A Fresh Face in Hell with Jonathan Evison
A Fresh Face in Hell with Jonathan Evison is an hour long, conversation based, format. Guests include writers, musicians, athletes, actors, philosophers, scientists, and an array of experts across a variety of disciplines. Subjects will be treated with a mix of humor and thoughtful exploration.
A Fresh Face in Hell with Jonathan Evison
Richard Cooke - Wikipedia Will Save Us
Use Left/Right to seek, Home/End to jump to start or end. Hold shift to jump forward or backward.
What makes Wikipedia work, and what does its success reveal about human nature? In The Last Best Place on the Internet, journalist Richard Cooke explores the unlikely rise of the world’s largest free encyclopedia and the community of volunteers who built it. We discuss the conflicts, biases, and imperfections that threaten the project, along with the extraordinary collaboration that keeps it going. In an internet increasingly dominated by corporations, algorithms, and AI, Wikipedia represents a different vision of how knowledge can be created and shared. We talk about whether this messy, imperfect experiment might be one of the last best places left on the internet.
linktr.ee: https://linktr.ee/freshfacepods.evison
Welcome to a fresh face in hell. I'm your host, Jonathan Evanson. Every week we talk to someone new and interesting, somebody intellectually curious, somebody who will hopefully renew our faith in this cold, cruel world we live in. A fresh face in hell. So we all use Wikipedia. It's an amazing source of information. Uh, and one that's really being put to the test in the post-truth era. The very idea of Wikipedia is remarkable. It's the largest repository of information ever assembled by humankind, bigger than the library at Alexandria, built from the ground up and maintained solely by volunteers. It's a great example of what humanity is capable of when we put our collective noggins together instead of fighting all the time. Today we talked to award-winning journalist and author Richard Cook, whose new book, The Last Best Place on the Internet, is a playful and provocative exploration of mighty Wikipedia, how it was built, why it's thrived, and what we lose as its ideals disintegrate. A lot of big questions here. How do we decide what's true? Who gets to control knowledge? And what happens when AI replaces human curiosity and participation? So let's dig into it. Please welcome this week's Fresh Face in Hell, Richard Cook. Richard, thanks for playing, man.
SPEAKER_00Jonathan, thanks for having me.
SPEAKER_01Um, so we all know what Wikipedia is, we all use it. But um just for context, let's start at the beginning. Who started this grand experiment, how and why? Give me some history.
SPEAKER_00So it was started by uh two people, uh Jimmy Wales and Larry Sanger. And uh there's a controversy between those two. Uh pretty much everybody in the world thinks that Larry Sanger is a co-founder of Wikipedia, except the other co-founder of Wikipedia, Jimmy Wales, who says that he isn't. Um an online encyclopedia was not just an obvious idea. It was something that already existed. Um, in fact, Encyclopedia Britannica had an online encyclopedia while Wikipedia was still a twinkle in someone's eye. Um, the difference is that they allowed anyone to edit, uh, which was a terrible idea. Um, that meant that it is now the largest encyclopedia that's ever existed.
SPEAKER_01It's a terrible idea, but I like it. I mean, I like the idea of it as long as it's being and we'll get we'll dig into that. Like it's gotta be monitored. So uh I'm a guy who used to have a lot of faith in people. Uh the great tragedy of my life, personally, but I'm sure many share it, is that now I have about 50% less faith in people. Uh 20 years ago, the idea of consolidating a database of knowledge collectively sounded like a brilliant idea, but now we live in a world where we're forced to view information in what amounts to a broken mirror or something. So anybody can edit Wikipedia, right?
SPEAKER_00Yeah, that's right. And look, this is uh a difficulty that we have now. Um Wikipedia is founded on something called, explicitly called, a good faith collaboration. And it can be hard to have good faith collaboration without the good faith part. Uh I kind of share that sensation that you're talking about where, you know, maybe people you thought were better than that weren't better than that in the end. Um, but as everything else gets worse, uh, Wikipedia can look a little better by comparison because there's so there are so few other examples of what you might call digital civic virtue sticking around.
SPEAKER_01So I mean I've seen where Wikipedia puts up notices um like at the top of certain things. They're just kind of like caveats, like this information needs citations or whatever. And I'm not picking on Wikipedia here because generally I found it to be more reliable than most sources. But how can we in theory know that the information we seek there is legit?
SPEAKER_00Look, we can never know a hundred percent, but I feel like that those kind of uh those signs that you're talking about, those banners, um, at first glance they have a bit of a, you know, abandon hope all year here, enter here feeling. You know, they might say that, you know, the article lacks citations. They might say that the article reads like it was written by the subject of the article, but I think those warnings are actually a sign that you're on firmer ground. You know, um the nobody else proffers information that says we're not sure about this. Uh it's it's contingent, uh, tread carefully. And I think that that is itself not a guarantee of quality, but certainly an example of self-reflection that maybe is becoming rarer.
SPEAKER_01That's actually an interesting point because I've seen on my own Wikipedia page. Yes, I've been there, um, that it has one of those citations notice. And no, I don't have anything to do with writing it. In fact, I can't. I I wish I could edit one thing about my Wikipedia. Maybe you can help me. My picture, which is the first picture anybody who googles me shows up. Somebody put whoever put my Wikipedia page up or whoever edited it, put a picture of me that looks like I look like I'm suffering from Bell's palsy and simultaneously swallowing an oyster. I mean, it's a really, really bad picture. And I found out, I posted about it once on Facebook, and I'm like, hey, does anybody know how to get rid of this photo? And a bunch of other authors go, Yeah, I have a terrible photo of me too. And we kind of got our heads together and we figured out it was this one guy at the Texas Book Festival that takes bad pictures of authors. I think he's trolling us because other authors are like, I've been wondering the same thing. I've got a terrible picture of me. How do I get rid of it? So there's this one guy, which is a pretty great draw, but like, if you like you check out my Wikipedia page, I'm like, I mean, it's really bad. And so proof that I don't know how to edit the thing or I didn't write it myself, but it does have that citation thing. So that's interesting. That's actually a good point. That means somebody's looking at it.
SPEAKER_00Yeah, Jonathan, I feel your pain. There's a lot of bad photos of people on Wikipedia. Uh, it's partly a rights issue that's uh, you know, copyright photos. I'm sure your author's headshot is a copyrighted photo. So instead, what happens is a worthy uh Wikipedian goes around uh sticking their iPhone in people's faces with the flash on and getting what they can from the side of a signing. Um, so it's not pretty, but it is you, I'm afraid to say.
