Andy Carvin's Waste of Bandwidth

December 15, 2005

Nature’s Wikipedia-Britannica Death Match; Adding By-Lines & Trust Rankings Next?

Filed under: Media & Politics — Andy Carvin @ 3:58 pm

The latest round of the Wikipedia accuracy wars turns out to be a tie. According to an investigation by the science journal Nature, Wikipedia fared well when compared head-to-head with Encyclopedia Britannica on science-related entries.
The magazine asked experts in various scientific fields to review 42 topics in both Wikipedia and Britannica. The result: both sources had a similar number of mistakes. On average, Wikipedia entries had four errors or ommissions, while Britannica had three. When you add these up with misleading statements, 162 were found in Wikipedia, while 123 appeared in Britannica.
“People will find it shocking to see how many errors there are in Britannica,” Nature quotes information scientist Michael Twidale. “Print encyclopaedias are often set up as the gold standards of information quality against which the failings of faster or cheaper resources can be compared. These findings remind us that we have an 18-carat standard, not a 24-carat one.”
The most error-prone article? Dmitry Mendeleev. If you go to the Wikipedia entry, there’s a note at the top saying that Nature identified certain problems with it and they’re working to correct the article. But when you access the Britannica version, there’s no mention of any dispute. Perhaps it’s because Britannica’s online entry for Mendeleev is so brief there’s little room for error, and the errors are only in the most recent paper version of the encyclopedia. There’s no way to tell. With Wikipedia, at least, there’s a paper trail you can follow to see who change the entry and how it was changed.
My guess is that the Nature study will do more to harm Britannica’s rep than improve Wikipedia’s. Chances are, there will be another instance similar to the Siegenthaler controversy that led to so much bad media attention for Wikipedia. They’ve change the rules so that anonymous participants can only edit articles rather than create new ones. But to me, that doesn’t go far enough. While I’m concerned about protecting the anonymity of Wikipedians posting on sensitive topics, particularly from countries that oppress free expression, this issue only affects a very small minority of entries. Isn’t it more important for Wikipedia to build confidence among the online public? If that’s the case, the anonymity policy needs to be assessed more radically. I love Wikipedia, but I’d feel a hell of a lot more comfortable with what I read if there were also a transparent paper trail for the Wikipedians editing articles.
Perhaps a solution would be to strongly discourage anonymity. For those entries that have edits posted by an anonymous Wikipedian, place an icon prominently on the page warning us that the entry was edited anonymously, so readers can make an informed judgment. For those that have been edited by people willing to log in with their names, have those names appear on the entry’s page as a by-line, with links to their biographies. I know that all of this information can be found in the entry’s history page, but the average Internet user who doesn’t know the inner workings of Wikipedia won’t realize this. By placing the names of the contributors on the marquee of each entry, the authors are forced to stand up and take credit for it – for better or worse – just like a scientific journal.
Then, perhaps we need to add an eBay-like rating system for Wikipedians. For Wikipedians whose work is judged as accurate, let readers award them a point, or perhaps 1-10 rating system (ie, a perfect 10 for stellar wikipedians with strong credentials who cite primary source materials obsessively, and a 1 for those whose work is clearly incompetent). That way, when you go to a wikipedia entry, you can judge it on the rating of the Wikipedians. You could even do the same for the articles themselves: wouldn’t it be useful to know if 83% of the readers of one particular entry found it lacking in one way or another?
On eBay, I’d rather buy from someone with a high trust ranking; wouldn’t we all want to find the same level in trust among those who are creating the knowledge we’re consuming?
Anyway, it’s just an idea…. -andy

No Comments »

No comments yet.

RSS feed for comments on this post. TrackBack URL

Leave a comment

Powered by WordPress