Vision is one of the most powerful forms of long-term thinking. Jimmy Wales, founder and president of the all-embracing online encyclopedia Wikipedia, examines how vision drives and defines that project and its strategy--- and how it fits into the even larger world and prospects of "free culture."
"The design of Wikipedia," said its founder and president Jimmy Wales, "is the design of community."
When Wikipedia was started in 2001, all of its technology and software elements had been around since 1995. Its innovation was entirely social--- free licensing of content, neutral point of view, and total openness to participants, especially new ones. The core engine of Wikipedia, as a result, is "a community of thoughtful users, a few hundred volunteers who know each other and work to guarantee the quality and integrity of the work."
Wikipedia, already enormous, continues to accelerate its growth. It is one of the top 20 websites, with 5 billion page views monthly. As an encyclopedia, it is larger than Britannica and Encarta combined and is now in so many languages, only 1/3 of the total Wikipedia is in English. When Wales went to Taiwan last week, strangers recognized him on the train, and 1,200 came to his talk. (One attraction to a Chinese audience is that Wikipedia takes the position of "no compromise with censors, ever.")
The free licensing of Wikipedia content means that it is free to copy, free to modify, free to redistribute, and free to redistribute in modified forms, with attribution links. This is in service to the Wikipedia vision "to create and distribute a free encyclopedia of the highest possible quality to every single person on the planet in their own language." One byproduct is that Wikipedia's success is helping shift the terms of the copyright debate, in a public-good direction.
The secret of Wikipedia's content-generating process, Wales explained, is the nurturing and shaping of trust, instead building everything around distrust. He said that most social software systems are designed around expected problems. "Suppose you ran a restaurant that way. If you serve steak, that means steak knives, which are really dangerous in the wrong hands, so you need to put barriers between the tables."
"If you prevent people from doing bad things, you prevent them from doing good things, and it eliminates opportunities for trust."
Thus every page of Wikipedia has an open invitation to edit it, and the operational motto is "Be bold." The expectation is that most edits will be improvements, and they are. Problems are dealt with completely post facto. There is an all-recent-changes page watched by hundreds of people, and another page proposing "Articles for Deletion." Regular users set up watch lists for Wikipedia articles they care about, so they are notified immediately of new edits. Besides the edit history and text comparison features of the wiki itself, many users employ IRC (Internet Relay Chat) to discuss ongoing issues, from article details to general policy. The court of last resort to resolve fraught issues is a benign emperor, Jimbo Wales.
Wales continually fights the programmers to keep them from automating matters he thinks must remain social. Issues are decided not by voting but by dialogue, in which some voices have more weight because they are recognized to have earned it. Yet users do not get formal ratings. "Suppose you had to go around wearing a badge that says how many people like you." In support of the Wikipedia rule to welcome new contributors, programmers would like to install the ability to automatically send a welcome note to anyone who has made eight contributions. Wales insists that only people can welcome people. The best way to keep Wikipedia deeply radical, Wales feels, is to keep its process deeply conservative.
Wikipedia is a window into further realms of free culture. What else can be done with wiki-enhanced communities? "A library is bigger than an encyclopedia." So alongside the nonprofit Wikipedia Wales has set up the for-profit Wikia--- a general purpose wiki community enabler, drawing its income from Google ads.
Most leaders, in my experience, focus on their organization's product. Jimmy Wales focuses with exceptional clarity and insight on Wikipedia's process, and therein lies its magic.
--Stewart Brand
--
This talk took place at The Cowell Theater, Fort Mason Center, San Francisco
Last edited by bkeating on Jan 09, 02007 6:35 pm; edited 1 time in total
Craig_Hubley Joined: 05 Oct 2006 Posts: 7 Location: Nova Scotia, Canada
Posted: Oct 10, 02006 5:52 am
How much of Wikipedia's vision was originally Wales' is open to question. For one thing, it was Larry Sanger who defined the original epistemics and tirelessly enforced namespace conventions and prohibitions on originality - these are the difficult problems. Sanger's PhD thesis is about how to do Wikipedia, basically. And Sanger never accepted powers to coerce any other user, and even in the ugliest debates with others he never asked for anyone to be excluded. He was the one who rejected coercive means so conclusively. He made only one major policy error, that being, to fail to put content absolutely above an extremely tenous "virtual community".
However, the leakiness of Wikipedia seems to have mostly corrected those problems: no one can actually be excluded, and many different (and fighting) "communities" can be hosted. The introduction of portals probably helped - there is now no one page everyone is expected to be reading all the time.
A lot of very foolish people make very silly noises about anonymity as if all those fly-by users would continue to contribute if they had to "log in". For myself I can say that I don't "register and log in" to correct errors I see on any web page. Letting the fly-by edit is obviously the right policy (and it would do LongNow a world of good, too - can we replace this BB with a mediawiki please - or better the getwiki software that Wikinfo uses). A couple of things seem to me to have led to Wikipedia's usefulness.
First, there was Sanger's good foundation - without good names and good dispute resolution you have nothing, because you won't be able to write a new article with links to the old articles unless the naming is near perfect. They knew enough in the beginning to not be afraid of open links, which mattered a great deal, because it meant articles didn't have to be written ten times over as many related issues got covered. Each open link was an invitation to edit, and when there were lots of them, it was very inviting. Another key thing was that you could REMEMBER a Wikipedia URL, as it was just "en.wikipedia.org/wiki/WHATEVER" and if WHATEVER was always the most common or neutral name, that made it very easy to recall.
