 I hope you all had a good break. Welcome back. The next item on the menu is a conversation with Mark Graham. It's a great, great pleasure to have Mark here. He's really one of a kind. Currently, he's the director of the Internet Archives, Way Back Machine, where he's responsible for capturing, preserving, and helping people discover and use more than 1 billion new web captures each week. So they basically archive the web. It's an amazing feat. Previously, Mark held positions at NBC News, iVillage, Roho Networks, The Well, and at AOL, where he worked on Gopher. Yes, that's Gopher for the dinosaurs and the historians. You still remember that thing. And right in fact, Mark also served in the US Air Force in the Pentagon, where he actually acquired many of his early computer skills. And he worked on a US-Soviet email service. That's a random fact. Maybe we can come back to that during our conversation. I'd definitely like to learn more about that. But just to kickstart the conversation, Mark, you had an illustrious career starting in the very earliest days of the web. Can you reflect a bit on the evolution of the web as you've seen it, and maybe with a particular focus on the role that Wikipedia has played in your eyes? Sure. First of all, yeah, it's just really great to be here with everyone. I was looking forward to this conference face to face. I've enjoyed many, many of the gatherings that we've had over the years. I met many of you at Stockholm last year. I was looking forward to Bangkok. But you know, this is good. These breakout room things, I've done a lot of zooms, but I haven't actually done the breakout room where you're randomly thrown in with some other people and have conversations. And I just love that, because it actually is like when we physically come together in that kind of semi-random kind of way that happens when people come together in a physical place. And it is also, I think, like the net too, at least the net that I don't know I was thinking about and have been working on and toward for a while now. I don't know if Illustrious is the right, it's just I've been doing it for a long time. Let's just put it like that. I'll just say, I didn't get into this space because I wanted to do things with the internet, per se. I mean, well, first of all, we didn't call the internet back in the early 80s. What did you call it? Getting online. You know, and but I, as you noted, I had spent some time in the Air Force. I worked in the Pentagon. I'd been, this is the time in the early 80s from 80 to 83 when the Cold War was really at one of its heights and there's a lot of activism in the United States and indeed all over the world. There was the nuclear weapons freeze movement, for example, the call to halt the arms race and other kind of efforts like that. So I, when I got out of the military, I moved to Berkeley, California. I was going to get a degree in peace studies and instead I decided to start a computer network for peace activists. To be honest, the bulletin board systems were with a rage at that time. Networks like FIDO, which was a distributed network of PCs, thousands of them all over the world. And I saw many other constituencies developing networks and I thought to myself, you know, why not the peace movement? There were thousands of peace groups in the US and around the world that could benefit from being in collaboration with each other. And that was one of the key ideas and so do I. So myself and a bunch of other people got together and that's what we did. It's still running. It's called Association for Progressive Communications, apc.org and many of the APC member NGOs around the world are active in the Wikipedia communities as well. So that's very gratifying. I'd say, you know, obviously I don't mean to go through the history of the ebbs and flows of the changes of the web and the internets and all the rest of that, but I think obviously we're coming back to maybe some of those earlier ideas of openness and collaboration and sharing and community. And for me at least Wikipedia and the Wikipedia communities plural and the Wikimedia foundations and associations they embody many of the best characteristics of those early ideals that maybe took a while to gain hold and you know, I'm sure many of us here in this meeting have had gatherings and conversations and you know, virtual dinner conversations over the last few weeks, we've been thinking about what these times of corona have to teach us and what are the opportunities here? And for me at least one of the opportunities is an amplification of many of the qualities and values and practices that we're engaged with in the Wikipedia community as we help to evolve what we can do together using web-based technologies. So maybe this crisis can help us refocus on these values that the community has and I guess even the fact that we have this remote conference is really in a Wikipedia spirit where a lot of people that collaborate probably never saw each other in real life and this mode of operation is now coming to the world kind of everyone is becoming a little bit of a Wikipedia that way I guess. Indeed, I also wanna say just for this start, please for me and my work and what I really love about physical conferences is I love talking to people. I love you know, what happens when people come together and they just like randomly talk about what they're interested in and then new things for me at least new things usually come of