With the announcement of “Wikiasari” or Search Wikia, Jimmy Wales is in the news again. The Wikipedia founder wants this to be the “project to create the search engine that changes everything.” OK, we would get nowhere without ambition. But this gives me the opportunity to publish the full transcript (only some minor technical edits) of an interview I did with Mr. Wales in Bergen last May. I have used excerpts before in stories about Wikipedia for Dagbladet.no and Axess.
In the interview, Wales talks about business models for the different free culture/open access movements, the need for reform of intellectual property legislation, the background to the “neutral point of view” regime in Wikipedia, how to build and sustain a positive community of active users, how to achieve a good debate and handle vandals in online forums – among other issues. Read the full text in the extended entry.
Undercurrent: Wikipedia has a lot in common with other initiatives that may amount to a kind of movement for free access to science and culture in the broad sense, and to make it possible and legal for people to produce and distribute knowledge. Creative Commons, open access in scientific publishing, all kinds of user-generated content, citizen media, maybe also Google and others’ scanning of books. It’s easy to sympathize with these projects – who can be against making knowledge available to more people – all people. But all this is very new, and often challenges old business models and regulatory regimes. How can a sustainable model emerge out of this free culture movement?
Jimmy Wales: “It’s a very big question. As you just mentioned there are many different kinds of organizations that are in some sense loosely part of the same movement. For some parts of, like Wikipedia for example, we have a business model, so to speak. We’re a non-profit organization, but I think you would agree that even a non-profit organization needs some kind of a business model, in the sense of somehow you need to get some money in to keep the servers running and do things. Which is based on donations from the general public and having a very low cost production method using volunteers, and that seems sustainable for Wikipedia. That model doesn’t work for everything. It works very well for an encyclopedia that’s neutral and therefore can get very broad support. People feel comfortable supporting Wikipedia because you know it isn’t serving any particular agenda – we try to be very uncontroversial and mainstream about most things.
Other parts of free culture would be political sites or just pop culture, fan things. In that context I think an advertising-based model works better. It seems more acceptable for people to say, let’s get some advertising in. That’s what I’m doing with Wikia, my new company.
Then we have things like the open access movement in science. I’ve been talking to a lot of people at the Public Library of Science which is science journals – PLoS Biology, PLoS Medicine – and what they are doing is a very traditional editorial process, just like any scientific journal with peer review and traditional editors, and yet they release everything they do under a Creative Commons licence. Their basic concept is to say, well, there’s actually a pretty small number of institutions who pay for that publishing process right now. Right now they pay for it through subscription, but instead they could pay for it at a different stage in the process, could be the submission of papers stage, could be part of a grant – the money is associated with a grant. The same institutions end up paying, but if they pay for it at a different stage in the process we don’t have this bottleneck of open access. The problem of it as I see it is that the PLoS business model hasn’t proven itself so far. They’re dependent very much on grant money at the moment, though they have managed to be producing some very impressive high quality journals that are getting a lot of respect. They still spend more money than they bring in, so it’s not clear how sustainable that is in the long run. And they’re thinking about different, more innovative ways to have less centralized peer review and processes, but it’s very difficult because the quality control they need is very different from the kind of quality control we need in Wikipedia. We’re very tolerant of the flow convergence to better quality. If you go into the Norwegian Wikipedia today and start a new article, maybe it won’t be very good for a while. Maybe it will only be one paragraph, hopefully it won’t be too far wrong. But a lot of articles in Wikipedia start out as short, not very helpful articles, and we’re tolerant of that. For scientific publishing it’s a very different kind of environment. You really need finished published papers that have gone through a quality process.
So I think it’s really interesting to see within this movement, where the areas are where change is going to be more difficult and change is going to be easier. I don’t think we’ve seen the end of innovation that are not necessarily technological innovations, but I would call social innovations or organizational innovations. Wikipedia for example, I always point out that all of the technology for Wikipedia existed six years prior to Wikipedia. It’s very simple technology – a database and a website. But what didn’t exist was the idea of the social organization and how you could actually get people organized in the right way. I think the same questions could be asked about scientific publishing. We know that the scientists who are involved don’t get paid now, so they could still not get paid, but organize themselves in a different way (…) But nobody has figured out how to do that yet.
