Jimmy Wales on Wikipedia

Good evening. I'm Stuart Brand from the Long Now Foundation, how's our sound? That okay? How many people here use Wikipedia routinely? I love that breath of air that comes up when you wave around like that. I'm curious how many here have not been to one of these seminars about long-term thinking before, have not been? Yeah, the ooh from the the audience, that's a measure of the viral nature of the internet, because the only way word gets out on these things is by email. So you either got email from me or somebody forwarded it to you. That's pretty interesting. Wikipedia is some kind of revolution in progress, and it's already yanked the rug out from a lot of theories of human behavior. And you know that people only do things in their very own personal interest. And some kind of phenomenon is going on, some kind of thing that people apparently want to do with knowledge that we didn't find out until it was set in motion. So please welcome the man who set it in motion, Jimmy Wales. Thank you. Thank you thank you. So just a few months ago, I ran into Stuart and he asked me to come and give a talk about vision, and I thought to myself, well that's getting to be typical these days. Even though I remember three years ago I was just this guy in my pajamas at home, typing on the internet. And so now I'm asked to all kinds of big picture visionary questions, but the truth is I'm really more of a carpenter than an architect. I do my practical work and it seems to work okay. But I do have some thoughts about vision and the role that vision plays in Wikipedia. In 1962, Charles van Doren, who was later a senior editor at Britannica said, "The ideal encyclopedia should be radical. It should stop being safe." But if you know the history of Britannica since 1962, it's been anything but radical. It's still a very safe and stodgy old institution. Wikipedia, on the other hand, begins with a very radical idea. And that's for all of us to imagine a world in which every single person is given free access to the sum of all human knowledge. And that's what we're doing. Alieu Mamadu Yalo knows something. Kiera Wales, my daughter, wants to learn. Alieu, though, lives in Mali. And Kiera lives in Florida. So we have a problem. But Guaca, who's a Wikipedia volunteer, is here to help. So the point of this story here, what Guaca does, he goes in Mali out to local people and he helps them write articles in Wikipedia. He speaks French with them, they speak French and Bambara, he types in to the Bambara Wikipedia, he translates into French, and people translate the articles from French into English, and thereby we're able to learn things about local politics and customs in that area. The point of this is that when we talk about our work in the developing world, we don't think of what we're doing as building an encyclopedia and, you know, here's something fancy that some nice rich people made for you. Instead, we think about empowering everyone on the planet to get involved. Everybody to join that global conversation. So what is Wikipedia? Wikipedia is a freely licensed encyclopedia. It's written by thousands of volunteers in many languages. So the thing that's really, to me, for an audience like this, everybody I'm sure has used Wikipedia, is that correct? Yes? Yes? When you've just seen Wikipedia and you just see what it's like and you just use it as a reader, you may not understand some of the philosophy behind it, and the freely licensed bit is one of the most important things. So what I mean by free. I mean free as in speech, not free as in beer. This is an all-sanctioned free software movement. And we basically what it means is we give people four freedoms. You get the freedom to copy our work. You can modify it. You can redistribute. And you can redistribute modified versions. And you can do all of these things commercially or non-commercial. So when we talk about Wikipedia being a free encyclopedia, what we're really talking about is not the price that it takes to access it, but rather the freedom that you have to take it and adapt it and use it however you like. And that's really really core to our mission, and it's really core to the vision of Wikipedia that inspires people to work so hard on it. So how big is Wikipedia? It's now extremely big. We have, the English Wikipedia's the largest, it has nearly one billion words. The-- this is larger than Britannica and Encarta combined. It's actually, that's an out-of-date comparison, because it's several times larger than Britannica and Encarta combined. It's really quite a large body of work. And the German Wikipedia, which is the second largest, is equal in size to Brockhaus. So how big is Wikipedia globally? We've got over a million articles in English, but one of the things that's interesting is that English is less than one third of our total work. We're truly a global project, in many many many languages. So we have 380,000 articles in German, over 200,000 in French, Polish, Japanese. There's five more languages that have at least a hundred thousand articles. One of the more interesting ones is Chinese. Chinese, we have 62,500 articles. Up until last October, we were completely accessible in mainland China, and we've been blocked since that time. I was just in Taipei, Taiwan last week, and I met with a bunch of Wikipedians there, and it's really exciting to see them working on it. Of course, Taiwan is not blocked, so they're able to keep working. But one of the cool things that's going on is that the Taiwanese Wikipedians communicate with the mainlanders by Skype and they-- yeah, by Skype. So they're all on Skype together, and they're basically, they're communicating information to the mainland Wikipedians about how to get around the blocks. And so about half-- we've lost about half of our mainland editors, but about half are still coming in through various mysterious means, which is really cool to see this community coming together to do that. This has been a lot in the news lately, with Google's new thing, and this that and the other. One of the things that we say is that Wikipedia will never compromise with censors. Period. So. What this means is, we're trying to get unblocked in China, but we have this very firm position that our work is already neutral. That's one of the core values of Wikipedia. 