Visit Your Local PBS Station PBS Home PBS Home Programs A-Z TV Schedules Watch Video Donate Shop PBS Search PBS

 

Main NerdTV Page NerdTV Archives
 
send feedback rss/podcasts I, Cringely Sponsored by Datapipe Creative Commons License
transcript

NerdTV #4: Brewster Kahle

Bob: So, Brewster, welcome to NerdTV.

Brewster: Thanks. Good to be here.

Bob: What - you know, most people haven't heard your name. People in Silicon Valley in the internet world know you very well, but people who aren't in Silicon Valley or aren't in the internet world don't. So what are your - what's your claim to nerd-dom? What makes you a nerd?

Brewster: Certainly not fame. (Laughter) What makes me a nerd? Well, we call them hackers back at - at school and as those who take the technologies and for it's own sake where can it go, what can we do that's really good to build up a new world using our technology. And so it's not about having your signature all over things. It's not about fame or fortune. It's about what can we do with technology that would be interesting, fun, something that people would have noticed.

Bob: Okay.

Brewster: And - and noticing the technology, not necessarily the people behind it. And so I guess that's - that direction has been constant for me for 20, 25 years, and so, I guess that's what makes me a geek or a hacker or a nerd.

Bob: Well, can you - can you go back 20 years and sort of tell us what you've been doing for the last 20 years?

Brewster: Oh gosh. Sure. I was back at MIT and there was this whole idea of what can you do with technology. This is the late 70_s and so it's still hang over the 60_s where there's still some sort of idealism of some sort of okay let's - let's see if we can end up with a better place at the end of our stay here than when we started out. And there's two big ideas that were floating around. One was encryption, so to protect privacy, which people seem to be perfectly happy to throw away, you know, for any reason whatsoever. And the other was digital libraries. The idea of going - making good on the promise of - of having information at your fingertips. Everything out there, the Library of Congress on your desk, whatever the - sort of expression -

Bob: Yeah.

Brewster: - of that as we may think type of idea of having the information there. And it seemed like the digital library thing was just too obvious somebody else was going to do that, so I went off to work on the encryption piece and try to make that possible to encrypt telephone conversations, really cheaply and easily so that it wouldn't - you would probably have to get a privacy two people with some other feature than privacy, because people aren't really that motivated it seems to protect their own privacy.

But when - so I learned how to build chips and how to do that. But I couldn't figure out how to do it cheaply enough to make that such that it would be a Radio Shack $5.00 thing, it would be a $2,000.00 thing and then you're serving people that really didn't need to be served. Either the large corporations, the mafia, you know, the - the secret sides of the government, which didn't need my help. So, said okay, well, it's going to take too - it's too hard to do the encryption thing, so let's go off and do this digital library thing. And Danny Hillis had this really great idea of building a machine that was fast enough to be able to do something. And the - so around 1980 there's, you know, there were Vaxes around.

They were good at doing accounting and sort of numbers, but the idea of actually storing books, much less images and also movies was still pretty far off and crunching on these was really, really slow. These are machines that if you spend a couple hundred thousand dollars, you've got a MIP machine, a one megahertz machine ________. And so these were pathetic, and it was - they weren't going to get there. And so Danny Hillis had this idea making a super computer. It didn't seem that hard, you put together these chips and - and make it go. So I - I said okay, let's - let's spend some time seeing if we can get a computer that could actually serve this stuff. So I did it all with chips and boards with this connection machine. Then we spun off a company called Thinking Machines, which made these parallel super computers. And that went for about ten years, that - that company. And it did actually quite well for a while and then they basically had a hard time making it programmable. And it - and the - the company collapsed.

But we were able to see the future. And that's what a Xerox park idea give people too much computing power. So whatever the computers in five years are going to be, give it to them now. So we had these workstations that were million dollar workstations, and it allowed us to play with things that were going to happen. And so we put on search engine on this super computer, so we could actually index a gigabyte of - of text which was unheard of at the time. And we were able to do some of these search engines and put them in the Dow Jones and - and - and the like. And found that well, even these search engines that we could build on these super computers had lots of great information in them, but it didn't cause the sun to come up a different color. (Laughter)

Right? Finally, people could go and do this matching of going and finding things across all sorts of things. We could do all these things, but it wasn't widely used. The question was, why not? And there were lots of different theories. Well, it cost a dollar a minute to use those systems. A dollar a minute is more than you pay your shrink, so its, you know, it's just know something you want to just spend time babbling with. Another is, maybe you didn't have the right the information in it. We didn't know. So at the - the end of the '80's, we said okay how do we go and rethink this thing? And so a couple of us set out to - to build on top of the internet service, you know, still defense department, university structure in the late '80's, a publishing system that allowed people to integrate their personal information, corporate information and wide area information in one interface and be able to search this and browse it easily. The idea, which was, you know, sort of looked a little ludicrous at the time - but back in the late '80's you said why don't we have executives turn to their computer to help them answer questions. Which actually, you know, was not the way it was done.

Bob: No.

Brewster: Usually you - you call somebody up, you know, and wrangle them and they'd figure it out and they'd plow through libraries or through these databases and come up with things. So we - we put this together as a joint project between thinking machines, Apple computer, KPMG and Dow Jones in - in 1989. It was the first Internet publishing system, it was called WAIS. And the idea was to try to get people to publish online. Because only if we could get people to publish online, could we then get a library built. And it worked, people were thrilled with it. And so in 1991 we made an internet freeware version. I've really liked this idea of open source. So I really wanted to play with that to sorta see how far could it go. And these general purpose computers that - you know, with the Apple computers and the PCs that have become ubiquitous by the late '80's, you could go and give away software and make somebody else's computer into your application, which is fantastic. Again, it's sort of - you - you kind of don't remember back before that the used to be to get things to lots of computer. You had to break into a retail market, you had to go and figure out how to get it through distribution channels. It was a nightmare. But we can actually make a free piece of software, have people jut use it, and then suddenly our application gets out there. And this - was something we wanted to play with. So we made a free version of the - of the client software, and we posted - and also on the server software, made a free version, and we posted one note to one email address once in 1991 and I just watched it spread. This was - and it just went nuts. And people started setting up WAIS databases that are a lot like Gopher servers or web servers, but they're earlier versions -

Bob: Sure.

Brewster: - and they're based on open protocols and - and alike. And it did really well. So we basically got some - some traction there. Apple was very slow in doing their interfaces, unfortunately. So the free versions were sort of what were majorly used and when NCSA, the super computer center, came out of Mosaic - Mosaic was named that because it was a mosaic of different protocols. The WAIS protocol, the Gopher protocol, and the web protocol.

Bob: Yeah.

