I got a response from Jimmy Wales to a note about Wikipedia on Scripting News on a private mail list. He may if he wants, make his response public. Imho, it was unnecessarily personal. But he showed me something about Wikipedia that I hadn't seen before. My first note wasn't really a bug report to Jimmy Wales, or a request for an investigation of the change history of any single article on Wikipedia, but I certainly understand why he might have intererpeted it that way. But now that he has responded, I feel the issue has grown a little larger, if only temporarily.
This is much like the discussions we had when we first met the advocates of open source. They, like Wales, maintained that they had cracked a large problem, a bug in human nature, they had found a way to create a hive that worked cooperatively in almost a utopian fashion to develop software for free that was perfectly tuned to human needs. People who didn't develop software, who were somewhat mystified and frustrated by the process of asking for features but never getting as much as a courteous response from the people they paid for software, hoped they could get better service from programmers who worked for free.
Why should working for free be the big difference, we puzzled. We were already working for free, and also giving away most of our source, but human nature hadn't changed in our world.
Well, it turns out it hadn't changed in theirs either -- most open source projects treated their users' check-ins as feature requests. Projects were pretty much as they were in the commercial world, or on Compuserve communities (dating back to the early 80s) or in communities we participated in in the 70s on Unix systems at universities, without the formality of open source, without license agreements. Basically programmers can be generous people, but to get a crafted user interface required the work of a single mind, or a very small group of people who worked really well together.
Enter Wikiipedia, making the same kind of claims about reference work that Eric Raymond and others made for open source in the mid-90s. It turns out it's as much snake oil as what ESR was selling. They hadn't cracked some basic code in human nature. People are still people, and they're going to produce biased work and there are gatekeepers, only power isn't expressed in money these days, it's in time.
The supposed technical methods they have to guarantee integrity fall down. You can see the results in a fairly technical area, where there really is very little doubt about the facts, all the people are alive, a researcher could, if he or she wanted to, build a pretty accurate picture. I read histories as a hobby, I'm not a historian, but I love history, I have since a child. These people are making a mockery of the process of doing research, much as the open source advocates were mocking the very serious art of developing clean, usable and reliable software, something I am an expert in, not as a hobby.
The result in software was a worsening of an already bad situation. We were making progress in the 80s on ease of use, on crafting user interfaces that really worked for people. A lot was lost that we should be getting back to, and maybe now that the hype for open source is completely over, we are getting back to work on these very important problems. There's real cause for hope. But please, let's not go through the same thing here. Wales is a promoter, clearly, selling his own kind of snake oil. He's really smooth, had me going there for a bit, until he made it personal. Then I was thinking here, no -- he's not trying to build a great reference work, he's hyping something. Not sure what his gain is, and frankly I don't care.
People have asked me off-list -- what should be done here? (I guess they don't want to get in Wales's cross-hairs as I have apparently.) There's nothing to do, I say, other than set expectatiosn realistically. Wikipedia is not a revolution. The concerns of librarians and professional researchers are valid. Let's not throw the baby out this time. This isn't about the promoters, because Wikipedia isn't something they control, they'll tell you that all the time, and it's true
# Posted by Dave Winer on 6/13/05; 4:06:07 AM - --
