Skip to main content

Full text of "Brewster Kahle Interview with Daniel Erasmus 2-2010"

See other formats

-PJ ^y - N J 

February 20 I 

Brewster Kahle 

Internet Archive - Co-Founder & Director 

Interview by: Daniel Erasmus, Managing Director, The DTN 

Copyright Reserved, DTN B.V., 20 1 Page I of I 

_ D J ^? JLl 

I'm Brewster Kahle, founder and digital librarian of the Internet Archive, 
and have been involved in a lot of the technologies and companies that 
have created the Internet that it is today. Started my career after MIT 
at Thinking Machines, building big computers so that we could actually 
do something with them, then made the first search engine that was a 
broad-based search engine, then made that available on the Internet as 
a distributed publishing platform called WAIS. I started that as a 
company, sold that to AOL Then, once we had publishing starting to 
move on the Internet, then we could start to catalogue and collect all 
of the information that is the Internet, that is available on the Internet, 
to build a great library. Built Alexa Internet, which catalogues all 
websites, and the Internet Archive, which is the idea of a precursor to a 
Great Library of Alexandria, version 2.0. 

So we're still in the midst of this. This is 20 1 0. We're still in the 
transition of putting everything online, not just this organisation but the 
whole world, and starting to put it together in such a way that anybody 
can have access to anything at anytime forever. And so, all published 
information is starting to be available to people at any time, any place. 
And that's a very exciting place to be, and worthy of sort of a 
generation's efforts, I would say. 


You can take the business models that have sort of evolved over how 
do you get information out there? And there's advertising. There's 
royalty-based systems like books, and there's subscription-based 
systems, sort of who you know, a little club to go in and to be a part 
of. I like the royalty-based systems because it means that there's a 
simple buying structure. Anybody can buy anything, and there's a very 

easy method of distributing the proceeds of that. Advertising-based 
industries tend to consolidate, so you get things like newspapers, 
magazines, television. Those are sort of the advertising-based. And 
now you get these search engines that suck the air out of the room for 
other people because they can control the ad-selling network. And it's 
always been that way, so you end up with something quite different 
than you thought you were going to get, very consolidated, centralized 


In the information age - so, we're not talking about just railroads or 
software, where actually it is important to have distributed ownership 
of transportation and our software infrastructure. And so, we wrestled 
with the monopoly structures in those areas before. But in the 
information area, it's even more important that if somebody really is 
able to control what people think, how people are educated, then I'm 
not sure you can get your way back out of it, that if you have just a few 
players being able to control things at that kind of level, such that how 
thoughts evolve is influenced, even if it's not diabolically controlled by 
one or two corporations, it will influence the evolution of our society 
and who we are as people in a very corporatist, centralised model. 


If we're trying to figure out sort of where is the ICT infrastructure 
information technology going in 1 years, sometimes it helps to go 
back 1 years and try to see where we came from. Let me pose a 
question. Who owns the Internet? Who owns the Internet? And I 

Copyright Reserved, DTN B.V., 20 1 

Page 2 of 2 

_ D J ^V - N J 

think our answer to that question - both our perceived answer and 
maybe what's real, and how that might be changing, our perceptions 
might be changing - could give a really good mirror onto how we see 
what's going to happen. And how we see things going to happen 
might actually influence what's going on. 

So, 1 years ago, the answer to, "Who owns the Internet?" was 
laughed at. There was no one that owned the Internet. It was always 
a very - it's a scripted answer. It was like it's a network of networks. 
It's an open structure that anybody can go and create things on. You 
didn't need permission to go and build things on the Internet, and we 
had fantastic things to be created in that world, whether it's e-mail, the 
World Wide Web. The peer-to-peer networks in '99, 2000 were 
coming out so that you had the greatest music library ever made, 
which was built on top of Napster. People just poured their time into 
going, putting, making all of this available. It was the sort of the 
fundamental premise of the Internet. 

So, now we're 1 years later and trying to predict what's going to 
happen in the next 1 years. Let's take that question again, and I think 
it might be illuminating, and both as a perceived, "Who owns the 
Internet?" Is that question even more relevant now than before? I 
would say yes, that people are starting to answer that question that 
there is someone that owns the Internet, or some piece of the 
Internet, or some of it's controlled. Maybe it is, maybe it isn't, but 
people are starting to think that way. But let's take it as no, people 
don't own the Internet, and new and wonderful things are coming up 
because it's open, because there's a mechanism of getting to a 
distribution channel. You can get your ideas out there. 

I look at the top 1 websites, and I happen to know them from year 
2000, and I'm pretty sure what they are now. In the year 2000, the top 
1 websites were 20 percent of the attention on the Internet - the 
clicks. If you count all the clicks of all the people of the world, 20 
percent of them go to 10 organizations - pretty astonishing. That's a 
real concentration of power. They were mostly things like search 
engines. AOL was in there. I guess sort of the normal cast of 
characters that you'd kind of imagine - eBay was in that list. Amazon 
was in that list in the year 2000. 

Now, we have some of the same characters, but we have Facebook in 
there. We've got Wikipedia. We've got a search engine from China. 
So, there's some shifts that are going on in the top 1 websites, but 
what also has changed is the amount of time and attention people put 
towards the Internet. Television has continued to fall away as an 
influential medium, so people are spending more and more time on the 
Internet. So, while there's still - there are these important 
organisations in there, but they do change. They're not controlled by a 
few organisations that are going to stay in place forever, that the open 
nature of the Internet and ICT technology has made it so that there is 
upward mobility, even into the top ranks. 