Brewster: Sorta all in one interface, the ideas are just kinda being interfaced all of the systems. And that started taking off and the web interface, of course, just - just far superior and the protocols were - were easier for people to adapt to, because it wasn't a binary protocol. And that did really, really well. And so our job then turned from promoting a protocol in the like, to trying to get publishers online. We wanted to anchor the open internet with publishers that would make it such that when the proprietary guys came to town, they'd have to play the open game. So we worked with Encyclopedia Britannica, The Wall Street Journal, Dow Jones stayed in -

Bob: Yeah, sure.

Brewster: - New York Times, government printing office, the White House, we put them all online. This was in '91, '92, '93 to go and anchor the open - the open world. And it worked by - when - when Microsoft went and changed their strategy to respond to the internet as opposed to trying to kill AOL, that was sort of an indication that okay this open internet thing is doing pretty well by the monopolous and the AOL was also geared over somewhat towards to making gateways to the internet. So we thought that we had a pretty good shot. By '94 the open structure was working pretty well, and publishers were jumping on board and they're trying to make money based on whatever mechanisms they could. We were the first advertising - we put up the first advertising base system with CMP and the first restricted access service with Dow Jones, The Wall Street Journal to try to get subscription based revenues. To try to get people to make money by publishing on the net.

Bob: Yeah.

Brewster: By '94 that was sort of rolling, and so sold that company to AOL - that - the resulting companies of WAIS to AOL, and to try to get -

Bob: Why?

Brewster: Why? The thing I loved about AOL is they had a business model that could work. They - back in the '80's and '90's - early '90's, AOL would charge people about $6.00 an hour to be on their service, and 10% to 15% of that gross revenue went to the publishers that were making the experiences that the - that the people wanted.

Bob: Yeah.

Brewster: This is a royalty model.

Bob: Sure.

Brewster: It's not an advertising based model. It's not a - a - it's a royalty based model. So all you have to do in the AOL world is be popular and you get paid more. This is a terrific business model. I love the (Laugher) - the - the books world had that business model. It evolved it around 1,600 and I've always been a fan of libraries and books and sort of how did that all evolve. And until about World War II to a little bit afterwards, the - the books world worked really, really well. You had many authors, many printers, many publishers, many book sellers, there was nobody with control. And there was a mechanism for very easy user experience which is you buy a book and the little bit of money trickles back up stream.

Bob: Yeah.

Brewster: Books worked great. And I thought that basically we could do the same thing on the internet. And I liked the 10% to 15% model. And so we went and joined up with AOL to get more publishers on board - on to that general model.

Bob: Yeah.

Brewster: But this was just at the time that AOL was expanding colossally, and they made a change in direction. They said, maybe we shouldn't pay publishers to get to our - for their works, maybe they should pay us to get to our eyeballs, one of these great expression of the late '90's, eyeballs.

Bob: Yeah.

Brewster: So they flipped things around and they started to go towards more of this cable model is - you know this we're in control of enough people that you have to pay us to get to people.

Bob: Yeah.

Brewster: This blew the royalty model. And they stopped this royalty model, at least they didn't do it for anything that was on the web.

Bob: Yeah.

Brewster: So where we were trying to get the web and their business model put together to basically make something really great work. They could have been the internet. But they made a different decision. I can't say they did it wrong, I mean, you know, it's not for me to say that Steve Case, you know, should have learned from us and done it that way. But - but it wasn't what it is we were looking for. That wasn't what I had wanted out of the net. It wasn't what I wanted out of trying to get the publishers on board. And so I left that company after that sort of decision had been -

Bob: Yeah.

Brewster: - made and talked with Steve Case about are you sure you want do that?

Bob: Do you ever have a chance to go back and say I told you so?

Brewster: No, no. It's not an "I told you so". I mean these guys are really, really smart.

Bob: Yeah.

Brewster: And they did really, really well.

Bob: Yeah.

Brewster: I mean, you know, AOL blossomed into a fantastic set of services. I told you so? No, no, I mean - I just had a particular point of view and - no, these - these guys are much smarter than I am, so no.

Bob: Really?

Brewster: There's not - oh yeah.

Bob: Well, they seem to be a decline lately?

Brewster: Companies always do, they go up and down. I mean, so that - that - the idea that a company eventually goes down, doesn't go an argue against the people that were doing it.

Bob: Okay.

Brewster: So, you know, I think AOL's a fine company. Or - and - and they were really blossoming then. They've - I - I - you know, looking back on their Time Warner acquisition, I'm not sure - you know, that - that seemed to be just such a rocky road. But acquisitions always are. So, I don't know, we'll see.

Bob: So you left WAIS with - with AOL.

Brewster: AOL, and the idea is to go and build the library. So we built - gotten enough of the publishing and really left that era with one major sadness. We didn't put in place a really good business model on the internet. We blew it basically. We had ads, and ads ends up with things like television or newspapers or magazines. You get a very - and usually there are not very many companies that control those.

Bob: Yeah.

Brewster: So ad based systems don't make the many, to many, to many, to many, to many thing that we were really trying to get to happen on this internet. And so I'd say that's probably the biggest sadness when I was leaving that era of the internet, is we didn't put in place a better business model in advertising. And we're still suffering from that. It will make for colossal players as opposed to a distributed environment -

Bob: Yeah.

Brewster: - of - of people that are making money. You have a distributed environment of people not making money, and that's about the best we can say about the business model of the internet at the moment, which is - yeah, really sad.

Bob: A distributed model of people who are making money.

Brewster: Are - are not making money. (Laughter) We have a lot of big players making money.

Bob: Yeah.

Brewster: So we've - so that's kind of where we stood at that point, but I didn't feel like I was in a much - you know, a very good position to try to put - put in place in this model thing. Said all right, well, you know, what was the original point of this. The idea is to build a library. So I said all right, let's do that. So this is early 1996. So Netscape had gone public, AOL and Microsoft were adapting to this open internet, lots of publishers were on board, things were - things were moving along pretty well. So I said all right, why don't we start by going and collecting the __________ materials, things that aren't being collected very well. And so we started with the web. And we started two organizations at that time. One was the Internet archive and the other as Alexa Internet. We started a nonprofit and a for profit at the same time. Bing! And we started it with a contract between the two. That the internet archive would receive a copy of everything that Alexa Internet collects with a six month time delay.

Bob: Okay.

Brewster: So six months to knock off the commercial edge. This isn't life plus 70 or 100, you know.

Bob: Yeah.

Brewster: A hundred years of copyright.

Bob: Yeah.

Brewster: This is six months.

Bob: Yeah.

Brewster: Right, to go and take the commercial off of it, and then the information would go over to the internet archive. So the Alexa Internet was designed to be a catalog for the net. It was to try to basically help in the navigation when the search engines started to become, you know, worn out in terms of just handling too much stuff.

Bob: Yeah.

Brewster: And it turns out these - these search engines are still doing great -

Bob: Yeah.

