Wikipedia:Deletion reform/Proposals/Speedy delete for pages with no citations

This is a proposal originated by Bunchofgrapes regarding deletion reform. As of 22:21, 29 October 2005 (UTC) Bunchofgrapes is himself no longer a supporter of this proposal.

Speedy delete for new pages without citations

I believe there is consensus that citations should be provided for all facts in Wikipedia. Without citations, showing something to be non-verifiable is usually impossible (a negative proof) so it is reasonable for there to be a strong burden on editors to point the way to verification sources. In practice I doubt this is enforceable in the general case. One reason for this is because many edits fall into a gray area between adding new facts or simply copyediting or wikifying. Any policies that attempted to say "all factual edits without citations will be reverted" fail due to this gray area, since you'd then need a process to arbitrate whether any given edit made factual claims or not. There is, however, at least one form of edit that never falls into this gray area: the creation of a new page.

In almost every case, new pages must contain factual content to be useful. (Disambiguation pages are an exception and there may be some others.) Therefore, new pages should always contain citations. These citations might be footnotes, references, external links, or, in some limited cases (pages about movies and books, perhaps) just implicit citations against the topic at hand.

My proposal is this:

Add the following as a criterion for speedy deletion: "New articles with no citations or references to external sources, excepting disambiguation pages or other pages of a similar non-factual nature. A "new article" for this purpose is one created within the last three days."

Also, modify the "cite your sources" text in the edit page for new pages to indicate that pages without citations may be deleted.

Benefits and risks

• Fewer new articles needing to go through the full Afd process. This improves the overall quality of the Afd Process.
  • Not really. The AfD editors are often able to improve articles just because those who want them kept will want to improve the articles. If you just have those pages speedy deleted instead, then that prevents all those sets of eyes from seeing that article. --Idont Havaname 19:16, 9 September 2005 (UTC)
• Raising the bar a little for creating new pages. This might be a benefit or a problem, depending on your point of view. I think it is a good thing. One thing that has struck me as I try to do a little New Article Patrol is how much easier it is to create a new article than to nominate one to be deleted. Requiring new articles to provide at least one citation restores a little balance to that equation.
  • It will be a problem. Some topics aren't necessarily that able to be cited, but are nonetheless encyclopedic. Also, many great articles here began as stubs and substubs. --Idont Havaname 19:16, 9 September 2005 (UTC)
• New pages without citations aren't required to be speedily deleted under this proposal, so editors acting in good faith who notice an otherwise-good new article can add a reference or two rather then submitting it for speedy deletion. Common sense still applies.
  • You can't really assume that. There are admins here who are quite deletionist and will exercise that power. All in all, while I think this proposal may benefit the encyclopedia from a researcher's standpoint, there are still (as I said above) great articles that began as stubs, and some people who are in the business of stub expansions (like I am) don't know the topic even exists before they research it and expand it. --Idont Havaname 19:16, 9 September 2005 (UTC)
    • I'm not against stubs. Tell me about these sourceless stubs. Are they just placeholders, or do they contain factual material? If the former, why not just leave links to uncreated pages? If the latter, is it really too much to cite just one source (of any sort) for the page? The proposal isn't after meticulous sourcing for every fact; just a kind of token effort to point the way. Above, you say "Some topics aren't necessarily that able to be cited, but are nonetheless encyclopedic." I thought verifiability was a requirement for Wikipedia material? Aren't verifiability and citing inseparable? Can you give me an example of the kind of page you are thinking of? Bunchofgrapes 20:23, 9 September 2005 (UTC)

      • Matt Darey (was previously only informally cited), Walter Young (wasn't cited), Antestor (wasn't cited), Inaccessible Island (wasn't cited, and I didn't know it existed until I found the article on Special:Random one day), and Al Denson (wasn't cited and was tagged for a speedy before I saw somebody on my watchlist asking another user about it) are a few of the ones I've been working on since I've been here. (I'm not trying to show off my contributions here; these are just the examples that came to mind because collecting the information was a lot of work.) As for your question about verifiability, there are many articles that go to AfD because their verifiability is in question. On most AfD's where the information looks factual, there is usually some voter who will do the homework, expand the article, and add the external links as necessary. That's one of the nicer things about AfD as it is right now; it occasionally takes a poorly-done article, brings it to the community's attention, and gets expanded. Also, until the speedy deletion criteria were last amended, articles could be speedied just for having the same title as an article that had previously been deleted, so if somebody took an article with no citations, had it deleted, and then reposted the article with citations it still might be deleted anyway, regardless of whether or not the revised article would pass AfD as it stood. It's also quite a headache to expand CSD; with the CSD proposal I linked to earlier, there were many new criteria proposed but very few that were actually added to WP:CSD. I'm not sure that this would get a lot of traction among other users. --Idont Havaname 23:54, 9 September 2005 (UTC)