that. And so I put my email address in there mark at archive.org I would, if there's one thing that I want to come away from today with and that is some new ideas, some new connections, some new projects and new opportunities to collaborate with others either on this call or even people who say, oh, you know, this other project here that maybe might be of interest. So I encourage and invite direct emails my DMs are open on Twitter. I'm just Mark Graham on Twitter and in full disclosure I'm not really active on my talk page. So that is one thing. So Mark after I introduced you, maybe you can introduce all of us to the internet archive. What is it that you're gonna do? What's the mission? Yeah, you know, I'll go kind of quickly on this one here. I think the internet archive is a nonprofit. It's about 24 years old based in San Francisco, mostly although we're all over the world. It was started by a guy named Brewster Kale with a mission of universal access to all knowledge. Brewster was an early internet pioneer. He developed some of the initial publishing platforms, one called Waze based on a library standard for those library geeks out there, Z39.50. He sold that to AOL. His second company was called Alexa Internet and he sold that to Jeff Bezos at Amazon just before the crash, it was a pretty good exit. And it gave him the ability to basically pursue a dream that he had had. He calls himself a techno utopian. So this dream of universal access to all knowledge of basically building the library of Alexandria version 2.0. Could we take all of the works of humankind and make them available to everyone, wherever they are, and to continue that process into the future? And so that's what we've been working on. It's about, you know, I'd say before this happened that the coronavirus, we had about 150 people. Some of them are furloughed now in scanning centers because they physically can't go to their scanning centers or in libraries all over the world. Those libraries are shut down. And yeah, so for the last more than 20 years we've been working to take analog information, make it digital, preserve it, and then make it available to people and taking born digital information, preserving it and making it available to people. And I can give an overview. We work in a range of media. We think of like media types. So certainly books is a very significant media type for us. Television news is another one. We archive about 60 television news channels 24-7. Radio, we're now archiving more than 1,000 radio stations, 24-7. This, the music, vast quantities of music in a variety of different medium, including 78s. We've got this fabulous collection of about 180,000 digitized 78s. Journal literature, it's an experimental service right now called Fat Cat Wiki, as in like Fat Cat, like big catalog Wiki. And we have 22 million completely open journal articles there, and maybe metadata on about 100 million others. Yeah, so I find this focus on the analog really fascinating, I have to say, because for me the web epitomizes the digital and in that nearly paradoxical, the Internet Archive also has this huge focus on the analog, which makes a lot of sense the way you put it, because I guess otherwise it'll be gone. Yeah, well, it'll be gone or wouldn't be accessible to people, right? You know, certainly things like, well, certain media just have historically done somewhat ephemeral, like television news, right? I mean, yeah, unless you were, there are some examples, we actually worked with the estate of someone named Merriam Stokes. There's a movie about her, you could go Google Merriam Stokes who obsessively recorded television news on tapes, you know, hundreds of thousands of hours. So that's an anomaly. Other than that, I say produce out there, well, radios like that too. And it's not just for historical purpose, but how can you even know what the voices of humanity are right now? Much of the world's population gets their information with their sense of reality, what's true and important in the world, from radio. And, but unless you're capturing all of it and doing the speech to text translation and then be able to do the full text indexing, how can you do research on that? How can you know the spread of ideas? How can you know the correlations between the messages that people are getting and maybe, oh, I don't know, certain, you know, public health related behaviors that they might be expressing, things like that. So, so yeah, lots of opportunity for research, right? I'm starting to kind of want to seed some of those ideas for those of you out there who are researchers looking for data, we got some data. But I'd be remiss to, I'm sorry. How can we get access to that? For example, through the Merit and Stokes archives. That was pretty easy, honestly. You know, we tend to group things into what we call collections. So it's archive.org slash detail slash something. And that's something, there's millions of those somethings and some of those millions of those somethings are very large, I mean, billions of things. And some of them are smaller. One of the big ones with billions of things in it is the web collection. And the web collection is in turn broken down into tens of thousands of other collections. But that's pretty much where we put things that we archive from the web via HTTP protocols and then make available through the Wayback machine. The Wayback machine being the term that we use for this service that's been