Brewster: Dealing with billions of pages, which was a complete surprise to me. I'd been involved in the search stuff for years and years and years and years, and these search engines are still doing amazing things. Based on meta data, right, -

Bob: Yeah.

Brewster: - the next thing was - was meta data, which was a term - just claimed by Bill Dunn. He was one of my mentors.

Bob: I didn't know that.

Brewster: He was at Dow Jones.

Bob: Yeah.

Brewster: And he - he was at Dow Jones and always saying that the meta data's more important than data itself. This was back in the mid '80's when - so -

Bob: And meta data -

Brewster: - he's - he's a complete kook.

Bob: - meta data being the data about the data.

Brewster: Data about the data. So Google is - uses as a ranking mechanism how many people link to an article.

Bob: Yeah.

Brewster: Right? That - that's how important that - that's how they can find out whether something's important or not. Not what the words are in it.

Bob: Yeah.

Brewster: It's who thinks something about it. It's meta data.

Bob: Yeah.

Brewster: Alexa was also based around meta data. It was people who visited this site also visited this site.

Bob: Uh-huh.

Brewster: So using people's trails as a mechanism of finding what's important out of the net. So the first of the collaborative filtering things, before Amazon was doing it or before Google was even born -

Bob: Yeah.

Brewster: - Alexa was doing these paths and making these available to people through this toolbar.

Bob: Now how did you do that? How did you track the paths?

Brewster: We made a toolbar to try to give you an idea of if you're on this page, these are the other pages you might want to see.

Bob: Okay.

Brewster: And it was a give and take. So the give to people was, if you're on this website and the little toolbar would sort of update on every page turn that you went to - if you went to page, it would go and say okay you're here, you're on this site, these are the other places you might want to go to. So the way that we did that is it would make a request back to our server saying okay they're on this page, what should I display for them.

Bob: Uh-huh.

Brewster: As a side product we knew what page people went to. We didn't care who was who, -

Bob: Yeah.

Brewster: - we just wanted to know how many people went from one site to another. We found that people are very good at ditching crap. (Laughter). They just - they're not going to stay on a website long.

Bob: Yeah.

Brewster: That is pathetic. People actually are - are not dumb in that way.

Bob: Yeah, yeah.

Brewster: So if you watch people's trails, you can actually find what sites are worth staying on -

Bob: Yeah.

Brewster: - when they're on particular tasks.

Bob: Yeah.

Brewster: And going and data minding this stuff on the whole, you could make a service that was useful to people. And it - I - it's still around, it's still be used, I think, by a hundred thousand or a couple hundred thousand people to - to help them out.

Bob: Now is that a big enough sample to tell you something about the - the net overall?

Brewster: It seems to be. Some of the things we were able to do with the Alexa data is find out how concentrated are people's web activities. So we found that 20% of the web clicks are on only ten sites.

Bob: Wow.

Brewster: The - that was world wide too.

Bob: Yeah.

Brewster: So the level of concentration of power based on - on the web is like nothing we've ever seen since the Roman Empire.

Bob: Well, what are the ten sites?

Brewster: I'm sure they've changed slightly over - over -

Bob: Yeah.

Brewster: - time.

Bob: Well, what are some of them?

Brewster: They're basically the portholes, the search engines.

Bob: Okay.

Brewster: So even though those guys weren't making money, you wonder why did those sell so - for so much, and why were they valued so highly.

Bob: Yeah.

Brewster It's that.

Bob: Okay.

Brewster: It had so many people going through them. And then the next - the top hundred sites got 40% of the traffic, a thousand sites 60%, 10,000 sites 80%. It was in this semi log graph -

Bob: Yeah.

Brewster: - and very, very flat. The interesting thing about it, and one is there's a lot of concentration of power in a very few, but it also has a very long tail.

Bob: Yeah.

Brewster: To use the current term for it. It has a - to get to the last 20% of all traffic goes to hundreds of thousands of sites.

Bob: Yeah.

Brewster: Which is interesting, which is different from magazines or newspapers where you have basically you're either in or out.

Bob: Yeah.

Brewster: You know, they're only so many and they're not many or television or the like. So the - the - then that does have a long tail. The other thing that was interesting is a very flat curve. So if you had something good, you could move up the chain. Which is different from being in the retail distribution business, where you're either in or you're out.

Bob: Yeah.

Brewster: And it's very difficult to break. The net still makes it easy to break in. If you have a really good site, you can move up the path and - and move up popularity, as we saw Google do in the last few years for instance. So this is - so any way, by - by studying usage logs, we're able to get kind of an interesting overview of what was going on on the net and making sure, you know, it's anonomized so that, you know -

Bob: Now - now -

Brewster: - not spooky.

Bob: - what do you do with that knowledge?

Brewster: What Alexa did, is they put out this toolbar -

Bob: Yeah.

Brewster: - to go and say if you're on this page, you can go and see this other page.

Bob: Okay.

Brewster: And that - that was the circumstance, that's all.

Bob: Yeah, but did - did Alexa have a business model?

Brewster: The idea was on this toolbar, you'd be able to contact - content of specific ads.

Bob: Okay.

Brewster: We tried to go and sell ads -

Bob: Yeah.

Brewster: - and we just could not make it go.

Bob: Yeah.

Brewster: Couldn't - you know, couldn't figure it out. So we ended up selling that company to Amazon.com -

Bob: Okay.

Brewster: - that was mostly interested in the ability to do data mining for cheap. All right?

Bob: Sure.

Brewster: Because Amazon wanted to crunch larger data, and when they started going around to go and find who knew how to do data mining, well, Alexa did, because it knew how to do data mining at the scale of the web which is a lot more than the number of books out there.

Bob: Yeah.

Brewster: So they became interested in Alexa and bought the company, but kept it going as Alexa with its name as a separate organization and just tapping in on the - the expertise sort of as in-house consultants, if you will.

Bob: So is Alexa still in this building?

Brewster: Alexa is still in the Presidio, we're in the presidio here.

Bob: Okay.

Brewster: And this was the original building of both the internet archive and Alexa. And when Alexa grew, it grew to another building that's also still here in the - the national park of the presidio.

Bob: And you still get things on a six month delay?

Brewster: Absolutely.

Bob: Cool.

Brewster: It's completely great. And Jeff Bezos was - was wonderful when we were talking about being acquired. I said I tried this acquisition thing with AOL and stuff, you know. How do we - it's pretty hard to make an acquisition work. He said, well, I don't know, you've been through it, what do you suggest. I said well, I know how to run a company, but I don't know how to run a division.

Bob: Uh-huh.

Brewster: He said okay, let's structure it like a company.

Bob: Oh.

Brewster: And I said, well, and there's one other thing you should know. There's this contract. (Laughter) And he said, six months. Well, that might be a little bit short, but I'm okay with it.

Bob: Really?