SPEAKER_01Yeah, it is painfully. So this is this is interesting to me because uh it's like uh the blue check, you know, in the social networks. One day I woke up with this blue check on Facebook. People kept saying, hey, how did you get that blue check? And I didn't know what the hell they were talking about. And then I noticed by my picture I had a blue check. Apparently, you can apply for that and and and and they will grant it to you if they what I don't know what the you know, how they judge that. But anyway, I just woke up with mine. How does it can anybody actually have a Wikipedia page? Because it doesn't seem like just anybody can just slap up a Wikipedia page, they have to like how does that work? Who determines like personality-wise, who determines if you have a Wikipedia page?
SPEAKER_00So you to have a Wikipedia page, and I mean some people don't like having a Wikipedia page. A lot of people do. It's almost a little bit like being in who's who or something like that. Uh, in fact, uh, if you know what a Nepo baby is, you know, people who are kind of a a nepotistic hire from a showbiz family or whatever, one sure sign of a of a Nepo baby is supposed to be if you look up their parents on Wikipedia, they both their names are written in blue because that means that they will have pages de uh dedicated to themselves. So to have a Wikipedia page, you have to fulfill um a quality which is called notability. Um, it's not always easy to work out who is notable. Um, they tend to go for, you know, are you uh winning awards uh in your particular field? Um do you get interviewed about your work? Are there are there kind of media institutions who are interested in you? Is there a paper trail we can see uh that shows that your work is important? But it's probably harder to have a Wikipedia now uh page now than it used to be. There's definitely some people who kind of wrote their own and and they stuck around. Um, but uh now the threshold is is much, much higher. So if you've got one, uh you should feel good about it.
SPEAKER_01Oh, well, I I feel fine about it, except for the aforementioned photo. Um what I'm wondering is is say I again, I would love how to, you know, I I've seen information on wiki before, but I wouldn't know how to edit. I mean, not just about myself, about anybody. Um, like music stuff, I'll be looking up and I'll be, yeah, that's not true. I just happen to know personally or something. If it's about a Seattle man or something, I'll like, no, that's not true. Um, but I wouldn't know how to go about it. What if I what if I have uh I had a writer who I just thought was amazing, but they're very obscure, published, but nobody really knows them. You may not call them notable in a popular culture sense, but I went to Wikipedia and I wanted to, I'm thinking this person deserves a Wikipedia page. I want to spread awareness. What would I run into if I tried to start a Wikipedia page? If I mean who who will determine whether or not that person is notable enough?
SPEAKER_00That's a great question. And we should say that, you know, we still talk about Wikipedia as the encyclopedia that anyone can edit. That has a little bit of an asterisk next to it. There are some types of editing that are harder than others. Um, one of them is uh starting a new page for someone. That's not something that's so easy to do if you've never edited Wikipedia before, both for technical reasons but also for cultural reasons. Um, so there are people on Wikipedia, and again, these are volunteers, this is not like a job, um, and they just specialize in deleting stuff that they think is not notable. And there are some people who delete a huge amount of things, and once they nominate something for deletion, there's a process where people go and kind of argue about it.
SPEAKER_01Um, this is what I want to hear about. I want to hear what kind of dialogue. We're not just all operating in a vacuum and we're not all the arbiters of information. There still is this sort of egalitarian democratic theory behind this. I can't just get on there and start deleting people because I don't like them or something like that, or deleting history that I don't like. Tell me a little bit about how this back and forth, this process, this nomination process works, because it's all opaque to me as a wiki user.
SPEAKER_00Look, it's it's really opaque and it and it can get complicated quite quickly. So I'm probably gonna give a uh a summary of it rather than go through the gory details of the whole process. Um, but roughly what happens when a page is deleted is it goes into a process called articles for deletion. Um it will be put in a kind of clearinghouse where people will look at it and to decide whether it's gonna stick around or not. Um, the people who who tend to nominate it for deletion, some of them only do that. They will just look at newly created pages on Wikipedia and and nominate a whole lot of them for deletion. Um, that's partly because there's almost a sense that Wikipedia is close to as big as it should be, and that um there's a tendency now for people to especially write Wikipedia pages for themselves. So the moment that there's a there's a new Wikipedia page for a person coming up, they're not really someone you've heard of, they might not have those those media citations, there's a good chance that someone's going to uh kind of mark you for death, as it were. And um, once you're in the article for deletion hill, it can be quite hard to get out.
SPEAKER_01So, I mean, but who are these people? Like they're just, I mean, I guess if uh like if you read something like Malcolm Gladwell's tipping point, these would be be the people that would be call themselves the Mavens. They get to decide or they're the arbiters of what matter, and they're gonna say that's not important enough, that's not part. They're nominating it. You say it goes into this pool. Who's right? Who else is in the pool? Just anybody who just happens to look, somebody else who's interested in like having a decision in the say? How many people does that vary? Like it's just it's all so confusing. I want to I want to understand logistically how this is working.
SPEAKER_00Yeah, so I mean, a lot of the a lot of the stuff which is happens like this on Wikipedia, there are clear processes and there are clear policies. Anyone can read those, and in theory, anyone can participate in them. There are some exceptions to that. There are some hierarchical positions. Um, one of them is called an admin. You become an admin by getting voted in as an admin. Um, and there are some higher than that as well called stewards. And these are people who can do things like, you know, lock down an article if it's been vandalized or if a celebrity dies and they know that a huge number of people are going to go onto a page and start making changes, they can uh kind of put some oil on the water there to make sure not just anyone can say anything uh just during those uh particularly uh high traffic periods. Um but those the number of admins and stewards is very small. And the short answer This is what I'm wondering.
SPEAKER_01I mean, is that a hundred people? Is that ten people?
SPEAKER_00Is that a thousand people or um so the admins are in the hundreds and the the stewards uh I think in the dozens, um, but they're the in the low dozens. So these are very, very small numbers of people. Um and we're also talking about who actually turns up. So the the short answer is that Wikipedia is um is written by the people who are there, or certainly these kind of decisions are made by the people who are there. And there are a relatively small number of people. So um if we look at the overall picture of English Wikipedia, it's something like 2,000 people making more than a hundred edits a month. So for a project this size, which is, you know, in the billions of words, that's that's a pretty small number.