Second, there was the insistence on never requiring that anyone "log in". Because the unpopular can't be kept out, they can only be prevented from building up a positive reputation among the current users - which I see as a good thing, since reputation leads to less examined contributions, which leads to errors and all kinds of biases and problems. Examined edits are better edits, and untrusted users get their edits examined. In peer review we take the authors' names off so that no one will give them a bye, why is this any different? There's a strong argument to never have usernames at all, if you really want each edit to get the same even peer review treatment.
Third was the commitment to practice what it preached. The group of people that got attracted in early did care if they were being hypocrites mostly and did correct their own bad behaviour instead of justifying it as almost all power structures do. While there was no formal board, there were a lot of people who had the status of one. A lot of people wanted a more serious nonprofit structure in 2002-3, including Fred Bauder (who founded Wikinfo.org) who deserves more credit than Wales for many good policy decisions (Bauder remained involved in Wikipedia). But the person who stands out most in forming the present Wikipedia polity is Florence Devraux, who was most instrumental in creating its ArbCom. Plus a lot of people who never revealed their names, maybe to make some kind of point. Like "anonymous edits are good". As I say, they practiced what they preached. It's hard to think of a project that took peer review into all its processes so deeply. Certainly Creative Commons or LongNow have not done that, they are nowhere near as transparent as Wikimedia now is.
Fourth, one must credit Wales not with originating but with understanding most of the doctrine he got from others. It's true that "Wales continually fights the programmers to keep them from automating matters he thinks must remain social" though Wales by doing so is effectively maintaining power among a clique. A clique that he personally has a lot of influence over (see wikitruth.info for a critical view of the Wikipedians) - even though some of it is unintentional. Wales doesn't object to being called a "GodKing". And as if he were some kind of Norse god, he continues to use subjective and indefensible terms to describe users, like "trolls", which in response quite a few users adapt into their user names to thumb their nose at him (or make some point about exclusion). As just one indication of the groupthink factor, for a while anyone who adopted such a name was accused of being myself, as if I was somehow the only person who realized that "troll" was a power word and could be adopted by its victims. If anyone had bothered reading the dates on pages, they might have realized that my adaptation of the term post-dated Wikipedia's troll cult. I guess I could be thankful as this is what drew me into reading a lot of the power struggle dialogue, which is very often absolutely hilarious. It also turned my own name into a power word. Which can't be bad, really. Wales may be interesting as much for the people he has fought with - or thinks he has fought with - as for those he has collaborated with.
Fifth, Wikipedia is funny. Very funny. One thing about resolving problems purely socially, it's absolutely riotous. After reading Wikipedia:vandalism_in_progress I often can't even breathe. It's easily the funniest page on the entire net. Some of the SOLLOG episodes were worthy of Shakespeare.
And it's funny precisely because "issues are decided not by voting but by dialogue, in which some voices have more weight" but not "because they are recognized to have earned it" but because they are more popular and say things that more readily trigger the hypothalmus of the stupid - it's "politics as usual".
Wales admits this when he explains why "users do not get formal ratings" - when he says "suppose you had to go around wearing a badge that says how many people like you." This is a bald admission that it's nothing but a popularity contest.
So, sixth, and again to Wales' credit, it's good that there are no formal "ratings". If Wales himself likes you, that does not really count for much. Yes if you want some kind of power to arbitrarily exclude others or destroy their work (the so-called "sysop vandalism"), then Wales himself has to like you, and you have to be concerned that he keep liking you. This is a problem but a minor one - ArbCom sorts out most of it pretty well and may eventually lead to an independent nonprofit board of directors, without Wales or Sanger on it. Failing to have that board is probably why Wikipedia is getting by on a few hundred grand a year, while longnow can get millions. I told Sanger in 2002 that he needed this independent board. (I think the next few people to tell him that he needed one really annoyed him. It was hardly an unusual observation, as most such projects have unbiased board members from outside).
I can't really say that the process focus is all that good. It's far better at more political forums like openpolitics.ca or dkosopedia.com - but those don't have the user volume. It's odd to read Stewart saying that "Most leaders, in my experience, focus on their organization's product." That is certainly not modern quality management which explicitly focuses on the core business competences - the processes - that make the product.
To say that "Jimmy Wales focuses with exceptional clarity and insight on Wikipedia's process" is more than a bit overblown. He has learned from a lot of good observers and advisors, and no doubt continues to. But he lucked into the project, and has little to contribute to it now other than as a mouthpiece for those he personally likes and defends. Wikipedia needs an editorial board - if it wants to be conservative, it should start to be a bit more Britannica-like. That board need not do very much, it might only decide between versions to put on a disk or which references are most credible. But the time is long past when Jim Wales was capable of moving that project forward.
To refer to the project as coming from his personal "vision" just retards it.
I'd be quite interested in what Larry Sanger or Florence Devraux have to say, especially as Larry is off working on his Digital Universe project now. Jim Wales has the first word on Wikipedia, but he shouldn't have the last.
Uploading ....
