Search

[Tlogmer]

I wrote an article called setting the record straight about wikipedia (yes, this is another post to pimp out my blog).

But also -- what do people think about wikipedia out in places it might actually be used (like schools, etc.)?  Sometimes I think there's not enough information transferred between insiders who inhabit wikipedia and the public who reads it.  (This is bad for the insiders, too, because they're cut off in their own world.)

Er.  And could someone submit the article to digg?  Because I'm terrible -- fucking, fucking terrible -- at making digg posts.

[SpacePope]

Enjoy.

[Disco Stu]

I find Wikipedia useful as a first point of call for any research project. It gives a good background to most issues, and more importantly, almost always gives links to other sources which gives a more in-depth coverage of the subject. Any discrepancies should become apparent as you do more digging into the topic.

[Onebrain]

About half a dozen of the eighty or so undergraduate essays I marked last year tried to cite Wikipedia. It's become very pervasive, and I don't regard it as acceptable as a source because of the lack of authorial credit (which is very important) and the fact that there is no quality filter. I don't object to it as a starting point - when dealt with cautiously - but I don't accept it as a source in an academic essay.

The interesting issue is over what will happen in the next few years. This year, our department switched textbooks to a version available in hard copy and in a very nice online version (password access required). Since this was a specialist text in a similar format to Wikipedia, written entirely by experts and without regular changes to article content, it was a much more reliable resource and so more people used it. I wonder if, as textbooks shift increasingly online, whether the use of Wikipedia by students will decline somewhat.

I'm ambivalent on the subject of Wikipedia in schools because at that stage I'm not sure if the students are able to identify subtle vandalism as well as they should. I have taught pre-university students and been presented with some howlers in the past culled from wikipedia. I'd be more confident of them using Wikipedia if articles had a kind of 'health warning' that was easy to understand.

[Johnny Roastbeef]

One problem that I think wikipedia has in terms of acceptance by schools is their "No Self Censorship" policy.  What it results in are visual aides that you would not find in any other academic resource. 

Some NSFW examples.
Mammary Intercourse
Ejaculation, containing pictures and video of men ejaculating
Fellatio
Encyclopedia Dramatica (also very NSFW) has an extensive list

These aren't exactly medical diagrams.  And censorship or not, I think a good many parents of school children find this kind of material objectionable.  The article for "vagina" has a close up of a woman's vagina in a pose that could have been taken directly from porn.  While I can appreciate that the intent of inclusion is different, what parents and schools see is that the outcome is the same. 

[boonmeister]

Not totally in the vein of this thread but here is an interesting article about a law suit brought against wikipedia.

Tlogmer: Have you read much of The Register's coverage of Wikipedia? Admittedly they seem to be pretty anti the people who run the project but they bring up some good points. This is an interesting article that provides some insight into both Wikipedia and The Register's attitude towards it.

-------------------------------------------------------------------------------------------

As to the use of Wikipedia in schools it should be made clear to students that Wikipedia should never be used as anything more than a starting point for your studies. Sure, when I want to check out a few "facts" about late eighteenth/early nineteenth century warships (don't ask) I don't mind that there may be errors in Wikipedia's articles, but its convenience outweighs that issue for my purpose. If I was going to write a book (or even a paper if was back in education) I'd look for more credible sources, but I certainly wouldn't be above using Wikipedia as a starting point.

[jryan]

Another sobering exercise with Wikipedia is to go to a somewhat technical article and read the discussion sections.  It is amazing to see how often trained adult scientists get in bitch sessions over conflicting edits.

[Johnny Roastbeef]

Another sobering exercise with Wikipedia is to go to a somewhat technical article and read the discussion sedctions.  It is amazing to see how often trained adult scientists get in bitch sessions over conflicting edits.

I find it even more amazing to see how often high school kids with no expertise on the subject get treated with the same deference as those scientists.

[Greywar]

Another sobering exercise with Wikipedia is to go to a somewhat technical article and read the discussion sedctions.  It is amazing to see how often trained adult scientists get in bitch sessions over conflicting edits.

I find it even more amazing to see how often high school kids with no expertise on the subject get treated with the same deference as those scientists.

I only find that disturbing if the scientists can't muster arguments persuasive enough to win over the kids.

I find wikipedia to be amazingly reliable for some technical issues but for anything to do with historical events or politics the founders views (and thus the "insiders") views seem to sway the edits pretty dramatically.

[jryan]

Another sobering exercise with Wikipedia is to go to a somewhat technical article and read the discussion sedctions.  It is amazing to see how often trained adult scientists get in bitch sessions over conflicting edits.

I find it even more amazing to see how often high school kids with no expertise on the subject get treated with the same deference as those scientists.

