Wikipedia talk:Speedy deletion criterion for unsourced articles/Archive 2

Synopsis of arguments

Between this page, the archive and the discussion at WT:CSD, this proposal has accumulated over 400kb of commentary. The table below was created on 17 Nov 2006 to recap the arguments being made for and against this proposal. This is intended both as an aid to new participants in the debate (400kb of commentary is a lot to wade through) and to help limit the repetition of arguments.

Note: Since this table was first drafted, the original proposal has been forked into four separate proposals. Some of the arguments (for or against) may not apply to each specific proposal.

Please be concise and please don't change or delete an argument just because you don't agree with it. Rossami (talk) 22:00, 22 December 2006 (UTC)

Arguments in favor Unsourced articles are an urgent issue needing immediate resolution Unsourced articles jeopardize the credibility of the project Requests for sources often go unanswered Current processes default to keeping unsourced articles with no mechanism for mandating sources The threat of deletion will result in the sourcing of many articles This puts the burden of sourcing on the original contributor -it encourages editors to do the work of sourcing themselves, rather than expecting others to do it later. AFD is too slow or is ineffective at dealing with this problem We need to change the culture of Wikipedia in this area Topics which can be sourced can be easily undeleted It is a logical extension of the WP:V policy that states "The obligation to provide a reliable source lies with the editors wishing to include the material, not on those seeking to challenge and/or remove it."

Arguments opposed Unsourced content is a real but not urgent problem Deletion is too severe a remedy for the problem - lack of sources should be dealt with via cleanup "Speedy"-deletion does not put enough eyes on the problem Unsourceable or persistently unsourced articles can be deleted through the Prod or AFD processes - this new proposal is unnecessary Some users enjoy finding and adding academic sources, others find such work boring and will leave the project if forced to do such work Some topics are unnecessary to cite to a single source because they are widely known by any practitioner in the field If mistakes are made, undeletion is not easy and will confuse or drive away some proportion of new users Some articles will be deleted despite being sourced or sourceable, even with the best intentions Represents a negative shift in openness of Wikipedia's culture May produce inaccurate, careless, or false citations Will be used more aggressively than intended Encourages editors to demand work from other editors rather than encouraging them to do it themselves While the spirit of the plan is good, it is wrong to start with a CSD variant. It would be better to start with an AFD variant, as that will more effectively and less contentiously move the community culture toward sourcing. Incomplete articles are where new users start. Often new users won't feel comfortable editing a fully developed and referenced article.

Comments from Seraphimblade

Given all of the proposals and counterproposals I've seen above, I would tend to suggest the following:

• The edit page should be changed to reflect that Unsourced information may be removed or deleted without notice. The "removed" part of this is already policy in itself anyway-this policy is simply a logical extension, if the whole article's unsourced, its complete removal may be called for.

• If this were to be implemented, its initial duration before deletion should be 30-60 days. After "patrollers" have a chance to get set up and running in earnest, the time should be dropped to 14 days. This would allow time for that without a tremendous number of deletions at first. (This could be done by requiring the tag to be subst'd, much as with the prod tag, and changing its notice to "14 days" at 14 days before the expiration of the grace period.)

• This tag should not be allowed to be applied to any article created before the policy went into effect for a period of one year. Again, this would allow time to clear the backlog of {{unreferenced}}'s without a massive flurry of deletions. (Again, this could be solved in the tag itself-"This tag is invalid and may be removed if placed on an article created before xx November 2006, unless the tag was placed on or after xx November 2007.")

• The addition of any source, however questionable, would invalidate this tag-even if the source likely wouldn't qualify as reliable. (It might be a good idea to add that a clearly unrelated source, such as a cite to a web page on pit bulls in an article about moon rocks, wouldn't qualify, but any source that could be considered to have even the most marginal relation and even which would generally be considered highly unreliable would prevent deletion under this criteria).

• Like prod, anyone who can produce a source may have the article undeleted immediately if it's already been deleted under this. The deletion should also be "without prejudice", so that anyone providing a source may freely recreate.

Comments on previous objections:

• As to badlydrawnjeff's concerns that an admin might abuse this-an admin might abuse anything, they already have the delete button. If an admin is in the habit of making questionable deletions, that should be brought to an RfC or ArbCom-and deletion review exists for a reason. If an admin is constantly getting overturned there, that'll only bolster the case that they're misusing their authority. "An admin might abuse it" could also be used as an argument against-well, really against anything we allow admins to do. Blocking could be misused. Protection might be misused. Rollback might be misused. Hell, even editing could be misused, we'd stop vandalism 100% if we quit allowing anyone to do that! The point of having admins, is that they're people we trust not to do so-and who are accountable if they do. We should also make it clear that WP:SNOW and the like may never be applied to delete an article tagged with this before time's up.

