On Tue, 7 Jun 2005, Matt Brown wrote:
On 6/7/05, Geoff Burling llywrch@agora.rdrop.com wrote:
And I'm sure that there are other issues one could discuss. However, if we could agree that published sources -- either primary or secondary -- can be cited, but unpublished works can not be, this would solve most of the problem.
That is a position that I could stand behind.
Well, I've convinced one more person than I thought I would. ;-)
I would be opposed to a suggestion that we should limit ourselves to published sources that can be acquired through Amazon, Google and the average inter-library loan service, however.
My example was intended to point out the contrast, not as a guideline. In fact, I'm amazed at how easy it is to find a title that is not listed at Amazon or Advanced Book Exchange -- a few of which I've consulted in my writings. If I had to insist on a limit, it would be only to those works listed in the catalogs of the Library of Congress & the British Library -- & maybe one or two other similar institutions. Allowing for duplicates, I estimate that gives us 50-60 million titles, most of which would be available thru the average inter-library loan service.
Obviously, the easier the sources can be accessed the better, but I would not rule out using rare or old books, magazines and newspapers even if not commonly archived, small publications, and foreign language works.
Honestly, if I were to critique a given article for its sources, I am not going to get upset if they include one or two sources out of a dozen that are hard to get ahold of; it's when the article only uses such sources, & I have suspicions about the accuracy of the article.
That said, I think that outside a certain vanishingly small subset of very controversial articles, Wikipedia's problem is not the citing of hard-to-verify sources, but the absense of sources at all. Outside the context of Israel vs. Palestine, certain crackpot science theories, and a few other controversial places, I don't see much of a problem at present.
And I'd rather someone cite an unpublished source than none. It's always possible that a better reference than that unpublished source may be findable by someone else, or a published source that references that unpublished one.
Either case would be troubling to me. Wikipedia is supposed to be no more than a secondary work, not a place to do publish research. With the exception of facts that are either banally obvious or commonplace (e.g. "France is a country in Europe"), if you can't find a published source to cite, then you should consider whether you are writing about something that is not part of Wikipedia's scope; much as that word is dispised here, the subject may not be notable.
(Of course, an objection to this would be the matter of contemporary pop culture: AFAIK, there are no books about subjects like the Lost television series, Ken Jennings, Pokemon trivia. However, these topics have enough of an audience that we should trust that the Wiki method will provide a means to keep bad information from entering the article.)