archiving much of the web since 1996. And I think when I wrote that little introduction was a while ago, we're actually archiving more than a billion URLs a day now. At one point it was a week, it's now more than a billion URLs a day. But I also wanna truly underscore, I hope I don't use any superlatives in this share today because words like all, it's like we scratch the surface. We just do only some and we'd like to do more and we try to do what we can that we think will be most useful. How do you decide what to collect? I kind of set myself up for that question, didn't I? Well, first of all, it's a work in progress and I'm always open to suggestions from people about how we can do better. And it's not just me, it's a lot of other people. So the answer is a whole variety of ways. So I'm gonna speak to the web for a little bit because say in each medium is a little different. I'll speak to books too, I'll get to that. But in the case of web, there's kind of, I kind of think of things as two worlds. One world is where we probably can get almost everything if we try. And so an example of that would be, can we archive every URL referenced in every Wikipedia article in all Wikipedia sites within a few seconds or a few minutes of that being entered and then archive all of the links on that particular page. Can we do that? And the answer is yeah, pretty much. We can pretty much do that. It's a couple of million a day. So we do that. Another one would be say Facebook, right? It's like, okay, we archive Facebook. And the answer is gosh, not even a little teeny little bit of it. So there we have to be selective. And so we use a variety of methods to listen to signals of the kind of things that we think might be useful to archive. But we also don't do it. Some of it we do ourselves internally where we direct, we come up with these lists. But in many, many, many cases, I mean, three other examples where it's a big collaborative process. One example is we do have a subscription service called Archive It. And the users of that service, you museums and governments and libraries and they pay us money. We're a nonprofit, but that's one of the ways that we bring in revenue is these institutions. They pay us money to do web archiving. And they're basically like various and then they will define what it is they want to be archiving. So we give them a whole turnkey environment they can define the things they want to archive. It's called the Archive It service. Another example would be, and this is especially true over the last several weeks, is we collaborate with the many different research organizations, the Stanford Internet Observatory, Graphica with a K for a Straft and a variety of others, the internet research folks in Cambridge. And they do studies and they'll come up with their own list. They're also journalists. We're very actively engaged with many journalists. The New York Times in particular had been providing us with URLs, thousands of URLs that they have been asking us to archive relative to COVID. And then the third category is everyone who use the service because it's this great feature on the bottom right of web.archive.org called Save Page Now. And anyone can enter a URL into that feature and they can then even do a check-off box that says capture outlinks and they enter and then we will archive that URL and all of the outlinks immediately. Right now that is archiving about 50 million URLs a day. Now, obviously that's not 50 million times people are entering a single URL, some of it is scripted. Plus if you enter one URL and it happens to be like a front page of a news site, that could easily generate thousands of URLs to archive. Yeah. It seems like an amazing symbiosis between the Wayback Machine and Wikipedia because Wikipedia is all about documenting. Some sources go away, but you want to have them permanently. That's why Wikipedia articles also have a permanent ID and it just seems to fit together so nicely. It does. And so let me shift a little bit to our work more closely with Wikipedia. So about books. We've been digitizing books for a long time. Before coronavirus, we were doing about 2,000 books a day. We get much of our books now from an organization called Better World Books. Better World Books is a used bookstore, basically, like a big online used bookstore. And one of our nonprofits bought it last year. We wanted more of the books that we wanted. We were having a hard time getting them. So we bought one of the world's largest used bookstores to get it. It's about a 400 person company. And then we prioritize the books that we want to get from that bookstore based upon whether they're referenced to Wikipedia articles and whether they're cited in college and university courses. And also whether they're popular in certain library lists. So Wikipedia is a huge driver for us in terms of identifying the books that we want to digitize. Why do we want to do that? Well, there's a couple of very specific projects that we've done with Wikipedia over the years. The first one many of you have heard about is where we fix broken links. Fixing broken links is kind of remedial work. And it's important. I wish we didn't have to do it. I actually think it would have been better if when we build things, things can be born archived so that you wouldn't have link rot. You wouldn't have the risk that a given URL could just go away or content drift. Maybe even more difficult than link rot. Link rot's when