system. 

William Hewlett and David Packard founded Hewlett- 
Packard (HP) as a partnership in January 1939 and 
incorporated it as a company in 1947. It was initially a 
manufacturer of electronic test equipment. HP began 
evaluating computers to automate their instrument 
measurement systems in September 1964. This resulted in 
the development of the Model 2116 instrument controller 
in 1966. This was the company's first computer and HP 

Personal Computing in the 1960's 2/9 

sold many units as stand-alone minicomputers. The HP 
2114A minicomputer was released shortly after at a price 
of $9,950. HP also started development of the 9100 
series of electronic desktop calculators in 1966. The 
company released the HP 9100A in 1968, that has been 
described as a "computing calculator." It was a 
predecessor to the programmable calculators released in 
the early 1970' s . 

Other companies competed with DEC, HP and IBM for 
a share of the small system market. Some of these 
companies were: Computer Control Corporation (3C) , 
Control Data Corporation (CDC) , Data Machines, 
Honeywell, Scientific Control Systems (SCS), Scientific 
Data Systems (SDS) and Systems Engineering Laboratories. 
A group of people from DEC established Data General in 
April 1968 and announced their first 16-bit minicomputer 
called the Nova later that year. 

Time sharing, simple high-level programming 
languages and small low cost computer systems were 
significantly increasing the number of users. However 
the interaction between the user and the computer was in 
most cases an awkward process through a teleprinter type 
of console. The 1960's was a decade of significant early 
improvements to the user interface 

2.5... Graphics and the User Interface 

Douglas C. Engelbart developed concepts for 
augmenting the human intellect at the Stanford Research 
Institute (SRI) in Menlo Park, California between 1957 
and 1960. In March 1960, J.C.R. Licklider wrote an 
important paper entitled "Man-Computer Symbiosis" [12, 
pp. 306-318] . The paper contained ideas similar to 
Vannevar Bush's Memex concept, but extended them through 
the use of interactive computer technology. In late 
1962, Engelbart presented his SRI report entitled 
"Augmenting Human Intellect: A Conceptual Framework" and 
a proposal to the newly formed Information Processing 
Technigues Office (IPTO) . Licklider, who was the first 
IPTO director, approved the proposal that provided early 

2/10 Part 1 The Historical Background 

government funding for an Augmented Research Center 
(ARC) at SRI in 1963. 

Various graduate students at MIT and staff at 
General Motors Corporation did research on computer 
assisted drafting in the early 1960's. However a turning 
point occurred when Ivan E. Sutherland conducted 
significant research on computer graphics for a doctoral 
thesis in 1962 at MIT. Sutherland created an interactive 
graphics system, that enabled a user to create graphical 
figures on a video display using a light pen. The 
geometrical shapes could be copied, expanded, moved, 
rotated and shrunk. This also resulted in the 
development of the first user interface that 
incorporated a split screen (two-tiled windows), menus 
and the use of icons for such things as constraints to 
limit line lengths. Sutherland published the results of 
this research in an article entitled "Sketchpad: A Man- 
Machine Graphical Communications System" in May 1963. 
This became the basis for computer assisted drafting 
(CAD) and computer assisted engineering (CAE) software 
systems. In the late 1960's various companies such as 
Applicon, Calma and Computervision offered turnkey CAD 
systems . 

The ARC research on facilitating the use of 
computers to extend human knowledge and intellect, 
resulted in the development of significant improvements 
to the user interface. Engelbart concentrated his 
research on an interactive graphics environment as 
compared to the then prevalent teletype communication 
interface. Engelbart evaluated various methods of 
interacting with the screen display. During this 
research he developed the "mouse" in 1964, for which he 
holds the patent. The interactive graphics system also 
used a small five-key keyset that supplemented the 
selection capabilities of the mouse. Another principal 
in the research was William K. English. 

Engelbart's group also developed an on-line system 
with new capabilities called NLS . NLS was an acronym for 
"on-line system." He demonstrated the new system at the 
ACM/IEEE-CS Fall Joint Computer Conference in San 
Francisco, California in December 1968. The NLS system 
incorporated the capabilities for mixing text and 

Personal Computing in the 1960's 2/11 

graphics on the screen display. It also used a split 
screen (two-tiled windows) that led to the development 
of multiple tiled windows in 1969. The group also 
developed an electronic-mail (e-mail) system and 
incorporated it into NLS. 

