Freedom of Anonymous Speech: Ross Mayfield writes of the Wikipedia debate....
Beyond this attempt to weaken anonymity on the Net, Wikipedia's open nature is also under attack. Adam Curry edited podcasting history in his favor. Big deal. It's a wiki, just edit it if you disagree and let the community's practice work over time.
Consider regulating against graffiti. You have two options:
Guard every wall in town to prevent the infraction from occurring
Paint over infractions and enforce the law by chasing down perpetrators
The former is not just prohibitively expensive, it kills creativity and culture. The later is the status quo and generally works, especially where communities flourish.
So what would have Wikipedia do? Lock down contributions through a fact checking process with rigid policy? Or let people contribute, leverage revision history and let the group revert infractions.
Ross is wrong when he says "Big Deal. It's a wiki...." The tools don't add value, the user does. If the content of a wiki is essentially different because of the the tool used to create and edit it, I've yet to see it documented and that would be a good piece of research for someone to undertake.
As I see it, the content of a wiki is a direct reflection of the human behind the words, data, images or media. So, we do need to take seriously the quality and veracity of information wherever we find it—Wikipedia is no exception. Critical reading is an important skill to carry forward. When you are talking about creating a reference work like Wikipedia, it is important that the reader have the information to judge the "facts" they are presented with so that they can decide whether they will take it as correct or continue to do research.
If the purpose of Wikipedia is only to create a forum for changing entries, I suppose it would be sufficient to say "Big deal, I can change this entry." That, however, is not a viable reference system, it's just more work for people who need access to information quickly and conveniently. If this is the purpose of Wikipedia—to provide a venue for anonymous debates about facts and interpretations of facts—then the company should stop licensing the content to other sites as a "reference" source.
If we are not thinking about that reader using the Wikipedia as a reference, who will suffer from misinformation they adopt as fact, then we are not understanding the social importance of capturing and organizing data.
We don't buy an almanac to use it to begin research into [picking up my The New York Times 2005 Almanac and flipping randomly to page 512] the average inflation rate in Belarus between 1990 and 2001, I want to find that it is 318.1 percent. I can accept this "fact" based on my relationship with the source, which is the Times Co. and the World Bank, which is the source cited by the almanac. Those are sources I can rely on or, at least, cite when challenged by someone who insists that Belarussian economy experienced no inflation during those years.
The important thing is that when we are talking about sources of shared information, as compared to graffiti (a spurious comparison, as the accurate analogy is to an encyclopedia or almanac or dictionary), we can debate on the record. The reader has recourse to a source with a name and a reputation rather than trusting the editorial process alone. If I wonder whether this World Bank has the Belarus inflation rate wrong, I can check another source. If the other source has an anonymous attribution for a different number, it is of no value to me, but if I find that the International Monetary Fund says the Belarussion inflation rate between 1999 and 2001 was 165 percent, I can judge that source against the World Bank and Times Co. Then, I do more research.
I found the other day that there is a Wikipedia biographical reference to me. Apparently machine-generated, it was incorrect and incomplete. I have no interest in writing PR about myself, but I do mind that it identified my home incorrectly. It said I live in suburban Seattle. I don't. and I changed it to say that I live in a suburb of Tacoma, which is a separate city with a history and character all its own. In point of fact, I live in Lakewood, which is also in Wikipedia with the stunning detail that the Fox Network show, COPS, has filmed more segments with our local police force than any other. There is no external link to use to verify that "fact."
That COPS "fact" was contributed by an anonymous source at 126.96.36.199, who has also contributed to the "list of songs in English labeled the worst ever" and the entry on the Ford Crown Victoria (a police vehicle, which may explain the COPS connection). That is not a source I can trust, but for the uncritical reader this commonly repeated bit of trivia about the city, which may or may not be true (and I have no interest in finding out), is one of the only notable details in an entry otherwise limited to demographic and geographic information.
Okay, in the course of following one link, to me, we have a whole slew of unverified or incorrect information. That's inefficient.