• I don't understand any criticism that this comment is "too harsh"-the solution to this now (as in, "Prod: No sources") only requires 5 days before deletion, and doesn't specifically alert anyone that all that's needed is a source! This proposal would allow 14 days, and would allow those who are very good at finding sources to do so. It would remove clutter on prod (and allow prod patrollers to concentrate on those articles prodded for questionable notability and such).

• I don't think this is "biting the newbies", any more then userfying their vanity bio or speedying their company ad is. If what the newbie wants is to "put up whatever they want"-well, they mistyped the Anarchopedia address. If we don't provide teeth to policies, we may as well not have the policies-and this isn't even a very sharp pair of teeth for WP:V. But it's long past due that policy grew -some- kind of choppers, and this is a good middle ground between those who would allow "My Aunt Sally is really nice" as a full article, and those who would delete anything that comes in unsourced on sight.

To conclude: While I understand that some people want nothing (or nearly nothing) to be deleted, and others want anything that's not in line with every policy deleted the second it hits the server, it's about time we did some compromising. This proposal is an excellent compromise. Seraphimblade 04:30, 20 November 2006 (UTC)

Approximately 99.6% of Wikipedia is currently unsourced, by my rough estimate. Do you really think anyone wants to delete 99.6% of Wikipedia's current content? dryguy 05:53, 20 November 2006 (UTC)

Absolutely not (though I've heard figures ranging from 80% and upwards, I'm not honestly sure but there's certainly quite a bit). That's why I suggested that this tag shouldn't be applicable to current articles for a year. However, as the importance of sourcing begins to become clearer, I think we'll see more newer articles become sourced, and more older articles get sources added. I would propose a counter question-if there really is that much unsourced material (be it 80% or 99.6%), does that in itself not indicate a serious problem? Should that situation be allowed to remain? Seraphimblade 06:18, 20 November 2006 (UTC)

We can come back to it in a year; for now, I do not think that this should go beyond new articles, where we have some hope of reaching the creator; else, what's the point? Septentrionalis 18:54, 20 November 2006 (UTC)

That was exactly what I was proposing-if hypothetically this went into effect today, the tag couldn't be applied to an article created yesterday for a year. That's also why I proposed a longer "grace period" at the beginning, so that the "verifiability patrol", as such, has time to get set up and going with a longer limit at first. Seraphimblade 20:54, 20 November 2006 (UTC)

Not exactly; I would prefer both to limit this, if it went into effect today, to new articles; and I would much rather say nothing about future plans. Let's not plan until we see if this works acceptably even for new BLPs. Septentrionalis 21:03, 20 November 2006 (UTC)

Fair enough, I'd go for that. Seraphimblade 21:05, 20 November 2006 (UTC)

What counts as a source or reference in this proposal?

Higher up in the discussion Taxman says: "This involves nothing more than a single judgement call: is there a source or isn't there.". Sounds simple enough, but I still want to dispute it. What if the reference section contains things along the lines of one of these:

1. Recent research.
2. Recent research by Professor Doe at the University of Springfield
3. Recent research by Professor Doe at the University of Springfield, published in the letters section of The Journal of Everything, October 2005
4. A book
5. A book by Professor Doe at the University of Springfield
6. The book of everything by Professor Doe at the University of Springfield
7. A newspaper
8. The Chicago Tribune
9. The Chicago Tribune, September 26th, 2006
10. The Chicago Tribune, September 26th, 2006 Editorial

The way I see it, examples 1, 4 and 7 are not sources. There is no information to guide the user on the way. Something along the lines of examples 3, 5 and 10 are clearly sources, as it provides enough information so that people could check them with reasonable ease. But what about examples 2, 5, 8, and 9? They are clearly lacking significant information, but would they count as a source/reference for the purpose of avoiding this speedy deletion criterion?

My point is that saying that we can easily discriminate between those articles having and those not having references is not as black and white as it might sound at first. Saying, "if there are any references at all, this CSD is not applicable" becomes nonsensical if "a book" is "accepted" as the sole reference, i.e. not listed for speedy deletion. People seeing "A book" listed as the sole reference would probably use common sense and book the article on the 14-day countdown to deletion, and, for the record, I wouldn't blame them. But where do we draw the line? As long as we have a gray-zone here we risk getting arguments over those who think "The Chicago Tribune, September 26th" is adequate and those who think it is not. Sjakkalle (Check!) 15:39, 20 November 2006 (UTC)

And is a book a source in an article on itself?