a web page just goes away. Content drift is when that URL resolves but what is there is different than what it was before where you'd expect it to be the same. Like you wouldn't expect front pages CNN.com to be the same every day but you would expect say an article to be the same or at least to be able to know the differences like you can in Wikipedia. So we started working several years ago as I noted earlier archiving outlinks from Wikipedia and then outlinks from those pages. And then we also began working, going through Wikipedia sites, identifying broken links and where we could edit those to correct them, to fix them, to restore them. And I think today we've restored about 14 million links on 30 Wikipedia sites. Obviously 30 is not 300 and a push for me this year is to get that number up so that our software which is called Internet Archive Bot is running on all 300 Wikipedia sites. I work with a great team of people doing this work. Many of them longtime Wikipedians, Maximilian Dorr is one of them, Stephen Balback, James Hare, very active in the Wikidata space, Edward Betts who used to work. Oh my God, this is so weird. I said Edward Betts and his name just zoomed by on my screen. So that was strange, it was spooky. Okay, and I, Edward used to work for the Internet Archive but I didn't know he was such a heavy duty Wikipedian but I met him in Stockholm again last year. And so that led to a collaboration that worked. And then also Jake Horlitz and Jake was the founder of the Wiki Libraries Project at the Wikimedia Foundation. So that's the team that's currently working with me on this work. So we started with the Fixing of the Broken Links but what we've been working on now for almost a year is adding links to citations of books and other things like journal articles on Wikipedia articles, right? And so we've been working on 10 specific Wikipedia language editions so far. We've added more than a quarter of a million links to more than 125,000 books. We got slowed down a little bit because of COVID but we're cranking away at it. And I just wanna show you, I don't have any slides or anything but I'm just gonna try one little screen share just so with this, so I don't know about the rest of you but when this started I got books about viruses I went to Wikipedia articles and I wanted to watch movies about viruses. So I watched Contagion, right? Contagion was a 2011 movie and then, let's see if this works here, I'm gonna do a screen share, I'm gonna say screen share and I went to, is that working? So share. Okay, good. So that's the Google search and I saw Contagion and it says 2011 film Wikipedia and I went to it and I was looking around and there's an entry for a book and the book is Animal Viruses, Molecular Biology, Pendra and Nephaviruses, this is a chapter actually in this book and ISBN number right there. Well, that's really great. I wanna learn more about animal viruses so I'm gonna click on the link and I'm gonna be able to, oh, wait a second, I can't read what's there. Oh, interesting, yeah. They want me to buy the book for $319 but that's interesting. That's the book I wanna see, bats and viruses, I wanna read that chapter, I can't do that darn. So anyway, that was, there we go, okay. So that's an example of what happens when a book is cited in a Wikipedia article that you can't read and I wanna fix that. So here's the proposition, every single book, every single article, every single journal article, every single thing that's referenced in every single Wikipedia article should be a click away. That's the proposition, why not? And so that's what we're working toward. In this particular case, I found a copy of that book and I really wanted to have it ready for today but it's not ready yet. I bought it for $197 from a used bookstore. It wasn't Better World Books. I would have got a discount if I got it from Better World Books but so I bought the book, $197 plus shipping and the best part is the book was deaccessioned from a library. So what does that mean? That means a library threw it out, basically, right? It was deaccessioned, taken out of a library, taken out of circulation, some library somewhere, some taxpayer dollar or some donation. Somebody bought that book at some point and then it was no longer available. Now, I don't know if this book is, the be on the end all is gonna have the answer to the virus but I think it's an example of the value of access to information when and where you want it and so as soon as we have, as soon as I physically get the book, we're gonna digitize it but I'll link it up. Now, obviously I don't scale. If I just go around buying all the books that we want to link, it will take me a few centuries but two things, one, what can scale is if we can automate the process through our own acquisition process and through partnerships with libraries and foundations around the world and some actively pursuing relationships with foundations and libraries around the world and also maybe if we can accelerate more of that grassroots individual, Wikipedia style, barn raising effort to identify and digitize and link primary sources to Wikipedia articles. Do you read your books before after scanning? Do you read them on paper or on screen? Well, I actually, I don't really like reading paper anymore personally so the answer is digital, always digital. For me, digital. I buy the hard copy and I donate. I mean, that's just, once again, that's just like the one offs but