Between 19 67 and 1969, Alan C. Kay and Edward 
Cheadle built a computer called FLEX at the University 
of Utah. Kay and Cheadle also developed a user interface 
that included multiple tiled windows and sguare icons 
representing data and programs. 

The graphics and user interface research at ARC, 
MIT and the University of Utah led to significant 
developments later at Xerox PARC, Apple Computer and 
Microsoft. SRI transferred activities of the ARC 
research group to Tymshare, Inc., of Cupertino, 
California in 1977. Tymshare renamed the NLS system 
AUGMENT and provided marketing for the product. 
McDonnell Douglas Corporation acguired Tymshare in 1984. 

2.6 ... Software 


The CODASYL group had started development of COBOL 
in 1959. In 1960 the United States government advised 
that they would not accept computer eguipment without a 
COBOL compiler. The development groups defined the 
language and compilers were operating by December of 


IBM decided to create an advance common 
programming language that would meet the reguirements of 
both the scientific, commercial and system users in 
1963. The SHARE user's group assisted in the language 
development. IBM initially intended to release the 
language with IBM's new line of System/360 computers. 
The first description of the language was in March 1964. 

The initial name assigned to the new language was 
NPL - New Programming Language. However in 1965 the name 
became PL/I - Programming Language /I. Some ambiguity 
exists in the suffix (I) . The pronunciation is 

2/12 Part 1 The Historical Background 

"Programming Language One," however the suffix is the 
Roman character I and not the Arabic numeral 1. IBM 
released the PL/I compiler in 1966. 

The American National Standards Institute (ANSI) 
subseguently defined a version of the language with a 
reduction in the number of features. ANSI named this 
version, the "G" Subset of PL/I. 


Bolt Beranek and Newman Inc., (BB&N) in Cambridge, 
Massachusetts started developing Logo in 1966. They then 
pilot-tested the language in the summer of 1967. 
Principals in the development were Seymour Papert, 
Daniel Bobrow, Richard Grant and Wallace Feurzeig who 
gave Logo its name. Charles R. Morgan and Michael Levin 
developed an extended version of Logo during 1967-68. 
The language is designed for use in education by 


Charles H. Moore developed the FORTH programming 
language around 1968. The language was designed by Moore 
to improve programming productivity. Additional features 
were that it was easy to move to a different machine and 
it reguired a small amount of memory. 


UNIX is an operating system developed by Kenneth 
L. Thompson and Dennis M. Ritchie at AT&T's Bell 
Laboratories. The main feature of the operating system 
was its portability that enabled it to run on almost any 
computer. Bell Laboratories released UNIX in 1969 and 
provided a free license to educational users. This 
resulted in its widespread use at academic institutions. 


In the 1960's most computers were large mainframes 
with restricted user access. Some had time sharing and 
remote terminals. However remote communication by a 
phone line and data transfer between different computers 
was difficult. 

Personal Computing in the 1960's 2/13 

Paul Baran conducted distributed communications 
research at the RAND Corporation as early as 1962, and 
published details in 1964. Then between 1965 and 1966, 
Donald Davies from the National Physical Laboratory 

(NPL) in the United Kingdom wrote papers describing 
concepts of digital communications using short messages 
or "packets." J.C.R. Licklider, who was the first 
director of the Information Processing Techniques Office 

(IPTO) of the Advanced Research Projects Agency (ARPA) , 
had also been promoting the concept of an "Intergalactic 
Computer Network." A concept that tried to define the 
benefits and problems of computer networking. 

Around this time period the IPTO was funding 
computer research projects and wanted to improve data 
communications for time sharing and networking. Bob 
Taylor who was the current director of the IPTO and a 
proponent of computer communications, recruited Lawrence 
G. Roberts to lead this networking project. 

Roberts began the experimental computer network 
research in 1966. He then received an appointment to 
manage the IPTO programs for the ARPA in 1967. This led 
to an initial plan for an ARPANET being published in 
October of that year. This network plan would enable 
load sharing, message service, data sharing to link 
university computers and researchers. It would also use 
the "packet" concept, interface message processors 

(IMP's) and leased telecommunication lines. ARPA awarded 
a contract to develop the network to Bolt Beranek and 
Newman (BB&N) in January 1969. Robert Kahn was a 
principal in the overall system design. By December of 
that year they had four nodes of the network installed 
and operating. This packet switching data communication 
system became highly successful in connecting many major 
universities, government organizations and research 
institutions. The ARPANET formed the basis for what 
subsequently became known as the Internet. 