I just deprodded an article on a children's book by Gahan Wilson which had been prodded essentially on these grounds. (I also took the small trouble of looking up the ISBN, which neither the original author nor the prodder had thought to do.) Septentrionalis 19:03, 20 November 2006 (UTC)

I think this should be similar to the "non-notability" speedy deletion criteria-any assertion of notability under those is enough to avoid speedy, even if it's highly questionable. (Of course, if the assertion is dubious enough, it's unlikely to survive AfD.) That is as it should be-any doubt whatsoever should be left to consensus to resolve. Similarly, here, any citation of even a marginally-relevant source should require a full discussion at AfD on the merits if deletion is requested. Only clearly and totally sourceless articles should be subject to this. Seraphimblade 20:58, 20 November 2006 (UTC)

But in practice I've seen admins disregard claims of notability that look dubious, even though the wording does not distinguish between hoaxes, minor claims, and valid claims. There's a reason people fear scope creep if this becomes a CSD. -- nae'blis 21:00, 20 November 2006 (UTC)

If admins are abusing their powers, there are procedures in place to deal with that. As I said earlier, I can certainly imagine that blocking, protection, deletion, and all other administrative powers could be (and likely have been) abused. That's not an argument to disallow these things, only to monitor them carefully. Seraphimblade 21:09, 20 November 2006 (UTC)

I'd agree with you, except that in the case of A7, the original version and the current version are very different. It's a Camel's nose problem; if we say "no sources", how will that be interpreted on the ground? Deletion Review is chock full of examples of variable interpretation of the CSDs, and a lot of people don't have a problem with those expansions. -- nae'blis 21:36, 20 November 2006 (UTC)

I'd also agree, however, the procedures in place to deal with admin misuse of tools do very very little in all but the most extreme cases. You can "monitor" all you want, the fact is, admins are free to do what they will, as long as they think a half dozen or so friends will support them. I don't really see a problem with that unless the admin tools become admin powers. Every new "speedy" proposal is a large step in that direction. Unfocused 01:43, 22 November 2006 (UTC)

Here's one version I find massively annoying when I see it: <ref />

Kid you not, people do this. Avt tor

A new idea

The fundamental idea here is that articles should either be sourced or out of the encyclopedia. There is another way to achieve this. We already expect to provide copies in userspace of any such article to its authors, so they can source it; why not cut out the middle admin?

Any unsourced article may be moved to userspace until it is [subject to the sort of limitations we are imposing anyway].

This applies to articles which would be good if only they were sourced; cruft shouldn't be moved, it should be prodded or AfD'd. I suppose both could apply; we might need a new XfD.

The proceedure would be:

• Tag article with {{please source}}
• Move it to userspace of the major contributor.
• Add {{unsourcedarticle}} to the talk page of all substantial contributors.

The only other thing we'd need is a discussion page for the appeals "No, no, this really is sourced." and "This doesn't need a source, because...." it's a dab, or whatever.

Comments? Septentrionalis 22:11, 20 November 2006 (UTC)

Understand where you're coming from, but I think the deletion proposal would be more beneficial-the lag between tagging and deletion would put the article in front of many editors, any of whom may source it. Userfying would put it in front of fewer people (likely just its creator), and would also require "spectatoring" of any links, tags, templates, etc., to avoid cluttering the categories. That'd be time consuming unless a bot simply went around and looked for tagged articles and userfied them, which of course presents a host of potential problems in itself. If the article is of potential benefit, and really might be verifiable, putting it in front of a lot of people (with what is really a pretty generous time limit, even if these articles were all to be prodded that'd only give 5 days) sees a lot better chance of having it verified. It also raises the issue of what happens if an unsourced article is created several times on the same subject, and then different users find different sources and try to merge "their" version back in. Also, many users seem to start a stub and never look at it again. Finally, the deletion proposal solves the issue of articles which really do turn out to be unverifiable. Seraphimblade 22:28, 20 November 2006 (UTC)

• How about userfication instad of deletion in the present policy, at the end of the fourteen days?
• Articles which are in fact unverifiable would either linger in user space, where they use up no more resources than if they were deleted; or they can be MfD'd. Septentrionalis 01:27, 21 November 2006 (UTC)

The copies in project space would need to be deleted if that happens, and I could see dense/ignorant editors constantly recreating the article/moving the article back into project space and having the article G4/protected. ColourBurst 02:08, 21 November 2006 (UTC)

We could widen G4 if this proves to be a problem. (Amd technically it's a problem with the present proposal too; X can request a copy of his article and then move it back in a couple months later.) Septentrionalis 04:38, 21 November 2006 (UTC)

ColourBurst brings up a valid problem though-an admin's intervention is needed in any case, to delete the article out of encyclopedia space (unless we just blanked the page, and then AntiVandalBot's likely to start getting a ton of false positives). If someone actually bothers to say "Hey let me have a userspace copy of this, I think I can find sources and I want to work on it", that accomplishes three things-

• • • • It is ensured that at least one person is interested sourcing the article. There's no use making tons of userspace copies that no one even wants.
      • The administrator can ensure that the person requesting the userspace copy is clearly informed that the article is not to be recreated unti at least one source is found. (A template could even be made for this purpose.) That would eliminate any question of "I didn't know" if the user recreates the article.
      • The user will be able to contact the admin for help if (s)he is unclear on policies and the like, and the admin will already know what's going on. This will help new users understand how to source, lessening the problem in the future.