Brewster: Yeah.

Bob: He's one of the good guys.

Brewster: Jeff Bezos is an inspiring man in many, many ways. And even though, you know, there was - we acquired - the Alexa Internet was acquired in - in 1999, I stayed on _til - for another three years. So, you know, and it's sometimes hard when you're running your own show to go be part of something else.

Bob: Yeah.

Brewster: But he left enough slack to make it interesting and inspiring. He's - he's an interesting man.

Bob: Yeah. Yeah, I've - I have an old interview with him. I'll have to find - do a new one.

Brewster: Yes, oh he's - he's fantastic.

Bob: Yeah.

Brewster: So yeah, I - I - I've been very happy to have gotten to work with some of just the top -

Bob: Sure.

Brewster: - people in this - in this field. It's - it's been great.

Bob: But you're now you're devoting all your attention then to the internet archive?

Brewster: Yes, now is the real time to just make the nonprofit sector of work again. It's kind of odd. In 1991, I'd be in these meetings where I'd hold up my hand and say I'm the token dot com in the room. I'm here to help publishers make money by publishing on the net. And this was - oh well, you know - because this was, you know, in the land of when it was academic.

Bob: Yeah.

Brewster: And it was just government people in this whole for profit angle was new. Now it seems to be completely the opposite. I'm the token - token dot org in the room. (Laughter) I - I'm trying to make it safe to have a commons. All right, to make it so that there's old traditions of the computer industry and also society in general are upheld like libraries. And so we're working a long to try to get the library system really built up. And so the internet archive has been actually fantastic. It's - working in a nonprofit has been also quite interesting. You're allowed to have a more complex motivation that just shareholder value.

Bob: Uh-huh.

Brewster: Like you're trying to - you know, that - that's about all -

Bob: Oh there is no - there is no shareholder value.

Brewster: There is no shareholder. There are no shareholders. You're trying to basically serve the mission of the organization and you're given tax incentives to do it.

Bob: Yeah.

Brewster: Which is - it's really pretty great. So you're allowed to have sort of broad interests in trying to help things be better, whatever your idea of better is.

Bob: Yeah.

Brewster: And so the internet archives are just fantastic. Well - so we started with the web and we made that available through the Wayback Machine.

Bob: Yeah.

Brewster: Then we spun off another organization called the television archive, to archive television, because it turned out nobody did - is really doing a widespread idea of television. And then we started doing books, music, and video. And our offer has been to give people that - that - that have materials that they want to share and broadly define belong in a library, unlimited storage, unlimited bandwidth forever, for free.

Bob: Heck of a deal.

Brewster: Heck of a deal. So the idea is in this country, at least, it shouldn't cost you to give something away. (Laughter) You know, if you give something to the public or, you know, nonprofit or - not only does it not cost you to give something away, you get - you know a pat on the back and you get a tax incentive, except on the internet.

Bob: Uh-huh.

Brewster: You give something away, and if it gets popular, you can go under.

Bob: Yeah.

Brewster: You can - you might lose your house, that's with bandwidth fees. All right one Slashdot can be the worse thing that ever happened to you. It's just dumb. It's no way to run a culture.

Bob: Yeah.

Brewster: Right? If we had a royalty system, such that if your thing became popular, you started to get paid based on -

Bob: Yeah.

Brewster: - all those usage charges on the internet, that would be wonderful.

Bob: Yeah.

Brewster: But it goes completely, insidiously the wrong way. So we're trying to plug that hole by basically making the library a fee- a free place to post things.

Bob: How - well, how do you do that? I mean, there are two - two issues here. One is a technical issue.

Brewster: Yeah.

Bob: This is an enormous endeavor.

Brewster: It's enormous.

Bob: The other is - is a financial issue. How do you pay for it?

Brewster: Our general approach to the finances to be incredibly frugal. To just be really, really inexpensive where people, you know, go and trod out their expensive enterprise software whatever it is. We try to do things for 1% of the price of what other people do. And we're able to get down to 10% almost always.

Bob: Sure.

Brewster: We can drop a - a factor of ten. And when we're really good, we can get it down by a factor of a hundred. So our - our key survival trait is actually frugality.

Bob: Yeah.

Brewster: Which is something I learned based on the early internet. Doing a bootstrap nonprofit - bootstrap company, which is WAIS was -

Bob: Yeah.

Brewster: - in a non market -

Bob: Yeah.

Brewster: - was just incredibly insightful as to how to keep costs down. You don't have a lot of, you know, rah, rah, opening parties.

Bob: Yeah.

Brewster: You - you know, you stick to making sure that the people want you.

Bob: Sure.

Brewster: And doing whatever it is. So learned that and kept - kept that going. So then we had to apply this to an ever broadening amount of stuff. If you want to archive the web, how the heck do you do that cheaply?

Bob: Yeah, yeah.

Brewster: And we've always been investigating cheaper and cheaper technology.

Bob: Such as?

Brewster: Well, start out with tape. The tape is just a dismal technology -

Bob: Yeah.

Brewster: - and so we started using hard drives.

Bob: Uh-huh.

Brewster: And they - the hard - how to use hard drives and keep them spinning in computers, but cheaply.

Bob: Okay.

Brewster: And so we've basically built our own computers every couple of years. So the stuff that Google ended up using in - in their stuff, we've been using for years before. The clusters are very inexpensive technology.

Bob: Uh-huh.

Brewster: So taking Danny Hillis' idea of the connection machine, take the commodity stuff and stack it up and just get used to having things go wrong.

Bob: Sure.

Brewster: And that's been our technique. So a couple years ago we actually - we needed to get to a petabyte. So it goes mega, giga, tera, peta.

Bob: Yeah.

Brewster: So a million gigabytes. We need petabyte system. So we went to our friends at Sun, HP and IBM and said what do you have for us? And they said nothing really. And Sun actually was building a system based on having discussed things with us a couple years before, but they weren't ready to get it up.

Bob: Yeah.

Brewster: So we had to go and build our own.

Bob: Uh-huh.

Brewster: And so we built our own that is always measured at the how much overhead on top of the cost of the hard drives is it?

Bob: Okay.

Brewster: Right. So if you can't get down below the cost of the hard drives, -

Bob: Yeah.

Brewster: - but we're always around 40% or 50% overhead on top of that -

Bob: Yeah.

Brewster: - for everything else.

Bob: Sure.

Brewster: The CPU, the networking, the racks, the - the whole shebang. And that's what we - we design and it's call the Petabox. And we then made that open source, so the actual computer itself - all the metal, all those parts -

Bob: Yeah.

Brewster: - all open source. So we spun off a tiny little company to go and - and make these. And it's now actually growing along this little company making these petaboxes for people. _Cause it's something that other people didn't know how to do. So our key thing is to be frugal.

Bob: Yeah.

