Wikipedia founder admits some quality-control problems


Via robot wisdom:

Last Monday, an essay critical of Wikipedia prompted [founder] Jimmy Wales to raise the issue of how to improve the quality of writing in Wikipedia articles, conceding that there were significant problems in some areas.

The essay in question was posted by business journalist and author Nicholas Carr on his blog October 3. Its focus was actually on the Web 2.0 concept, and Carr gave it the title, "The amorality of Web 2.0". His reflections were prompted by media coverage leading up to this past week's Web 2.0 Conference, and the idealistic notions of people like conference organizer Tim O'Reilly. Carr argued that although the technology behind the Web is fundamentally amoral, the glowing rhetoric around it is creating a quasi-religious fervor and contributing to the "cult of the amateur".

As an example of this phenomenon, Carr turned to Wikipedia, saying, "If you read anything about Web 2.0, you'll inevitably find praise heaped upon Wikipedia as a glorious manifestation of 'the age of participation.'" His own assessment: "In reality, though, Wikipedia isn't very good at all." To support this, he quoted passages from the articles on Bill Gates and Jane Fonda that he described as "an incoherent hodgepodge of dubious factoids", adding that these were representative of much of Wikipedia's content ...

More at Wikipedia.


I think for me what was telling was a comment from David Gerard quoted by Carr in a follow-up post: "if we want a good encyclopedia in ten years, it's going to have to be a good Wikipedia." It sounded like the Wikipedians had brought Carr around. David Gerard is a Wikipedia administrator (with Blake-like powers, only less important;)). So definitely in the consider-the-source category.

Jimmy Wales does want Wikipedia to be better that Britannica, god bless him. I think there is still room for both Wikipedia and vetted print encyclopediae. I am unsure of the economics.

"who am I to make definitions?"To bring in something from another thread today: You're an authority.BoingBoing does CRAZY huge numbers. Last month they served 23,661,998 pages to 2,222,793 "unique visitors." That says nothing of their full text feeds either.pomegranates to potatoes is probably about it, we can only compare the numbers we have, as imperfect as they are, it's something I guess.Hits Pageviews Sessions IPs. Which is why it's best to look at as many web stats as you can to see the full picture.e.g. The other site on the LISNews server (There's just 2 domains on this server) does 1/10th the number of sessions and 1/25th the number of IPs but does about 30x more hits.

Looking at Bloglines' most subscribed feeds, of the top 25, 13 are professionally-generated content. So there are 12 blogs up there.

But most interesting is some of the statistics that Bloglines offered on numbers of subscribers per feed. They now have nearly 1.4 million feeds with at least one subscriber. Only 36,930 feeds have at least 20 subscribers. And only 437 feeds have more than 1000 subscribers. Bloglines represents, of course, a subset (maybe 5%?) of feed subscribers. But clearly the vast majority of blog feeds are read by only a few hundred people.

For those people, and the audiences they've found, it's great. And only possible online. But the reach of professionals is generally speaking far greater.

I don't think of either slashdot or LISNews as a blog; they're huge collaborative fora, along with Kuro5hin (OK, LISNews ain't that huge).

But then, some people continue to call Cites & Insights a blog, so who am I to make definitions?

Technorati calls Boing Boing the biggest blog. I'd bet Boing Boing doesn't average 1.5 million unique readers a month (but I could be wrong).

Comparing "80 million hits a month" to a magazine's subscription base is, of course, comparing pomegranates to potatoes, particularly given that each access of a different page or message counts as a hit. By that calculus, actually, magazines have phenomenal hit rates: After all, each story looked at, each ad looked at, should count as a separate hit.

No kiddin'? What would you call it? Or, why isn't it a blog, and what's a blog?I think of it as a blog, not like my niece who writes about what she had for breakfast, but it's a collaboritive blog with a helluve big audience.

OK, if you can stretch the definition of "blog" far enough to include slashdot (and if you can demonstrate that the 80 million supposed hits represents more than 1.5 million different people, as opposed to those who are there 10-20 times a day), you have a point.

It would never occur to me to call slashdot a blog.

I forgot the old adage: Who controls the definitions, controls the argument.

My bad.

Agreed. The problem is that many websites are still playing the old game of counting page loads rather then sheer viewers. My own site looks to have huge numbers until you seriously look at how that traffic corelates to real live people looking at the thing. The bulk of the web has nothing on magazines. About the only thing the web can do well is niche markets.

Slashdot reports they do 3 million pages a day, or that was a year ago at least. That's about 10x what we serve (we do 40k), so if I had to make an educated guess, they do about 10x everything else as well, which means they probably have about 50,000 readers a day, that's probably a low guess. Actualy unique readers a day, I'd say, is around 60k, maybe a few more.What that means in comparison to a print magazine, I can't say for sure. My guess would be Slashdot has a larger audience (in their niche) than most print magazines.

>Want to compare the reach of the most widely-read blog with, say, the 1000th most widely-read magazine?Yes, yes I do.Slashdot, a popular blog (maybe not the most popular, but it's gotta be up there), gets (as of over a year ago) 80 million hits a monthBased on a chart from "membership representing roughly 85 percent of consumer magazine advertising volume in the U.S." the magazine with the 100th highest ad revenue (~popularity) so far this year is "Popular Science," which Ulrich's reports a circulation of 1,550,000 for.Given all that fuzzy math, the 1000th is going to have less of a reach than the #1 blog.I don't publish for revenue, so I don't see how that's relevant. :)Someone needs a taste of John Stewart....

Big if. Want to compare the reach of the most widely-read blog with, say, the 1000th most widely-read magazine? Or, even better, the revenue of the 5th most widely-read blog with the revenue of the 500th most widely-read magazine?

Somehow, "old-fashioned magazines" have no more withered up and died than have old-fashioned books.

The analogy between encyclopedias (and other reference works for which bound books were never an ideal form) and other print media is, to be blunt, horsepucky.

Well, our newest hottest selling item is the "Mrs. Carver" Tee, outselling all our old stuff.

Carr's piece is a really good read. He notes: "promoters of Web 2.0 venerate the amateur and distrust the professional." Not an original idea, but still an interesting one. I'm not sure I buy it, especially the way he explains it.It's good to see some interesting critical writing. It's important we remember the limits of the things we praise.The quote that relates most to libraries, and things I've been thinking of is this:" trumps quality all the time. So what happens to those poor saps who write encyclopedias for a living? They wither and die. The same thing happens when blogs and other free on-line content go up against old-fashioned newspapers and magazines."So, if that's true, what does that say for the future of libraries?

I have powers? Wow! Remind me to tell my wife tonite, maybe she'll be impressed :-)

I'm an authority? Where was I mentioned as an authority?

In fact, I made a bad comparison, given that the power law is much more active in weblogs than in magazines. I should have said, "compare the 20th most widely-read blog [that defines itself as a blog] with the 100th most widely-read magazine." At the very tippy top of the power law, some (few) blogs may approach magazine readership.

Of course, even the most widely-read magazine doesn't reach as many eyeballs as the most widely-viewed TV shows, but it reaches them in a different way.

Subscribe to Comments for "Wikipedia founder admits some quality-control problems"