As to creating redirects from encyclopedia space to user space-that's not something I'd like to see started, and could be very confusing to readers-most users don't even know how redirects work, and will figure they found the article they looked for. Seraphimblade 05:06, 21 November 2006 (UTC)

I like this idea as a useful middle ground. The only modification would be this. You say "This applies to articles which would be good if only they were sourced; cruft shouldn't be moved, it should be prodded or AfD'd. I suppose both could apply; we might need a new XfD." Why not just say any article may be moved, but that such moved articles in userspace may be subject to prod or AfD like articles? This is simpler and makes more sense to me. Dmcdevit·t 04:53, 21 November 2006 (UTC)

Also, we would have to delete the redirects to userspace from article space which would be created, no? That should be a much less controversial CSD, I suppose. Dmcdevit·t 05:00, 21 November 2006 (UTC)

I wasn't sure whehter to leave the redirect for any of authors who looked for the article; but a waiting period would take care of that. Septentrionalis 06:14, 21 November 2006 (UTC)

• This variant of the proposal has fewer negative consequences but adds significantly to the bureaucracy of the project. It will still put off a significant number of good contributors. I'm not convinced that such a rule would really result in a better encyclopedia. Rossami (talk) 05:30, 21 November 2006 (UTC)
  • Why do you think that? One of the advantages of this process is that it seems to have considerably less bureaucracy than the original proposal. Dmcdevit·t 11:13, 21 November 2006 (UTC)

I've put in a draft to answer some of these concerns. If other details are more popular, please edit. The basic idea of userify is the only interest I have here. Septentrionalis 06:25, 21 November 2006 (UTC)

Waiting period

One of the chief questions of proposal #2 is whether there would be a waiting period between tagging and moving to userspace. Please discuss. Septentrionalis 06:28, 21 November 2006 (UTC)

Hm, maybe. Well, isn't the fact of moving it to userspace rather than getting rid of it outright supposed to cut down potential biting? This is the same reason for the waiting period in the original. I like the elegant, bureaucracy-free solution that anyone can just move an article they spot to userspace as long as they tag it and alert the authors. If they do that, it seems like waiting is unnecessary: what are you waiting for? They're not bitten, since they still have the article, and they add references and move it back whenever they want. Dmcdevit·t 11:00, 21 November 2006 (UTC)

I think moving immediately is reasonable, as long as the article remains tagged with something that puts it in a category (e.g. {{unsourced userfied article|MONTH}} or something that puts its articles into Category:Unsourced userfied articles from MONTH. I think it's important that other editors should be able to find such articles in order to improve them; it could be the original author has no interest in sourcing it, or doesn't know how to.

Such a date-ordered category system could also allow somebody to periodically {{prod}} articles that have been unsourced for a long period of time, which might be a useful housekeeping task.

In fact, I think it would be important even if there is a waiting period... JulesH 11:32, 21 November 2006 (UTC)

I'll put it both ways. Thinking about a waiting period more made me realize that what I'm really thinking about is a Request for Sources, which should be written up separately. I like the low bureaucracy too. Septentrionalis 17:27, 21 November 2006 (UTC)

One problem I can see is that a user might just remove the unsourced tag without actually sourcing it and moving it back. Something will have to be done to address that. ColourBurst 22:44, 21 November 2006 (UTC)

I fear this would overload MFD (which is where user subpages almost invariably go) or cause a substantial redefinition of "article for deletion" (which I'd actually be okay with, for obviously userfied articles). I'm concerned, though, that userfying them first will make them less likely to end up sourced and good articles, due to less eyes on them (and you run the risk of gaming the system or walled gardens where userfied articles connect to other userfied articles). I do think we can strengthen unreferenced by dating it, though; see its talk page. Then we have good evidence when we send an article to AFD for complete lack of sources after a reasonable period of time. -- nae'blis 21:53, 21 November 2006 (UTC)

Thoughts by Elaragirl

I see a lot of conflicted opinion on this issue here. And it really boils down to the old conflict of what it is as opposed to what it could be. Many people feel that stubs should be allowed to hang around without sources since a mysterious someone will come along and source them later. Translation: sources really don't matter, all that matters is adding more articles, no matter how skeletal. If you disagree, you're a filthy deletionist trying to destroy the encyclopedia.

The proposal, as it stands, suggests that unsourced articles shouldn't be here. Some people say that this is a deletionist proposal, and that it's up to editors to add sources, and that stubs are fine. To me, this is hardly a deletionist proposal, but since people want to say it is, I'll give you the deletionist version of this policy:

"Any article lacking reliable, verifiable sources may be speedily deleted at any time. Articles created without appropiate sourcing should be automatically deleted via bot. It is the responsbility of the article's creator to provide initial sources (at least ONE) for any article created. Users who routinely created unsourced articles may be considered vandals and banned."