99.99% of Wikipedia is of no interest to the Chinese government. No article about the moon or tigers or whatever. What they used to do is just filter certain pages. We wouldn't approve of that, but if they went back to doing that, that would be a lot better than what they're doing now. So we're trying to get unblocked, but we won't compromise, so we'll just see what happens. We have 3 ½ million articles across 200 languages. So as I said, English is only about a little less than 1/3 of the total work. But saying 200 languages isn't really all that accurate, I don't like to really brag and say we have 200 languages, what we do have, we have 38 language versions of Wikipedia that have at least 10,000 articles. And we have actually 85 now that have at least 1,000 articles. So a thousand articles is not really an encyclopedia, that's just a beginning. But I like to monitor that number because that's, once we have a thousand articles, I know there's a community there. There's five or six people, they're getting started, they're starting to build, there's regulars there, and that's when it really starts to move. So how popular is Wikipedia? Wikipedia has become a real internet phenomenon in the last couple of years in particular. We're now, according to Alexa, we're a top 20 website. And we now have a broader reach, for example, than the New York Times. So that by reach, I mean the number of unique IP numbers that we see in a given day. We see more people, or more people see us, I should say, than the New York Times, we see more people than the LA Times, more than the Wall Street Journal, more than MSNBC.com, but the really cool thing is, we see-- oh, and the Chicago Tribune. The really cool thing is, we see more unique visitors in a day than all of these sites combined. So that's pretty good. So these days when people from the mainstream media call me up, I'm now one of the people on the short list, if you're a journalist, these are the people you call when you want to know what the internet thinks. So they call me up, and they say what does the internet think, and what does the internet think of mainstream media. And I said, You mean Wikipedia? Because...who is the mainstream media anymore. We've got around 5 billion page views monthly. I say about that many. One of the interesting things about the way we're organized as an all-volunteer organization. We have 3 employees, we have a lead software developer, I have an assistant in the office, and we have, my assistant has an assistant, but we have... there's a lot of noise around me, believe me. We have all volunteer developers and so forth, and the kinds of metrics that most people in a dot com, a top 20 website, would know, like how many pages we have, we don't really need to know that. We don't have advertising or whatever. So we're really bad about getting these numbers. I'm the only one who actually cares, and I only care because I'm coming out and giving speeches. So from time to time I beg them to please tell me how many page views, and somebody grudgingly goes and calculates it, so. It's about 5 billion but I don't actually know. This little picture just shows the growth of Wikipedia. My favorite part here is in early December, there was a bit of a scandal you may have heard about. There was apparently a mistake in Wikipedia, believe it or not, and-- who knew, and the media went crazy and they called me on CNN to yell at me, and so forth, and the net result as you can see is that our traffic almost triples. So. You know, what are you going to do. So what is the amazing technology behind Wikipedia? And this is what I want to lead into, is one of the main themes I really like to hit upon. The technology required for Wikipedia is essentially, you need a database, you need a web server, you need a web browser, and you need the Wiki editing concept. Now the Wiki concept was invented in 1995 by Ward Cunningham. Wikipedia didn't start until 2001. So all of the technology, including the idea of a Wiki, which is a website that anyone can edit, has existed since 1995. Why, if it's a technological innovation, then why not have it earlier. And the answer is, it isn't a technological innovation at all, it's a social innovation. Wikipedia is a social innovation in the sense that what we figured out between 95 and 2001 was no new technology. We had the web and all that stuff. basically the basic idea of how to organize a community. What are the social rules that you need. What are the social norms, values, practices, that you need within a community. And a lot of that is the design of Wikipedia is the design of the community. Some of the core features that really make Wikipedia work are the free licensing... so everything is under free license. This is really empowering to all of the people working on the site. Most websites, when you visit the website, you'll see that, if you read the terms and conditions, they're really abusive. They basically say, anything you enter on the website belongs to us. Sorry. And people have put up with that for a long time, but it