Brewster: The funding for this place comes from foundations, government grants, and contracts and in any kind of donations from corporations.

Bob: Uh-huh.

Brewster: So if people like Amazon that are continuing with - through their Alexa -

Bob: Yeah.

Brewster: -subsidiary to donate things to the archive is the way that we survive.

Bob: Uh-huh.

Brewster: And the foundations help with start up funding -

Bob: Uh-huh.

Brewster: - for projects, and government contracts we're working with the Library of Congress, the National Archives, the British Library, the French National Library, all of these libraries are contracting with the internet archive to crawl the web for them.

Bob: Okay.

Brewster: So France wanted all the dot FR domains. So the internet archive has crawled these and has delivered them on petaboxes that are not sitting in Paris. So that's - that's been working. So now the web side of it, is sustainable based on the library system. And we're starting to work towards having books movies and music and kind of do the same - go through the same set of progressions.

Bob: Now I - I do Nerd TV, and we push out these relatively large chunks of data -

Brewster: Yes.

Bob: - and to relatively large numbers of people, not in your standards, but you know, a couple hundred thousand people getting a couple - getting a hundred megabyte file is a lot -

Brewster: Yes.

Bob: - and - and our issue is bandwidth.

Brewster: Yes.

Bob: How - how do you get cheap bandwidth?

Brewster: Well, in the United States, because we've pretty much lost the concept of the commons, unfortunately - we haven't gotten it back. You have to buy it.

Bob: Okay.

Brewster: In Europe, there's still sort of the old style of having peering and having a commons, -

Bob: Uh-huh.

Brewster: - which is really - you know, I don't know what happened to the United States in the last - but anyway, in Europe we have peering agreements.

Bob: Okay.

Brewster: But in the United States we actually have to buy it. We just buy it in bulk. We've gotten good at the storage technology, and we by gigabytes at time, gigabytes per second. So it puts us in the, according to Alexa, we're at the 100th most popular website.

Bob: Okay.

Brewster: And we're an archive.

Bob: Yeah.

Brewster: That's kinda cool.

Bob: Yeah.

Brewster: And it's sort of non obvious.

Bob: Yeah.

Brewster: But anyway, we're - we're - we're 100 in the Alexa list of - of top websites. And it's - so it's buying in bulk, being frugal, finding other people that want us to win. So for instance, we're - in the internet we now have fiber that can move 80 gigabytes on - on a single wave length.

Bob: Yeah.

Brewster: 80 gigabytes.

Bob: Yeah.

Brewster: But we're not using any where near that capacity.

Bob: Yeah.

Brewster: And that's a lot of the pricing model around the ISPs and they're having trouble, the telcos are - you know, are famously in trouble, but we have to break through so that we start actually using enormous amounts of bandwidth.

Bob: Yeah.

Brewster: How do we do that without having those guys go under? Either you lower the prices and then people will start doing it more, but then there's this gap. If they lower their prices, will people buy more? Or will their revenues drop?

Bob: Uh-huh.

Brewster: And so we're trying to show with video, with music that actually there is a forward looking business model, -

Bob: Okay.

Brewster: - and we're playing a role for those guys that want to play with high bandwidth applications legal content that - that people actually want that doesn't damage their existing for profit customers. So it's finding things like that -

Bob: Yeah.

Brewster: - that work for people. But basically it's being - by being frugal, by being straightforward about what it is we're trying to do and why, and lots of people help. Steve Case helped, Jeff Bezos helped, we're getting help from all sorts of folks. We'd like more help. We hope the Google guys come on board at some point.

Bob: What's wrong with them?

Brewster: What's wrong with Google? I think they're doing great. Gosh, their services are - are terrific. Holy crow!

Bob: In - in fact to - to a certain extent they're - they're almost running a parallel to you given that they cached -

Brewster: Oh yeah.

Bob: - they cached the web and - and -

Brewster: Yes.

Bob: - and they're doing a library project.

Brewster: Yeah, but they're in - they're mostly an index as opposed to an archive.

Bob: Uh-huh.

Brewster: Not that they don't - they can't be an archive or a library.

Bob: Sure.

Brewster: It's just - in general it's not cost effective to be a librarian archive, in the long haul. Right now we've got - you know, they've got a stock price that's stratospheric -

Bob: Yeah, yeah.

Brewster: They're - they're sitting on top of the world. They don't have to worry about anything. They're sort of the Netscape of now.

Bob: Yeah.

Brewster: Or the Yahoo of now, right?

Bob: Yeah, yeah.

Brewster: I mean, we've seen this - this movie before. So at some point they're going to have to screw it all back down again.

Bob: Right.

Brewster: In the meantime, it's a heck of a lot of fun in Google, I imagine. (Laughter) But they're - and they're off doing really interesting things. But in the long haul a libraries and archives tend to be supported not through the commercial players -

Bob: Yeah.

Brewster: But by society that has a longer term view. And so I expect that to happen as well. You're - you're absolutely right. They're - they - they've got lots of money and they've got lots of smart people and they say gosh what can we do with all of this. And they're doing a lot of things that actually we'd love to do more as well. Our points of view are slightly different, in that they're trying to index things so that they keep you on their website and their ad based mode.

Bob: Yeah.

Brewster: And we - when we digitize a book, we'd like to be able to have so that people can download, print and bind it, and walk away with.

Bob: Yeah, yeah.

Brewster: But that's not part of their business model, but it is part of our tradition.

Bob: And yet, you know, in speaking in business model terms, they seem to work out the ad side better than it was in - in the '90's. I mean they're making a lot of money from ads.

Brewster: Yes.

Bob: Where that was difficult before.

Brewster: Yes.

Bob: What - what are they doing differently?

Brewster: Well, at least when - at Alexa when we were trying to - to sell ads, we tried with a sorta straightforward ad sales force.

Bob: Yeah.

Brewster: And you basically go and find individual people that would talk to a salesman to buy your ads.

Bob: Uh-huh.

Brewster: What Google seems to have done beautifully is make it so that you don't have to talk to anybody. You can go online and transparently understand what the value of the ad is that you're putting online, and you don't have to talk to somebody. They've automated the ad selling process.

Bob: Yeah.

Brewster: And that's one of their major insights. And they've done a beautiful job of it, you know, obviously by their revenues that they've done great.

Bob: So that - you - you go there and you're buying a bucket of eyeballs. That's interesting.

Brewster: Yes, or a bucket of clicks.

Bob: Bucket of clicks.

Brewster: So that you know a little bit of what it is you're actually buying.

Bob: Yeah.

Brewster: And they've made something that works, and they made it so that it's no so offensive to have people come to your site.

Bob: Now one of the things that I find still so limiting about the search engines is that I can find something on a topic that I'm interested in, but it turns out, you know, I'm - I'm half way into it when I realize it was written in 1997, you know. What if I wanted something from last week? And - and to a certain extent, if I go to Google News, you know -

Brewster: Right, right.