Now, obviously, the above is completely insane. It would definately cut down on the amount of junk that got into Wikipedia but would also destroy thousands of perfectly good articles and ruin hundreds of hardworking editors. The current proposal is hardly that reactionary. If you belive that stubs are viable, then we should discuss a cutoff period for how long stubs can remain unsourced.

I, for one, would like to see something in this proposal that prevents against abuse against new editors. A real deletionist does not delete unless the article violates policy. Stubs that were created a week ago with no sources, despite the clutter they make, do not harm the encyclopedia. A stub that's been unsourced and unworked on for months, however, is clearly not being utilized and should be deleted. IF someone wants to then recreate it, they can find some sources for the blasted thing. Those of you who act as if this proposal is the equivlent of my hyperbole above are engaging in histronics.

User is too stupid to sign her own statement. (sigh) --Elaragirl^Talk|Count 18:42, 21 November 2006 (UTC)

As I wrote higher up the page but you might have missed:

The verification that the information in a stub is valid can come from links to other Wikipidia articles, and internal links do not count as sources. For example think of an timeline which starts as little more than a disambiguation page, providing the links to the items include the date put into the timeline, there is no need to provide an alternative source. BTW would this suggestion as it stands mark disambiguation pages for deletion as they do not have sources? --Philip Baird Shearer 20:46, 21 November 2006 (UTC)

To answer your question 1) Disambigulation pages are just that, not stubs. 2) I belive that I said it best earlier : links to another page to formulate a timeline or something where the stub is CLEARLY being worked on should count -- as long as it hasn't been months and months. --Elaragirl^Talk|Count 22:00, 21 November 2006 (UTC)

I cannot see anyone's intending this to apply either to disambiguation pages, nor to pages written in summary style, provided that they contain only basic facts that are needed to direct a reader to the proper page. That should, perhaps, be explicitly mentioned. Robert A.West (Talk) 22:27, 21 November 2006 (UTC)

Then again, there are people out there who do slap speedy tags on everything without carefully looking...hrm. Phillip, I'll have to rethink this a bit. I might leave a message on your talk page once I figure it all out. --Elaragirl^Talk|Count 22:36, 21 November 2006 (UTC)

A stub that's been unsourced and unworked on for months, however, is clearly not being utilized and should be deleted. I disagree. I've expanded on stubs that have existed in stub form for 6 months or more in the past. I've added sources to unsourced articles that have existed for 2 years or more. Unsourced articles, whether stub or not, are useful to editors. Whether they're useful to readers or not is a little more debatable. JulesH 08:58, 22 November 2006 (UTC)

Like JulesH, I not only expand old stubs but also deliberately search for stubs to find topics that I want to learn more about, then expand them. Stubs are useful to readers, for what they are. If you think readers don't know what a stub article is or why it's permitted, then the solution may be to simply edit the stub templates to further clarify that they're works in progress. Stub articles are what gets people to join! Who would feel comfortable making their first serious edit in Yom Kippur War when it's already so fully developed? Unfocused 14:08, 22 November 2006 (UTC)

Question about userfy variant

If someone wanted to delete a poorly sourced article, could s/he game the system by challenging and then removing sources, claiming they were the main editor, move to their own user space and then speedy delete? Addhoc 18:19, 21 November 2006 (UTC)

• No. They can't "claim" they were the main editor because anybody can see the revision history which will point this out. Besides, if they want to speedy delete their own article which they were the main editor, they can already do this through CSD. ColourBurst 18:34, 21 November 2006 (UTC)
• On the other hand, if the admin is clueless enough not to notice that the article was moved and is being deleted by the same user, they can do it now. A note on U1 may be in order. Also, a defender need only add a source (they're in the history) and move back. Septentrionalis 18:41, 21 November 2006 (UTC)

UfD

A name consistent with our conventions (which DfV was not) has been placed as option #4 on the project page. I have already made the case for an AfD-style process above. I think that having the opportunity for discussion and isolating the pure verifiability questions from the more nebulous issues of notability and so on will be useful. Robert A.West (Talk) 23:58, 21 November 2006 (UTC)

While well-intentioned, this is a classic example of instruction creep. Rossami (talk) 01:57, 22 November 2006 (UTC)

The instruction creep charge makes sense only if you believe that (1) five days is sufficient time to find sources for an abandoned article and (2) admins by-and-large actually ignore responses of the form, "Keep, it's interesting" or "Keep, and look harder for the sources that I know must exist. I haven't looked and won't lift a finger to help, but no matter how hard you look, you could have looked harder, so my response will be the same." My experience is that five days often does not suffice, and the matter is left hanging -- not enough good sources to really support the article, but enough to raise the hope that more exist. Also, too many AfD's get treated like votes if that will lead to a "keep" result. Robert A.West (Talk) 13:05, 22 November 2006 (UTC)

