Wikipedia today, Citizendium tomorrow

Sanger now believes that the world deserves something better than his former start-up when it comes to online research. Citizendium, a new project being launched this month, abolishes posting on wikis anonymously, brings in experts to edit submissions, and enforces strict reviewing procedures.

Wikipedia has more than 400 million registered users, is offered in 250 languages and features almost 2 million articles in English alone. Sanger parted ways with Wikipedia in early 2003, but has remained committed to creating a trustworthy Wiki-based forum. With the launch of Citizendium approaching, Sanger talked with CNET News.com about how his company plans to compete with the giant Wikipedia and change the way the world seeks information online--for good.

Q: You are better known as someone who co-founded Wikipedia. Why did you choose to leave Wikipedia?
Sanger: Well it was awhile ago now, and there were two different reasons that I left. First, I was laid off. The company that founded Wikipedia and Nupedia lost the ability to pay me with the economic downturn. But then I distanced myself from the project, essentially, because the project managers were really unwilling to rein in the troublemakers and also because there really wasn't any sort of special role made for experts. And that is what I told (Wikipedia co-founder) Jimmy Wales in the beginning of 2003 or so...explaining why I was leaving the project for good.

I think we absolutely need another wiki--first of all, simply because Wikipedia lacks credibility unfortunately.

So do you just feel that you had a different viewpoint, and do you feel that you're embracing something that's more of your own with Citizendium now?
Sanger: Well, I certainly have a difference of opinion about what the high-quality, free encyclopedia should look like. Citizendium is initially aimed to be a better competitor essentially to Wikipedia, but it isn't just that. We are going to be looking to aggregate a lot of different kinds of reliable information and we're already talking to different potential partners about how to do that. So, we've got greater ambitions than simply doing one better than Wikipedia.

Wikipedia has grown to be quite big. How do you plan on competing with them? Do you really think we need another wiki?
Sanger: I think we absolutely need another wiki--first of all, simply because Wikipedia lacks credibility, unfortunately. It's a good starting place, as people say--on some subjects anyway--but it isn't really what we want out of a reliable reference resource. And frankly, I don't think that the Wikipedia community is prepared to make the changes that I think need to be made in order to transform Wikipedia into something that's really reliable.

As to the other question, how can we possibly compete? I simply think that it will take some years before we have developed on the order of several 100,000 articles and we will grow in the same way that Wikipedia itself grew. Obviously, we're not going to be much of a competitor for some time, but just give us a few years and we will be equally useful for the most widely read topics, and actually more useful, of course, simply because our information will be more credible.

On Citizendium, you have "experts" and "constables." Can you explain where the experts come from? How do we know that they're credible?
Sanger: Well, experts have approached us mostly as a result of the press that we've gotten. We've done a little bit of recruitment, but for the most part, it's people who just show up, and there was actually quite a few of them. There are also a lot of people who have gotten frustrated with Wikipedia and have left it to join us.

Listen up

Larry Sanger looking to trump Wikipedia Wikipedia co-founder Larry Sanger speaks with CNET News.com's Neha Tiwari about the upcoming launch of his new site, Citizendium.

Download mp3 (1.0MB)

About the second part of the question, well there's a lot of different views that people take about the credibility of experts per se, right? I mean, I myself (don't get) all gaga over someone simply because he or she has got a Ph.D. I think, however, that the fact that someone has credentials of various sorts--not just degrees--is an indicator that they know unusual amounts about very specific subjects. So really, the question here is why should anybody think that people that are considered by society as experts are really credible or reliable? That's a question for each of us to answer, I think, on our own. If it has to come down to my believing someone when someone writes something on a Web site, I would rather believe someone who has made it his or her life work to study something than someone who's read a single book on the subject written by that expert.

Are you aiming Citizendium at an older, more mature audience than what Wikipedia has aimed at?
Sanger: Well, in so far as I'm aiming at all, which I really can't do because it's a wiki, but...you might be surprised to know that I would like the wiki to be as open and inclusive in terms of topics as possible. And all the people who are into, you know, videogames and fan literature and all together concerns that people make fun of Wikipedia for having all these articles about--I'd welcome them with open arms. Personally I think that it is the citizens' compendium and we need a lot of information about everything. The whole question is, can we maintain large (and comprehensive) sets of articles about different topics? I'm not going to go out of my way to attract the Trekkies, for example. When I sit down and actually take the time to do recruitment, which we really haven't had the time to do much so far, I'll definitely go after academics first.