So, we have these worlds. Let's take the answer as somebody does 
control the Internet, and then somebody doesn't control the Internet. I 
think this debate in 1 years, having a question, will be very important 
again, but even our perception in the next 1 years. If you answer that 
somebody does own the Internet, what does that mean to our 
organisations, our role in society and the like? And I think that that, 
companies becoming either defensive or open will start to really pivot 
on this idea, is there a man? Is there somebody that you're going to 
have to pay attention to or please to be able to be successful, or are 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 3 of 3 

_ D J ^V - N J 

we really in an area that a new idea can come about and have their 


Another thing that's changed in the last 1 years is we've shifted from a 
technology focus to a content focus. We used to be that really it's sort 
of who organised server farms and the like. It was who controlled the 
protocols, instant messaging 1 years ago, were the sort of the 
emerging technologies. Now, we've got more and more people going 
and figuring out how to own other people's information, whether it's 
YouTube or Facebook, where there are whole environments for 

Now, we're not avatars running around in a big, 3-D world. That's 
happening, but it's not the dominant force yet. It might be in 10 years, 
but we're starting to see whole environments that are built in platforms 
for people to interact in, but more importantly, put critical information 
about themselves. Data mining is all the buzz in the last 1 years within 
the government sectors and corporate sectors, but when we really 
look at it, very few people, very few organisations know how to use 
the information that is gushing out of our web servers, out of our IT 
technology. And those organisations may become more and more 
important, and more and more consolidated. 

More of the publishing infrastructure, what we think of as publishing - it 
might have been old newspapers or magazines or books in the old day 
- it might be something quite different that involves very large tracts of 
data, and organisations that control and know how to exploit these 
classes of information. 

So, what does it mean to have a social network where people know 
who all your friends are? What good is it? Who cares? What does it 
mean to be an organisation, to have something proprietary in a world 
where your organisation's information is going through somebody else's 
e-mail system? What does it mean when you're not using Excel and 
Word so much anymore as you're using these hosted services, where 
who knows quite where your business plans are going? And who's 
able to exploit that in interesting and new ways? 

Even if it's the threat of exploiting it, can help play into people's hopes 
and fears as to what it is our IT infrastructure is doing for us, or are we 
really just feeling like we're kind of a cog? And just the idea of freedom 
of open opportunity is an incredibly important myth towards creating 
an interesting world that's served us very, very well. We've had some 
really interesting technologies built. I think we have an opportunity to 
do much more of it, but I think it's going to be people's answer to the 
question, "Who owns the Internet?" that really will help divide the 


An amazing shift that's going on among all of us that are in our 40's is 
what our children are doing on the Internet, and how much trust there 
is in what's going on. And what they see is the benefit of going and 
having their social experiences be completely exposed. Their photos, 
who they're friends with, what they're doing all go through other 
people's computers. Those of us that were brought up by parents that 
had gone through World War II, the Depression, the McCarthy era of 
you can get in trouble, based on your associations - we were warned 
about these sorts of things, and it was just part of the air that we were 
brought up in. 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 4 of 4 

_ D J ^V - N J 

But this next generation is finding much more benefit to just sharing it 
all, to going and having your photos on Flickr, your Facebook. And 
there is just a - there's a trustingness that's going on at the populist 
level that may be warranted and may be not, but it's certainly going on. 
And there are organisations that are starting to learn how to exploit 
this trust that is built up, in ways that not right now seem like targeting 
of advertising. There's certain interactions with the government, where 
there's more and more information being shared with the 
government's intelligence and law enforcement. 

Since I run a digital library, we archive the World Wide Web, so 
people have - we have probably the only most public copy of the web 
that's existed over a decade's time. We collect a snapshot of the Web 
every two months, and it's used by, oh, about 500,000 people a day 
come in and use this resource for all sorts of things. And we get an 
increasing number of lawyers coming and demanding information from 
us, and it's kind of interesting because it's all publicly available. They 
could just use it, but they come and do demands with subpoenas and 
thuggish stuff that we have to go and do something for them. 

So, why is this interesting? One, it's just the growth curve, just the 
number of organisations that are using these resources to sue each 
other, that we're seeing this. And since we're a public entity, I have no 
problem going and telling you this. This must be going on with 
Facebook or Gmail or Yahoo! and whole teams of people are just 
dealing with law enforcement all of the time. Amazon will tell law 
enforcement, with a proper court order, what books you've read. 

So, those sorts of things that strike fear in us, that we really know in the 
20 th century is absolutely happening now. And we have a world that's 
very trusting. The good thing about this is all sorts of great things 

happen. You find all sorts of interesting people are a member of many 
different communities that overlap each other geographically, that you 
have several facets of yourself, and it's OK to be developing those at 
the same time. 

Our work environments are becoming much more fluid between 
what's your personal and work environment, and we're leveraging all of 
that so that - somebody yesterday was helping in sound design in our 
new building, but he's a programmer doing great things on cluster 
development, but he happens to also do a lot of things in clubs at night. 
So, he knows a lot about this and he's able to leverage his hobbies and 
his other interests that are highly developed, in large part because of 
the Internet and the other people that are able to interact in his work 
world. It makes him a much better employee, and it makes us much 
more interesting people because our information sources are so wide 
and diverse. 