SPEAKER_01And you say there, they're really all over the place. I mean, there's no central, there's no office building or there's no headquarters, right? I mean, these are just people sitting at their desk. Yeah, that's right.
SPEAKER_00And and it can be really hard to get your head around, you know. I I speak to people all the time who who tell me that they donate to Wikipedia and they say it's for the fact checkers or, you know, it's to pay for all this work. All the work is unpaid. Um, all of these people are volunteers. So where does my money go?
SPEAKER_01Because I've fallen for that a billion times. I've probably given I've probably given to Wikipedia like 10, 12 times.
SPEAKER_00Yeah, that's right. And look, that money does go to Wikipedia, but it goes to something called the Wikimedia Foundation. And that's that's a not-for-profit um which kind of owns Wikipedia, is the best way to understand it, but it doesn't control it. And that can be a surprise, uh, especially to the people um on Wikipedia uh who can kind of till the foundation to to go jump.
SPEAKER_01Okay. Well, thank you for being patient here with my questions. Because I'm really trying to understand the organizational structure of this, because it just seems, you know, when something as democratic as this, where everybody has a voice, now I have a much better understanding that there's only a couple of thousand people or that are administrators and a couple of hundred, I believe you said, that are actually stewards. So that's a lot for them to comb through. I mean, you're talking about millions and millions of entries. How does such a small workforce of volunteers cover that much ground? How how how how can they steward it?
SPEAKER_00How can they possibly do this? Yeah, yeah. I mean, it's fascinating, right? So, first of all, there are a lot of automated processes on Wikipedia, and there have been for a long time. So there are uh bots and scripts which run all over Wikipedia looking for common types of typos, looking for common types of vandalism, uh, looking for formatting that's out of whack in a predictable way. And this is a really, really important part of the Wikipedia ecosystem. One of the reasons that it's been so hard to replicate how Wikipedia works elsewhere is that you don't have this kind of symbiosis, if you like, um, between the people and the computers that you get elsewhere. It's a culture that that sticks them together and makes them work. And um making that from scratch is almost impossible. Um they triage it as well. So when you've been on Wikipedia a while, and and some of these people have been on Wikipedia for decades, that they know the kind of things which are gonna go wrong. They know that that um constantly you're gonna get a rush of people writing articles about themselves. They know that when a celebrity dies, their page is probably gonna be vandalized. Um, so this this means that they can uh apply resources uh where they're needed, as it were.
SPEAKER_01Okay. So the it's sort of a similar, similar discussion to identifying AI writing. You know, they're saying that we we uh there are now programs, I believe, that scan for AI writing. Like if there's too many M-dashes, it's a red flag because AI likes to use the M-dash or whatever I've heard separate. So that explains a lot. There's automated processes sort of scanning Wikipedia for obvious mistakes and vandalism all the time.
SPEAKER_00Yeah, that's right. And to be clear, um these are these are not uh large language models. This is not, you know, Chat GPT weighing in. This is stuff uh which has been, you know, it's much cruder than that. It's been around for a long time, and there are special policies to govern its use as well.
SPEAKER_01So in your book, you talk a lot about um ideological clashes, clashes and philosophical disputes among the founders. Anytime there's kind of a brain trust situation, I'm always curious what sways the direction of a project. Can you talk about some of these ideological clashes and philosophical disputes?
SPEAKER_00Yeah, definitely. So um you could say that you know, philosophical differences are what define Wikipedia. And um, what brought Jimmy Wales and Larry Sanger together was um philosophy and also arguing on the internet. Um, so Jimmy Wales was an objectivist. I'm sure some people listening to this show will know that means uh he was very into Ayn Rand books and and her ideas. Um but uh unlike a lot of objectivists, he was talking to more kind of you know card-carrying philosophers at at colleges and elsewhere. And that's how he bumped into into Larry Sanger. Um, Larry Sanger was a trained philosopher from uh working in Ohio, and um unlike his kind of college brethren, uh he really thought philosophy was was a life and death stuff. He he grew up in Alaska and he saw some of his kind of peer group get into drugs, and the the takeaway for him from that was that living a rational life was the most important thing you could do, and philosophy was a big part of that.
SPEAKER_01So he would talk this kind of reminded me of uh Ram Das or uh or or like um what's his name? Um late the 60 philosophy. I mean, Timothy Larry started experimenting with the drugs, and that was born out of philosophy and and yet uh philosophy of consciousness and uh Charlie Watts or not Charlie Watts, but uh Alan Watts. I mean, so is this time period we're talking about, like mid to late 60s?
SPEAKER_00Uh it's a little bit later than that. Um so Larisanger's not he's not a kind of a counterculture guy. If anything, he's a counter-counterculture guy. Um, he's quite conservative, and he winds up talking to these objectivists on the internet and arguing with them because ironically, it's only amateurs who take philosophy as seriously as as he does. When he goes around his philosophy department in Ohio, he's saying to people, you know, why why do you study philosophy? And they think, oh, it's interesting. And he's like, nah, this is this is life and death stuff. So that's what brings him and Wales together. Um, and what happens then is that um Wales makes some money as a trader. He has an an idea for a whole lot of internet businesses, some of which kind of start to stick, but it's a very volatile time around the dot-com boom. One of these ideas is an online encyclopedia, and he's got exactly enough money to make one hire, and he hires a philosopher. And that's not not something that I would do, not I think an intuitive thing to do. Um, but that turns out to to make a big difference to why Wikipedia works.
SPEAKER_01I think it makes sense. I mean, who were the stewards of knowledge in ancient times? I mean, if you talk about Carthage or Alexander, I mean, who who are the people? Well, the answer might actually be religious people. I don't know. Um, I still love this idea of a large scale cooperation that can produce something unique. Useful, durable, uh, and intellectually ambitious without conventional institutional control. I like that. I'm a bit of an iconoclast, but how do I have faith in this idea when it seems to inherently lack accountability in a way?
SPEAKER_00So I don't think that you need to be an optimist. It's a little bit like democracy, you know. Um, you you can be a cynic or a pessimist or a skeptic and and still believe in democracy because it's might be bad, but it's it's better than all the other systems. And Wikipedia is a little bit the same way, and not coincidentally, is a democracy. You know, I had a Wikipedian say to me, Wikipedia is terrible, it's just better than everything else. And I think there's something to that. Um, I mean, one of the things that the the alternatives have never been able to provide uh is the scale, which still has a human in the loop. You can have a small encyclopedia where people are paid to write it. Um, you can have a large encyclopedia where AI produces it all, but it's only Wikipedia that can have a large encyclopedia where humans write it. And and that's something very distinct. And it's not something that's been easy to copy or knock off its pedestal either.
