Universities Appraise WP for Their Students

[Randy from Boise]

Browsing the internets this morning I bumped into what I feel is a fairly levelheaded review of the plusses and minuses of Wikipedia on the website of Cornell University (T-H-L).

linkhttps://digitalliteracy.cornell.edu/tut ... l3222.html[/link]

It occurs to me that a thread collecting various university policies and official essays about Wikipedia and its use might make a useful thread here.

Here are some excerpts from the Cornell page:

As a tool for scholarly research, Wikipedia can be either a grade-killer or a valuable friend, depending on who you ask and what you hope to accomplish using it. What is fairly certain is that your professor won't let you cite it in a scholarly research paper.

There are a few common reasons why you can’t site Wikipedia:

* Wikipedia is a general encyclopedia. At the collegiate or university level, your professors are looking for more than general rudimentary material. General encyclopedias usually give baseline information, the type of common knowledge that isn't usually cited. Academic subject-specific encyclopedias will often provide more scholarly and citeable information.

* There is often no way to know who is editing the entries in Wikipedia or what his or her level of expertise is.

* You cannot be sure that the content is “permanent” (although you can look at the revision history on the History page).

* You cannot be sure that the content meets standards of academic rigor. One of Wikipedia’s main principles is that it strives for a neutral point of view (which it abbreviates to NPOV). This standard states that all articles should strive to “represent…all significant views on each topic fairly, proportionately, and without bias.” The problem is that in any knowledge endeavor, much less a collaborative and ad hoc venture like Wikipedia, deciding what's neutral and having something reviewed for NPOV can be controversial undertakings and too uncertain to meet standards of academic rigor. However, having such a debate take place publicly on Wikipedia makes for interesting talk-page reading and for a good pros-and-cons debate.

Arguments in the site's favor included:

Use Wikipedia to get a general overview, and follow the references it provides as far as they can take you.

Look at the Discussion tab to see if the article you’re reading is part of a WikiProject, meaning that a group of people who care about the subject area are working in concert on its content. They may not be experts on the subject, but signing onto a WikiProject implies a writer has more than a casual interest in it.

If it is part of a WikiProject, see if it has been rated. Articles in WikiProjects go through a type of peer review. This is not the same type of peer review your professor talks about regarding scholarly research, but even such a limited review does at least imply that someone from the WikiProject has looked at the article at some point and assigned a quality rating to it. In any case, to be fairly sure that a Wikipedia article expresses what laypeople might need to know to consider themselves reasonably informed, look for a rating of B/A or above.

I believe the third paragraph here is erroneous — ratings don't imply independent review up to the level of B (outside of military history); anybody, including a page's creator, can assign a page to projects and estimate quality ratings — it's actually desirable for this to happen, since it alerts others to the page's existence and makes future improvement more likely.

RfB

[Kingsindian]

The quoted parts seem decent.

I generally believe that if Wikipedia had a much more open editing system and, simultaneously, a much more robust "rating" system, it would work much better. Ideally one wants ratings from as many viewpoints as possible. Obviously the reputation and past record of the "rating agencies" would matter here as well. Forks of articles can also help to provide different viewpoints, elaborations and so on.

[Randy from Boise]

I'm using WP with confidence in association with a book project, but I'm just chasing things like birth and death dates and general facts about a person's career for one or two line footnotes.

I think Wikipedia becomes more useful if a person knows how to read edit histories and to compare diffs. This strikes me as a basic skill that could be taught in ten or fifteen minutes in a classroom. I think the chances of accidentally absorbing vandalism from a bad Wikipedia entry can be reduced by about 90% if a person just had a grasp of using edit history and taking a look at earlier versions if something is in doubt. It also helps to have an "inside baseball" knowledge of the players, but that is something that can't be taught beyond "IP Editors are high risk."

Of course, for highly controversial topics I wouldn't use Wikipedia at all, unless it was to answer a question like "What day did the 1967 war start?" (June 5)

RfB

[Randy from Boise]

The quoted parts seem decent.

I generally believe that if Wikipedia had a much more open editing system and, simultaneously, a much more robust "rating" system, it would work much better. Ideally one wants ratings from as many viewpoints as possible. Obviously the reputation and past record of the "rating agencies" would matter here as well. Forks of articles can also help to provide different viewpoints, elaborations and so on.

WMF was fucking around with a user rating system for a while, or pretending to, but it turned out to be just a scam to trick people into making a first edit... They seemed to think that a person adding a comma or removing an extra space would suddenly transform people into Wikipedians, which is akin to thinking that washing a hubcap on a vintage Volvo will turn a person into an interstate truck driver. I hope that kind of magical thinking is gone now. It seems like it has dissipated over the last five years at least...

So what would a good ratings system look like?