does discourage people from feeling really empowered to take control of the site and really care for it, whereas under the free licensing they realize that, hey if the organization running this, if the company running this, does a botched job, we can all leave. We can take our content and go. It really does belong to us. And also, you know, if you're going to spend hours and hours and hours contributing knowledge to the world, I think it's really important to have that feeling that it will always be available. Lots of things come and go in the world, but as long as you put it out there under free license, that you've collaborated with other people, you know it will always be there as a base for someone to move forward on. So that's really important. There have traditionally been two views of how Wikipedia works. One view is that Wikipedia is an emergent phenomenon, pseudo-Darwinian, so you'll hear a lot of phrases like "the wisdom of crowds," "swarm intelligence," those kinds of phrases are very common. "Hive mind." But the other view is that and this is my view is that there's a community. A community of thoughtful users. And that's what really makes Wikipedia work. A former Britannica editor had a really nice way of putting the emergent phenomenon view. He was writing a very negative and critical article about Wikipedia and he said, sarcastically, "By some unspecified Darwinian process will assure that those writings and editings by contributors of greatest expertise will survive. Articles will eventually reach a steady state that corresponds to the highest degree of accuracy. Does someone actually believe this? Evidently so." Well, when I first read this, I thought, that's really interesting, because within our community you will only rarely or occasionally hear people talk this way. Normally within our community, we're actually talking about very old-fashioned types of things, like references. You know, good writing. Neutrality. How to re-write things. We're talking about people's behavior in the community. All those kinds of things. We're not talking about some kind of magic process. The emergent model view would have us think that Wikipedia's thousands and thousands and thousands of individual users, they don't know each other, they each contributed a little bit, and somehow out of this emerges a coherent body of work. It's like we're all ants, right? The alternative view, the view of a community, is that we're a dedicated group of a few hundred volunteers, we know each other and we all work together to guarantee the quality and integrity of the content. You can see, obviously, what my view is, since I put pictures up of a bunch of my friends on the slides. But when I first thought about this and heard this, I thought, you know this is really interesting because there are implications for these two views, implications on how do you manage a project like this. How do you get something like this to work? If you really believe in the hive mind type of view, well individual users aren't that important. They're like ants in a colony. If you step on a few of them it really doesn't matter. And the other thing is that you should have some sort of metrics. Programmers always love to have lots and lots of metrics, like the ebay rating system. The ebay rating system works really really well because most of the interactions on ebay are not necessarily community interactions. When I go to buy something on ebay, why do I care about that number? I care about the number because I don't know that person and I don't know anybody who knows them, right? And so I need some kind of a metric. And that works very very well for a site like ebay. But in our case, we don't have ratings of users, and the reason we don't have ratings of users is that the people who are working together and editing articles, they come to know each other, and they usually become friends. Even if they disagree very strongly, they do become friends or at least get some grudging respect for each other. They duke it out for several months and they come to some compromise. The kinds of judgments that we use to make the work be good have nothing to do with rating metrics and numbers. I mean, you can imagine, suppose you went to work at a company, and every day you have sort of a badge and it has a number on it that tells how many people like you or not. That's not really where you want to work, I think, right. Instead, what do you want. You want people to judge you on a wide variety of characteristics and that's what we do with Wikipedia. We have people who we say, you know this guy does fantastic work in biology, but whatever you do, don't let him get near Israel-Palestine, he goes bonkers. So how do you capture that in a number, you really can't capture that in a number. It's real human judgments about real human people. This person has, you know, they do good work but they're difficult personality, all those kinds of things. So what is the vision in software? How do we look at the software and how do we think about making the software so that people can do this kind of good work. How do we assure quality. So how does that work. Well, there's a lot of details to it. A lot of details to how the software works. And I'll just go into a couple of those. But before I do that, I want to talk about my overall philosophy of the design of social software. So imagine that you're going to design a restaurant. Okay? And so just think about the problem of design for a restaurant. You want to design a restaurant, and you think to yourself, well in this restaurant we're going to be serving steak. Okay? And since we're going to be serving steak, we're going to have steak knives, and