SPEAKER_01Okay, so I mean, before the internet as a writer, I was always an avid collector of encyclopedias. Like I'd buy them at book sales and stuff like that, because in those days you couldn't just fact-check on your laptop. You had to go to the library or the research library, and sometimes the research books were like, you know, you know, you have to blow the dust and they won't let you check them out. And you may need to lean on this book pretty hard. And so Wikipedia opened up a lot of uh a lot of fact-checking possibilities. I'm trying to get a sense of the scope. So I mean, I owned like I owned a world book encyclopedia. I owned uh, and it was always 27 volumes, you know, like one for every letter, and then like a glossary. And you know, each of the books was probably 1,500, 12 to 1500 pages long or something, tiny type. What does what does Wikipedia printed out look like scale-wise? Like how much bigger is it than uh than than those sets of encyclopedia that took up two bookshelves?
SPEAKER_00Like, so it's a lot, a lot, lot bigger. So if we take Wikipedia in every language in which it exists, there's there's 300 or so. Um that would be around 400, maybe 450 times uh the size of the biggest ever paper edition of Britannica. So people have uh they haven't done it for a while, but um, when Wikipedia became the biggest encyclopedia ever, there were some people who experimented, you know, with as like artworks of printing it out. Um and it was big enough that even if they had a delicated gallery space, um, it was too expensive for them to completely fill it. Um so they had, you know, hundreds of volumes of this size. Um, you're looking at around, if you just look at English, it's something like 60 times um the size of that uh world book you might remember. So this is, you know, an order of magnitude bigger, really. Um and difference of scale is also difference of kind. Um, the fact that it's it's you can't print it out in quite the same way because it's changing all of the time. You know, it's changing very rapidly. Uh, you can only get a snapshot of a moment in time, not a sort of final edition in the way that you got with paper encyclopedias.
SPEAKER_01Yeah, and well, that's the other thing. Uh, wiki evolves. Like, I I mean, obviously, I had to be careful, especially in things like science and and and things that are exponentially evolving, that my because I'd be buying a 12-year-old, you know, used version of World Book or Britannica. So I had to be careful to make sure now, and Wikipedia takes care of this itself, theoretically. Like it updates as the technology or whatever you're talking about here or history, as these things evolve, it evolves with it. Whereas I have to buy a new set. That's why they were encyclopedia salesmen, I guess, because after a few years, you got to buy a new set.
SPEAKER_00Yeah, that's right. Um, and look, the uh a lot of encyclopedias were sold to people who never opened them at all or over only open them once a year. You know, it was more like um something to have around the house, like a piano that shows that you you care about your kids' education. Um status symbol.
SPEAKER_01Exactly.
SPEAKER_00Yeah, a bit of a middle class marker. Um so I think books in general are like that.
SPEAKER_01I think there's a lot of people that just have books in their house that don't open them.
SPEAKER_00Yeah, definitely. Um, in fact, if you if you go back, um there's some evidence that with the first kind of modern encyclopedias, uh, there were people buying them who couldn't read. Um, and it was just like having a fancy vase or a nice painting as as something to stick around the house.
SPEAKER_01And that's why they were so readily available for a dollar a dollar a dollar a dark, this dark side to Wikipedia, the conflicts, exclusions, power struggles, deliberate deceptions that accompany any open system, really. But just because Wiki's committed to broad participation doesn't mean it automatically produces fairness, right?
SPEAKER_00Yeah, and I mean probably the lasting um criticism of Wikipedia around this is its representation of women. You know, for a very long time they've been trying to have the same number of female editors as male editors, and and that hasn't happened. You know, they really put their shoulder to the wheel to try and make this happen uh and and they've never got there. Um I'm not sure that they will for the foreseeable future. Um there's also the case that, you know, it tends to cover European and American topics better than it does in the rest of the world, um, you know, especially in places like Sub-Saharan Africa. For a while there were more Wikipedia articles about Antarctica than there were about uh sub-Saharan Africa. So it's still uh very much a lopsided view that it may never be fully corrected, but people are trying.
SPEAKER_01Yeah, this is speaks to the larger issue of globalism in general. Uh like I guess there's the fundamental if there's no governing body, like you say they're trying to get more female editors or trying to even that out. We're again, we're talking about this sort of nebulous day. This is just the stewards and the what do you call them? The the advertising vote on everything.
SPEAKER_00Yeah, so that this is uh again, it's uh it's a little bit of a tricky dynamic because you do have this this foundation, this not-for-profit. Um, and they uh their job is to try and make Wikipedia as good as it possibly can be. But they can't issue edicts, you know, they they can't fire people who work on Wikipedia. They they can control some, you know, really bad actors under limited circumstances. They can just make sort of adjustments to the culture. But if those adjustments don't stick or don't work, you know, it it's not like you then uh clear house. Um, it's all self-nominated. You you can't kind of push these volunteers out unless they're doing really bad stuff. So the best that they can do is kind of try to steer, um, but they're not even holding the steering wheel all of the time.
SPEAKER_01Uh it kind of begs the question. I mean, because you're talking about how there's this sort of western, you know, slant to a lot of the information or the amount of the information. I mean, is there any demographics available to who these people are? I mean, do we do we is there any accountability in the sense that okay, we have 1,400 people or whatever voting on this stuff, stewards and the the administrators. What if it's you know 1,390 white guys? I mean, middle-aged white guys. I mean, do we know who these people are?
SPEAKER_00Are they just yeah, so um so it's best guesses, um, is one way to think about it. Uh a lot of them still work anonymously. And there are people who say that, you know, there are more women on Wikipedia than it seems because they don't tend to answer these surveys, um, or they don't they don't self-nominate um as women. I mean, that that might speak to a a bigger problem or a different problem that just kind of kicks the can down the road. Um, but yeah, I mean, overwhelmingly it seems to be from from what we know, it seems to be men, and they're still concentrated in in uh the global north, as people call it.
SPEAKER_01Okay, so I mean, I guess the fundamental question about this whole Wikipedia experience for me is a question about well, it's several questions. Several things come into play expertise, identity, ideology. You have to consider all these things of the people who were the arbiters of that information. How how how how did old history work compared to this?