        • In each of your examples, it seems trivial to discover a citation or two (and the articles now have plenty). I was asking for an example of an encyclopedic article which nonetheless is isn't able to have citations. I don't think such a thing exists. Where we seem to disagree is on the acceptable citation burden placed upon page creators. Typical Deletionist vs. Inclusionist head-butting I guess. :-) No doubt you are correct that there doesn't exist a consensus to change the status quo in this area.Bunchofgrapes 00:06, 10 September 2005 (UTC)
          • A lot of our pages that start as (sub)stubs are created by anonymous editors who just want to create a page, and they frequently don't know much about our policies... yet new article patrollers are not perfect, so sometimes these articles don't get any improvement until somebody clicks on Special:Random or sees it in AfD. I was just somewhat concerned that given this new proposed CSD, the admins who work on speedy deletion would just delete it no questions asked (they usually have quite a workload and not much time to look up articles in the speedy deletion queue). (and a side note - no need to reply to this) Also, if you're even more deletionist than I am, stand and be counted! :-) --Idont Havaname 01:13, 10 September 2005 (UTC)

Discussion

I would strongly oppose this proposal. Many articles start out with limited or no sources cited, and souces are eventually added. This is what we have the {{unreferenced}} tag for. if you find an article with too few sources, insert that tag -- if none are addded in a month or two, propose it on AfD. I woulkd maybe suypport a proposal that if an articel has has the {{unreferenced}} for a sizable period, say a month, and no references have been added, it sould be speedy deleted. For an example look at the article Thomas Shipp . I found this on new page patrol as a fairly bare stub. I added a good deal of content and several external links. Another editor felt it was not referenced enough and asked me to improve it. It is now very throughly refernced. That is how wikipedia grows and improves. If your proposal had been in place, it might well have been deleted before I ever saw it. DES ^(talk) 18:03, 12 September 2005 (UTC)

• I also strongly oppose. Yes, articles should have references. But most articles don't have references. Half of wikipedia could be speedied if this became policy. It's like using {{delete}} instead of {{cleanup}}. ··gracefool |☺ 18:15, 25 September 2005 (UTC)
  • The proposal specifically only applied to newly-created articles. Bunchofgrapes 22:34, 25 September 2005 (UTC)

I think this proposal is a fantastic idea, and fully endorse it. Sadly, this page has no citations, so I'm afraid we will have to speedy delete it with greatest haste. -Silence 04:54, 11 October 2005 (UTC)

Hee Hee, good one. The word "article", to me, has a specific meaning, i.e., pages in the main article namespace. So, sadly, it doesn't apply. Bunchofgrapes (talk) 14:55, 11 October 2005 (UTC)

Opposed. I am sure there are some people who love searching out references and adding them to a text. But there are lots of people who know something about a topic but could not give a reference. This proposal is saying they could not start an article despite being knowledgeable on the subject, just because they have no references handy. I know perfectly well that people may be willing to sit at a computer and type something they know about, but would not be willing to go out and find references. Articles are needed in an encyclopedia. They may be better articles if they have references, but articles without references are very much better than none.Sandpiper 11:55, 19 October 2005 (UTC)

Opposed. For me all this proposal does is bring up larger problems with the deletion policy. Deletion has gained too much importance through the technical fact that the history is lost. If the history could be preserved, then we wouldn't need voting or admins to be involved because un-deleting would be available to all editors, same as any other revert now. If that can be changed, then there's not need to speedy deletion non-sourced articles, just put a note on the talk page of why and delete it. It can then be un-deleted and sources added.

Opposed. There is stuff like Common knowledge which doesn't need to be cited. Under your proposal, if someone were to start a stub on the United States of America and say "It is a country on the North American continent, and it has 50 states," it would have to be deleted since they did not cite where they got their information. It can end up with some good articles being deleted by users who may written an article on something that they love very much, and have spent many years doing much research on, and may simply not be able to go and find where they got all their facts from. What about articles that are about a song. Describing the structure of a song can't possibly be cited (you could in theory, cite the song but that would be repititiously redundant) unless you got it from another source. But anyone knowledgable in music can explain the structure of a song. --THollan 14:34, 27 October 2005 (UTC)

Strongly opposed. The proposal is contrary the "Assume good faith" guideline. This fundamental wiki guideline applies to the general behaviour of a user, and so it applies as well to the act of creating a new article. When someone creates an article, whether with references or not, we should first assume that the user acted in good faith and that the content is correct or likely to be correct, unless of course anyone can see in a direct and unambiguous way that the content is not correct, or patent nonsense. Adding references to an article is truly important, but IMHO speedying new article without reference would amount to assuming bad faith and would often bite the newcomers. --Edcolins 11:33, 29 October 2005 (UTC)

Well, that's the straw that breaks this camel's back. As time has gone by, I have liked my proposal less and less, and this argument completely convinces me. I've noted at the top of the page that I no longer support it. —Bunchofgrapes (talk) 22:21, 29 October 2005 (UTC)

Oppose, this proposal will require all lists to become awkward and twice as big as each item is matched to a citation. Endomion 06:40, 25 December 2005 (UTC)