since we're going to have steak knives, people might stab each other. So how do we solve this problem? So what we're going to have to do is we're going to have to build cages and keep everybody in a cage to make sure no one stabs each other. All right. Well this makes for a bad society. We reject this kind of thinking in restaurant design, and yet this is the predominant paradigm for social software design. What we do is we sit down and think, we're going to design a website, we think of all of the bad things people might do, and we make sure that we have controls and permission, everything to prevent people from doing the bad things, right? And this does, this has two effects. One effect, the first effect is, when you prevent people from doing bad things, there's often very obvious and direct side effects that prevent them from doing good things. If I look at a web page and I see a little spelling error, but I don't have permission to edit that page, I can't fix it. That's the first order of fact, that by having complex permission models, you make it very hard for people to spontaneously do good. But the second effect has to do with how human interactions are organized. How do people feel about each other. How do people feel about society. Many many people report when they've been involved in some online mailing list or things like this is that gee, it's so hostile. There's so many hostile communities on the internet. One of the reasons is, is this kind of philosophy of trying to make sure that no one can hurt each other is actually eliminates all the opportunities for trust. We had a situation in Wikipedia where somebody proposed on the mailing list, they said well, everybody has a user page in Wikipedia, and most people just use their user page to put a picture of themselves and some information about who they are, and it really isn't for other people to edit, so why don't we just lock all the user pages. Right? Because they are subject to vandalism and things like this. Well, a lot of us, the old-time Wikipedians, we went around and we put little statements on the user page, and mine is still there as far as I know, I haven't looked lately, and it says "I trust you. You can edit this page. But please, if you do, make it better. Do something nice." In fact, there was a picture of me that I hated, it's been on there for months, and last night I went to dinner with some Wikipedians and they took a picture, a very nice picture, and now that's on my user page. And yes, he's sitting right out here. So that's the kind of thing that just happens spontaneously, and it couldn't happen if we didn't have that culture of trust to say, Yeah knock yourself out, that's fine, and if anybody does anything bad, we have the real time peer review. So this is into the nitty gritty of how the software actually works. All the good intentions in the world, saying "Oh we love everybody, and Wiki love, doesn't get you very far if you don't really have the software tools to make it really work. So how does it actually work. Well, some of the core components are that every edit on the website goes under the recent changes page. And this is watched by hundreds of people daily. So if anybody does anything bad on the site, it gets reverted very quickly. It gets reverted because we store every single version of every article that's ever existed. So if an article is deteriorated in any way and starts to go downhill, anybody can revert back to the good previous versions, and they do that by keeping an eye on the recent changes. Nowadays, in English Wikipedia in particular but also in German and French and Japanese, the recent changes move so fast that it's really not practical to monitor directly, so within the community, people have built these pool software tools to actually monitor recent changes for various kinds of problematic entries and things like that. And then additionally people can set up their own personal watch list. So when you're editing a Wikipedia article, most of the power users have it set up so that anything they touch automatically goes on their watch list. So you can fix something in Wikipedia and then you can keep an eye on it for several days. You can come back every day and you can see if anything has changed. And you can actually see how that works. And then another element of it is the real time communication. So we use IRC, which is Internet Relay Chat, this is very old school chat rooms, predates Instant Messenger. It looks like this. It's sort of ugly. But there's tons and tons of people. We have, I'm not even sure how many IRC channels we have, we have them in many many languages, in fact the reason that the Chinese use Skype instead of IRC is that IRC is blocked for Mandarin Chinese. So the entire Chinese-- I always wondered why I could never find any Chinese people in IRC, and the reason is, they're all on Skype instead. So this was just a screen shot from the other day, and as you can see, just on a random morning, there were 238 people sitting in the english Wikipedia channel. So at any moment in time when you're on the website and you do something, if you start misbehaving yourself, 238 people are going to hear about it pretty quickly. And they're going to come in and they're going to start watching you. Normally we have policies of, we try not to bite people, so first we leave you a nice message saying, you know, thank you for your "fuck you" comment, but we're trying to write an encyclopedia. After a little bit, though, you can get blocked. The admins can block IP numbers for misbehavior, and that does happen. Then when you look at the page history, our software has a really nice-- compared to a lot of Wikis, although a