Check the history of the shortcut WP:UFD; it's been used in the past. At the moment, it goes to Wikipedia:User categories for discussion, disambiguated to Wikipedia:Userboxes for discussion (now merged with TfD, so the link goes there); I'm not at all sure if this abbreviation is free. --ais523 13:41, 24 November 2006 (UTC)

Thanks. I should have checked. VfD (Verifability for Discussion/Deletion) would be wrong for similar reasons. UAfD? I'll have to think about this. Robert A.West (Talk) 14:19, 24 November 2006 (UTC)

If the userboxes had an active process, this could be a problem. But since it's a redirect, just add a dab header. Septentrionalis 15:34, 24 November 2006 (UTC)

Sample group

I have created a random sample of a existing articles that have no external links and lack any evident formal references. If one is discussing unsourced articles, it is probably a good idea to consider the sorts of articles that people have historically neglected to include sources for, and also to think about how liberally one should be in defining a source. (For example, several are articles describing works of fiction. Though no explicit source is given, the work itself is likely the implied source.) I would invite people to use this to draw their own conclusions. Dragons flight 03:21, 22 November 2006 (UTC)

Interestingly, Awful Mess Mystery is one that I created. in this instance, as it's a stub about an album, one expects that it's information from the album, and if anyone challenged it, I'd merely slap a references section on there and refer to the liner notes. And what happens? The article is no better than it was before (it's only a stub, after all), but at least it's "sourced." Of course, we need a one-size-fits-all mentality if we want this to work, but we also know that a one-size-fits-all mentality makes no sense when dealing with, say, album, book, or movie stubs. Now, say, Column of the Goddess for instance, would be a prime candidate for some crazy vandal to come around and slap on a reference called, say, Architecture and France: Surrendering to the Art by Archibald Tect, and no one would be any wiser, since the large majority of people who will be reviewing these will have no clue regarding this sort of information. This list is extremely interesting, although not for the reasons I think it was compiled. --badlydrawnjeff talk 03:33, 22 November 2006 (UTC)

I mean no offense to Badlydrawnjeff, but Awful Mess Mystery is a great example of an article that needs better sourcing. That article says it was "released in 2000 via Mud Records," the Wolfie article lists "Awful Mess Mystery (CD/LP) - Mud Records - 1999," my library has the December 23, 1998, Riverfront Times "Sound Checks" column, which describes "Wolfie's debut, Awful Mess Mystery (Mud) ..." and Amazon lists the album as "Original Release Date: 1998" [1] This seems to be an obvious instance where incorrect information about basic facts could be easily corrected by checking some sources and then referencing them so other editors can double-check. Also, some sourced analysis from that review and others would make for a much more informative article about this album. -- Dragonfiend 06:13, 22 November 2006 (UTC)

Agree with Dragonfiend, and would also tend to question whether an album should have its own article at all, if one can find no better "sources" on it then the liner notes. I know WP:ALBUMS currently has that "notability by association" criteria in it (if the artist who created it was notable the album must be too!), but I disagree. Sometimes, an album is verifiable enough to deserve a mention on the artist's page, but nothing notable at all in its own right. Dragonfiend's example also is a perfect example of why everything should be sourced. Seraphimblade 07:35, 22 November 2006 (UTC)

No offense taken, it's my fault for doing one of the articles in one place and one in the other. I don't claim to be perfect, but I'll also say that even if I did put a random source in there, it wouldn't have fixed this error (which I'll fix when I get home and actually look at the album again). And yes, the album should have an article of its own per a completely workable and uncontroversial status quo, but that's not a discussion for here. If anything, this situation shows that, if we're looking for incontrovertable accuracy, the amount or quality of sourcing isn't going to matter on the lesser-known articles because people will still occasionally get it wrong. --badlydrawnjeff talk 13:12, 22 November 2006 (UTC)

To be fair, most of the WP:MUSIC criteria was created because the people who created the criteria probably thought that the articles would likely have reliable sources written about them. I don't like the notability by association criteria, mostly because it seems to be solely a WP:MUSIC thing and in any other article, we wouldn't be able to use that criteria to keep. ColourBurst 01:36, 23 November 2006 (UTC)

I have been classifying these on the talk page of the list. A large proportion of them are one and two-sentence non-controversial stubs, like Issum: Issum is a village and a municipality in the district of Cleves in the state of North Rhine-Westphalia, Germany. It is located approx. 5 km east of Geldern. " +stub label. Do we want to massively delete stubs like this? Is the redlink preferable? How long before it is replaced by another (probably unsourced and untagged) stub? Septentrionalis 00:55, 24 November 2006 (UTC)