SPEAKER_00Yeah, um, look, it can be one of the surprises in in writing this book, Jonathan, is how hard it was to remember what looking things up was like before the internet. Um there's a there's a great section in uh Chuck Kloisterman's history of the of the 1990s, where he talks about, you know, kind of sitting around a tavern with your friends, having a beer, and someone asks a kind of question about a fact, you know, who won the Super Bowl in in 1960 or whatever, 1970. And if no one can remember, that's kind of the end of the conversation. You know, it's not like someone runs out and goes and gets a football almanac. They just go, oh, well, we don't know.
SPEAKER_01Um that that's Yeah, those questions could linger for months and months and months. And then like one night you're at a party and somebody has a definitive answer and can produce evidence, and then you have no way to tell all those people you were arguing with then. I remember very distinctly, like I was mentioning earlier, like being a writer and having to fact-check stuff. Like I, you know, if I was writing about a certain automobile or something, I had to make sure what kind of engine it had or how many windows it had, or was it a two-door or a four-door? And I would have to go. I mean, sometimes I'd have to drive, you know, 20 minutes each way to the library just to get one little factoid, you know? And now I can do that instantly, even if I cross-reference it and check numerous sources. I'm pretty careful about that. I don't just go to one source. I, you know, I try to dig a little deeper just so I'd not, especially now, because it's so easy to be misinformed. Um, what what would you consider a reliable source for information these days? I mean, how do you personally discern a reliable source?
SPEAKER_00Honestly, it's gotten much more difficult, even in the time that I've been writing this book. You know, the difference between Google Search when I started and Google search when I finished was very, very noticeable. I did, um, I didn't use AI at all early on. And I thought, you know what, I'm I'm gonna ask it just a few of the most difficult questions. I'll I'll check everything naturally. Um, but I've got these these kind of lines of inquiry. They I haven't been able to find this stuff out otherwise. I'll see what AI has to say. So one of the questions I asked it was, when was Wikipedia first edited from Antarctica? We know that it was. I knew some people who had edited it from there, but I couldn't tell if they were the first. And it gave me this rolled gold answer that looked perfect. It looked super plausible. It uh seemed to have great citations and sources, but because I knew enough about the topic, I knew that it was out by at least, you know, five years. And it was only when I kind of looked into it a bit more carefully that I realized, okay, this is just this is just totally wrong. And it's totally wrong in a way which is very plausible. And if it wasn't on a topic I didn't know inside out, I may not have checked. Um, I think a lot of people have experiences like that using AI. Um, they're also finding things like, you know, Google Books or the news sites that they used to rely on are harder to find and access. So yeah, I mean, in some ways, the information picture, despite seeming so rich and and wonderful, might be a little bit uh worse than it was even three or four years ago. So I'm I'm struggling to answer that question myself, even though it's what I do for work.
SPEAKER_01Well, I mean, just philosophically, the large language model is sort of problematic to me because anytime you've got a whole bunch of people doing agreeing on something, if you're i i if the idea is based on consensus, you're gonna make some sausage. I mean, this is how it works with art. The more cooks you have in the kitchen, the worse the art gets. So if I'm understanding large language models right, is the more people sort of agree that this is how it happened, this is where hallucinations would be produced, right? If the if there's a bunch of people from Pittsburgh that are convinced that the Pittsburgh Steelers won the 1971 Super Bowl and they didn't, like won't that be kind of the first answer that's produced? Like if the mistake just keeps repeating itself, doesn't the mistake become by consensus the truth? Because that isn't isn't that what the large women model is doing? Is like scanning everything all at once and looking for the most commonalities and deciding that that's the truth?
SPEAKER_00Um we don't a hundred percent know how they work because the processes in most of them are uh commercial secrets. But we do know that they do give things different weightings. So, you know, if it says in the New York Times on the front page that the uh the Pittsburgh Steelers won the Super Bowl, um, then that's gonna be uh a different story to if it's our proverbial man in the tavern.
SPEAKER_01Joe Stankowski from Pittsburgh says it. Okay, I got you.
SPEAKER_00That makes sense. That sounds like like Joe, something Joe Stankowski from Pittsburgh would say.
SPEAKER_01So let's talk specifically about something that I I I would imagine the last eight years or so has become a much bigger issue with Wikipedia, which would be disruptive contributors. Like uh how how how are they identified? How are they assessed? Uh does Wikipedia, it sounds like Wikipedia actually has trolls, right? I mean, are there people who just troll Wikipedia and that's all they do?
SPEAKER_00Yeah, look, it's getting harder to do that. Um uh partly because these these bots that we were talking about that do stuff like find trolls on Wikipedia have gotten a bit more sophisticated. Um, but there's definitely people who made it their life's work to get particular forms of vandalism uh on Wikipedia, for example. Uh there was a guy uh called Willie on Wheels who would just add the words on wheels to hundreds, thousands of things in Wikipedia. Uh and he did this for years and years. So other people just had to go around deleting every time he he put on wheels there. Um and he actually looks kind of mischievous.
SPEAKER_01I don't that's kind of funny. Like, good for him, but like I I'm just thinking more about people with like strong political agendas and you know, trying to rewrite history.
SPEAKER_00Yeah, I think that there's always been a bit of that. Um it's it's still, you know, something you would expect to be a kind of a perennial problem on Wikipedia. It's a little difficult to say, you know, it's gotten worse because so many other things um which are a bit more unstable, like the rise of you know, AI editing on Wikipedia, have overshadowed that. But I think as kind of people have lost faith in institutions and lost faith in authority, um, what's happened is rather than go and trying to get their point of view on Wikipedia necessarily, some of them have said, you know, I'm giving up on Wikipedia and and we need an alternative. And that's where things like Grocopedia have come from, uh, thanks to our friend Elon Musk.
SPEAKER_01Tell tell me a little bit about that. I don't even know about it.
