This page is older archived content from an older version of the Emerald Publishing website.

As such, it may not display exactly as originally intended.

An interview with Jimmy Wales

Interview by: Margaret Adolphus

Options:     PDF Version - An interview with Jimmy Wales Print view

Photo: Jimmy WalesJimmy Wales is an American Internet entrepreneur and former options trader. In 2001 he founded Wikipedia and in 2003 he set up the Wikimedia Foundation, a charity, of which he is Emeritus chair. He is also, along with Angela Beesley, the founder of Wikia, which is working on a search engine project.

Jimmy travels widely to lecture and meet other wikipedians and, this year alone, has visited Moscow, London, Adelaide, Perth, Sydney and Melbourne. He was interviewed at the Online Information 2007 conference on December 4 by Margaret Adolphus.

What was the impetus that got Wikipedia started?

I was inspired by the growth of free, open source software, the great collaborative work being done by volunteers from all over the world, for example Apache, the Perl programming language, and the php server side scripting language – all the great software that runs the Internet and which is available under free licences. I realized that this kind of collaboration would be very important in the future and that it could be extended beyond just software into many kinds of cultural works, including an encyclopedia.

What are the strategy and tactics behind Wikipedia?

To be honest, I don’t think I had any strategy or tactics, we just started working. Prior to Wikipedia I had a project called Nupedia, which was very top down, very controlled and a complete failure, because it was just not possible for people to get involved so it was not much fun. And so, from the very beginning of Wikipedia, I really focused a lot on openness. A very important principle was to have as many people participating as possible. In a sense that strategy was born in reaction to the failed previous attempt at creating a free encyclopedia.

Other than that we thought a lot about what the users need to get their work done – what are the problems they face and how do we help them resolve those problems – so a lot of the early software development was really wrapped up in that kind of very user-centric design.

Did you achieve what you set out to achieve, or did you create something different?

No, it’s pretty much what I imagined, but bigger! I’ve shifted my own personal focus more and more towards the languages of the developing world, because while English Wikipedia’s quite large, as are all the European languages, we still have a lot of work to do in other parts of the world. I’m really interested in thinking about particular groups of users, what their needs are and how they may be different from those of people who have had broadband Internet access for some time.

You described in your talk for Online Information 2007 how you visited a school in India where they wanted things in English as English is the language of education. Is there a danger that English becomes the dominant language of the Internet?

What we’re seeing is a real rebirth of a lot of languages on the Internet. The lower cost of collaboration is making it possible for people to put together resources they couldn’t have before. Welsh is a prime example of a language where virtually all Welsh speakers also speak English. But yet there’s a group of people who are very passionate about the Welsh language and they are creating Welsh Wikipedia. It’s actually a little bit telling when you realize that the Luxembourgish Wikipedia is larger than Swahili Wikipedia, that tells you the possibilities for language preservation.

It’s just an observation, but what seems to me quite successful is what I would call the Dutch model, in which the Dutch language is in absolutely no danger of dying out; it’s a very robust, healthy language, with newspapers, TV shows, and so on, and yet, if you visit the Netherlands the amount of English spoken there is extraordinarily high. I think that is a model which can work for a lot of places in the world. If you’ve got a good education system, you of course continue to live and breathe your own culture and your own language, but recognize the importance of English as a second language, or perhaps French in some places.

Can you tell us a bit about that?

We are developing our search engine according to four principles.

The first is transparency: we are open about the algorithms and the API [application programming interface]. The second is collaboration: we want all sorts of people, either as individuals or organizations, to be able to contribute in some way to the project. We want to get participation from researchers, academics, programmers, hackers, anybody who wants to contribute, but people can also contribute in more editorial ways, giving us feedback on URLs. For example, this one is nonsense, spam, pornography and so on. It’s always been difficult for a search engine to distinguish between adult content and content that’s suitable for a more general audience, so this sort of editorial evaluation is very important.

Third, quality: if you don’t match the quality of Google and Yahoo, consumers won’t be interested. It can be a different sort of quality though; just as by some measures Wikipedia is higher quality then Britannica. For example, if your interest is in a narrow academic topic the very refined articles in Britannica are probably better, but if you want to find out about, say, pop culture, Wikipedia’s far superior. In terms of the search engine, we want to be of the standard of Google, or better quality, but that quality might be better in some areas and worse in others.

Finally, privacy is very important. When people are participating in a system that involves giving opinions about URLs and ranking things, we want to make sure that they are aware what data they’re sharing and what they’re not. So we put a lot of thought into making sure that what we are doing matches people’s natural intuition. For example, when you edit a Wikipedia article, you’re doing something public for which you’re accountable. If you are just reading that’s your own private business. It’s really important people should be able to read where ever they like without others necessarily knowing about them. Similarly for the search engine, if we have technology that allows people to say this is a good URL, we need to make sure that people understand that’s a public statement. If you’re just browsing the Web you might be sharing your browsing history with us, which will give us ideas about the patterns of what you were doing, but that’s private and you wouldn’t want it published. But you might be prepared to share that with Wikia for statistical purposes (finding out which were the most popular domain names on the Web) if there was a strong privacy policy.

You have an open editorial policy. How do you ensure that your editors really do control quality, make authors accountable, prevent vandalism, extreme viewpoints and downright abuse, such as that perpetrated by the CIA a few years ago?

We police the community by having the community police itself, rather in the same way that the academic community polices itself. Our editors are a group of peers who really care passionately about quality, and if somebody comes in and does something wrong they’ll be very quickly educated, supported, and kicked out if they don’t behave themselves. It’s a peer review kind of thing where people are free to be different, but everything’s visible so you’re always accountable and you can get blocked from editing if you are too bad.

How can the librarian advise patrons on use of Wikipedia, and themselves make us of wikis?

The most important task for librarians is to help their patrons use Wikipedia in a way that’s appropriate, by giving them a little more information. They might say, Wikipedia is usually pretty good, but be aware of how it’s created, these are some of the problems, and check the sources. It would be bad for the librarian to say, don’t look in Wikipedia, it’s written by crazy people on the Internet. I think the nuanced approach is the right one.

As for the organizational use of wikis, there are all kinds of potential uses from writing up company policy, or institutional memory knowledge, to scheduling lunch meetings. The last one I thought a little odd when I first heard it but it actually makes a lot of sense because instead of simply saying yes or no to the calendar invitation you can actually sort of debate about where you eat and whom should be invited. Although I think we’ve sort of fallen into, because of the hype around wikis, the attitude of when you’ve a hammer everything looks like a nail, so people try to use wikis for everything, it’s a great tool for some things, but sometimes it’s not the right tool.

If people are really given the feel that the wiki is theirs to do whatever they need, you can get a lot of innovative work which is specific to their domain. They need to be empowered to share information: all organizations are plagued with people who gain power by hoarding information, which is unhealthy and must be dealt with. If people are encouraged to use a wiki as a collaborative tool, even if it’s just sharing lunch, and not as the fabulous internal tool for collaborating on a new vision, then at least that gets them started and it can grow organically according to their needs.

I don’t know what that might look like to a library, maybe sharing notes on the magazine collection. Hopefully libraries are the antithesis of the hoarding knowledge culture.

We’ve adopted a new paradigm for the Internet – social networking. At what time did content as king move to the social web?

It’s absolutely true that the existence of free software and a lot of the tools that came out of that made all of this possible. But one of the reasons Wikipedia is so popular is that to some extent content is still king. Wikipedia is lots of content, but it’s also socially constructed by lots and lots of people, so it’s interesting to reflect whether it’s social or not.