Study of featured article quality, Facebook integrates Wikipedia pages, and more

This month's issue of the research journal First Monday contains an article entitled Evaluating quality control of Wikipedia's feature articles by David Lindsey (a student at Georgetown University and one of the two authors of the OnWikipedia blog mentioned previously in the Signpost). Based on the assessments of 22 featured articles by subject experts, the author concluded: "In expert evaluations, nearly one–third of the featured articles assessed were found to fail Wikipedia’s own featured article criteria."

In a discussion about the paper Shimgray analyzed the score that articles received in the evaluation versus the time that had passed since their FA promotion, and speculated that the low scores might be due to a decay in quality in "unmaintained" articles (rather than problems in the review process). Lindsey replied that drawing solid conclusions about this from the study was "absolutely impossible" because of the small sample size.

In a 19 April blog posting, social networking site Facebook announced the launch of what it calls "Community Pages":

"Community Pages are a new type of Facebook Page dedicated to a topic or experience that is owned collectively by the community connected to it. [....] Community Pages are still in beta, but our long-term goal is to make them the best collection of shared knowledge on a topic. We're starting by showing Wikipedia information, but we're also looking for people who are passionate about any of these topics to sign up to contribute to the Page."

The example depicted in the announcement shows the lede of Wikipedia's cooking article. On each Community Page, a "Wikipedia" tab links to the full text of the Wikipedia article (if it exists).

On the Foundation-l mailing list, the Wikimedia Foundation's Head of Business Development, Kul Wadhwa, announced the development to Wikipedians, adding:

"Our hope is that many Facebook users (if they are not already) will also be inclined to join the large community of Wikipedia contributors. Facebook will follow the free licenses (CC-BY-SA) and help us find more ways people can share knowledge. Furthermore, we will be looking at other ways that both parties can cooperate in the future."

Replying to concerns that Facebook would draw potential editors away from Wikipedia, Wadhwa said that "we did give this a lot of thought. Facebook wanted to do this anyway (and they could take the content as long as they follow the license(s)) but we thought that, in the end, it would be better if we work with them on this to influence them to do it in a positive [way]."

Facebook has already initiated over 6.5 million of these pages, according to TechCrunch, which observed that the new feature seems to be Facebook's reaction to the problem that the "Pages" feature, introduced last year for brands and celebrities to present information about themselves, was increasingly used by Facebook members to create unofficial pages, also about other topics like baseball or yoga. TechCrunch also noted that adding content to Community Pages is not yet possible.

To CNET ("Facebookipedia?"), Facebook appears to be "actively treading into a territory that few companies have explored other than Wikipedia and Google", in "one of the boldest steps that the social-networking site has taken toward, well, consuming your life".

In a blog post for the San Francisco Chronicle, Craig Newmark, the founder of Craigslist and a member of the Wikimedia advisory board, says he will be spending some time addressing complaints about biographies of living people. He writes:

I've volunteered to help out with short term problems, as part of my normal daily customer service work. (That is, it's part of what I feel is my personal public service mission. This is on my own initiative, not strictly as part of my role on the advisory board.) If you see a bio that's been attacked, please let me know, and normally I'll find a way to get it fixed, have already done so in the case of a sitting US Senator.

• The 11 April Bangalore meetup was covered by the two largest Indian daily newspapers, The Hindu (And now, Wikipedia India's new face) and The Times of India (Wikipedia cries for desi content), both focusing on new Wikimedia Foundation Board member Bishakha Datta. A video recording of the event (90 minutes, low quality) is available online.
• Vanity Fair examined the version history of the article about teenage singer Justin Bieber ("Wikipedia Editors and Justin Bieber Fans Battle for Control of Justin Bieber Wikipedia Page")
• U.S. newspaper St. Louis Post-Dispatch reported on recent hoax edits in the article about Grubville, Missouri, involving a nonexistent school and other false information.

← Previous "In the news"

Next "In the news" →

In this issue

19 April 2010 (all comments)

News and notes

In the news

WikiProject report

Features and admins

Arbitration report

+ Add a comment

Discuss this story

These comments are automatically transcluded from this article's talk page. To follow comments, add the page to your watchlist. If your comment has not appeared here, you can try purging the cache.

I think that Lindsey article is interesting, but has some major shortcomings. The biggest thing that stood out to me is the reliance on only one expert per article, meaning that each data point could be just as much a reflection of the reviewer's own standards and personal tastes. Amongst some academics, there is a tendency to rip any written work to shreds, just to show that you can do it. Other academics might be inclined to rate an article highly just because they are jazzed that someone else cares about the subject. This study has no way of controlling for reviewer bias.

I still think this study is useful, and given the nature of it, its a bit much to expect much more. Still, I think drawing sweeping conclusions from such a limited study is a mistake. Bonewah (talk) 18:13, 20 April 2010 (UTC)[reply]

Agreed, which is why I think it's great that the reviewer himself admitted as much. In terms of knowing that some older FAs would have trouble passing the current standards, I don't think that's news to anyone. Standards keep getting raised and there are always going to be some articles that are behind the curve. Matt Deres (talk) 22:00, 20 April 2010 (UTC)[reply]

Yes, the study did not examine the "age" of an article's featured status, which is essential for evaluating how well the featured article process is working right now. Sure, the Max Weber article may not meet the current featured article criteria, but it was promoted in 2006, so it's never been formally evaluated using the current, more rigorous, criteria. In my opinion, we've only had "serious" featured article criteria for about one year, when the "well-researched" requirement was added. It's good to see that none of the articles that were promoted in 2009 or later scored less than a 7 on the 10 point scale.

It's also nice to see that some older featured articles did relatively well in the study's review. Did they do well because they were in good condition when they were promoted to featured status? Or did they do well because they were improved over the years, after having reached featured status? The study doesn't tell us, but this information is essential for evaluating the featured article promotion process. Lindsey appears to have overlooked in this point, and so his "conclusions and recommendations" do not really follow from his findings. The study evaluated the current quality of a random selection of featured articles. But it did not examine the quality of promotion process; to do that, one would have to examine the quality of the version of the article that was actually promoted. Perhaps the results would be about the same, but I'm not sure that we should assume this.

But I suspect that the major recommendation of the study is correct: the process would greatly benefit from input by experts. How to do that is an open question. We at least have an example of how not to do it. —Kevin Myers 15:54, 21 April 2010 (UTC)[reply]

I have to agree with the assertion that time makes a difference. It is telling that none of the most recent FAs were considered of poor quality by experts. Also, that the Global Warming article, since it is contentious, has received more collaboration than most other articles; I think that it received a rating of 10 shows that our crazy little system of collaboration works quite well. The current FA process involves a lot of factual and referencing nit-picking, so I find a 2008 study on length as a criteria for FA to be less relevant to the current situation. The problem is, this site is so organic that research becomes out-of-date practically from the moment it is published. かんぱい！ Scapler (talk) 07:29, 22 April 2010 (UTC)[reply]

Interesting study, but I find the criticism that articles may be written at a high school/undergrad pretty worthless. The same can be said of most newspapers and magazines. Just because Wikipedia is not written in the utterly dull, repetitive, and jargon-filled language that many academics prefer is not an automatic negative; it's one of our strengths. Joshdboz (talk) 14:11, 26 April 2010 (UTC)[reply]