SPEAKER_00Yeah, so um uh Elon Musk decided that um Wikipedia was biased. There are different theories as to why. One of them is that um his Wikipedia page calls him an investor instead of an inventor, and apparently he finds that very upsetting. Um, another is that uh his Wikipedia page repeats the story that ran in Business Insider, um, that he offered a flight attendment uh a horse in exchange for a hand job. Um, so it it's either one of those or or maybe something else. Um but he decided that um after being a long time Wikipedia supporter, by the way, that uh Wikipedia was woke, Wikipedia had gone woke, and that uh he was going to create an AI alternative, um, which would be linked to his um AI model, uh, which runs on X, which is called Grok. So it's called Grokopedia. Most of it was just a copy of Wikipedia. So something like 80% of Grocopedia was just Wikipedia, but there was some definitely some political differences and differences of emphasis. Uh that was a while ago now, and like many Wikipedia killers, it doesn't seem to have killed Wikipedia, or in this case, so far, uh, even wounded it.
SPEAKER_01Well, it sounds like Wikipedia has some safeguards. I uh with egalitarianism being such a big part of the ethos. I mean, I just can't I can understand why a philosopher was hired. I mean, because it just brings up so many philosophical basic questions. Like, what happens when we fundamentally disagree on facts? You know, there's not supposed to be any disagreement on facts, but here we are in the post-truth era where people cannot agree on facts. What what what happens to that? And and how does how does Wikipedia help us find uh clarity?
SPEAKER_00I think there's there's always been some disagreement about facts. And a lot of what appear to be disagreements about facts are really disagreements about values, you know, they're they're disagreements about what those facts mean. Um I think a big difference now is that while previously a lot of people sought out media to be informed, a lot of people are now seeking out media to be affirmed. And and that means that they will kind of disqualify almost automatically um the stuff that doesn't fit their worldview. One thing that I think is hopeful about Wikipedia is that thanks to its processes, it has been able to get together people who disagree about things often very violently. And even though they might uh not like each other particularly or disagree in very fundamental ways, they hash stuff out through these arguments, through these editing processes. And there's there's some studies that show that articles on con on contentious topics, on culture war topics, even, can be more accurate than other kinds of articles because of this process, because they're these, you know, two teams kind of holding each other's feet to the fire, that can actually make it better. I mean, that's in some ways.
SPEAKER_01That's how it used to work with partisan politics. Yes, that's exactly right.
SPEAKER_00Um, yeah, in in some ways, that's how democracy is supposed to work. That it's this system of mutual accountability that um when one party screws up, the other party uh gets in. But when you have this entrenched tribalism, that becomes much more difficult.
SPEAKER_01Yeah, I mean, I think I have some other questions here that I think we've kind of answered, and I'm kind of surprised by the answer. Like I wanted to know, you know, how is Wikipedia supposed to preserve freedom of participation while maintaining discerning standards and resisting manipulation, propaganda, et cetera? And I guess the answer is largely these AI editing or either these programs, no, not AI, that are that are that are forever scanning Whip Wikipedia looking for red flags.
SPEAKER_00Yeah, this this vigilance is part of it, but uh a bigger part of it, I think, and the real reason that that Wikipedia um worked is that it created a culture. Uh it inherited part of that culture. Um, but the culture was um one of the reasons that I wanted to write this book is that I've worked as a as a foreign correspondent um in the United States. I'd seen the changes politically close up, and I love America, and that didn't leave a very pleasant taste in my mouth. And I wanted to write about an aspect of America which I had very strong admiration for. And Wikipedia is a very American project. Um, it's a constitutional experiment in radical freedom, um, not always applied in quite the way that it should be. Uh, it's very difficult to imagine uh Wikipedia coming out of another culture. And in fact, Jimmy Wales said to me that he had a German Wikipedian say, you know, Wikipedia could never have started in Germany because if you tried to start it there, um, people would just tell you why it wouldn't work. So it's that that culture of participation that um Culture of kind of democratic belief that anyone is welcome and anyone can make a contribution is much, much more important um than those little lines of script that take it out when someone writes on wheels next to something it shouldn't be on.
SPEAKER_01I love that guy. He's my new hero. I that's that's what's so painful about being American is that there are some basic fundamental ideals, like pluralism, great idea, everybody from everywhere. Let's absorb all these cultures and and sort of, you know, let let the good parts of each of these cultures rise to the top, you know, philosophically or whatever. And but like it just never manages to live up to its ideal. So like I like this idea that Wikipedia is an example of something that largely is doing it, but how does it survive the technological and political transformations that are reshaping the production of knowledge in general now? Like, or or does it? I hope it does.
SPEAKER_00Like what does it I mean this is this is kind of um some people have taken the the book to be uh a little bit more of an elegy than it was intended as, you know. Um I'm I'm not saying that Wikipedia is dead or dying, uh, only that it's under threat. Um if anything, it's a little bit like one of those uh living eulogies where you have a party sitting in your in a coffin and all all your friends tell you about how wonderful you are. Um so there is an institutional and and an existential threat to Wikipedia from AI, but there's an existential threat from AI to almost every everywhere else that that people try and create meaning in this way. Um, a friend of mine uh emailed Terence Tao, the mathematician, to ask him about this. And he got an email reply which said, you know, unfortunately, I'm completely occupied by dealing with the crisis in my field. Um there's a crisis in a lot of fields right now who are asking themselves, what does it mean to create something? What does it mean to create knowledge when a computer can do it all for us? So I think Wikipedia will pull through. It it's just having to ask itself the same questions that a lot of other people and places are at the moment.
SPEAKER_01Well, and that scares me because I find that largely I feel like 80% of people are sort of mentally lazy. So many people are just like on board with AI. I just I I think a fundamental question is whether societal society at large will continue to even value a human-authored, publicly accountable resource in an environment that's increasingly dominated by automated information systems. Like I I I believe the intellectual part of the most more thoughtful part of the culture are, but like, I mean, you know, I mean, look at popular music, right? A bunch of auto-tune crap. And I I think anybody like that's an officiato of music is usually offended by it. But here, 80% of people like it. Or look at, you know, the biggest box office movies that gross the most money, and you're like, ugh, what a you know, what a stinker. So, like, how how how is it gonna make it? I mean, or tell me, are the robots gonna win?
SPEAKER_00I don't think so. Um, I I think that they might win some battles, but won't win the war. Uh, in a way, Wikipedia is too important, um, and even too important to the robots um to just keel over or be taken for granted. All those models still have to be trained on something, and they can't just be trained on things that already exist because otherwise they'll only know how to interpret the world as it was, you know, before the uh the great uh robots uh kind of wave hit. Um they they still need people to do jobs that they can't do. Even writing a book like this, you know, I was able to go and interview all of these people for hundreds of hours. Uh I don't know that you could do that with an AI and not even really simulate that process with an AI, or at least I hope so for the time being. Um and that's why I think that Wikipedia is going to stick around. It's gonna fulfill a function, um, first of all, for people who don't want their knowledge in a homogenized form. You know, they still want to fall down that that wiki hole and have an immersive and pleasurable experience, which is not just a robot telling them that they want to hear.