That's true.  Though I find that in most cases where the argument is that lopsided, the immunization from idiocy eventually kicks in.  I was more talking about how silly these competing experts can look when they start deleting each others contributions to an article.

[Tlogmer]

Have you read much of The Register's coverage of Wikipedia?

Argh.  I hate the register's wikipedia coverage.  They hone in on messed-up stories and use them to draw conclusions about all of wikipedia.  That's like deciding chicago is terrible based on the fact that someone got hit by a car there.  This is the best article about wikipedia ever written.

Onebrain: I'd be interested to see some examples of papers citing ridiculous misinformation.  (I continually forget how stupid teenagers are.)

I don't regard it as acceptable as a source because of the lack of authorial credit (which is very important)

Hm.  Do you know about the history tab (showing all authors and revisions) and the "cite this page" link at the bottom of the sidebar (which creates a static link to the versions you're reading)?

[Johnny Roastbeef]

I don't regard it as acceptable as a source because of the lack of authorial credit (which is very important)

Hm.  Do you know about the history tab (showing all authors and revisions) and the "cite this page" link at the bottom of the sidebar (which creates a static link to the versions you're reading)?

I don't really see how that changes anything.  This may not be what onebrain was going for, but to me, authorial credit is important for credibility.  All you get from the history page is usernames like Zorro19367 and Spechter965.  There's no authority added to the text due to the lack of credentials associated with it. 

Short of spending months at wikipedia to learn which editors know their shit, it's not a practical method.

[Tlogmer]

Yeah, I think that's reasonable (you'd be surprised how many academics don't know about the revision history, though, so I thought I'd throw it out).

Google's Knol project -- intended to be a wikipedia for experts -- has a seamless identity verification system (you click a button, enter your phone number, get a call), but they messed up the project in so many other ways that it probably won't take off.

I've sometimes thought about starting a database matching verified college professors with their reviews of wikipedia articles.

[Onebrain]

Onebrain: I'd be interested to see some examples of papers citing ridiculous misinformation.  (I continually forget how stupid teenagers are.)

Mostly they involve anonymous trolls inserting false facts or amending dates (e.g. 1812 becomes 1912). These edits are changed quickly but junior students aren't aware of how to evaluate wiki version histories so on vandal heavy days often pull up incorrect information and then never re-check information.

I don't regard it as acceptable as a source because of the lack of authorial credit (which is very important)

Hm.  Do you know about the history tab (showing all authors and revisions) and the "cite this page" link at the bottom of the sidebar (which creates a static link to the versions you're reading)?

I actually spent more than a year editing wikipedia and so I'm familiar with these processes. However, authorial credit is important for identifying quality and the relationship between works - you want to know who wrote what, and when, and that is inherently difficult on wikipedia.

Credibility is also difficult to parcel out on wikipedia because of the changes that occur over time. For instance, the evolution article went through a mangled series of changes a while ago owing to the clumsy and inaccurate opening passages. The quality of the article degraded because nobody could work out how to integrate the new information. It eventually got done, but in the meantime the article looked as though a lot of it was influenced by a small number of respectable editors, while still being an awful article. At the time, knowing their credibility didn't matter because small, new additions were making the article worse.

[DevHyfes]

My favorite part of Wikipedia is picking some sort of ancient battle (Such as the Battle of Tours) and just clicking on links. It is amazing how many articles people create. Even better, historical experts have done a lot to fill in small definitions that you don't often see in regular encyclopedias. For example, Wikipedia has tended to have some great overviews of historical weapons, including pictures and details. Often these little points are missed in an encyclopedia where the experts had limited resources.

That said, I do have a couple of problems with it.

1) Specifically around historical research, there is often way, way too much import given to revisionist historians. You see a little of it in the Battle of Tours wikipedia article cited above. But if one author makes a striking historical claim, it can derail an entire article into a discussion of the pros and cons of both, giving an indication that there is heavy dispute when indeed the matter is generally settled.

2) Stubs: Often times people fill out stubs a little more- kind of drawing up a strawman article. These can be extremely inaccurate and a careless person can mistake them for a real article if they have trained themselves not to watch the top of the page for notifications.

3) Legitimacy: I find myself using wikipedia a lot more for informal cites, especially in forums such as HB. I think that there are appropriate times for this- such as when you are giving people a high level overview of a subject. Even when trying to buttress controversial claims, wikipedia can be useful because some articles have already aggregated quite a bit of research. So long as the person citing the article keeps these points in mind, Wikipedia can be very useful. But too often, people don't drill down the extra bit to verify the veracity of the supporting cites, and as a result they can find themselves relying on Wikipedia (not the underlying citations) as a source of truth. Wikipedia should be considered the Poor Man's Graduate Student. Often, professors writing papers have grad students go out and locate, organize and even summarize materials for research. Wikipedia is good for that, but like a good professor, you need to be reminded of its fallibility.

