A conversation with Brewster Kahle

It used to be a Christian Science church, founded by Mary Baker Eddy in the late 1800s. When we first got this space about 15 years ago, we asked ourselves, ‘What is a library? ’We don’t need stacks. We don’t need rows of books. We need a place for people. The Internet Archive in Canada is in this former bank that we’ve transformed into a venue. We also have a separate library in Holland.

Well, what usually happens to libraries? They’re burned. They’re actively destroyed by those in power, right? It may not happen soon, but that’s what happens to us. It can be traced way back to cuneiform tablets; you can see in the history of libraries that they’re destroyedor disbanded or crushed due to resistance. If that’s what happens, we can design around it by having multiple copies in different regimes, to make sure the Archive survives. There’s also a partial copy at the Library of Alexandria in Egypt.

It was mostly just inspired by libraries, and the Library of Alexandria is the Great Library. As a former librarian at the Library of Congress said, ‘We all live in the shadow of the Great Library’. That was the first instance where someone tried to build a library of everything. 

No, I just love them. But I come from a technology side. In college, there was this question of, ‘What do you do with your life?’ I thought, ‘OK, what can I do with my technology background that could be for good?’ This was 1980, before the Mac or the Internet or anything like that. But we knew it was all coming, right? Especially because I was 19 at MIT. When I was there, it was possible to see the future. It was all there, but it was just going to cost millions of dollars. So the thought was, let’s build the library of everything. It’d been promised for a long time—Vannevar Bush, Ted Nelson. And I thought, ‘OK, I can do that. How hard could it be?’ 

Yes, it’s not something you’re gonna finish, but we should be further than we are now.

Because it has long since been technologically feasible and it’s good for our society. But there are these institutions and monopoly-sized companies that are trying to stop it. 

Well all of that was to get the library built. Alexa is also named after the Library of Alexandria. 

If you wanted to go and build the library in 1980, you have to remember the computers didn’t hold much data, right? A supercomputer had 32 megabytes. But we needed supercomputers, so I helped build a supercomputer. We spun it off from MIT, called the Thinking Machines Corporation. So, now we had a computer and we could make it search so people could interact with it just by typing. That was new. Then we needed to get it online so it could be accessible to people who wanted to use it and also add their own data. So we made a network called WAIS. It was the first internet publishing system. It was all to try to get the library. So we got that going, but the worldwide web won, right? That was a better system, but it was open, so perfectly happy to go attend the system. Then, we needed to get publishers on board. So I helped get the New York Times Wall Street Journal through WAIS Incorporated, getting those guys online to get the open world to work. And what we did was show these guys how they could make money by publishing on the net. Then websites took off and by 1996 there was enough publishing online where I could then turn my attention to just building the library. So we had to make the precursors to then build the library, building blocks basically. I thought we’d be done with building the library by the year 2000. That’s at least what I budgeted for. But basically it’s been slowed down by large, powerful players not wanting people to have access to information—even if they’re paid.

There’s actually two of them. There’s one from book publishers on e-books and there’s the music guys, who are suing for 400 million dollars for just 78rpm records. 

It’s really based around the New York and Washington copyright cartel, which is the Authors Guild and the American Association of Publishers. When the Internet first happened, the United States decided to support information access. The Information Superhighway and law and policy allowed search engines to happen and for user-posted content to exist. Because of that, the dot-com boom happened in the United States. It didn’t happen in Europe or Japan, they didn’t have their own search engines at the time. So, why did it happen in the US and not in other places? Well what happened is they sided with the publishers, and the publishers crushed it in those other places. You’re seeing it with AI now, in Japan, it’s OK for people to use their computers and to make their AI models. And Europe deserves half and half right, but in the United States, it’s all lawsuits.

The biggest thing that I think I got wrong in my initial imagining was I didn’t expect that the United States would give up completely on antitrust policies. Basically since the 1980s we stopped enforcing any form of antitrust that favoured competition. We ended up with just a few big companies controlling every field, not just within the US but also globally, you know? It didn’t used to be that way because we had law and policy that favoured competition.  

I’d say the biggest real threat to the Internet Archive is if there’s just a few newspapers and book publishers left or if everything is a Netflix. This means that things don’t necessarily get onto these platforms and you have basically constrained the information creation ecosystem. We need an open Internet. We need an open web. We need open access materials or things that you can buy rather than get a license to. If libraries don’t have collections, what are they? What I want to know is: Can we build another generation of web technology that actually works, that looks a lot more like how publishing worked in the old days? In the past, a publisher would go and make something, then sell it or give it away, and that thing would go on to live in lots of different places. And people would read it from lots of different places—it was more robust, and in a sense more private.

This was something influenced by Thinking Machines Corporation, which was run by a woman named Sheryl Handler. The company was housed in this mansion called the Lyman Estate in this forest outside of Boston, and every room was different. And I think if you have your offices in suite 201 you’re gonna start thinking like suite 201. But if you are in a grand space, then you kind of have to live up to your environment. So our first building was in the Presidio. We were like the second leaseholder in this national park. 

We told the National Park Service, ‘We want to build a library of everything’, and they said, ‘Well, you’re either on to something or you’re crazy. Either way, you’re one of us. Come on aboard’. So then we got to be in a national park. But we kind of outgrew it. They also started charging more. So we said, OK, let’s move to where we’re based now. When you walk up those steps to go to work, you’re not thinking about how to eke out a few more percentages from advertising to people. That would be inappropriate for a place like this. We want to live up to the building. That’s why we got it—they don’t want offices or anything like that in this place. 