I'll tell you about another book. And this book came out last, it came out today, actually. The book is called Active Measures and it was written by Thomas Redd and it's a book about the history of disinformation. We're gonna make a distinction here between misinformation and disinformation as like when it's a deliberate attempt to mislead as opposed to just being in error like misinformation. So Active Measures came out, I bought four versions of it yesterday. I bought a hardcover version, I bought a Kindle, a Google Play and an Apple Books version. I normally don't get so obsessed but the reason I did is this book may be the most highly archived, most well archived and cited book ever published. They are useless to Prelative. Okay, so there are 67 links in the book to Wayback Machine links and there are 75 links in the book to specialcollections.archive.org with thousands of pages of primary source documents. Many of these are formerly classified government documents from Russian archives, US archives and other places around the world. So 142 links are in the book which means that they're, first of all, you can go deeper if you want to, you can go a lot deeper but also because the links in the book are archived links, they're not gonna go away. At least there's a much, much, much better probability that they'll stick around for a long time than if they weren't archived. Mark, I wanna make sure that we stay on time and we have about 15, 16 minutes left but I wanna open it up also to the audience questions. Before we move to that, maybe one less question from my side, Wikipedia has run by volunteers, many of whom we have in the audience. How important are volunteers for the internet archive? Hugely important. I'd say, I work with many volunteers, we use Slack pretty extensively so I invite people in but researchers, I mean, you had a researcher on your team, Bob that did some work with us so I would say we have data, we have applications, we have projects, there's a storytelling project that I'm trying to bring to fruition around using web-based archives to tell stories. So we've got project ideas, we've got data, we've got APIs, and we've got people who could help also. So I would just invite anyone who thinks they might want to be able to contribute in whatever capacity to please reach out to me at mark at archive.org. Great, thanks a lot Mark. Now over to Isaac who I think has curated some questions. Yep, thank you, thank you Mark, thank you Bob. First question comes from Zebra, and they say, first great strategy for picking which books to digitize and all strategies would have some downside. However, rare books are hard to find by Wikipedia's and would then likely be overlooked. Is there an equivalent to archive this link for rare books or other rare media? No, there isn't. And we really don't focus on rare books just because of the physical aspects of them and the expense. All the books that we work with, actually we own, we purchase them where they're donated to us. But having said that, there are other projects that do focus on rare books. And that's a great question because that's caused me to think what I could do as partner with some of those projects that have already done the digitization. I know Dan Brown, he paid for a whole lot of books to be archived about feminist spirituality and history. And I'm just thinking to myself, I bet there's an opportunity there to link some of those books to Wikipedia, that those particular people working on that library maybe haven't thought of yet. So thank you for that question. Next one comes from Nicholas, and they say, how do you deal with Facebook scraping not allowed policy? Not allowed. Proceed until apprehended. You know, we just do the best that we can. I mean, I'm not sure if that was a technical question or a policy question, but it's, yeah. We do the best that we can. But the other thing too is if people are using our safe pays now feature it's them doing it. I mean, they're initiating, individual people are initiating a request. Great. Next one comes from Omer, and they say, who are the other techno utopians of the pioneer generation like Brewster whose projects are still around as non-profits and not for-profit corporations? Well, I mean, in the Bay Area there's Creative Commons, for example. You know, certainly Wikimedia. I mean, big time public knowledge is another one. There's Mozilla, you know, Found Foundation. There are, gosh, I don't know. I'm sure there's others that people, Association for Progressive Communications, APC.org, as I noted earlier, Amnesty International. You know, not necessarily thought of as a tech organization, but certainly more of access now that people do that the great conference that was gonna be held in Costa Rica, I think, this year that's not gonna be be held. But it's called, what's it called? The Internet Rights Conference. Yeah. There's many, many to the study of knowledge management education that do OER Commons, open education resource organizations, Wiki Education Foundation, many. Great. That's actually really nice to hear. From Andrew, they asked, are there any plans to expose archive resources via API? Absolutely. If you go to archive.org, I'm done on the bottom. I think there's, we list all of our APIs there. If you go to web.archive.org, which is the way back machine on the bottom left, there's a link to APIs. Our APIs can always be better, but we've got a great Python library that you