lot of them are getting nicer and nicer of course, we have a really nice comparison between the old version and the new version. So if you think about, how could I monitor a long article, I would have to read the whole article every single day, that isn't what you really have to do. You just come and you look at the changes, and you see, did this change improve the article or did it make it worse. And you can see that at a glance in most cases. And so if this is one of your articles that you keep an eye on, you would come in and you would say, oh yeah this edit's good, this one I don't like so much, I actually think I can do something else, and you can just continuously keep improving it. Another key feature, and this goes back to the whole restaurant analogy. Is that we have hardcore, we have organization by the community. So one of the-- just to give an example of this, we have the articles for deletion pages. So the idea of the articles for deletion page is that occasionally articles, new articles get created and they need to be deleted. So why would you ever delete anything, there's lots of disk space, the simplest example would be something like, somebody makes an article with the title of ASDF, ASDF, and the article is "Hi, Mom." Well, when that happens, probably that's never going to become an encyclopedia article. You know, you can think about it for a minute and say, yeah I don't think that's ever going to be an encyclopedia article, in a case like that, that would be a speedy delete. Any admin could just come in and delete it and it would just go like that. But there are other cases where it's borderline. Where you might say, I'm not sure if this is a hoax, if this is real, is this not real, and the example here was a film called Twisted Issues, an obscure underground punk film from 1988. The funny thing is, I gave a talk just two days ago at the University of Florida, and the next day somebody wrote me and said, "Do you know I played on the soundtrack for Twisted Issues." I said, wow really, go ahead and edit the article, really, so anyway, so the first person says it's supposedly an underground punk film, but it miserably fails the Google test. So what's the Google test. You look something up in Google, and if you can't find it, then it probably doesn't exist. It's-- this is not a foolproof test, but it's pretty good. Right? There are still a few things on the planet that are not in Google. But it's pretty good. And so it fails the Google test, and it doesn't have any listing, so a couple people say, "delete, delete." And then somebody says "Hey wait wait wait wait, I found something. It's in the Film Threat Video Guide to 20 Underground Films You Must See. So maybe it has some notability. Next person down says, complete it. Next person says, it's a real movie, it's in IMDB, keep keep." So at the end of a discussion like this, this would have been kept. In fact it was kept, and the article's still there. But the interesting thing about this process is that although people are voting "delete" and "keep," right, this is just a Wiki page like any other, this is just free form text, you can go and click on edit at the top of the page, you can change this, you can add a comment, whatever. There's no actual automation of the voting process, and this is one that's always popular with programmers who say, This is a lot of work, why don't we just put a little button on the page, nominate for delete, after two weeks we tally the votes and it'll automatically gone, that's not a good idea, because just voting is something that we don't do very much, we really discourage voting. This is a human dialogue, this is a conversation about whether or not this should exist or not, and in fact, in theory, this could be 28 votes to delete and 3 votes to keep. And as long as the deletes are all like, "Delete, non-notable," "Delete non-notable," with no information or argument, these three at the bottom who say, "Wait, I found it in a book. Maybe people didn't realize this. I know something about it, it's actually on IMDB," that would make a big difference. The other thing that would make a big difference here is who the voters are. We're a community where we generally know each other. We recognize names. We can see people. I can say, "Okay, look, Rick Case says keep it, I trust Rick Case, if Rick Case says it's in IMDB, it's in IMDB, I don't have to question that." Whereas some of these other people, I just say, "I don't know who they are, really. They're newcomers." And so certain people in the community do carry more weight than others. So the most important thing though about this is that all of the rules are social. The software itself does not determine the rules of Wikipedia. Almost everything is completely open-ended in terms of what the software does. There's very very little in the software that is actually rule enforcement. It's all about dialogue, it's all about conversation, it's all about humans making decisions. So that's really really important. So what comes next. What am I working on next is what I'm actually going to tell you about. And what I'm working on these days is Wikea. Wikea is a new company, and basically what we're doing at Wikea is we're bringing the community ideals, the community models, all of the things that I just talked about in relation to Wikipedia, we're bringing to a broader context. So one of the conceptual ideas is to think about that an encyclopedia is big, right? It's a really big set of books. But a library is much much bigger. So what we're trying to do is empower people to say, Let's think beyond the encyclopedia, let's think about all the different kinds of things that people might want to