Going down to the specifics or the processes of how Citizendium works, we at CNET here, we try to submit links on Wikipedia to our own legitimate stories on relevant wiki topics, but they were blocked and immediately pulled down. At Citizendium would that kind of thing be allowed or would that not also be seen as legitimate?
Sanger: Well, I think the details of that policy have yet to be worked out. Generally, we don't like people to use general information resources to promote their own material. On the other hand, someone like CNET or maybe National Geographic or the Smithsonian or one of these sources of credible information can actually be doing the world some good by carefully placing links when they are really relevant and when they're as good as the ones that are up there already.

I think what we can end up having to do is make use of what we call our topic informant work group; these are people who will essentially be our liaison with people who are experts about themselves, essentially, or who are trying to promote their own causes and what not on, say, a Web site. That work group is probably going to have to carefully manage certain marketing efforts and so forth. I can see nothing wrong with CNET linking to some of its more important articles that really explain something very well or for that matter articles about current events that we have, but on the other hand don't want to be a link repository for every link that anyone might possibly want to put up. It's just that actually reduces the quality. So, it has to be managed, and how it's going to be managed is a problem that we're going to look at very carefully.

First, anonymity tends to make people into jerks if they have any tendencies in that direction.

In some ways the Net that we know is based on the idea of being anonymous. Why doesn't Citizendium allow this? Why do you have to reveal who you are and use your real name to use Citizendium?
Sanger: Well, there's a couple of different reasons. First, anonymity tends to make people into jerks if they have any tendencies in that direction. They lack accountability and because they lack accountability that enables some people to disrupt the process. And there's some other reasons too, though. The biggest other reason is that if we use real names, the whole of the project looks a lot more credible. If I look at the page history for an article and I see nothing but real names, I have some confidence that if someone has put in some really egregious error, they're going to be reined in. I don't have any such assurance if there are a bunch of pseudonyms and mere IP addresses.

By using your real name, you have a sense of responsibility to what you post?
Sanger: I think it certainly increases the sense of responsibility and the actual responsibility that other people can hold you to.

But that won't stop all offensive or untrue content from being posted, right?
Sanger: No, of course not.

How long does the article approval process take?
Sanger: Well, it depends on when you consider the article approval process to begin. From the time that someone submits an article for approval, someone nominates an article for approval, our minimum is one week. So in other words, other people have to have the opportunity to comment on an article for at least a week. The approval process is not meant to be rapid, because when we put our approval on something it's actually supposed to be meaningful and important. Right now we are doing all of our approval by hand, and what we are going to be doing is changing the software so that editors can simply click a button and an article will be approved. They really can't do that right now, and that will of course make the process a little bit faster and easier.

Citizendium is a huge project. Where are you getting the bandwidth?
Sanger: Well, Steadfast Networks of Chicago has generously donated the bandwidth to us, the literal bandwidth as well as two of our five servers so, we're very grateful to them for that, but we're also paying a monthly bill and it ain't cheap. We are soliciting donations and just yesterday I put out feelers to my own network, telling people about the launch and so forth, and hopefully we'll get some extra infusions of cash that way. And there's various sort of things that we can do. One thing that we have thought seriously about doing is adding information about our donors to the bottom of pages. I think that could be a nice incentive for some companies and foundations.

We're here at the beginning of Citizendium. Where do you see yourself in the next five years?
Sanger: I would hope five years down the road that we would have over a million articles, that we would have over a 100,000 approved articles, that we would have on the order of tens of thousands of quite active people and, say, something like 10,000 editors who are reasonably active. I would hope also that we would be one of the highest ranked Web sites. I also would hope that we would be launched in many different languages and each language having its own editor in chief. And I would hope that all of those people at least would be making a living as editors in chief.

And there's a number of other projects that we are, different people in our orbit are working on, for example, a sort of encyclopedia of debate, a debate guide so to speak--something like Debatepedia, but done our way--and summaries of the news and a few other things. Actually, five years down the road, by then, I hope that I will no longer be the editor in chief--indeed, I've already committed not to being the editor in chief--and that someone else would have taken over, and in fact that person should have been replaced by a third editor in chief by five years from now. And so the whole community should be, as it were, a sovereign entity beholden only to the community members and not to any single individual.