can do magic with. And that's one of Brewster's obsession. I should say the guy who founded this still runs it. So Brewster typically would be in the office every day. Now he's just on Zoom every day. And he's an engineer. He went to MIT and he was a software developer first. And so we're totally into, I'd say one of their top five strategic directions is trying to make it easier for other people to use our services, contribute to the services through the APIs. Great. Let's continue with this quick fire. From Hussama Dean, they say, have you thought of using tools for automatic semantic annotations to link books with Wikipedia references? Yeah, we have. And I think that was, I would like to, could you contact me directly? markuparchive.org? I'd say, we are doing a little bit of, fuzzy logic matching or whatever, for books that don't have ISBN numbers, for example. There's a bunch of challenges with books. One challenge is books that don't have ISBN numbers. How do you know which edition of a book sounds you're referring to? The other thing is, in many cases, the books we're talking about here are, these are books that are still being sold, right? So we just can't take a book and just make a digital copy of it and say, hey, anyone can have the whole book? That's just, that's not what we do. And in the case of linking to these Martin books from Wikipedia articles, we try to, we link to a preview, so two pages of the book. And so the trick is, how do you get the right two pages? And so, and if a page number is referenced in the article already, then we try to match that to the page numbers, the physical page number, the logical page number. And so, depending on how the book was scanned and how accurate the metadata is, that can be a little tricky. In many of these cases, if people wanna go deeper and they wanna see the whole book, then they can borrow a digital version of the book with digital rights management for up to two weeks. And so we lend out books equal to the number that we own. If we own three copies of a book, a paper, then we will lend out three versions of it in a digital format. We call that controlled digital lending. Then you can read about all of that at controlleddigitallending.org. Is that something that, this controlled digital lending, is that something that came out of your organization or is this a broader movement? It evolved, you know, we had created something, actually, the late Aaron Schwartz worked with us, and Aaron created an open library back in the day. And it was very similar to Wikipedia. The idea was one, it was basically a wiki, one page for every book ever published. And that was the idea, metadata about that. And now for about four million of those pages, I think there's 20 million pages, clearly there are many more books published. The actual book itself is available, at least in a preview fashion, if not a complete borrow fashion, house.archive.org. Very nice. One question that I noted down here earlier was kind of putting together the two different angles from Wikipedia and Internet Archive. In Wikipedia, there's this big debate about inclusionism versus deletionism. So some people wanna have Wikipedia articles about everything. There's nothing that does not deserve Wikipedia article and others are more this curationists. Some things are not noteworthy enough to go to Wikipedia. So I'm wondering how the, is the Internet Archive's attitude more like a hoarder's attitude? We wanna get everything, or is there stuff that you think it's just not worthy of being included? Yeah, it's not really, it's not an either or. I think certainly our staff, our team and volunteers who work with do focus on things that what, we focus on certain things that obviously we don't focus on others. So I already mentioned radio, about 1200 radio stations. So we had to pick those 1200 radio stations. And we didn't go for popular music radio stations, we went for talk radio. So that's just kind of at that level. Having said that, just like with Wikipedia, a big part of what makes the Internet Archive what it is is the contributions of just anyone who wants to contribute to it. And so there were thousands and thousands of submissions every day of text, of audio, of video, old time radio, people that are just, I mean, we had one, I just, let me come to the end here, but in late February, no, in late, late March, we launched something called the National Emergency Library. So the National Emergency Library is what we took about 1.4 million of our digitalized books and we lifted the waiting list on them. They're still available through DRM, but we allowed more copies to be available than we physically owned. Then we did this because almost every library in the world was shut down. And certainly in the United States, almost every single library, so all these books, hundreds of millions of books are sitting on shelves. People can't get access to them. So we launched that. And yeah, the... Would it be a great way of keeping people in the libraries employed, giving them all a scanner and start scanning A to Z? Yeah, yeah, so. What if, for example, let's say in 1999, I created a webpage and it has like this really embarrassing photo of myself and I, it was snapshot it for whatever reason by the internet archive. Is there any way I can get it off or? Yeah, look, Bob, if you just write to me and I'll take care of you, say info.archive.org we have no interest in embarrassing. Oh well, if you're