• On the other hand, what would we lose much if your example were deleted? Absent a source, I don't know if the information is correct. Yes, I can look it up myself, but then Wikipedia has added no value. I could then add the source I found, but so could the person who created the stub in the first place. I've created unsourced stubs intending to get back to them "real soon now" and, guess what? In a lot of cases, I got interested in something else and never sourced them. Did I really improve Wikipedia that much by adding them? Had I gotten a note on my talk page that they were on a 14-day clock, I probably would have done the library work within that period, saved the stub and improved Wikipedia. Robert A.West (Talk) 12:29, 24 November 2006 (UTC)
  • Lose much? No. Do we gain anything? And if so, what? This is probably everything most people want to know about Issum, as my favorite English major comments.
  • What will happened if Issum is deleted? It will be a red-link for a while; then some Westphalian will recreate it. Like most of our half-million stubs, it probably will be recreated as a couple sentences, by a passing anon, without stub-tag or cats. It will then be a random process whether the stub project catches it before CSD does; if they do, it will be back here pretty much as it is now (but not G4). If they don't, the net effect will be to remove the stub-tag. This is a waste of time, our one limited resource. Septentrionalis 15:32, 24 November 2006 (UTC)
    • But, that is not realistic. First of all, thirty seconds with Google] yielded a municipal website, which is surely an acceptable (albeit not ideal) source, and which yielded the noteworthy information that the town contains a 13th century church and a synagogue that survived the Nazi era. It would take less effort to add the source (and the additional information) than to nominate for CSD, or XfD, so it probably will never get nominated. If nominated, someone will source it and drop a note on the nominator's talk page grumbling about WP:POINT. Schloß damit! Robert A.West (Talk) 16:09, 24 November 2006 (UTC)
      • If UfD became, in effect, a project to source articles, that would be a good thing; I would support it. The poll above was on proposal 1, which would either delete Issum or leave it as it is. I've looked at half the articles. Most are at least as worthwhile as Issum: one definite cruft and one article prodded before I got to it. Septentrionalis 17:05, 24 November 2006 (UTC)
        • No, the project to source (I mean really source) articles is a different, and complementary process. Every project needs a dumpster, and UfD (or whatever) is that dumpster. Robert A.West (Talk) 17:26, 24 November 2006 (UTC)

I've skimmed your list. I see one article which refers to a topic certainly notable in Canadian history, Peggys Cove, Nova Scotia, and a couple of others that I would consider notable offhand. Each of the articles I looked at had between around seven and twelve named editors, of which I would conservatively assume at least a quarter of which are actively watching. If an article I was watching had an unsourced-will-delete tag put on it, I'd be darn sure to find a few references (I know I could find plenty of sources about Peggys Cove and UML). The 14-day delay seems adequate to save the good articles from the reaperbot. Avt tor 21:25, 31 January 2007 (UTC)

What about projects?

I noticed that Dragons flight's random sample contains some articles that have been claimed as part of wikiprojects. The projects are designed to improve all of the articles in its "jurisdiction". I work on Wikipedia:WikiProject Films and know that it is a massive task that will take a long time to get anywhere. But, it would be very disheartening for me, and I'm sure a great many other project workers, to know that articles could be deleted simply because I haven't got to them yet and had an opportunity to improve them. To my knowledge there is no one looking for "orphaned" articles to add them to estiblished projects. Would it not be better to assign unsourced articles, wherever possible, to a project where it is very likely to be improved? This would be especially true for stubs. Mallanox 02:30, 24 November 2006 (UTC)

• Would all projects welcome having an outside editor "assign" articles to them? Except for biographies of living persons, if we use a CSD for this, then I certainly would support adoption by an active project in lieu of sourcing. (There are inactive projects that would be of little or no value in this regard.)
• I'll continue to plug my idee fixe as an option where it would be in-process to note that the article has been adopted by an active project and should be exempted from deletion. I would expect that when a project determines that an article is unsalvagable (such as being unverifiable) they nominate it for deletion. Robert A.West (Talk) 12:21, 24 November 2006 (UTC)
  • Ok, this is a minor point but I just wonder who would be regarded as an editor with a good reputation? Mallanox 02:26, 26 November 2006 (UTC)
    • One who doesn't have a bad one. <grin> Seriously, that would have to be up to the consensus. I am sure a better phrasing can be found, but there are some editors whom I would expect to gladly volunteer for the purpose of disrupting the process. Perhaps "Volunteer acceptable to the consensus," would be better? Robert A.West (Talk) 18:43, 26 November 2006 (UTC)

Strongly Oppose This Proposal

This proposed criterion is inconsistent with the purpose of the CSD, which is to allow the rapid and efficient deletion of articles that would almost certainly be deleted as a result of the AFD process, without the burden of holding AFD discussions on every such article. If someone creates a page full of typewriter salad, a blatant attack page, or a page filled with links to offshore pharmacies, it is quite justified for administrators to unilaterally delete such pages on sight, without the needless extra work of an AFD nomination and discussion. Indeed, the header on Wikipedia talk:Criteria for speedy deletion states that new CSD proposals should only be used to effectuate uncontroversial deletions:

Contributors frequently propose new criteria for speedy deletion. If you have a proposal to offer, please keep a few guidelines in mind...The criterion should be uncontestable: it should be the case that almost all articles that can be deleted using the rule, should be deleted, according to general consensus. If a rule paves the path for deletions that will cause controversy, it probably needs to be restricted. In particular, don't propose a CSD in order to overrule keep votes that might otherwise occur in AfD...