build or do. What made me think of this, why do we want to do this. Well, I have certain beliefs about genuine communities, and the future. This is where I want to talk about the future of free culture. One of the things of course that's going on in our society is the struggle between proprietary and free culture. So you have things like the battles going on about copyrighting the movie industry and this and that. One of the things that makes me really proud about Wikipedia is, we're starting to change the terms of the copyright debate. We're starting to change the terms and say, "Look. The copyright debate is not all about kids stealing music. When the law gets in the way of Wikipedia's work, it's about good-hearted people sharing their work, trying to do good things, and if we have these really restrictive copyrights, it gets in our way." It not only gets in our way, but it isn't necessary. The standard argument for really strong long copyright terms is, you need it in order to economically incentivize people to do good work. Well, Wikipedia proves that isn't true. People do good work because it's the good thing to do. Thank you. So my idea now, is okay we've got this first big shining example, but let's build this out into a much bigger cultural force. Let's take these ideals of Wikipedia and bring them out to lots and lots of people in lots and lots of areas far beyond just encyclopedias. I think the genuine communities, like Wikipedia, will be built on love and respect. I get a lot of talks at tech offices, and I always like to say I'm the only guy who's going to get up at a tech conference and talk about love, right? But it's really important because of all the things I've been saying, Wikipedia is not about the technology, it's about people. It's about leaving things open ended, it's about trusting people, it's about encouraging people to do good, that these communities, I believe, are going to be the norm on the internet. I think people have seen that some of the old models are really unhealthy, they really don't work. And I think that Wikipedia shows us a really powerful means to move forward to empower lots of people to do good work, cooperatively. That people, when they're empowered by the software and they're set free, both by the license but also by the terms of the site and the way that it works, they're free within communities to decide their own rules and think about things, they can do great things. And so we're really looking at, what are all the kinds of great things that people can do. So a lot of people say, "Well an encyclopedia covers everything, so what else do you really need" and so I wanted to give some of the examples of the things that we're seeing so far. This is the Lost Wiki. Do people here watch the show Lost? Lots of people? A few people? You should watch it, it's really great, but it's really really really complicated. And there's lots and lots and lots of characters, and it's really hard to figure out, and the cool thing is, the Lost Wiki, the fans of the show are coming in and they're documenting all of this complicated information, so you can go in here and you can read about this. This is far beyond what you would want an encyclopedia, for one thing. Second, it's not necessarily neutral. The people who are writing the site are fans of the show, so they don't need to be neutral. They don't need to include criticisms and controversies about whatever. They may choose to, but they don't have to. It can be a very favorable review because it's written by fans who really love it. These are the kinds of things that people can do and build. In the old days, if you think back on the old Geocities days, and lots of people were building fan pages and things like this, but what would happen. People will build a fan page and after a while, they get bored with it. It's really not that fun to do all by yourself. The idea here is, people whatever passion or hobby or interest they have, and they can build something, they're going to give it away under free license, and they're going to come together with love and respect and build all kinds of cool things. On any kind of topic. So we have some of the most successful things we have so far are of course, geek culture. So things like Star Trek, Star Wars, this is a Doom Wiki, you can see this little picture here on the search box, the guy looks a little angry. I don't know if they really know about the Wiki Love idea here. But they seem like nice people. And then so there's one that does cover Wiki Love, this is the personals. The personals at wikea.com. One of the cool things about my job is that if you found an encyclopedia, you somehow get this reputation of a guy who knows about everything, right, so tonight I'm going to be on Free FM, I'm going to be talking about personals ads, and dating on the internet. Why me, I don't know, but I'm going to talk about it and I'm going to talk about the Personals Wiki. And the idea here is, there are all kinds of personals sites, and how do they work. Well they work, you think about the software principles I talked about earlier. They think about all of the bad things people might do, and then the software doesn't let them do that. Whereas the idea here is, let's be free flowing, let's let this community think about it and negotiate and figure out, how would you properly do it as a community. What are all the kinds of things you really want to do. And the software's really open-ended, so people can form whatever they like. So I think that really has great potential, anyway we'll see what happens. So this then brings me to my final point, which is my basic philosophy is for us all to make Wiki Love, not war. Thank you.