Student hackers and academic staff developed some 
of the earliest games at the Massachusetts Institute of 
Technology (MIT) on the TX-0 and PDP-1 computers. One of 
the games was for a mouse, that would poke its way 

2/14 Part 1 The Historical Background 

through a maze constructed by a light pen to find a blip 
in the shape of cheese. 

Then in 1961 Stephen Russell who was a science 
fiction enthusiast and a student hacker created an 
interactive game called "Space Wars" on the PDP-1 
computer. The game displayed rocket ships that could 
fire missiles in a celestial battlefield. MIT displayed 
the game at the annual Science Open House in May 1962. 
The software was free and received wide distribution. J. 
M. Graetz, Peter Samson and others contributed various 
enhancements to Space Wars that provided additional 
challenges to participants. 

Ralph Baer of Sanders Associates in Nashua, New 
Hampshire received a patent for a ball-and-paddle video 
game using a TV set. Development started in September 
1966. Baer had basic ball-and-paddle games working in 
early 1967 and a hockey game by September. Magnavox 
marketed the game system as Odyssey 100, the world's 
first home video game. 

2. 7 ... Hobby & Amateur Computing 

This aspect of personal computing started from an 
interest by many enthusiasts in the building of their 
own computer. These people were both amateurs and 
professionals with a strong technical interest in 
hardware and software. Prior to the development of large 
scale integrated (LSI) memory chips and the 
microprocessor, it was not easy to build a computer. It 
required a knowledge of vacuum tube or transistor 
circuitry, digital logic, core memory, peripherals and 
other areas. It could also be a costly investment to 
create a complete system. 

In May 1966 Stephen B. Gray founded the Amateur 
Computer Society (ACS) . In August of that year he 
published the first ACS Newsletter devoted to hobby 
computing. The society and newsletter were a significant 
source of information for building a computer in the 
late 1960's and early 1970's. 

The November 1967 issue of the ACS Newsletter 
included a survey requesting details of each member's 

Personal Computing in the 1960's 2/15 

computer. The January 1968 issue reported the following 
results. Clock speeds ranged from 500 kHz to 1 MHz, with 
the average 500 kHz. Instruction sets were small ranging 
from 11 to 34 instructions. The number of registers 
ranged from 2 to 11, with three being the most common. 
Word sizes were from 4 to 32 bits, with 12-bits being 
the average. Memory size ranged from 4 to 8K, all 
magnetic core. Most computers used discrete transistors 
and a few reported the use of integrated circuits. A 
Teletype terminal was the most common input/output 
device. Cost ranged from zero to $1,500, with an average 
of $650. 

The April 1968 issue of Popular Mechanics included 
an article entitled "A Computer in the Basement?" The 
article described the ECHO IV, one of the few home-built 
computers actually completed. ECHO was an acronym for 
Electronic Computing Home Operator. It had a designation 
of IV because it used surplus boards from a Westinghouse 
PRODAC IV computer. James F. Sutherland designed and 
built the computer between 1966 and 1968. ECHO IV was 
seven feet long, six feet high and 18 inches deep. A 
console desk included an electric typewriter keyboard, 
surplus teletype printer, 8-channel paper tape punch and 
an 8-channel paper tape reader. The unit had 120 circuit 
boards using 2N404 transistors and NOR logic elements. 
It had four registers, used 18 instructions with a clock 
speed of 160 kHz. Memory was an 8K surplus core unit. 
The computer is now located at The Computer Museum in 

Many amateurs copied existing designs of small 
computer systems. Some based their designs on 
instruction sets from IBM 1620 or DEC PDP-8 computers. 
Enthusiasts started many computers. However only the 
most determined completed them. 

2/16 Part 1 The Historical Background 

Time sharing, small computer systems and to some 
extent, amateur computing started personal computing in 
the 1960's. Research had also improved the human 
interface to the computer. However the personal computer 
was still complex and expensive to construct. New 
integrated circuit developments that led to the creation 
of the microprocessor significantly reduced this