First, in a Nov. 29 op-ed piece in USA Today, a former administrative assistant to Robert Kennedy lambasted the free online reference work for an article that suggested he may have been involved in the assassinations of both Robert F. Kennedy and John F. Kennedy.
Then, on Dec. 1, a new flurry of attention came when former MTV VJ and podcasting pioneer Adam Curry was accused of anonymously editing out references to other people's seminal podcasting work in an article about the hot new digital medium.
To critics of Wikipedia--which, in a spin on the open-source model, lets anyone create and edit entries--the news was further proof that the service has no accountability and no place in the world of serious information gathering.
"Wales, in a recent C-SPAN interview...insisted that his Web site is accountable and that his community of thousands of volunteer editors...corrects mistakes within minutes," former Robert Kennedy aide John Seigenthaler wrote in USA Today. "My experience refutes that...For four months, Wikipedia depicted me as a suspected assassin."
Wales has dealt with criticism for years, and he's sensitive to it. He knows that many people worry that Wikipedia's self-policing process can't possibly keep up with the massive number of new articles that crop up on the site, and the edits that appear in existing entries. The cybertome, after all, is home to millions of articles--nearly 850,000 in English alone, with many other entries in dozens of additional languages. In October, the English-language site hosted 1,515 new articles per day.
History by committee
But Wales said the Seigenthaler incident was an aberration.
"The system failed in this case," Wales said. "A bad entry was kept for some time until (Seigenthaler) actually fixed it himself. Basically, what I would say is we're looking right now, and over the weekend, at this particular incident and what went wrong. It seems like the key issue is we're having some growing pains."
When Wikipedia articles are first published, they show up on a special page, and volunteers--so-called new-page patrollers--monitor entries in their area of interest.
Wales said the Seigenthaler article not only escaped the notice of this corps of watchdogs, but it also became a kind of needle in a haystack: The page remained unchanged for so long because it wasn't linked to from any other Wikipedia articles, depriving it of traffic that might have led to closer scrutiny.
Also, Wales said, the entry was unusual in that it was posted by an anonymous user--most new articles are published by registered members, who are more likely to be held responsible for what they write.
Thus, to avoid future problems, Wales plans to bar anonymous users from creating new articles; only registered members will be able to do so. That change will go into effect Monday, he said, adding that anonymous users will still be able to edit existing entries.
That's less of a problem, Wales suggested, because changes are frequently vetted by members who keep watch lists of articles they want to ensure remain accurate--perhaps even articles they've written themselves.
The change is one of the first that would specifically limit what anonymous users can do on Wikipedia. And some may see that as a significant step for a service that's traditionally prided itself on letting anyone participate. But Wales said the move is not a major one because, as mentioned, most new articles are already written by registered Wikipedia members, and most anonymous users' actions are edits to published entries.
Currying disfavor
Meanwhile, the brouhaha surrounding Curry and the podcasting article raises new questions about whether people should be allowed to create or edit Wikipedia articles about themselves or projects they've been involved with.
"Wikipedia is so often considered authoritative. That must stop now, surely. Every fact in there must be considered partisan, written by someone with a conflict of interest," blogging and podcasting pioneer Dave Winer wrote in his blog. "Further, we need to determine what authority means in the age of Internet scholarship."
Curry deleted references to work presented by Technorati principal engineer Kevin Marks at the 2003 BloggerCon at Harvard University. But from Curry's perspective, conflict of interest had nothing to do with it; he simply believed the references were inaccurate.
And when he discovered they weren't, he explained in an e-mail to CNET News.com, he realized he'd made a mistake.
But that "doesn't mean I'm not allowed to have an opinion of the facts and change Wikipedia to represent my viewpoint," Curry said.
Wales said he's not sure how to approach the question of whether people should be allowed to post on subjects in which they have a personal interest.
"That's an interesting philosophical issue," Wales said. "Because on the one hand, particularly with things like podcasting, the people involved are people who know a lot about it, and on the other hand, when people are editing something they've been personally involved in, it can be hard for them to be neutral."
He added that traditionally, Wikipedia has discouraged users from participating in such entries and asks them to be mature and serious when they do.
"But we don't have a rule about it, because it's too hard to enforce, and it may not be a good idea."
In the blogosphere, however, Curry is getting beaten up for having edited out the Marks references as well as a sentence in which Stephen Downes had been credited with delivering MP3 files via RSS feeds.
But Curry bristles at the accusation that he was intentionally trying to deprive anyone of due credit.
"That I'm trying to inflate my role in the history of podcasting is a mean-spirited claim," he said, "and not based on the facts of my (Wikipedia) edits and entries. But the meme took, and now I'm the asshole of the week."