It's a trusting environment, which is very, very good, but might really 
come back around in some interesting way. These next 1 years are 
going to be interesting in this way. How do corporations deal with it? 
How do governments deal with it, when there's an export of their 
information and is some level of control? I live in Silicon Valley. It all 
seems pretty good to me that the music industry really looks like it's 
being majorly contorted by Apple Computer. Google has got 
information on people from all over the world. We have Yahoo! here. 
We have Facebook. It all seems kind of good, but when you're at the 
centre. How does this feel if you're not feeling like you're at the 
centre? What if it feels like I don't really trust these guys to help out? 
How is it going to evolve? Is it going to be more centralised, less 
centralised? Is it going to come back to bite us? 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 5 of 5 

_ D J ^V - N J 

And it's at the content level that I think is important, not so much at 
the technological level. Yes, it's important who owns and controls the 
transport systems, but who controls the filters? What are the new 
applications that are coming out? Does everybody - is there going to 
be a new way of communicating that's able to be built on iPhones, or is 
Apple going to control that, such that it's not going to be kind of what 
they want to have out there? 

The Macintosh and the PC allowed anybody to write a piece of 
software and distribute it for free. It was hard to make money on it in 
the early '90s, but you could get it in front of millions of people. That 
used to be very difficult to do through the retail chains and controlled 
distribution structure, but you could get this thing out there. 

In these new worlds that are really dominated by things like video 
games, where it's really controlled environment - phones, really 
controlled environments, there's some hand wave about, "Oh, well, 
there's this network we have to protect." Is that really the issue? Are 
we in an era where new applications, new companies, new 
technologies for managing social interactions are going to be able to 
blossom? Are we going to build new and different navigation 
structures for taking this vast wealth of information, of personal, 
corporate and wide area of information that's becoming available, and 
making new answer engines that go and learn from these and extracts 
facts, finds contradictions? Those are some of the new things that are 
coming out. 

The Wolfram Alpha engine that is really what the Al community 30 
years ago was dreaming of, is now starting to come about. The reason 
why they're coming about now is the information is available, that you 
can go and just have curators that go and pull information from a wide 

variety of sources to make new information sources. 1 years from 
now, are there going to be more of these or less of these? Our 
answer of what's our perception of that will separate us. But also, the 
reality of it, I think, is our opportunity to shape our future. 

And I do believe that it's - the open thing just continues to surprise us, 
to shock us, that it's the kid in Lithuania that comes up with something 
that seems a little risque, a little bit odd, but is what is going to blow 
things up and make it a very interesting 1 years. How do we position 
ourselves with that kind of change? The same way we've looked back 
- how were we doing it the last 1 years? Did we do a good job? If 
not, maybe look at your kids and how trusting they are. 


So, some of the world changes that are going on, the role that Asia is 
increasingly playing, Europe, Middle East, how is it evolving? I get a bit 
of a point of view from where I sit by running a digital library and 
working with basically information flows. Where is it coming from? 
Who's doing what? Who's investing in the content layer of the 
Internet, not so much the technology layer, but the content layer? And 
where are ideas coming from? And increasingly China is where at least 
the hardware technology is really shifting, so that if you really want to 
know what the future of tablets look like, go hang out in Taiwan. 
Everything's so dependent on them. They're really setting the pace. 
Yes, maybe it still has brand names that still look American, but it's 
over. It's shifted, and when the hardware manufacturing, then the 
design follows and it really wraps around, what it is we can do next. 
And you get changes such as the touch screens that will really radically 
change what's going on, but it will be dominated by the Asian 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 6 of 6 

_ D J ^7 _Nj 

And I think that's great Let's spread it out somewhat The Silicon 
Valley just had a major win in the last two years. Wikipedia moved to 
San Francisco - interesting. So, it used to be in Florida - great idea 
blossomed - but they felt kind of out in the woods. So, the non-profit 
content organisations still are finding their way to San Francisco, and I 
think it's because that there's a sort of balance between openness and 
proprietary control that we still see explored in the Bay Area. 

The Grateful Dead gave away their concert recordings and did very 
well that way. Would that have happened in Los Angeles? Would that 
have happened in London? I don't think so. It happened in San 
Francisco 30, 40 years ago. The World Wide Web, CERN, Tim 
Berners-Lee went to Boston to go and do the World Wide Web 
Consortium in Boston, but it was really Silicon Valley that really took it 
and ran with it. But the Electronic Frontier Foundation, Internet 
Archive, Wikipedia, the Mozilla Foundation -these organisations that 
are building infrastructure, new operating systems - Linux Foundation - 
these things are in the Bay Area. 

So, does it mean that all the work is going on here? Absolutely not. 
More and more of these communities are worldwide, and more and 
more of our organisations have got interacting communities that are 
everywhere. But there's still something about the Bay area's ability to 
do this balance. Is it shifting to other places? Of course, all eyes are on 
China, Taiwan, Japan. How that will shape and influence, I don't know, 
but I think more and more of us are going to start learning Mandarin. 
More and more of us are going to spend time there. Where we might 
have spent time in Paris, we're going to put on our schedules going to 
Hong Guo [SP] or going to Beijing, and just to hang out, understand, go 
to conferences, learn, absorb. What does the world look like from 
over there? 