SPEAKER_01Um, but you and I do, but I just don't know that everybody does. I just see how many people are just willing to throw up some AI slop, like, look, I animated this old picture of my grandma and my grandpa, or look, you know, here's a here's a you know, a bear with a flamethrower. I mean, people are just the novelty of it is so irresistible.
SPEAKER_00We've always had novelty, Jonathan. We we've always had cheese. On this scale, it's so easy though. We've we've always had people who, you know, the moment that that uh TV hit, uh, you lost a percentage of the population um who sunk into the into the couch and never really got back up. Um But instead, I think that slop will hit, slop will hit hard. Everything that that um can become slop will become slop. But things that are not slop, you know, live music, um, a bit more organically made art, maybe with some proof of work to it, I think there's going to be a return to that. Um, because not everyone wants to be knee deep in slop all the time.
SPEAKER_01God, I hope you're right. I really do. I mean, each each each week and month, it just seems to get I mean, I in the publishing industry, I mean, they're the you know, the the bean counters at the top, the general management, I mean, they they just would prefer to just cut the artist out. Like this is easily more producible for us. We just publish AI books, lower the standard culturally for people. And I mean, we've seen it happen in Hollywood, and that's not automated. I mean, we've just seen it under where every studio is making 75 films a year to now there's like you know, 40 films a year that even have theatrical distribution or something, you know.
SPEAKER_00It's like that's that a big change is gonna hit. Um, at the kind of low end of the market, what might have once been mass market paperbacks or musac um i is gonna become slob. So uh you might be starting to hear about what are called micro dramas. Um that's uh that's where you watch like a you know a one-minute long soap opera on your phone, which is all AI. The biggest company that makes um micro dramas last year made two million hours of content. Um, but they're already starting to find that there's a bit of a drop-off in this because there's just too much and it's too similar. And eventually people do tire of it. So we just got to hold on until then.
SPEAKER_01Yeah, that's actually a very hopeful take on it. You basically just said that the people that really are loving the slop have always loved the slop. You brought up the mass market paperbacking, but I mean just the lowest common denominator. So as long as it's not infiltrating the more discerning cultural uh then then maybe we're safe. But let's talk about uh the threat that authoritarianism presents to something like Wikipedia.
SPEAKER_00Yeah, I mean, uh if you've got an inherently open project and an inherently decentralized one, um that is a threat to to authority and and a threat to centralized authority and to some governments, you know. There are lots of countries in the world where Wikipedia is banned, or where the government, like in Russia, um, put it under severe legal restrictions and then promote alternatives um which have a version of events more like what what they would like. There have been some attempts um by people um in Congress and and elsewhere to do things like strip Wikipedia of its not-for-profit status, um, which would cause it real headaches. Um but so far it's been a little more resilient than I was expecting. That might change, um, but it still has this oddly like beneath the radar quality um that it it it's seen as as a little bit too confusing almost for governments quite to know what to do with it. So I'm hoping it can stay that way. So nobody's gonna just unplug it. Yeah, I mean it can't be unplugged. Like one of the hedges against um this kind of repression is that it's a genuinely international project. You know, its servers are in multiple different locations in the world. Its editors can edit it from everywhere, um, or most places where it's not blocked anyway. And uh yeah, it's almost like if it's if it's switched off in one place, then people can still edit about that country and in that language, um, and then you just have to find it if you've got a VPN or some other workaround.
SPEAKER_01So if AI can generate millions of plausible articles almost instantly, what are the forms of human judgment, accountability, or collective effort that will remain essential to producing trustworthy knowledge? What what what are we going to what specific values are going to help us overcome the tide of slop and misinformation? One of That's a question for a philosopher. I don't know. Sorry. It's a little open. Yeah, okay.
SPEAKER_00No, no, no. I mean that's that's that's exactly the the questions where the the rubber meets the road. Um, you know, long before Wikipedia uh people started to imagine things like, you know, there's uh Borges, the Argentinian writer, um, wrote short stories about what would happen if you had a map that was the same size as the territory that it describes. What would happen if you had a library um which had every possible book in existence, but there was no way to to search for them. So when when you have you know the idea of an AI written encyclopedia, which might be, you know, trillions of words long, uh that's not about writing anymore in a way. It's about reading. Um it's about, okay, well, how can you how can you possibly ever check anything? And we in some ways have to do it the old-fashioned way, you know, we we have to go back through these chains of evidence that we have um uh with, you know, even hard copies or archives or still even poor old us, uh the shoe leather reporter uh turning up to check something in person. We might get a return to the days where you had to drive 20 minutes each way to the library, Jonathan. So I hope you're uh I hope you're kinda ready for the journey.
SPEAKER_01I keep hearing about the backlash, but I've yet to see it. I like, I mean, I remember, you know, we had people on two years ago on the podcast saying, oh, this the AI thing is it's the bubble's about to burst and stuff. And I keep, I'm hopeful, but like it just hasn't happened yet. I just I I I another philosophical one. I'm sorry to keep doing this to you because a lot of these answers sort of overlap, but I'm just trying to see what emerges from all of it. Is so knowledge versus information. We have more information available to us right now than any other time in history by you know, tenfold, a hundredfold. Yet public trust in fact seems to be increasingly fragile for obvious reasons. What does Wikipedia reveal about the difference between accumulating information and creating knowledge? I'll repeat that last part because that's the hardest question. No, no, no. What does Wikipedia reveal about the difference between accumulating information and creating knowledge?
SPEAKER_00I mean, Jonathan, these are great questions. You're are you are you kind of in this existential? I got them from AI.
SPEAKER_01Joking.
SPEAKER_00Look on that will be something, but but it feels uh a little bit like you're are you are you looking for some hope here? Are you feeling this like a lot of us?