There is a lot of dissatisfaction with the established system. The dissatisfaction is different in different fields. I know that in high energy physics a preprint service became very influential and important in a certain subset of the physics world, and one of the reasons was that the scientists working in that area were very frustrated with the slow pace of traditional publication. So they just needed for their own purposes a place they could put papers as they’re written, that they can refer to and learn from each other more quickly, instead of having to wait 18 months until something finally comes out in a journal, and then read it and realize you just wasted a year’s work because somebody already did it. But as a side effect, what’s interesting about that is that it seems that preprint services have become influential and the value of a traditional publication has declined. Because the scientists simply review the work in the preprint services themselves, people get reputations based on that, and it’s a completely different process that accidentally grew up. That was unique to one particular field, different fields have different requirements and different issues, but I think that where we will see the most changes if there are fields right now where the traditional process is not working very well, it’s not really meeting the needs of scientists, and that’s where the opportunities are for people to come in and say, let’s do it in a different way that will be more open and cheaper and get broader access to knowledge.”
Undercurrent: Is it possible to talk about main counterforces to this movement in general – political, commercial?
Wales: “Yeah, I think again it varies in different ways. One area would be if we talk about Creative Commons – it focuses a lot on the world of music, entertainment, video, movies… A lot of what happens there are actually side effects. For example Disney has a very strong interest in protecting financially their big money-making franchise, properties. And it’s an unfortunate side effect that when we have these copyright extensions that were designed to protect Disney’s money, that the side effect is that it affects things that Wikipedia wants to do that really have nothing to do with Disney.
So a simple example would be: Suppose that there is a economics textbook that was written in 1992. I choose that because that’s 14 years ago, which in the US was the traditional term of copyright. So an economics textbook that was written in 1992 but was only published for a year wasn’t very popular and it’s been out of print ever since. That book currently has no economic value, but it stays under copyright for now the life of the author plus 95 years. So that author’s still alive, so maybe in a 120 years the Wikipedians will be able to take that book and update it. Well, OK, that’s a long time to wait. If we had access right now to a 1992 economics textbook, that’s not so long ago and we could actually update it fairly easily. I could find ten or twenty economists to quickly go through each chapter and make some changes and feedback and get a lot of Wikipedians to help with copy editing, and we could update that book very quickly. That’s a book that has currently no economic value, and yet why was the copyright on it extended 95 years? Well, it was a side effect. Disney doesn’t really care about that, it cares about early Disney movies. Although of course there are the traditional textbook publishers who do care about that. They don’t want us to have access to part of print materials from the 60s, because if we do we’re going to use those materials to compete with their contemporary offerings. But when you think about that you begin to realize, this is no longer serving the purpose of being a necessary evil or a necessary economic incentive to produce textbooks, it’s actually standing in the way of production of knowledge. And copyright really should be about the idea of what’s encouraged more creation of knowledge by creating a limited time monopoly over certain types of work so that the authors can recoup their costs and make some money and have an incentive to do it. But as soon as we realize that oh, these really long copyright terms are really not serving that purpose, they’re just locking up our culture, for unreasonable periods of time. So those are some of the tensions of who’s on the other side, and what’s interesting about it is that I think some of the bigger players like Disney, they don’t intend to be… they have no interest in copyright laws that hinder Wikipedia, that isn’t really their field of business. They’re just concerned about Snow White falling into the public domain. They got it from the public domain, but that’s a different story!”
Undercurrent: I would like to ask some questions about Wikipedia. When I first heard about it, the idea of neutral point of view surprised me, because it sounds so outdated. The idea of a neutral point of view must share its roots with positivism in science, objectivity in journalism – ideas that have been under attack for decades. Blogs and forums with their often ferocious partisanship and individuality seem to be more in line with the Zeitgeist. How do you then explain that Wikipedia and the idea of neutrality still is so popular?
Wales: “I think it’s old fashioned in a certain sense. But there are a few things about it that are important to understand. One is that we actually use the term neutral point of view or NPOV as people always write it in the community, as a term or jargon that’s in the community. It really refers to a social model of cooperation rather than an epistemological or knowledge-based concept, although obviously it impacts in both areas. But the idea is that in order to facilitate people who may disagree being able to work together in a way that all feel is productive, we can just say the text we are working on in Wikipedia shouldn’t take a stand on controversial issues, it should just describe those issues. And it turns out to be a very powerful technique.The simple example that I give that people immediately understand is, if you imagine a very thoughtful Catholic priest, who takes a very traditional position of the Catholic church on abortion, and then you take some activist on the other side of that issue who takes a very different view, but is also a reasonable, thoughtful person, you begin to think about, oh well, yeah there’s a lot of hot air on both sides, but there are actually reasonable people who can say, “ok, well, I know you and I don’t agree, but let’s write this down in a way that people can understand what the debate is, and we can say things like, the Catholic church has argued thusly and people have responded thusly.”