And certainly I've spent a couple months this last year, just spending 
time over there. And I've spent time over in Egypt and in Europe as 
just we have to, to just broaden up, to get an idea of the influences that 
are going on, because it's influencing our culture, it's our technology, 
the content, and the point of view on lives is all coming from these 
other places, as well now. 


So, we're prehistory in the sense that there's going to be a gap in our 
knowledge, because the writers are writing on word processors that 
you throw out. Do you have your electronic mail from 1 years ago? 
Probably not. It's too hard to keep these things now, but what does 
that - that might mean there's a real gap in the record, or it's a very 
selective, weird record. It's very deep in some ways and not in others. 
So, if it's in other people's corporate memory banks, it might be there 
by now. 

One of the great early websites of homesteaders, the Facebook of the 
day, was called Geocities, and it shut down. Yahoo! shut it down. 
They bought it for about $400 million back in the dot-com days, and 
last year they shut it down. And they, to their credit, they let us know 
at the Internet Archive that they were going to shut it down in nine 
months, and so they said, "Crawl like nuts." And we said, "Can you 
give us a list of all of the homesteads on it?" And they said no, because 
that would sort of violate what they thought the privacy interaction 
with their user was at the time. But if we knew of the site, if we knew 
of it from any which way, if there was any links into it, then feel free to 
crawl, to be able to make sure that there's original content set. 

And that's not just sort of historically interesting. It's documents that 
are referenced all over the web all the time. The Internet is vastly 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 7 of 7 

_ D J ^V - N J 

trusting of other people's sites. Websites have so many external links 
that we're all dependent on each other to survive. And when a big 
hunk of the web like Geocities goes away all at once, we have troubles, 
in terms of how the Internet's built. So, this is what some people are 
thinking of as a digital dark age. 


The Internet Archive is a digital library in the modern age. We're 
becoming a little less odd to people - I don't know, maybe as we're 
getting older. It's I 3 years old. Been archiving the World Wide Web 
with a snapshot every two months for the last almost 1 5 years. And 
this, we've made into the Wayback Machine, so that people can go and 
use a URL to find old websites. It's the out-of-print web. Statistically, 
we have 1 50 billion web pages from 50 to 1 00 million websites, that 
the average life of a web page is about 1 00 days before it's either 
changed or it goes away. So, it's a really evolving mush out there. And 
being able to have access to the older material means that you actually 
have a reliable web. And it's used by, as I said, about a half a million 
people a day. 

We also started collecting television, so as much as it sort of seems like 
yesterday's medium, it's still very important, and people had not been 
doing a very good job archiving. So, we've been archiving 20 channels 
24 hours a day- Russian, Chinese, Japanese, Iraqi, Al Jazeera, BBC, 
CNN, ABC, NBC, CBS. And it's sitting dark. It's not accessible, mostly 
from a money perspective. We're trying to get it available. 

We also collect movies and music, and people often upload this, or we 
digitise them and make them available permanently as reference 
collections. And we're also archiving books. We've got - we're 
scanning about 1 ,000 books a day in 20 scanning centres, including a 

Copyright Reserved, The DTN B.V, 2010 ( 

scanning centre at the Library of Congress, one in London, Scotland - 
20 of these things in five countries to go and make it so that books are 
available again. 

So, mostly we have now 1 .8, 1 ,800,000 out-of-copyright books, so 
anybody can download those. And we're also scanning things that are 
in copyright and making those increasingly available to the blind, 
because there's copyright exemption for that, and we will be launching 
a lending library so that kind of like when we were growing up, you 
could go and get materials from the library and then bring them back. 
We're focusing on the out-of-print, so that the in print, the publishers 
have sort of room to try to build a business model of books on the 

So, the idea of universal access to all knowledge is within our grasp, and 
we're making progress on that, as well as some other organisations. 
Google very dominantly is working towards this, but that idea of 
building the Library of Alexandria Version 2.0 is no longer just a hand 
wave. Significant percentages of all the published materials ever are 
now in digital form, and in different ways, are available to people. 


So, as we're making huge collections of information online, we're 
starting to see an evolution of a couple different approaches to it. 
There's one that comes from a tradition of libraries and one that 
comes from a tradition of products. And neither one of these are 
quite right. Libraries in the past didn't have information like social 
network information in them, and corporations had products that they 
designed, built, and owned, and now they're starting to accumulate 
other people's information into them. 

Page 8 of I 

_ D J ^V - N J 

And these two different points of view are coming out with very 
different futures. My favourite, we're very much involved in right now, 
and the big tussle is around books. As books are going online, Google 
is digitising these books, and they're doing it very fast in such a way that 
they can put it into their search engine and exploit it and sell it. And 
they're really taking sort of atop-down, "We own it all" control. 
Everything is contractually regulated. All of the user's trails are audited 
and kept, so that what everybody reads is owned by a corporation, the 
terms of service. 

