On 5/18/06, Steve Bennett stevagewp@gmail.com wrote:
On 5/18/06, Anthony DiPierro wikilegal@inbox.org wrote:
Nowadays you don't even have to go to a library to look up journal articles. Wikipedians must be getting the information they put into articles from somewhere, and I find it hard to believe that more than a miniscule portion of it is straight from their memory.
I know heaps of stuff that isn't in Wikipedia. And I didn't get it from reading "reliable sources". To take another example, [[Bouchon]]. Any person who has spent more than a couple of days in Lyon knows what a bouchon is (a typically Lyonnais restaurant). Yet Wikipedia had nothing on them. I didn't feel like I was going out on a limb creating a stub about a bouchon saying that they serve very meaty dishes that many foreigners wouldn't necessarily appreciate. Of course for the details on exactly what they do and don't serve, I went to a book.
I'm offering this account not as what should be done, but what I believe is current practice - but the use of a book is even more stringent than most, I suspect.
Maybe my estimate is wrong, in which case this would be a lot more work than I suspect. But then again, it wasn't that hard for me to find a few decent sources to use when writing about bouchons, using Google books: The Concise Larousse Gastronomique: The World's Greatest Cookery Encyclopedia has a short entry, and France by Lonely Planet Publications has half a page. That should be enough for a stub.
Now what if the software didn't let you create a stub right off the bat? What if you had to list your sources first, then take info from those sources, *then* you could start a stub? I don't think writing an article would be incredibly harder, it'd just be radically different. Getting away with plagiarism would be a lot harder too, probably harder than just writing a decent article yourself.
This isn't to say that I think *Wikipedia* should work that way. After all, Wikipedia has a long history of running completely different from that, and a relatively successful history too (certainly volume-wise, and the average quality isn't too bad). What I'm saying is that providing enough sources to write a decent encyclopedia article isn't very hard for any but the most obscure subjects, whether you've been to university or not. Providing sources for an encyclopedia article that's already written, on the other hand - that can be hard or nearly impossible (or even impossible if the article is unverifiable).
Anyway, to bring this out of theory and back to reality, is there anything I can do with those two references I found on bouchon? Put them in the ==References== section, even though no one actually used the source? Put them on the talk page? Maybe we need a "notes" page which can be organized a little bit better than the talk page.
The interpretation of WP:V that says that any information that is not accompanied by a citation should never enter Wikipedia is, IMHO, novel. I would have nothing against that becoming the official interpretation, but it's not even close to being the dominant one at the moment.
I don't interpret WP:V that way. The way I interpret WP:V (as of a few months ago), it means that a citation should be easily added to any information upon request, which in turn means that anything not cited, at least in hidden comments, is subject to potential removal. What I'd like to see Wikipedia move toward is a situation where all the information is sourced somewhere, though for practicality purposes the details would probably not be in the article itself (just a list of sources in ==References== is acceptable for unextraordinary claims).
But let me be clear that I don't think this can be achieved simply by a change of policy. I'm not even sure there is a critical mass of Wikipedians that want this in order for it to happen.
The ad hoc system in place now is completely backwards. You're
supposed to get your sources first, *then* write the article. Believe it or not I'm completely in agreement with Jimbo that unsourced material should not be in Wikipedia articles. But just telling people to do a better job or "be kicked out of the project just for being lousy writers" is not a very productive way of achieving that.
If we don't want unsourced material, why have we tolerated it so long?
I never said "we" don't want unsourced material. Jimbo said *he* didn't want it. And I said I agreed: I don't think there should be unsourced material *in Wikipedia articles*. If you want to create a stub from your memory, I think it should go on the talk page (yeah, even if there isn't anything on the article page).
But I'm not claiming to speak for all Wikipedians, just giving my opinion.
Why have Wikipedians tolerated unsourced material so long? Because not tolerating unsourced material is *exceedingly* anti-wiki, and Wikipedia evolved from a wiki.
Anthony