SPEAKER_01I am I'm looking yes, I'm definitely I am. I mean, this is the this is why we call it a fresh face in hell. I'm always looking for some hope. You know what I mean? These are these are pretty hopeless times in a lot of ways. I mean, like just financially for people. I got teenagers now, and and listen, I know it was always difficult being a teenager. It was never good, but man, I'm telling you, it's nothing for my generation like it was for them. I mean, there are similarities, but theirs is exponentially more complex and and I think more daunting. So yeah, I am looking for hope. And I I want to know, I feel like Wikipedia's a uh kind of a beacon of hope in a way. Like if the if the Wikipedia experiment can survive the onslaught of authoritarianism, of artificial intelligence and large language models, of the general dumbing down of the global population, well, then it seems like a beacon.
SPEAKER_00I think it seems like a beacon as well. That that's a huge part of why I write this book. You're you're absolutely right that it sucks as a time to be a teenager. And the reason that it sucks is because of this. That's that's uh, you know, it's because of smartphones. Um, it is because we have this intrusion, uh, this surveillance that has saturated every part of our life, destroying our ability to concentrate. There's um some people came out a couple of weeks ago and found that just getting rid of your smartphone, if you're depressed, can for some people be as effective as taking antidepressants. I mean, that is a phenomenal, phenomenal thing to be true.
SPEAKER_01Um, it makes sense though, because you're saying, you know, the blue screen and all that, that dopamine hits. That's I mean, your body basically just produces dopamine. Everything else is like a trigger for it, right? So, like, take that away, and all of a sudden, maybe your body sort of you know regroups and starts releasing the dopamine when it naturally should in a natural environment.
SPEAKER_00That's right. And and you're not just, you know, you can see something again without imagining it as an Instagram photo. Uh, you can feel a human feeling, which isn't just waiting for a for a like or a thumbs up or a click on your content. Um there's this moment where Wikipedia, the Wikimedia Foundation um moved from Florida, St. Petersburg, Florida, to San Francisco. And when they went there, this is in the early 2000s, they really thought, and they and they weren't being arrogant in this, that they were going to be kind of the vanguard of a civic internet, you know, that there was this, we'd had the dot-com era and we were gonna have the dot org era. There would all be all these not-for-profits um doing things like, you know, enriching and supporting our political processes and making them more democratic and um, you know, helping people with free access to knowledge. And um, you know, they were even looked at stuff like uh couch surfing um or or file sharing as ways that we might have this, you know, this digital commons that would be for everybody's benefit. And very quickly they realized this is it. You know, that there's nobody else coming. We're we're the last of the dinosaurs and we're only six years old. So part of this book is about trying to think about that that promise. And what I would say if you're looking to hope for hope, as as I am as well, is that that was deliberate. You know, they they created this culture by very explicitly saying we want to do something democratic. We want computers to empower ordinary people. You know, they were really talking about it like in empowering nurses and teachers and and parents and uh and children to learn and to do things themselves. So they they put together not a manifesto, but certainly a set of ideas and principles that they could then enact, and I think to a degree successfully. Um but what they did find is that and what we've found through a kind of massive natural experiment, you know, there's there's this joke like a smartphone is also a device that you have in your pocket that has access to all of the knowledge of humanity, and we use it to look at pictures of cats. That that's that's the hardest thing to explain to someone 50 years ago, you know, I I can read any book that I want. Um, but instead what I do is uh what was your example before? Uh uh look at a cartoon of a bear with a flamethrower.
SPEAKER_01No, it's not even a cartoon anymore. It's like a real grizzly bear with a flamethrower. Like, they treat that to do that. My son's like, Dad, that's AI. You know what I mean? I'm I'm just like everything you just described, what what what is consistent to me about me ideologically is that is a bottom-up solution. I just think that there's no way we talk about politics and we talk about um voting and we talk about elected officials and and and it matters, don't get me wrong, I'm not saying it doesn't, but like actual change, I feel like is going to have to be bottom-up, not top down. You're not gonna fix broken systems. They're already, you know, they've they're they've worn a nice solid groove with some grease, you know, the wheels are grease to work a certain way. So when you're talking about these democratic processes, it's kind of like the internet version of the farmer's market or craft brewing or the more localized, it it makes this thing that's usually a top-down thing fed to us as a bottom-up solution. So, I mean, I guess, I guess that to me is probably the most hopeful idea. If we can find the kindred spirits, if we can find the people who agree with us and we can move the needle that way by starting our movements in the underground, which is how all great counterculture always started, but maybe it's time to start looking at some of this stuff as um counterculture rather than mainstream at this point.
SPEAKER_00Definitely. Um, I mean, the the Washington state government didn't get together and say, Oh, we're gonna promote a musical movement um from Seattle called Grunge, where everyone wears a flannelette shirt uh and plays a particular kind of music. It just arose organically from people colliding with each other and a and a set of influences which are very hard to control. Um it just comes when the time is right. And it's harder to break through that now because you know you've got these algorithms kind of pressing on you in a way, like in some ways, what's popular is is determined by um sets of code which are written elsewhere. But I I don't think that it's gonna be like that forever. These are these are dead ends as well, and and things do change, and and I'm hopeful because of that.
SPEAKER_01Well, and that is why I like the cut of your jib, Richard. I appreciate that. I appreciate a little infusing a little hope into what could be a depressing dialogue. Um, we're running out of time, but in closing, I just have to call out if there's anybody out there that does know how to edit Wikipedia, could they please replace that photo of me that looks like I'm swallowing the oyster? Come on, hit me up. I'll you know, I'll buy you, I'll owe you beers for life. I I just can't I can't handle it anymore. I get a Google alert every morning. I click on that thing, and there I am, just looking like I'm it's just I can't. So one of you smart people out there, get rid of that photo for me, will you? I'm sorry, I'm vain, but I need help here. Um, he is Richard Cook. Uh, he is much smarter than me. Check out his new book, The Last Best Place on the Internet. Um, Richard, again, thanks for your your hope. Thanks for bearing with my sometimes long-winded labyrinthine repetitive questions. And um, most of all, thanks for being our Fresh Face in Hell this week.
SPEAKER_00Uh, it was really my pleasure, and uh and there are great questions. Uh I don't think they're a labyrinth line. Um, they are getting to the heart of something which is uh affecting us right now. So thank you very much for having me.
SPEAKER_01A Fresh Face in Hell with Jonathan Evan is brought to you by Fresh FacePods and produced by Jason Botkin and yours truly. Check us out on Instagram, YouTube, or at our website at FreshFacePods.tv. Don't forget to subscribe to our newsletter for updates and special content. And if you love the podcast, by all means run up and buy all of Jonathan Evison's award-winning novels wherever books are sold.