Fernanda Viega has shown in her research on the "chocolate" entry at Wikipedia that incorrect facts are repeatedly reinserted, over-writing correct information until the person who had the correct facts simply gave up [she talked about this at a Microsoft Research event earlier this year]. The group doesn't "revert infractions," it also reinforces groupthink, as Nicholas Carr has written.
If the only measure of Wikipedia is the ability to change the text, then we are serving the authors of information. We are not serving the audience, who may labor under perverted interpretations of the world and increase the number of bad decisions made based on that incorrect information. If Wikipedia is essentially inefficient, which it is if the purpose of the reference system is only to change and not to provide information as its core value, then it will fail. And I don't want to see it fail.
It's the person who comes along seeking information rather than seeking to contribute to the discussion who we have to keep firmly at the center of the value calculation, since we're talking about bodies of work that will eventually outlive us. If it can be inaccurate because of the format and tools it is presented in, but no one is available with direct knowledge to correct errors in the work, then our primary contribution to the future will be deeply flawed records of the past. That's troubling.
Ross is fighting the wrong fight about the right tool for a particular kind of work. The tool doesn't bless the content or the contributors with the right to anonymous tinkering with history. Wikipedia with some process can be a profound leap forward in capturing our species' story, but only when we approach it critically and with an eye to the strengths and weaknesses of the contributors, too.
UPDATE: Ross replies, saying:
Cnet has a pretty good article on the liability reform [Ross' posting has the incorrect link at the moment, which is fixed here] sought by Seigenthaler, the first argument I made. Mitch Ratcliffe takes issue with my second argument, about how a wiki works and how best to regulate it. Mitch, you keep trying to fit Wikipedia into your model of how an encyclopedia should be instead of recognizing how it is different. A print version of Wikipedia should have an editorial process bolted on to emergent practice, as it is a comparable product, frozen in time. But instead, the evolving nature of Wikipedia needs to be recognized and celebrated for what it is. Help people understand what it is, not what it is not.
Here's my reply:
I am absolutely not trying to fit the Wikipedia into my model of how an encyclopedia works! The encyclopedia doesn't work anymore, but the process of making judgments about information that people use today—and have always used since the dawn of rational thought and recorded ideas—is essentially the same. That process is not changed by the gathering of information, but in its publication. Wikipedia publishes its information as a reference work and gets paid for it, without providing any cautionary alerts that the content may be flawed, because, apparently, it's no big deal since the flaws can be erased. But what about the person who was misled by transitory bad information?
The argument that wiki makes the information product different is specious, because, even if the content is constantly in transition, each instance of a person using the Wikipedia as a point of reference must fulfill the criteria for judging the veracity of information.
How Wikipedia fulfills those criteria is exactly about what it is. Comparing Wikipedia contributions to graffiti is a poor analogy, but saying so doesn't imply that Wikipedia must fit into the analogies I suggested. I'd rather see you reply to my counter-argument with a better idea.
The publisher's exemption for liability issues only underscores the importance of bringing the contributors themselves into the public record so that their ideas can be judged based on conflicts of interest and general reliability over time. It would be a better world if no liability lawsuits ever happened, but we first have to come to terms with the essential contentiousness of the marketplace of ideas if we're going to move past hurt feelings and reputations based on incorrect information that has been quickly corrected.
The anonymity issue is very important in many circumstances, but it has nothing to do with the disruptive nature of information gathering represented by wiki in the context of the organization or markets, which are predicated, even if non-traditional models, on the exchange of value. An anonymous marketplace of ideas is one without any foundation in reputation which makes judging something or someone very difficult without being able to refer to the opinion of others—it's far more valuable to arrive at a shared practice, to use your term, through knowledge of one another than to treat the information as devoid of human influence by insisting anonymity is necessary to statements of truth. If we are saying that now all social interaction must be anonymous then we are erasing the accountability on which the social contract—any social contract—is built, and that's just plain bad for everyone.
UPDATE2(tangent): This is very funny.
[Disclosure: I am on Socialtext's board of advisors. I believe wiki is an excellent tool for collaborative editing and have used them in a variety of settings that proved their value to me. All those environments involved some process and accountability that contributed to rapid communal thinking. This posting should be considered part of the discussion, not a dismissal of wiki technology.]