4) Slippery Slope: As Wikipedia articles start getting more play, it is extremely troubling to see them cited as authoritative. A recent blog posting on the libertarian law site, Volokh Conspiracy, notes that there are over 300 cites of Wikipedia by courts in the US. The post goes into depth about one, specific case where the court used a Wikipedia article to supplement or modify a Webster's Dictionary definition for "Wear and Tear". This becomes extremely troubling, because Wikipedia is not permanent and is prone to manipulation, especially for obscure articles that people don't often check.

5) Accuracy: Many people often talk about that Nature article (linked in the OP's original article), but it is rarely mentioned that Encyclopedia Britannica has released a 20 page rebuttal, mentioned here and available here. Further, it is important to note that the Nature article found Wikipedia to have 33% more inaccuracies than EB.

From the EB Response:

The “article” on “aldol reaction” that the journal sent its reviewer consisted
of passages taken selectively from two different Encyclopædia Britannica articles
and joined together with text evidently written by Nature’s editors.
[...]

Nature’s review of the Britannica article “Pythagorean theorem” claimed the Britannica misspelled
an Italian town that the reviewer said should be spelled “Crotona.” However, according
to the U.S. Board of Geographic Names, the preferred spelling is in fact “Crotone,” as
Britannica spelled it. Other reliable sources also give “Crotone” as the right spelling. For
Nature’s editors to have ruled this an error on one reviewer’s say-so, without confirming the
spelling, was inexcusable.

This is dangerous, because the Nature article has been used by many to support the impression that Wikipedia is as valid a source of knowledge as EB. Not only is this destructive to our overall intellectual capital, but it was catastrophic to EB's bottom line.

Uses in School:

i) I think a great class idea is to get kids to find a subject that hasn't been written about and to write their own article. Sure, the common stuff is already taken, but there are some more obscure or less consequential articles out there which still need work. Just look up the list of wikipedia stubs for details.

ii) As others have said, wikipedia is a great place to start your research, providing 80% of what you need to get an overview of your subject before going down any major paths.

[Tlogmer]

but it is rarely mentioned that Encyclopedia Britannica has released a 20 page rebuttal

You're right, that study kind of sucked as a rigorous comparison -- but it did demonstrate that scientists were pretty impressed with Wikipedia's quality.  I think it's a mistake to say Britannica went broke because of that study, though.  Britannica has always had financial problems, and it's been very slow to adapt to the internet.

[kelvinc]

Excellent article.  As someone that edits Wikipedia regularly, it's nice to see some balance in explaining how Wikipedia works.  What's weird is that Wikipedia is better than its most extreme naysayers say, but only because it's a lot less utopian than what its most extreme fans suggest.

I cite Wikipedia only in informal contexts, and I'd never submit something to a professor that used a Wikipedia article as a source.  The best use for it is as a aggregator of common information that would be available in other areas but dispersed.

The kicker is that I usually trim the hedges and revert vandalism and remove copyright violations, which require almost no technical expertise.  Most articles in my (supposed) area of expertise (economics) have been edited by people far more knowledgeable than myself on the subject, which is why I trust them pretty well.  It may also explain why some people are so skeptical of Wikipedia: it takes a certain amount of technical education to appreciate a well-written Wikipedia article, but the skills you've explained are sufficient to mark out complete bullocks.

[The Furlinator]

As a student myself, I know that some teachers at my high school are vehemently anti-Wikipedia, and some won't let us use the site for research even if we don't cite it in our paper. They give us the obvious explanation of "Anyone can edit it, so it's unreliable". Then something occurred to me. Sure, most anyone can edit an article if they want, but that doesn't mean they are going to. Because you know you can edit Wikipedia, are you going to? What loser goes "Hahahaha! I'm going to spend three hours changing every major fact in this article!!!" only to have it reversed ten seconds later.

Teachers at my high school make it seem like Wikipedia is some unorganized social get-together of people who think they know something.

Tlogmer,  your article should be standard reading for all teachers. Nice job.

[kelvinc]

What loser goes "Hahahaha! I'm going to spend three hours changing every major fact in this article!!!" only to have it reversed ten seconds later.

Oh you'd be surprised.

Also, it's not just the vandalism.  It's the crackpots and conspiracy theorists with a point to push: they're way more persistent.  Reversion is common, but only with article popularity.  And sometimes you'd have big ones that bomb: you'd be surprised how long the Gulf War article lasted with crap about Syrians and Jordanians fighting beside the Iraqis before someone (namely myself) caught it.
http://en.wikipedia.org/w/index.php?title=Gulf_War&diff=prev&oldid=193884995

[Tlogmer]

Well, it lasted a week.  That is long by vandalism standards.