I think the Internet’s vision of access to all information is deeply rooted in the post-war American ideal. When I was studying in Boston I asked people where I should put the Internet, because I knew the Internet was coming in 1988/1989. My favourite advice was ‘Go someplace people don’t think you’re crazy’. It was from Bill Dunn who was the head of Dow Jones electronics and came up with the term ‘metadata’. Anyways, that’s San Francisco. Here, people will give you a listen. So when I came out here in 1989 I didn’t want to start a company, I wanted to start an industry. I needed a  beehive of information servers—I needed everyone to be in the same place and doing their own thing for the Archive to work. Coming from MIT, we had the right machines and the Internet, but out here they barely had the Macintosh. You had New York, but that was all finance; you had Washington D.C. where AOL was based, but they had people suing everyone and their grandmothers for using the Internet; Los Angeles was a good candidate but too big; Tokyo and Boston were also options, but with San Francisco, I think tech was drawn to it for the same reason why the Grateful Dead, with their sharing ethos of their concert recordings, were drawn to it. There were new ideas coming out of San Francisco, and people didn’t call it crazy. I think there’s a reason why the Internet was centred out here, it didn’t necessarily look obvious at the time, and it wasn’t all because of me. But I certainly helped, and I did push like hell.

Well in October 2025 we hit 1 trillion webpages, which is an indication, I think, that people are awesome. There are some pretty dreadful things going on, what with corporations and governments who are trying to suppress information. But people are up for sharing and we have a billion voices in our library.

Yes, people are putting up gates. The big platforms are becoming more like publishers. Reddit and Yelp have gone and demanded not to be archived by us. I think AI tools are great, but I don’t think it should be concentrated in the hands of the few. We need to build a public AI. 

Big organisations love big organisations, so big, muscular, authoritarian governments like big, muscular, authoritarian corporations, right? But if we’re going to have an innovative ecosystem of startups and flowering new ideas, it’s going to be based on lots of small experiments.

It’s Colonel’s housing, actually. I’ve lived there for 27 years. They prioritised housing in those early days for people who worked in the Presidio—they’d only just started fixing up the houses, since it had recently been turned from an army base. I can bring you over now, it’ll be helpful for your article. 

I take my little Smart Car. 

It’s great. Let’s go, I’ll drive you, if that sounds good to you. It’s like a street-legal golf cart. I have to call my wife to let her know we’re coming. 

Yeah, welcome. So right now you’re at a Full Bird Colonel’s house. All this was built in the 1840s by the military. At the end is the Commanding General’s house. It’s still owned by the government. It’s public housing, and we can’t renovate or really change anything about it. This is my wife, Mary.

Mary Austin: You’ve caught me in the middle of making lunch, but I’ll go turn on the flamingo for you guys.

Brewster: Something I find interesting about the architecture of this place is that a Full Bird Colonel has to be able to entertain. So this dining room is designed to have sit-down dinners for 20, and we have a sit-down dinner for 14 every week. We have been for 35 years. Here’s a book that Mary made of all the menus we’ve collected for those dinners over the past several years. We don’t have a menu every week, but they kind of serve as a conversation-starter. Each dinner has a theme and a question of sorts to kind of get people to talk. For example one was ‘Tell me a story about apples’ and you can see people have gone around the table and written on this menu a summary of their stories. 

The book brings together menus from 2010 onwards, but we started having these Thursday dinners in the mid-80s, back when we were in Boston. 

Mary: I do! I run the San Francisco Center for the Book. My business partner is a typographer, and I’ve got lots of friends that are designers. But you know, I’ve never been trained as a designer, but if you look closely you’ll see the progression from when we first started. The first dinner menu didn’t even have anything on the back. And then I was like, ‘Oh, well, we should put something on the back’. So I started putting numbers there. It’s definitely developed a lot more. I think the Kumquat May and the Blood Moon-themed dinners were my two favourite designs. But we’ve done a lot over the years. It’s just, you know, a way to punctuate our dinners. 

Brewster: It’s also about asking ‘How do you build community?’ When you’re out of college, how do you meet new people? What’s the next step after you meet someone? A meal. And how do you do that? We only have so much time. And I want to decommercialise more things. Let’s decentralise food, housing, information access, yes, but also community building, as opposed to always trying to figure out, ‘How do we make everybody pay for something?’

Not really, but what we’ll do is we will plan the menu and buy the ingredients and have people over. Mary puts everyone to work, everyone’s cooking. That’s the best way to have dinner.

Brewster: I think Mary will have a different answer for that. My answer is people. You’ve met Daniel, the futurist back at the Archive, and Mitra just now. It’s people—having people freely coming and going in and out of here.

Mary: Making a home is my superpower. I have a cousin who actually tears down walls and builds extensions, and that terrifies me. But give me a space, a little colour, some interesting art… It’s not super fancy, but yes, most of the things on the wall are made by people we know. The neon flamingo in the living room is from my brother-in-law. He’s a neon artist, and he was one artist that worked on—I don’t know if you go through Chicago much, but between the two United terminals there’s this neon tunnel. The person who came up with the concept called in every neon artist in the country, so he was one of the people who worked on that. But yeah, like I said, we can’t change much otherwise. My dad was in the military, so I actually grew up in government housing, but it’s maybe the best government housing I’ve ever seen.

In college. I dated her best friend in college. She wasn’t that interested in me, but she moved out to San Francisco first. When I moved out here, straight men were a little more rare. So I think my star shined a little.