Obviously it’s easy to say it in a way that sounds kind and thoughtful, and sometimes people do get into big arguments – being human beings – but it turns out to work reasonably well, that people from very different viewpoints are happy to work together. And say let’s present all the sides, and it works very well. So we’ve always thought of it as a means to empowering people to work together.
But also there is the other point, which is that if I approach an encyclopedia I really want basic, background knowledge. And then I want to go from there and then be able to form my own polemical point of view. Both kinds of writing have value, of course. And it can be very interesting and healthy to read very diverse viewpoints on something like the war in Iraq, right? But it’s also very helpful to say, OK but I also want to know what are the facts that people on all sides agree on. Just the basic background. Where is Iraq, right? Who was in charge, what is the history of Iraq? And I want that in a non-contentious way. I want to know, what are people on both sides willing to say, yes actually everyone agrees that the following events happened in the past and led us to the point where we are today. And beyond that, you also want other things.”
Undercurrent: It’s not possible to have two different articles with different viewpoints on the same subject?
Wales: “We don’t normally do that. Sometimes we will have – not exactly that, but we’ll have articles that will be about a particular perspective, but that are not advocating that perspective. One recent example that comes up would be evolution versus something like intelligent design. In our normal, scientific articles about the theory of evolution, basically we don’t include any information about intelligent design. We’ll just say at the beginning of the article “in the modern theory of evolution bla bla bla”. But then we’ll also have an article about intelligent design. And then we’ll have another article that will be about the controversy within our culture. But none of those articles should really advocate for any point of view. And that’s not the sort of thing that we normally would do.”
Undercurrent: I’ve listened to one of those talks you have given about how to make the community work in a good way. Can any of these insights be transformed into how to make debates function on newspaper websites? Some Norwegian newspapers are really good at experimenting with blogs, and they have forums where you can participate. But how do you create a good discussion? Because there are so many people who are just out to destroy it, it all risks ending up as noise. So how can they really crack this problem?
Wales: “It’s a really complicated problem. I think I know some of the answers but not all. A part of it has to do with really generating an atmosphere of genuine community, a community of respect. Which is really local, I don’t mean necessarily geographically local, but it’s local meaning it’s a small number of people… Normally what you might find on a newspaper message board is, here’s a very interesting thoughtful commentary by someone, and here’s someone on the other side who also has an interesting, thoughtful commentary, and they’re debating each other with some respect, but then the trolls come in and their just sort of inflammatory remarks. Of course even good people sometimes get into very serious argument, but that’s understandable, we’re human beings so those things happen, but what I believe is that the techniques that work best are to in various ways empower the good voices to shape and guide the discussion, and to even exclude people. And to say, look we’re all for open, free dialogue and debate, but it has to be respectful and constructive, and of course it’s a very complex matter of where do you draw the line and things like that, it’s a matter of judgment and you need thoughtful people and sometimes they’ll get it wrong. A lot of this is driven in my belief by the software tools that are at our disposal. So we’ve had software tools that allow either completely anonymous random wild posting, and it’s completely uncontrolled and evolves into random potshots, or very top-down tightly moderated and controlled, which can generate quality of discussion but it’s also very expensive for a newspaper or something like that to manage. Someone has to actually supervise it. But somewhere in the middle there should be tools, and this is how Wikipedia works, there are tools that the community can use to moderate themselves, and for consensus to emerge over time. I think that what we’ll see in the future is increasing discoveries and understandings about how to generate healthy debate environments, and I think a lot of it actually mirrors what can go on in the real world. That we can have riots in the streets or we can have closed meetings of people who all agree with each other, or sometimes you can actully have a public forum where you bring together two sides and they have a civilized discussion and an intelligent audience, and it’s the same sort of thing online. It isn’t that we have only two choices of a riot or a complete lockdown. Maybe somewhere in the middle we have to learn to accept a little bit of noise but hopefully have more signal than noise.”
Undercurrent: But it’s essential to actually have participants moderate themselves or take responsibility for the discussion?