Bob: - I - I can do that because it ages quicker, you know.

Brewster: Yes.

Bob: But if I - if I'm looking for a long term content that's out there, it's hard.

Brewster: Yes.

Bob: I'm - I'm just - don't understand why someone hasn't __________ that yet and cracked that problem.

Brewster: It's - it - the amount of data comes large is one of the problems. So the time axis is you're absolutely right, is going to be the next frontier of search engines. So we've exploited the link structure. I don't think we've exploited the - the trails as much as we - we could in terms of metadata to augment search and navigation in general. But the timaxes - often I want to know what's new and different.

Bob: Yes.

Brewster: Right? It's like I don't want to know what's popular, I want to know what's - wasn't popular last month, but is popular now. I want things on the ascendancy. And that requires time based knowledge. But the amounts of data comes, well, large. We're collecting about 25 terabytes a month.

Bob: Okay.

Brewster: So just even handling that kind of inflow is daunting, much less doing and trying compare that against the other petabyte that we already have saved up.

Bob: Sure.

Brewster: And so even for organizations as large as Yahoo or Microsoft or Google, that type of computation is not trivial. And it - a lot of the science hasn't even been done. So we've done some of the work on that, and they'll copy it. You know, they'll - they'll absorb it.

Bob: Well, the interesting thing is - is that in order to make it work, you really have to have a - an archival data set.

Brewster: Yes,

Bob: Which they don't have and you have.

Brewster: They do, as I understand it -

Bob: Really?

Brewster: - those - those companies don't throw things away.

Bob: They don't throw things away?

Brewster: No, they don't - they may not make them publicly accessible, but it's - they have a lot of data, a lot of data. And having those on - all in a commercial company does bring up some issues about trust -

Bob: Uh-huh.

Brewster: - that I think in the longer term we should be a little bit more attentive to, to get back to that privacy aspect.

Bob: Well, tell me about that. Why should we be worried that they have all this data?

Brewster: The motivation structure of - of companies in the United States are very straightforward. If you're a public company, you have to pursue shareholder value. That's the only thing that you're allowed to pursue. You're not allowed to go and pursue other goals. That's it. As it - that - you're - by law, if you're a fiduciary of a company that's all you can do. So it really limits what you can trade off. There are times that you sort of given some slack, but at some point, shareholders can go and sue the company if they're not directly pursuing their financial interests. It's very limited. I mean it's sort of - what's interesting about it is it's very straightforward -

Bob: Yeah.

Brewster: So you can predict people very easily _cause they're just going for money.

Bob: Yeah.

Brewster: So - but if you're - if you've got some other things you want to trade off against let's say, you know, employee happiness or trust with your customers or things like that, those get a little bit hazier and often lost in the mix. And over time things change, leaders change, you know, these - these companies last a long, long time.

Bob: Yeah.

Brewster: And if they're just collecting information that's not only the web pages that are on the net -

Bob: Yeah.

Brewster: - but who searched on what?

Bob: Yeah.

Brewster: One of the dirty little secrets of the internet is the internet - the IP address is personally identifiable. So if you know that somebody - an IP address searched on this, in a large number of cases, you could figure out who is that person that searched on that. That's a lot of information. We're typing things into these search engines all the time.

Bob: Of course we are.

Brewster: What I'm thinking about, I don't know, exploration, their not mining it in the way that could really come back and haunt us -

Bob: But they're saving it all.

Brewster: - as much, but they're saving it all. So it's - yeah, I think we should be very conscious of that. Alexa is very careful to anonomize its usage logs. By - by either eliminating the IP address or throwing away most of it to try to make it so that it wasn't personally identifiable. I don't know if other groups are doing that.

Bob: Well, probably not. If it takes effort, they - they wouldn't do it.

Brewster: In general, unless they, you know, would feel threatened by honing it.

Bob: Yeah.

Brewster: In - in general, the - the general propensity is keep everything.

Bob: Yep.

Brewster: So EFF is working on usage log anonomization and tools to help make these available. We participated in some of these workshops to come up with standards for how to get rid of data in your usage logs, because it's poisonous. It's just too much. It's - it's - it's too - it's too spooky.

Bob: Yeah.

Brewster: It's not like the library where you can read anonymously. You're - you're trails are tracked. So what do we do about that? I think we just start to come to terms with the amounts of information that are being collected and start in an organized way throwing the data away. Such that we can still do the marketing studies -

Bob: Yeah.

Brewster: - that we want to be able to do, or you know, know what's more popular than - or even be able to track down abuse.

Bob: Sure.

Brewster: Right. Keep high-res versions of our data so if somebody's badgering you, you can try to figure out who it is and turn it off. But after 30 days, get rid of it, get rid of it. So that -

Bob: Yeah. Or you can even like desample it.

Brewster: Desample it.

Bob: Yeah.

Brewster: Yeah. So that you have the marketing data -

Bob: Yeah.

Brewster: - in this sorta overview that you're looking for. But you don't have the sorta of the poisonous toxic wastes that is accumulating on our computers, and is a lawyer's hayday.

Bob: Well, that's not going to happen unless it's mandated somehow.

Brewster: Correct.

Bob: So what do we do?

Brewster: Find some leaders. Find some folks that are willing to step up and say this was working for me, I did these experiments, do it very much in the open. And I think that will happen largely in the nonprofit sector.

Bob: Uh-huh.

Brewster: The - the new ideas come from the commons. You know, there's this whole idea that it all has to be locked up infinitely forever with copyright before anybody's going to come up with a new idea. That just doesn't bear out when it comes to history.

Bob: Yeah.

Brewster: So I'm interested in some of the rise of the amount of profits, the technical nonprofits. Think they're - they're fundamentally interesting. So there's Alexa - there's internet archive, there's EFF, there's creative commons, there's public knowledge, there's Mozilla, there's Apache. There's a - there's a new generation of people that are working within the nonprofits here -

Bob: Yeah.

Brewster: - that have been traditionally in corporate environments.

Bob: Yeah.

Brewster: What's interesting about working with these organizations, and we work with them all the time, is they're very open.

Bob: Yeah.

Brewster: We say sure come on in. Yeah, you can have this stuff. Yes, it's all open source. This is basically not a big barrier towards trading things or sharing information. We're extremely open. And I find that, in some sense, more efficient towards creating new ideas -

Bob: Of course.

Brewster: - than some of the - the corporations that currently exist, which are kinda like exoskeletons.

Bob: Yeah.

Brewster: They're sorta soft and gooey on the inside, but they have a thick shell on the outside. And if you're on the inside, it's very hard to get through. You have to do nondisclosure agreements and all this crap, which really slows down your company. And so you get these - these little exoskeletons running around, and I'd say the - the - the nonprofits are a little bit more like vertebrates.