So, that's one approach, which is really coming from a product design 
background. It just expands to be the library of all books. Then, 
there's, you take another approach which is the Open Content 
Alliance, the Internet Archive. We're working with about 1 50 libraries 
to digitise books from all over the world. We're digitising about 1 ,000 
books a day, ourselves. We're working with projects in India and 
China. It's the open world. It's a very open, exchange-oriented, a lot 
less top-town control. We don't keep usage information. We don't 
know who read what. We know how many people read something, 
but we don't know who they are. We never kept it because it's part 
of the tradition of coming from a library, and it gets spooky when you 
can go and track who reads exactly what. 

The ownership is much less dictatorial. We share things a lot more. 
We'll push a million books back and forth from continent to continent 
just to make sure there are backups in other organisations, because we 
see longevity as part of us. So, corporate control will live forever - 
trust us - versus the distributed, open world. And so, as books are 
going online, we're seeing a competition of these two approaches. 
Which will win? I don't know. Will it be a hybrid? I doubt it. I think it 

will really be dominantly one way or the other, and it's playing out now 
in a fairly dramatic way. 

Music, when it went online, really went to one corporation: Apple. 
And yes, there's third-party sellers that are trying to play a catch-up 
role, but it's really owned and controlled at this point, the distribution 
of commercial music. And the only way around that is a bottom-up 
reinventing music completely. So, it never went to a distributed World 
Wide Web of music. It went to a centralised store of music. 

I'd like to see a distributed world of books, a World Wide Web of 
books, a World Wide Web of information, educational materials. But 
the temptation of a single organisation, the opportunity of a single 
organisation, taking and controlling whole media types is one that can 
shape the next 1 years in a very interesting and important way. 


Cloud computing is an evolution of this networked world, which is 
fantastically efficient. The idea of software as service, or as outsourcing 
of different parts of organisations, or allowing interactions, is causing 
people to be very excited. But pulling it apart a little bit does make 
sense. So, software as service - it's a lot easier to make an application 
that runs on a server than it is to try to make an application that is a 
compiled piece of code that distributes and runs in everybody's 
Windows machines or Macintoshes or Linux boxes. You can rev it 
much more quickly. You can put new ideas into a service so that if you 
were running a company, trying to put ideas out there, putting up a 
website - even a sophisticated one that has software that downloads 
part of your browser- is so much more efficient. And that's sort of 
thought of as cloud computing - software as service. 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 9 of 9 

_ D J ^7 _Nj 

There's also other things called cloud computing, which is like the 
Amazon service of being able to be an outsourced IT system, and most 
of the start-ups these days are using those services so that they don't 
have to manage hardware. Hardware is really hard to manage, and it's 
hard to know how much you should get It sort of seems all very old- 
world that you have to go and pre-invest in hundreds of thousands, 
millions of dollars worth of hardware infrastructure, when you could 
just use as much as you want to, this utility model. 

The downside of that is well, it's on somebody else's server, and what 
does that exactly mean? But it's starting to be even more than just a 
software service or outsourced IT. There's starting to be these cloud - 
people are using these cloud services, meaning that there's these 
collaborative environments, shared information environments that are 
hosted, whether it's Google Docs, Facebook, Flickr. You're putting 
things in the cloud, and it's not just putting in the cloud because it's 
more efficient to go and store your things on somebody else's IT 
infrastructure. It's because there's new and added features that you get 
by having crowd sourcing of information connectivity, that you're 
putting your stuff with everybody else's stuff out there. 

But the metaphor of a cloud, I think, is deeply deceiving, the idea that 
it's sort of everywhere and it belongs to everybody is absolutely not 
true. It actually does belong to somebody, and it is going to be 
exploited within their own mental model. And we haven't seen the 
rise of non-commercial clouds yet, or things that are a little bit more 
trusted, things that don't give you the heebee jeebees if you really think 
about what information they have in them, and what level of influence 
they could have by having this information. 

So, clouds are starting to mean way too much, but I'm seeing it as 
these different areas, and the most powerful one is the interactivity 
between people that's enabled by having our software and services and 
our information in these combined mush out there, that then can be 
data-mined for good and for inappropriate reasons. 



I was involved in the earliest cluster computing search engines back 
before there was the Internet, helping the Internet come about. So, 
I've been involved in a lot of these things and seen a bit of perspective. 
What's surprising to me is how slowly things have changed. I think that 
really shows my failure more than - just how long it takes. The idea 
that we're in 20 1 now and we don't have textbooks online is just, 
how could we possibly be here? What does this mean that the 
maturity of the technology is still so lame? That there isn't at least 
some base level of free Internet access to all wireless devices now in 
our cities is, I find, really hard to believe. 

And it's happening, but so slowly and in such fits and starts and in 
clunky ways, that the government has really taken a backseat role 
where 20 years ago, a lot of this was happening in universities and 
national labs. A lot of it now is happening in either corporations or 
independent non-profits. What happened to the government? I mean, 
what happened to the next generation of information superhighway, 
and how do you level the playing field and sort of keep things under 

The government is just absolutely checked out in every - in Europe 
and in the United States, there's not an investment in large-scale, 
interesting infrastructure projects in the universities. They're just 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 1 of 1 

_ D J ^V - N J 

playing around the edges of how do you build services on top of 
Google in some putzy way? The new generation of supercomputers 
aren't coming out of these places. The national labs are trying to rent 
them from the Googles and Yahoos! and Microsofts out there. What's 
this about? 