a politician, a public figure and you do things and then it's public, if it's a public event, then we're not going to. They may be a little different, but generally speaking. My question is actually a serious one because in Europe we have this right to be forgotten. So there are lawsuits now about people wanting, like their pages off Google for example. I'm wondering have you been... Yes, certainly, any request that's a legitimate request under the law of the United States is the Digital Manual Copyright Act about things that you'll own, but also under the European right to be forgotten. We generally honor those requests. If we didn't, it would be very hard for us to. But I wanna go back to the National Immersion Library and say we noticed some interesting behavior at one point. We wrote a blog about this. So if you go to blog.archive.org, we wrote a blog about this yesterday. We talked about kind of obscure things. It's like, well, how do you know, one person's used the word hoarding. We tend not to do that, but something one person is interested in may not be of interest. So we noticed that there were thousands of books that were being downloaded from the National Immersion Library every hour. And then put back, people were borrowing them and putting them back where someone was over and over and over again. We thought, oh, someone's abusing the system, right? They're trying to, I don't know, whatever. But it turns out it was someone who was interested in Isaac Asimov. And what they wanted to do was they had built a dataset of every reference that they can find to Isaac Asimov in every book they could find. And that's massive collection of references to Isaac Asimov. So that we didn't expect that anyone would have that particular use case based upon what we've done. I just wanna say, things that I'm interested in is, I'm interested in having ensure that everything that people add to every Wikipedia article is archived in real time and that there is a born digital, a born archived link available to people. I wanna make it easier for Wikipedians to write better articles by having easier access to more useful data sources, including books. I mentioned earlier an interest in Easter Island. My personal collection of Easter Island books is pretty extensive. Only a small number of them are currently referenced in the current Wikipedia article about Easter Island. So I wanna make it easier for editors to get access to the reference material that they can use to write better articles. I'm also interested in how different topics are treated in different languages. And so I spent a fair amount of time about Chernobyl, for example, and collected an archive library of books about Chernobyl. So that the Ukrainian and the Belarusian and the English and the Russian, the Japanese articles about Chernobyl at least are able to benefit from common data sources. I could go on about that, but it's kind of like looking forward, moving forward. We'd like to see how archives and libraries can be of greater use to Wikipedia editors and Wikipedians writ-march. Isaac Asimov would have been the perfect segue back to Isaac because I think he has one more question from the audience. Okay, thank you. One final question. I think ties together what you just said with Dr. Wade's keynote as well. It's from Lodovic. And they say, what are some of the best ways for people to help make the internet archive more diverse, thinking in the context of culture, country, gender, and so on? I would say, bring yourself to the archive, right? You are part of that diversity, or if you're part of that diversity, but I think it's not just what we're doing, it's what we're all doing together. And so follow your particular passion and your particular expertise and contribute. Contribute those books, those materials, those audio recordings. I'll just close with, for myself, there was a children's author that I wrote in, that I wrote that I read when I was a little boy. And it was a series of books I've written in the 1920s to the 1930s. And I actually went on eBay and I bought all the books. I think it was like 37 of the books that this author wrote. It took me a couple of years to get all the books, but I got all the books, 37 of them, digitized all, there's a collection now, the Leo Edwards collection on archive.org. But then I thought to myself, that's interesting. I mean, I could do that. I could spend the money to buy a used copy of all of these books that were written so long ago. But why not all the books? Why not have an archive and make available every children's book ever written, ever published in every language, and make all of that available? Many of them are so old that they're no longer in copyright at all. So that's not even an issue, right? And so there's really no reason. And we have the technology, it's not a lot of money, and it would be a benefit and value to the world. So that's another example where people can come together and contribute their own particular point of view. Mark, this barrage of superlatives came exactly when my cell phone reminded me that time's up. So I think it's the perfect time to stop. Thank you so much for this conversation. I had a blast. And I think it was illuminating for a lot of the people. Great, great for everyone. For me, certainly to learn so much about the Inner Archive. And I hope you will have a lot of people writing to mark at archive.org. Thanks a lot. Excellent. Thank you very much.