Wales: “Yes, and also for them to actually have the tools available to moderate themselves. One of the things that we’ve seen on the internet was, I’ll call it the rise and the fall of Usenet. The rise of Usenet was fuelled by, it was really open to participation by anyone. And particularly for some parts of Usenet there was no controls at all, and no possibility of having control. Simply because the architecture, design of it, there was no one organization which controlled it, so to be able to kick a problem user out was impossible. And it killed it. I was in a small Usenet group and there was good discussions for a while, and then there was this guy who came on and he would post in all capital letters about the Armenian genocide. Daily, every day, hundreds of… It completely destroyed the forum because we weren’t discussing his issue and we didn’t care, but he was crazy or something and there was nothing we could do about it. Those of us who were having a political debate and were maybe on different sides of the question, before we were all like saying we’re trying to have a discussion here. If we had had the tools, then we wouldn’t have used the tools against each other, even if we disagreed politically. I actually think even in the world of blogs it has progressed in some areas to where, speaking very loosely, you’ll have bloggers on the right and bloggers on the left, but who have been in a dialogue for a couple of years and who have gained respect for each other. Even if they don’t agree, they may be very polemical, but they can at least say, you know, he’s wrong about everything but he’s a good writer, or something like this.”
Undercurrent: I heard you mentioned that Wikipedia turns into a meritocracy.
Wales: “We hope so”.
Undercurrent: So this echoes in a way the open source software movement?
Wales: “In large part, yeah.”
Undercurrent: It’s not like anybody’s contribution there has the same value, and it’s probably the same thing with Wikipedia?
Wales: “I think a big part of it has to do with having mechanisms whereby people can establish reputation. Because if you have established something like a good reputation, then you know to behave yourself because you don’t want to ruin your reputatation. The reputation gives you some social influence that’s valuable. That’s a really important part of how things work. If we have a situation where maybe eight out of ten people think one way, and then two people think the other way and one of them is someone you’ve never heard of who seems to be not a very good writer, but the other one’s a very respected, thoughtful Wikipedian who normally… Then you stop and say that, why is Bob disagreeing, that’s really interesting because everybody seems to be… And when it’s working well, it’s a nice (…) to say that one person can disagree and, it’s not a voting process, it’s a process of consensus. One person can disagree and actually get a lot of people to rethink something. That’s really healthy. And difficult to preserve, of course, as you have a growing community.”
Undercurrent: Which aspect of Wikipedia are you the least pleased with?
Wales: “Recently, and primarily in the English Wikipedia, the community has gotten so large that some of the processes are not working as well as they used to. One example from yesterday that was very surprising to me. An absolutely excellent contributor who writes mostly in Wikinews, but is also very involved in the global foundation issues, and he does a lot of press work and things like this, a very valuable volunteer, was nominated to be an administrator in the English Wikipedia, which the community votes on, and he didn’t pass as an administrator, which was shocking to everyone who knew him. And we went and reread it, and we saw that the people who voted for him, which was a majority, the criteria is like two thirds, and the people who voted against him didn’t know him. So we thought there’s something wrong with this process, because these people who voted against him are basically saying, oh I checked your editing history and you only have 700 edits in the English Wikipedia, you don’t seem experienced enough. Well those of us who know him said he’s very experienced, he just works in Wikinews, not Wikipedia, he’s completely sane and he’s the kind of person we want to have as an administrator. So that was where I said, oh this is interesting because this is an example of the wrong conclusion was reached simply because the community has gotten large enough that people don’t know each other as much as they used to. That’s a real challenge to think about. How do we modify some of these processes so that decisions are being made within smaller groups where people can know each other and can have social reputation and all the things that we know are healthy and work.”
Undercurrent: Since you mentioned Wikinews, what are the experiences? It seems that it sometimes is overtaken by fast Wikipedia entries on major news stories?
Wales: “Still, Wikipedia is definitely stronger on big breaking news stories than Wikinews, which is interesting. That’s just because Wikipedia has a much larger group of contributors, and things like this. Wikinews though, has done a really interesting set of things with lesser news. They do a good job on big news stories too, but it’s hard to beat Wikipedia if the story is the tsunami or something like this. It’s hard to beat the Wikipedia crowd. But Wikinews has been growing, steadily, the number of contributors grow, I’ve been watching as the number of contributors grow the stories become more mainstream. One of the funny things about Wikinews in the early days is that it tended to be very quirky. It would just follow the interests of whoever happened to write something. And now they have enough sections for things, they break things down so that the front page normally has most of the big stories of the day summarized.”
Undercurrent: In English?
Wales: “Yes, and in German too, I think. In the English I remember in the early days I went and three of the five headlines that day were about Romania. There was nothing going on in Romania but they were all written by a Romanian who was were busy that day. There were only about five or six people working on it so Romania was a very important country for a few weeks. Now it tends to follow global headlines a little more.”