So, the things that I find surprising is the shift and roles between dot- 
edu, dot-gov, dot-com, dot-org. A thing that I found very surprising 
and very encouraging is the rise of the dot-org, the non-profit, that 
that's been the - it's nongovernmental, international organisations that 
are building infrastructure. That they operate on a set of principles of 
privacy, of transparency, that they are - people feel it's OK to go and 
invest their time and effort, whether it's hobby hours, to build a 
Wikipedia, or going and building Amazon's reviews. You want to be 
paid for that, so there's this new sector that I find very interesting, very 
surprising, and I'm very intrigued with where that's going to go. 

I think we may be building another generation of government, if you 
will. They have their own rules and regulations. It's in its infancy, but 
these non-profits are interacting with for-profit corporations in very 
symbiotic, positive ways that this is, I think, a very interesting evolution. 
And I think it's come about because we've had basically governments 
that have just been off doing dumb stuff for the last 1 5 years - in the 
United States, the last 30 years. It's been in relative collapse. 


So, the high-tech non-profits are starting to really take off. They started 
really with some of the open source software moment. So, Richard 

Stallman - the grandfather of this whole idea - he made his Free 
Software Foundation be a non-profit. Linux Foundation is, again, non- 
profit. So, in terms of those that are governance over open source 
software, they're in the non-profit sphere. There was an attempt with 
Red Hat, which is a commercial company, to be a commercial owner 
of Linux, and it fried. It failed. Java is within Sun and we'll see how that 
evolves. It's sort of an uncomfortable place. 

But these non-profits are doing very well in these open source 
software areas, and they're starting to build full on operating systems, 
or browsers like the Mozilla Foundation is a non-profit building a very 
viable contender to browsers. There were browser wars 10, 15 years 
ago between big commercial players pounding it out. Now, it's open 
source is really the contender. What does this mean? Big hunks of 
what Apple ships is open source, and so, where Apple is leveraging it, 
it's owned and controlled, if you will, governed by non-profits. 

We're seeing this now happen in the commercial sphere. Wikipedia, 
Internet Archive, the Public Library of Science - all of these are non- 
profits, and they come with rules. You can't pay people too much in 
them. They're not buyable, which is very important for longevity. I 
met with Mitch Kapor. Mitch Kapor started Lotus, the important - the 
spreadsheet - important guy. But I saw him at the 20 th anniversary of 
EFF. So, he was one of the founders of EFF and the founders of Lotus, 
and I asked him, "So, what else have you been involved in that's lasted 
20 years?" 

He said, "Well, I hope my marriage does." But Lotus lasted I 3 years 
before it was bought and made into a brand inside some other 
corporation. So, I 3 years was sort of a major software company. 20 
years for a non-profit, Electronic Frontier Foundation -there's 

Copyright Reserved, The DTN B.V, 2010 ( 

Page I I of I I 

_ D J ^V - N J 

something going on, that there's the long-term, stable - even in this 
economic downturn - all of these organisations, on the board of a 
bunch of them, are all doing very well. We're all growing along. 
There's a need for it. 

So, this high-tech non-profit as building infrastructure, I think, is a very 
important, new capability, and I think we'll see it grow. And 
organisations that know how to deal with the non-profits, whether 
they're a government or corporations, and leverage this new way of 
harnessing creative, innovative, participatory, transparent, open 
organisations, will be a critical form of whether it's a successful country, 
successful companies. 


One of the most interesting things I've seen out of the Internet, the 
growth of the Internet, is something that probably most people don't 
really see. It's the ontology. It's dividing things - the world - into dot- 
gov, dot-com, dot-edu, dot-mil, and dot-org. It's fascinating, right, to go 
and cut up organisations of people into those top-level ways of viewing 
the world. It was done by Jon Postel, and I think it's one of the most 
insightful things of the Internet. So, what we've seen come about is this 
interaction between the dot-com and the dot-org that it's, I think, 
potentially one of the most hopeful things to come out of this whole 

So, the gasoline of capitalism, right, you can get people really excited in 
money and power going towards making something happen. But at 
some point, it hits a level of maturity, and does it end up then into a 
monopoly organisation of a corporation that owns it, or did it actually 
diffuse out into the dot-org world, the non-profit? Does it become all 
of ours? And sometimes seeing one, sometimes seeing the other. I'm 

hoping that, for instance, in the books area - as books are going online, 
it could go so that it was really spearheaded and the power was put 
towards it with the dot-com, but then they belong to all of us once 
again. So, we had - a lot of the early development in book scanning 
was done with government funding, but really it hit stride when Google 
said, "We're gonna just go and digitise all of it." And that caused 
Yahoo! and Microsoft to go and say, "Oh, well, we better do 
something about this," and they invested in an open organisation. The 
Internet Archive - we worked with lots of libraries. Microsoft put $ 1 
million towards getting the deployment of book scanning out there, 
and this came after foundations had put in a couple million dollars into 
the R&D. 

So, R&D was the foundations. The VC - Microsoft put a power 
stroke, but they didn't want to own it. They just wanted somebody 
else not to own it - interesting. So, Google is powering through it. 
They're working very hard. But we might see - and I'm hoping that it 
does happen this way - that the enthusiasm and power of a Google to 
go and say, "We can build a future that has all the books online," 
doesn't mean that all the books are owned and controlled by Google 
at the end of the day, that they belong to all of us, that the technology 
- and more important probably, the idea that it's possible - is brought 
to us by a corporation. And now, it's time for the public sector to 
stand up to it. 

