Thursday, January 17, 2008

Risks in using Wikipedia?

In our research on Wikipedia, we have been using a broad framework published as an one-page article in CACM (Communication of the ACM) in 2005 by Denning et al. on the perceived risks in using Wikipedia contents. The question before researchers is how to mitigate these risks while enabling a vibrant social community who wants to get together to build a encyclopedia and help each other obtain knowledge. As Denning's article mentions: "But will this process actually yield a reliable, authoritative reference encompassing the entire range of human knowledge?"

In the framework in thinking about answers to this question, Denning's article suggests that numerous risks that we should consider:

  • Accuracy: how can you be sure that the information in the article was actually accurate and not some misrepresentation of the fact?

  • Motives: how can you be sure of the motive of the editors were to present the facts and only the facts, and not opinions? For example, as we have discussed before, how can we be sure that editors to political candidate pages are not there to simply push their political agenda? For example, what does User:Jasper23's editing history tell you about his potential political positions?

  • Uncertain Expertise: How do we determine the expertise levels of the people who are editing in Wikipedia? Some appears to really know what they're talking--for example, User:BillCl was the top editor of the NASA page, and appears to have done a bunch of edits around aviation related topics. On the other hand, the top editor (Wasted Time R) of the Hillary Clinton page appears to be also to be big fans of Bruce Springsteen, Elton John, Dixie Chicks, Tony_Bennett, and other musicians. So it is less clear this user is an expert on political positions of Hillary Clinton as compared to other candidates.

  • Volatility: If the topic was just in the middle of a huge debate, then the content of the article could be really unsettled.

  • Coverage: Coverage of certain topics appears to be better in Wikipedia than others. What are the inclusion and exclusion standards is less than clear.

  • Sources: Many articles do not cite authoritative sources, so it is hard to trace and find out if the information is actually accurate.


The article then goes on to say that "[WP] cannot attain the status of a true encyclopedia without more formal content-inclusion and expert review procedures." Our WikiDashboard tool is precisely designed to help with collaborative review of Wikipedia editing history and patterns.

7 comments:

Moulton said...

Wikipedia cannot attain the status of a true encyclopedia without more formal content-inclusion and expert review procedures.

This condition is unlikely to arise within the peculiar governance model established over the years by the entrenched Wikipedia culture.

Credentialed reviewers who are subject-matter experts in their field are routinely marginalized and excluded under complex and erratic guidelines that disfavor authoritative scholars.

The favorite means of disempowerment is the RfC (Request for Comment), a bizarre regulatory process that would be better characterized as a Pythonesque RfSI (Request for Spammish Inquisition), wherein the chief weapons are Fear, Uncertainty, and Doubt (FUD).

Scholarly editors who naively deign to edit Wikipedia might not expect to become ensnared in a Spammish Inquisition, but experience with the site soon proves otherwise.

Jon Awbrey said...

I can do no better than refer you to a series of previous comments that I collected in this post at The Wikipedia Review.

Ed H. Chi said...

Moulton:

Indeed, we have read and been puzzled by what some have called an "expertise-bias" in Wikipedia, in that, some Wikipedians appears to shun experts in various domain areas. There is almost a distrust of knowledgeable folks. I can only remember this kind of distrust happening in history during the Cultural Revolution in China. I've found this rather disturbing. Why did this happen?

Ed H. Chi said...

Awbrey:

Thanks for pointing to the collection. Indeed, the question of whether participation in the whole Wikipedia process teaches something about scholarship is a good one. There are a lot of interesting "hype" about Wikipedia: some of which perhaps correct, and most of which perhaps incorrect. What our role, as researchers, should try to do is to document what is happening and learn from it.

That's why we're studying "social cognition". Like non-social cognition, I believe social cognition has pitfalls, including groupthink, prejudices, and bootstrap problems.

Moulton said...

Subject matter experts are liable to overthrow popular misconceptions — a practice that tends to make one singularly unpopular.

Wikipedia tends toward popular culture rather than cutting edge research. Indeed, a scholar who is a leading researcher in a field is likely to be disqualified on the grounds that original research is not allowed on Wikipedia.

Moulton said...

Yet another take:

"Wikipedia is in part built by harnessing popular envy and resentment against the status of academics." —Seth Finkelstein in the UK Register

Jon Awbrey said...

«Why did this happen?»

There are several streams of research that may help to give a clue to the answer.

In logic and philosophy, there is the study of the "inquiry" process, as carried on in the pragmatic tradition of Peirce, James, Dewey, and Mead. From a cognitive standpoint, inquiry encompasses everything from concept formation and hypothesis generation to critical reflection on practice. Inquiry begins with what Peirce called the "irritation of doubt" and it proceeds until it succeeds in a state of belief or certainty — or until it is waylaid by any number of blocks, caltrops, or short-circuits.

Sadly, all too sadly, the blocks are the way of Wikipedia as we see it today.

To be continued …