Bob: Yeah.

Brewster: You know, they've got a backbone, which is their mission, but they're soft, they're easy to - to - to go and deal with and - and they're easy to work with other of their kind.

Bob: This Nerd TV is distributed under a - a - well, we originally aimed for to do it under GPL, but -

Brewster: Yes.

Bob: - then in turned a creative commons sort of came a long and - and so we used that. And you - you know go to a traditional television network, and you say oh yeah, we're going to give it away and by the way, they can copy it and change it and modify it and edit it and -

Brewster: Right.

Bob: - and they think you have three heads.

Brewster: Yes, it's - but the creative commons licenses have been great. So the - the lesson really stood on the shoulders of Richard Stallman who did the open source software idea.

Bob: Yeah.

Brewster: And he's absolutely right. I mean the - the real - you know, they're some great heroes in our -in our field. I think Richard Stallman is - is one of them. And -

Bob: Well, if you want to be viewed and heard, you know, you should do it that way.

Brewster: Yes, and it's actually returning to the way it used to be. People say oh it's a radical new idea. It turns out, that's not true at all. The copyright laws changed in such a radically different structure, that it's only really catching on of just how bad it is. So in _76, they suddenly made everything copyrighted.

Bob: Yes.

Brewster: So before that you had to assert copyright. You had to be a little "c" on it, and send a copy into the Library of Congress otherwise it's not copyrighted, so almost everything wasn't copyrighted.

Bob: Yeah.

Brewster: In fact, all the - the software that we were writing back in technical schools when we worked with Richard Stallman and al those, they were signed. It was all part of the hacker culture. We were building a technology together. It wasn't all this locked down stuff. Then the - then the copyright laws changed. And then MIT took all the work of all us and packaged it and sold it. And -

Bob: Damn them.

Brewster: Damn them. (Laughter) It was - it was - it was selling of the commons, it was properterization of the commons. It - it was nuts, and - and Richard Stallman blew a fuse.

Bob: Yeah.

Brewster: He just said this is wrong. And it's all because of the _76 copyright act. Before that, it wouldn't have been possible. But after that, it was. And then he started the open source movement, which was to just try to regain the balance we used to have, frankly what it is we grew up with.

Bob: Yeah.

Brewster: And so the open source idea is a - actually a retro action to get back to the way things were. And creative commons is building on top of that to try to get the flexibility, the slack, the stuff that it takes to have society work especially with ideas.

Bob: What - what did MIT sell?

Brewster: They sold license to the elisp machine system to a couple of companies. They licensed it to Symbolics and to Elisp Machines, Incorporated. So they took this machine that we had built, and participated in and licensed that software to a commercial company, which felt like a great betrayal of trust.

Bob: Yeah.

Brewster: And so Richard Stallman said, you know, not on my watch.

Bob: Yeah.

Brewster: And gloriously he came up with some interesting trade offs of how do you go and make a license that works. And there have been, you know, a bunch of different variations.

Bob: Oh yeah.

Brewster: But his basic idea was a huge step forward. In terms of other heroes, in this - in this -

Bob: Yeah, please.

Brewster: - well, I think of Alta Vista. Alta Vista - you know people point to Google now -

Bob: Sure.

Brewster: - and they pointed to Yahoo before, whatever it is, but Alta Vista was the first time for a couple thousand years - and somebody said we can collect it all. WAIS didn't go that far. WAIS the old publishing system -

Bob: Yeah.

Brewster: - said no, it's going to be distributed. Alta Vista said no, no, no. Let's just gather it all. And I remember going down and walking down into digital equipment corporations room and looking at it. And it was a size of a couple of - of coke machines, and it was about 30 gigabytes, and it was the web. It was all of it. (Laughter) It was sitting there. And so you could go and search the whole web. This was in 1995. And that was a real mind blower. And the last people that tried to collect it all were the folks at the Library of Alexandria. So that - so those guys had the guts and the insight to do something really very interesting. So I'd say Alta Vista is another of sorta the great heroes. Another is Tim Berners-Lee Tim Berners-Lee, I think, is the statesman of the net. He's the guy that when he could have gone off to be rich, -

Bob: Sure:

Brewster: - he said no, let's see if we can make the rules for other people to get rich and still have a world we want to live in. And he's just a - a phenomenal person that's selflessly gone to try to help keep a balance and keep an open system working, and - and it's doing quite well. Bill Dunn for me is another hero came - coming up with the idea of metadata.

Bob: Uh-huh.

Brewster: Completely unsung fellow. But incredibly insightful. There are these people that are more right that they ever deserve to be. If you look back, you know, you hear there's sorta wild rantings, but then you check back up on them in five years. It's like wait a minute, they were right.

Bob: Yeah.

Brewster: And he's another one of - of those. So the internet, I think is - is doing okay. I think we're doing - you know, it's - we've now geared around our idea that you should have access to information.

Bob: Yeah.

Brewster: We're turning to our machines to help answer questions.

Bob: Uh-huh.

Brewster: All right. So we kind have the question answering machine, the Mr. Peabody or whatever it is -

Bob: Yeah.

Brewster: - that goes and tries to answer - Mr. Know It All that - that is - is the internet.

Bob: Yeah.

Brewster: But we have some big steps to go. So with the big steps to go, is we need - now we need to get the rest of our great information on the net. And we need to do it in an open way. So most of the information on the web has been written in the last ten years. You know, you said incredibly antique in 1997.

Bob: Yeah.

Brewster: Sorta ___________ something doesn't it?

Bob: Yeah.

Brewster: (Laughter)

Bob: Yeah.

Brewster: So there's a lot of information that we can hear, from - from last hundreds of years.

Bob: Sure.

Brewster: And so now we need to go back and backfill the internet and put all of the old material on it. So Google is scanning books, we're scanning books with - with a - with libraries, and the idea of going and putting these online such that you can read them, download them, print and use them, is a fantastic opportunity. And I think we'll see some really terrific things come out of it. The copyright laws again, have been so badly screwed up.

Bob: Yeah.

Brewster: That -

Bob: And getting worse all the time.

Brewster: And getting worse all the time. So now is our chance in terms of the geeks of the world, the nerds of the world. We are in a place to go and shape what it is we want the future to look like. And there are some things that are really contorted. One is this copyright stuff, the privacy thing is - is a shambles. And a lot of it comes down to technical tradeoffs. And we're an - an empowered guild, if you will, we technologists.

Bob: Yeah.

Brewster: And it's up to us to go and put something reasonable in place. And if we look back after 20 years and we go gosh, oops, you know, sorry about your privacy, you know, that you're able to get knocked - somebody can kick down your doors based on something that you've searched on 20 years before that's now politically out of -

Bob: Yeah.

Brewster: You know, the 1950's weren't that long ago.

Bob: That's right.

