Larry Sanger, cofounder of Wikipedia, writes about concerns he has with the project (having left it some years back). He sees three major concerns with the project, summarized as follows:

1. People don't see Wikipedia as reliable because anyone can edit it (PR problem)
2. People tolerate the clueless and the trolls too much in the name of civility
3. People don't defer to experts, in the name of anti-elitism

You can read his posting for the fine details of the above. So far, this is the best critique of Wikipedia I've seen -- Larry backs up his points well, and one can see how he thinks and how it makes sense, even if one doesn't agree with him. Having been deeply involved in the project and the politics involved, I can see that the issues he's pointing out are quite real. Now, for my specific thoughts on each part of his article. The first part I don't see as so much of a problem -- while it'd be nice not to have the PR issue of being seen as unreliable, I don't think it's quite so bad. Of course, he addresses my reply, as it's a predictable one, and makes a case for perception of reliability being useful. I remain unconvinced that it's very important, but I can see it to be a plus to have good PR. I also don't find the SEP to be a particularly inspiring example. As for the second part, I really have mixed feelings. I'll disclose that I'm a candicate for the Mediation Committee, which works to try to help people work together, attempting to build compromise between editors who have trouble agreeing, and I was a (failed) candicate for the Arbitration Committee, which deals with applying judgement to users who cannot be dealt with other ways. Should such groups be necessary? Historically, on the internet, we've always had a variety of people, from full-fledged kooks such as a certain stamp collector with whom I once dealt to marginal people who would be fine for years or on most topics, but have a few ones where they were a bit unhinged, to people who knew each other personally and would not be friendly for any reason, topic-independent. The community has always had difficulty deciding how to deal with them -- in part we all had tools that would help make their posts disappear (blocklists *PLONK*) and similar, because Usenet consists of threads of discrete posts, with intelligent clients that dealt with the data as they saw fit. Wikipedia isn't shaped like that though -- we use generic clients that are handed the data in a way significantly closer to presentation-layer than data-layer, and it wouldn't really work to ask for all of Ralph Raver's contributions to Wikipedia to be filtered out for our view, because articles have many editors, and it's not really possible to ask for an article sans one of them. In other words, that solution doesn't fit here. One way to deal with the trolls is to have harsher punishments for bad actions. This sounds good to me -- it does seem to me that people are willing to give the benefit of the doubt too far on Wikipedia. It is important to be polite, and all other things being equal, to be welcoming, but not at the expense of the project. Wikipedia has its share of slogans that various policy factions use to push their point of view -- the one for this one is Wikipedia is not a social club. I think Arbitrators and Mediators play an important role in helping with civility though. As for the third point on anti-elitism, I'm not sure Larry is entirely right in his assessment on the project, or at least that it's so clear-cut. I've been involved with disputes where people pull sources from all over the place, dealing with historians, philosophers, and similar sources in order to justify their positions on what should be in an article (e.g. is X a theocracy, is Y totalitarian, etc). I'm not sure they consult experts directly, instead consulting their work to bolster arguments, but I'm also not sure if that's the same thing as Larry suggests. As for his suggestion for peer reviewed versions of Wikipedia articles, it has been brought up, and I think it's a good idea. It may need to be managed in a separate wiki from the main one, in order to keep those articles open for mainstream edits. In sum, social changes would be helpful with Wikipedia, to remove the idea of "the newbie is always right", and to move further in the direction of an open meritocracy (without the shame that the culture currently ascribes to its motions so far in that direction). I wish I had met Larry when I was at Ohio State -- it would've been interesting. I was involved in Nupedia, almost, when it was starting, but found their organization to be too exclusive, so I do appreciate that Larry's ideas for wikipedia now are minimally invasive.

Of course, this being posted on Kuro5hin, a slashdot clone, there are a few people who replied who have nothing to say and want to say it loudly. Larry has a reply to some of them that's kind of interesting.

In other news, I just got back from visiting my family for new years. It was enjoyable, although I really wish my grandparents had been around. N got to meet my family, and I got a large number of things that I lacked that my parents had spares of (mainly kitchen supplies). Among the things that happened in the gift exchange were my reciept of a 300G external USB2 hard drive. This is *very* useful -- I finally don't need to worry so much about hard drive space. Hurrah!