So, we now have France going and investing I billion euros to go and 
digitise their heritage. So, that seems like a large number, but it's 
actually very small in the course of how much they spend on their 
libraries. If each country were to go and do this, we would have - or 
even one-tenth of that - we would have all of the world's libraries 
available online within the public sphere. So, where we have dot-coms 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 1 2 of 1 2 

_ D J ^V - N J 

starting to really do some powerful things, it might be that the dot-orgs, 
the non-profit, the more distributed Internet, that is a possibility, is how 
it actually is rendered in the future, so there isn't centralised control. 
Will it happen? Don't know, but I think it's our opportunity. 


If I were to go to the Oracle at Delphi now and try to ask some 
questions to try to get some idea of what it is is going to happen and 
what it is we should do now, the question I'd really try to - and I do - 
try to understand is what can we do now within our own small worlds 
to build a future we want to live in? What are the tipping points that 
are going on now? If we were to go and work hard at something to go 
and make it so that the future really is within whatever view of the 
world it should be - my world is I want an open information 
environment that's creative and fun to be in. I want whacko people to 
be able to go and share information in wild and interesting ways. I 
want to be surprised and have things come about. 

Anytime I'm in a different location or a different community, I want 
them to be very different from each other. I want a diversity. I want 
language to be flowering. I want poetry and all sorts of different kinds 
of things. This is the world I want. This is not necessarily what other 
people want. So, if this is my Utopia, what should I be working on now 
to create that, and what are the pivot points? What are the things that 
are - if you just push that asteroid a couple degrees now, the world's 
very different later. And what actually doesn't matter? What is just 
redundant with what other people are doing? What's just playing out 
the roles, or being part of the cog in the wheel, that frankly it doesn't 
really matter if you just do another one of those - eh. But what are 
the important things now that are going to shape the future? 


The easiest pieces that I see right now that if we were to invest a small 
amount of time and effort, societally, if we were to just build an 
underlying infrastructure of a certain set of services that are free as air, 
that are just like government - what's the equivalent of the road 
system or clean water, but in the information age? And how do we go 
and organise ourselves to create those? I think it's open source 
software. I think it's things like Creative Commons licenses - really 
smart on Larry Lessig's part. What we need is a good set of licenses. It 
cost a couple million dollars to go and build those, and it's an 
organisation that costs about two million dollars a year. And we now 
have licenses that allow us to share. Everybody benefits because Larry 
Lessig and a small band of people set up a set of licenses. It's what 
government should have been doing - it wasn't. It's going on in non- 
profits now. 

But what are some more of those that are high leverage point parts of 
infrastructure? And it turns out that a lot of these don't cost very 
much. Some of it's regulatory, going and making sure that there are 
open bandwidth so that there's Net Neutrality, so that there's 
possibility of new entrance, so that there's swathes of the wireless 
spectrum that if you just use any - if you use WiFi, you're allowed to 
use it. 

WiFi came about in this awful little part of the spectrum that was 
where microwave ovens were, and that was the only reason why it 
wasn't licensed already. It wasn't packaged and sold to some 
corporation, and we have an amazing future because there was a piece 
of unlicensed spectrum. What's the equivalent now in the content 
area? What's the information that wants to be free? Is it government 
information with no copyrights, universally accessible? That's available 

Copyright Reserved, The DTN B.V, 2010 ( 

Page I 3 of 1 3 

_ D J ^V - N J 

in the United States, or at least, that's the law in the United States, but 
is it really happening or is it getting locked up? Is Europe going with 
open information? 

Regulatory information is not even free. Laws are not even open and 
free. I think there's an underlying bedrock of information that we need 
to operate as a society. It doesn't cost us very much to put up there 
and put out there. It doesn't require selling it off to an organisation, 
single corporation to own and control. I think small investments like 
that can help us create a fantastic and interesting world going future. 


There's this myth that openness dissolves and is anti-capitalist, and 
that's absolutely flip opposite. If you take capitalism, in its best sense, as 
a competitive environment where lots of people get to contend and 
play in, openness is absolutely critical. So, if we take WiFi as an 
example, where it's a very narrow piece of spectrum where we've had 
lots of interesting things happen, billions of dollars of organisations have 
been able to be created. Lots of interesting, unpredictable things 
happen, but you take a couple megahertz up, couple megahertz down, 
you have owned and controlled and it's a desert. 

So, in a very narrow - it's because it was open that we could do 
something interesting. We're seeing the same thing with software. 
The Internet, because it was - the protocols were open - it caused 
billion dollar corporations to come about. So, openness is the best 
thing for capitalism. Closed is the best thing for monopolies, and 
monopolies are a starved desert of information and innovation. 



So, in the 1 980s, there was this idea that you could build a question 
answering machine. You can make a machine that's interesting to hang 
out with, a machine that you wouldn't instruct. It would be fun to be 
with. It would know enough, and the computing has been going along 
on the right curve. The storage of knowledge has been absolutely on 
the right curve. And we're now starting to see the interaction between 
the computing and the knowledge bear fruit in ways that are somewhat 
similar to what it is was dreamed of back then, where you could say, 
"Computer, please tell me the answer to this." 