Brewster: You know, and going and getting in trouble for -

Bob: ___________________ meaning you went to, you know.

Brewster: Yes, when you're in college -

Bob: From ________________.

Brewster: - and it was called something completely else.

Bob: You just went there to get laid you know.

Brewster: That was it. That was it. You know, oh you made a visit to Afghanistan. The - these sorts of things are - can come back to haunt us in ways will be quite devastating. And so we're in a position now to make some of these changes. And as long as people feel empowered to start putting them in place, it'll take some leadership. It'll take leadership from the EFF's, the creative commons, the internet archives, but also the Yahoos the Googles, the governments, and I'm - you know, governments come and go. But there are some governments that actually do have a sort of balanced view of education and personal interest and rights. And I'm not sure we'll see it in the - the United States or even in Europe so much as - as hopefully places in Asia, the up and coming places that will start to say how - how should we use this technology to actually make a better society.

Bob: Well, now let's -

Male Voice: 53.

Bob: 53. We're 53 minutes in, we're almost - we're running out of time here. If we looked say five years and ten years into the future at the internet and - and thought of it in two terms, one is if we do everything right.

Brewster: Uh-huh.

Bob: And the other is if we don't. What are those points out there?

Brewster: What are the - what are the possible paths?

Bob: Yeah, what are the possible paths?

Brewster: What - what are we looking forward to?

Bob: Yeah.

Brewster: If we're to do things right, we would have a many to many to many to many to many system. So competition is not made up of Coke and Pepsi. Competition is not made of five record companies. Competition is made up of tens of thousands of different players at every point in the channel.

Bob: Sure.

Brewster: So if you have tens of thousands of people making things, websites and - and tens of thousands of ISPs, tens of thousands of different mechanisms of getting access to things so it's not all going through one browser type.

Bob: Yeah.

Brewster: It's - it's going through lots of different things, and it's going to lots of different communities. If - if there's not a choke point, then we've actually made something efficient. And actually that makes that good business sense.

Bob: Sure.

Brewster: So there's good reason for governments to not want to put in place monopolies. Then the short term it always looks better because you can go and, you know, get bigger campaign contributions from Verizon -

Bob: Yeah.

Brewster: - if they, you know, they're - or Microsoft or you know whatever somebody who's - who's trying to get a strangle hold.

Bob: Yeah.

Brewster: But if you've got a longer term view in making this an open environment, you can actually make it both open and make business sense. So that's how to make this stuff actually go right. How to make it go wrong, at least in this part is - is allow somebody to control some part of it. And once you - it seems, I don't really have the - the good data to back it up, but it seems that if you get - allow somebody to control one piece of it, then they'll strangle it and they'll work back up stream and down stream.

Bob: Really?

Brewster: Yes. So I - I'm not quite sure what happened in the book publishing industry.

Bob: Okay.

Brewster: But something went wrong. And it's now - I think it might have been the distribution - distribution chain in Ingram. So once somebody starts - most books went through one place, -

Bob: Yeah.

Brewster: - then it - it becomes easier to make a Barnes & Noble. All right, for that part and then back up stream, it all folds back down into a Bertelsmann something -

Bob: Yeah.

Brewster: - or just a few major scale book publishers.

Bob: Sure, sure.

Brewster: It's sort of just became constricted -

Bob: Yeah.

Brewster: - all the way up and down the chain. So the theory is, and I can't really back it up with - with stats, -

Bob: Yeah.

Brewster: - that if you allow somebody to control any part of it, they can use that control to build large other - other organizations. _Cause organizations like to deal with other organizations that are their same size.

Bob: Yeah.

Brewster: So we had certain threats of this, whether it's the browser, such that it's all going to go down to one operating system and one browser, -

Bob: Yeah.

Brewster: - then you could - then you have control over people. AOL was trying to get, you know, control over enough people such that everybody had to go through them. Therefore, they could go and change the business model -

Bob: Sure.

Brewster: - of the open internet -

Bob: Sure.

Brewster: - which is a little scary. Now we're seeing potential problems at the - at the tier one ISPs. So they're starting to be in a position of control and large organizations say the government or publishers, the publisher associations like dealing with them because then they can go chop off parts of the net.

Bob: Yeah.

Brewster: If they're, you know, doing something that they don't like. So we're seeing real potential problems sort of up and down the chain. But it's based on vigilance and great people like Larry LaSagin. People like public knowledge in DC that are trying to communicate these issues that are kinda complicated technical geeky issues to a broader populous and specifically to - to some of the government folks that can try to keep it actually safe to make businesses on the internet. Otherwise, we'll end up with a very controlled environment. Well, you know, if we end up with just cell phones all being sort of like a little personal AOL where it's all controlled by your provided as to what you see, what a tragedy. If we ended up building another television out of this, what a shame.

Bob: It's a waste.

Brewster: What a waste. We would have spent 20 years having the possibility of having done something great and having lost it. So I'm spending my time, my fortune, thanks to Steve Case, Jeff Bezos, on trying to make sure that we have a - a future that we actually want to live in. Something that we're actually proud of at the end of the day, having said yeah we built something as good as books. We took the Library of Alexandria idea of having all information available, and not only made it available in Alexandria, Egypt -

Bob: Yeah.

Brewster: - which they did for 500 years. We all say oh burn, yeah, but it was up and running for 500 years. They had 500 years of great materials available. But then take the Library of Alexandria and making it available to anybody all over the world, that's a worthwhile goal to get up and spring out of bed in the morning and go to work.

Bob: Hey, thank you. We're done.

Brewster: Thank you.

Bob: Baffo ending.

Brewster: Was that what you were looking for?

Bob: That was great idealistic. It was wonderful.

Brewster: Yeah, it was - I'm still hopeful.

Bob: There's - there's such - I - I write a column on the PBS website, and I've been writing it since 1997, so I'm old in that -

Brewster: Right. So you've seen the -

Bob: Yeah, but the thing is that we have on - on my page, we have an archive of everything I've written since '97. And the interesting thing we've noticed now is that I have more people read my archive than read my column.

Brewster: Really?

Bob: Yes.

Brewster: I'm surprised. _Cause - well, it might be that their web is so hard to find new stuff.

Bob: No, it's not that. It's - it's that - you know the - it's that - I like to think I write interesting things. And - and here's something that's 600,000 words, approximately -

Brewster: Right.

Bob: - on - on this - these sort of topics and people find them through search engines -

Brewster: Yep.

Bob: - and they - and they - and they go to them, but - but the nature of the audience is such that every week, I got like 200,000 people who read my new stuff, but I have 300,000 -

Be more connected. Help support your PBS station and projects like NerdTV...pledge online.

© 2005 NerdTV LLC and PBS. Offered under the Creative Commons license.

Privacy Policy | Terms of Use | Published on September 6, 2005