RfB

[Kingsindian]

I think a good rating system would have several characteristics. I don't know if they are achievable on Wikipedia:

(a) Ratings should have a coherent essay attached to explain the rating. Kind of like a movie critic review.
(b) The "rating agency" (probably a Wikiproject) would have records and reputations so that people can trust them (or not).
(c) Multiple rating agencies to give different viewpoints on, say, controversial stuff.
(d) Periodic review of ratings.
(e) One could also solicit outside expert opinions on particular matters.

Some of this is already done on Wikipedia, but half-heartedly and inconsistently. Anthony's efforts detailed in another thread go somewhat in this direction.

This is offtopic, but I have read some of the weekly updates on the Debs project. One of the things I remember, a link to David A. Shannon, Eugene V. Debs: Conservative Labor Editor.

[thekohser]

There are a few common reasons why you can’t site Wikipedia

Oh, Cornell...

That review has been around since at least 2010.

[thekohser]

I think the chances of accidentally absorbing vandalism from a bad Wikipedia entry can be reduced by about 90% if a person just had a grasp of using edit history and taking a look at earlier versions if something is in doubt.

So, if you looked at Inflammation on this day, which aspect of the edit history at that time would have caused you to doubt that rhyolite is a factor in the inflammation process? Because none of the previous nine edits, nor any of the 125,000 page views in the interim, seemed to prevent the accidental absorption of that vandalism.

[Poetlister]

Presumably, Cornell abides by WP:AGF and assumes that nobody would deliberately and knowingly add incorrect information.

[Wonderer]

For what it's worth, I think Wikipedia is a terrible place to assume good faith, because you have no idea what anyone's motivation might be.

To continue with the medical example: A page on inflammation on a pharmaceutical company website might be motivated by a potential increase of sales of their NSAIDs. A page on that same topic on a "holistic" healing website might be motivated by a potential decrease of big pharma NSAIDs. A small practice doctor might have a page on the topic just to show he knows what he's talking about.

On Wikipedia you can't even be sure of the motivation of people who deliberately insert false information. Are they trying to increase awareness of the fallacies of Wikipedia, and will they follow up months down the line, like our own Gregory Kohs? Or will they drunkenly insert the false information one night and then forget to follow up?

[Randy from Boise]

I think a good rating system would have several characteristics. I don't know if they are achievable on Wikipedia:

(a) Ratings should have a coherent essay attached to explain the rating. Kind of like a movie critic review.
(b) The "rating agency" (probably a Wikiproject) would have records and reputations so that people can trust them (or not).
(c) Multiple rating agencies to give different viewpoints on, say, controversial stuff.
(d) Periodic review of ratings.
(e) One could also solicit outside expert opinions on particular matters.

Some of this is already done on Wikipedia, but half-heartedly and inconsistently. Anthony's efforts detailed in another thread go somewhat in this direction.

This is offtopic, but I have read some of the weekly updates on the Debs project. One of the things I remember, a link to David A. Shannon, Eugene V. Debs: Conservative Labor Editor.

Yeah, I need to do an update of that. I'm just finishing up the first, tight proofreading of the 260K words of Debs content of volume 1, two weeks until that deadline. Then we get to turn around and do it all again for a second volume of early material, due the first of the year...

t

[thekohser]

Are they trying to increase awareness of the fallacies of Wikipedia, and will they follow up months down the line, like our own Gregory Kohs?

If you're suggesting an honor like "Most Conscientious Wikipedia Vandal", I'll be happy to accept.

[Bezdomni]

I think a good rating system would have several characteristics. I don't know if they are achievable on Wikipedia:

(a) Ratings should have a coherent essay attached to explain the rating. Kind of like a movie critic review.
(b) The "rating agency" (probably a Wikiproject) would have records and reputations so that people can trust them (or not).
(c) Multiple rating agencies to give different viewpoints on, say, controversial stuff.
(d) Periodic review of ratings.
(e) One could also solicit outside expert opinions on particular matters.

[...]

Moody's should get to work on mediating info-box scores straightaway so the agents de notation can get grading (two bits a mark).

This is offtopic, but I have read some of the weekly updates on the Debs project. One of the things I remember, a link to David A. Shannon, Eugene V. Debs: Conservative Labor Editor.

Hm... I should look at the dates on that again. But the NW territory states (Brexit States) and Indiana (in particular) were still wild wild northwests back then: Indiana History > The Golden Era (1900-1941). I've never seen such a title for the D.R. Stephenson and the Klan story.

This Jacobin article on DuBois & Debs may be of interest.

Back on topic:

Here's some people from Western Caroline U., CUNY, MIT, & UC San Diego writing for a journal at Ole Miss. I wonder if that place will accept anonymous collaborative submissions from folks not necessarily seeking a publication to add to their CVs? Their first author, like, ever, was A. Virtue... (!)