We're starting to see the first cell phone apps that can answer 
questions that you ask just that way, and the expectation of the 
computer in Star Trek or in 200 1 : A Space Odyssey is starting to come 
about. It's so slow and so subtle that it's hard to see it arrive at any 
particular birthday. They're not shocking and being announced all at 
once, but the search engines, I think, were one of the first biggies that 
really blew people's mind, where you could ask it a question. And it 
didn't really know the answer. It didn't really put together the answer. 
It just regurgitated an answer that somebody else came up with. 

Wikipedia really is basically that. I mean, you can almost type the titles 
of articles, and I know people that do. They don't search Wikipedia, 
because articles are designed around pretty much all the questions and 
answers that you might want to answer. And people put it in, and it's 
the idea of going and crowd sourcing, or everyone putting together 
that knowledge base was not what we were expecting 20, 30 years in 
the artificial intelligence labs around the world. 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 1 4 of 1 4 

_ D J ^7 _Nj 

But it was a pretty interesting answer, but we're starting to see 
computers start to put facts together from multiple sources, and be 
able to make those available to us in interesting and new ways. That, I 
think, we're going to become absolutely dependent on, so that it's 
going to be - the idea of being offline is going to make us sort of really 
be on vacation, or almost asleep. The idea of being very far away from 
our personal, digital assistants, it's going to be really difficult. I mean, 
we're going to be really almost going into a meditation or a trance 
that's separated from the world. 

I wonder- if you were to take a bank now, and you're going to have 
the choice, the draconian choice, of kill all the people or kill all the 
computer systems in the bank, and you were trying to maximize the 
stock value of the thing that's left over with, I think you'd probably get 
rid of all the people. The computers - and what's known by those 
computers - is much more important than what's actually known by 
the people. The people are more replaceable than the computers. 
That's a little odd. 

What does it mean if we do that with another level out? So, we're 
dependent on question-answering capabilities of our devices, of our car 
navigation systems, of the Internet search engines, whatever they 
become. And when that really evolves - education system is hopefully 
going to change enough. But it means that these computers are 
weaving into our lives in a way that we can look to the science fiction 
of 50 years ago and see it absolutely coming true. And I find it very 
exciting, very interesting. I'm finding it fun to play with my computer. 
They're interesting things you can play with, and I learn from my 
computer, and is it teaching me? In some fairly real way, yes. I see this 
area is going more and more in this direction as computing, the 

content, and the networks all come together to build a library that's 
worth talking to. 



So, the search engines of today, where you go and ask a question and 
you get a whole document that's kind of an answer, is very library-like. 
When I went to library school, we were taught a reference interaction. 
So, when a patron comes and asks a question, you don't just go and 
say, " 1 943." You say, "Here's an answer toward - to that," so that 
people can explore not just the answer, but around it, what's the 
context, who said so, that type of thing. 

That's what the search engines are. People's tolerance for this is low. 
They would really much rather have, " 1 943," or they'd like to know, "I 
should be doing this." And the evolution of those sorts of systems is 
starting to come about, that's much more general than going to the 
United Airlines site to go and find out whether your flight's delayed, 
that we're starting to see question-answering machines come about, 
whether it's Siri or Wolfram Alpha. They're really very sophisticated 
and so, people are going to want these much more quick interactions. 

So, I think we'll see the evolution that you'll use libraries and search 
engines kind of like libraries in the future, where that's when you're kind 
of exploring. You're not quite sure what. You want a more in-depth 
answer. You're not quite sure who to trust, but much more of our 
interactions with our technology is going to become much more fluid 
towards natural language and expecting precise answers coming back. 
Very exciting. And there are going to be people that try to - 

Copyright Reserved, The DTN B.V, 2010 ( 

Page 1 5 of 1 5 

_ D J ^7 _Nj 

organisations - that try to extract factual information from the fluid 
search engine-like stuff to go and build these crisply-interactive 

And we'll see, I think, some of the fruition of these systems and start to 
be pretty interesting and fun. I'd like to see how these things interact 
with each other, because right now they're in silos, these silos being 
different organisations where - can we go and build a World Wide 
Web of question-answering machines? We've built a World Wide 
Web of search engines and web-type things, but can we go and make 
it so that - I know how to answer questions about baseball better than 
anybody. So, can we make one of these answer engines go and probe 
me when there's an appropriate time, so that there's a new 
infrastructure of question answering, as opposed to just document 

Open question, we haven't even seen people try. Mostly they're just 
trying to get anything to work at all, and some of the early things like 
Wolfram Alpha are extremely exciting to me. 


Any place there are people will play a role in the future of the Internet, 
because fundamentally the Internet really is about the people. Now, 
they may go and give control to other organisations that'll Hoover 
them all together but I think, all in all, really this thing comes from us. 
And if we let it get out of control, if we have governments that are 
asleep, we can end up in some pretty bad times of focused, 
monopolistic control. But it's so inefficient to run an economy, based 
on monopolies that hopefully it will eventually dissolve. One of the 
best things is to just not have them happen in the first place, but will 
Europe play a role? Absolutely, as long as they're sort of alive and 

kicking, and we'll see India and China play- and microclimates of 
people that are distributed over the world will play a role, as long as 
we keep the Internet flat for new applications to be created and 
distributed without approval from somebody that controls either a pipe 
or a device. 

Copyright Reserved, The DTN B.V, 2010 ( 

Page I 6 of 1 6