Based on statements made by Dmcdevit, the author of this proposal, it is quite apparent that this proposed criterion was created for the specific purpose of circumventing the probable outcome of AFD discussions, and deleting articles against consensus:

References are erroneously regarded as a "cleanup" issue at AFD and PROD, and nominations for lack of sourcing are consistently shouted down, since they should be fixed, not deleted. This leaves us in the position of deciding to keep articles for which we idea or proof of their accuracy, neutrality, or existence. This is precisely not the reason that things like poor writing are not reasons for deletion, but our current deletion mechanisms don't handle this problem. Honestly, try nominating a recent article for deletion solely because of sourcing, regardless of its notability, and see what yu get.[2]

The creation of a criterion for speedy deletion for the specific purpose of effectuating controversial deletions in a manner that is concededly against consensus would probably cause extensive wheel warring over page deletions, thereby creating a severe disruption in Wikipedia's administration. Administrators unilaterally deleting unsourced articles could cite this proposed CSD as a justification; administrators unilaterally undeleting unsourced articles could utilize the uncertainty inherent in Wikipedia's policy making process to question whether this proposed CSD was actually valid policy, or could invoke WP:IAR to undelete articles that are deleted under a CSD specifically designed to allow articles to be deleted against consensus. Indeed, the articles for deletion process is designed to avoid wheel wars over page deletions, because the archived deletion discussions provide evidence of the existence of consensus for the deletions, thereby dissuading other administrators from performing unilateral undeletions. Unilateral deletions, by contrast, promote unilateral reversals by administrators who disagree with them; consequently, the present CSD only permit the unilateral deletion of articles under circumstances in which disagreements over the deletions are highly improbable. Attempting to enforce Wikipedia:Verifiability by deleting articles against consensus and causing wheel warring would be a serious violation of Wikipedia:Do not disrupt Wikipedia to illustrate a point. John254 05:23, 24 November 2006 (UTC)

This doesn't really allow for "unilateral" deletion-and I think applying it as a "speedy" deletion criterion is a bit misleading. If it were placed in CSD, the criterion would read something like "The article has been tagged with {{nosources}} (or whatever template would be used), the article was not created before (date proposal went into effect), and no sources have been added for 14 days since the tag was applied." I would, however, be up for simply making it a "deletion criterion" instead. However, consider this instead: "{{subst:prod|Unsourced.}}" That's really the only way to deal with that now, if sources can't be found and aren't forthcoming, that only gives 5 days to solve the problem, and no one looking through the prodded articles category would know that one's tagged for lack of sources. This would give a full 14 days in a category specifically devoted to sourceless articles.

This isn't "unilateral" at all, and unlike the CSD's isn't really a "shoot on sight" criterion. But to some degree it is (an administrator may "shoot on sight" an article with an expired prod tag, after all.) But allowing two weeks lag time hardly means no time to discuss anything controversial.

And, once again, "it might get abused" is potentially an argument against everything. I mean, we all know letting everyone edit gets abused terribly. Why are we still doing that, we could stop those vandals dead in their tracks! Seraphimblade 12:37, 24 November 2006 (UTC)

The statement that "This doesn't really allow for 'unilateral' deletion" is factually incorrect. If an article is deleted on the basis of an administrator's determination that the article qualifies for deletion, without any attempt to determine whether there is consensus for deletion of the specific article in question, then the deletion is unilateral, irrespective of whether the article has been tagged for 14 days prior to taking this action. Moreover, as I explained above, Dmcdevit created this proposal for the specific purpose of deleting articles against consensus, where such articles concededly would not be deleted as a result of AFD nominations.[3] I do not merely claim that this proposed criterion "might get abused." Rather, this criterion was proposed for an improper purpose (deletion against consensus of articles that could not be deleted as a result of the AFD process), and is quite likely to be used in a manner consistent with its improper purpose. The comparison to prods is inapposite, as a proposed deletion can be stopped if anyone removes the tag, and therefore does not operate in a manner contrary to consensus. The statement that "allowing two weeks lag time hardly means no time to discuss anything controversial" is misleading. While it is true that the proposed deletion of an article could be discussed for two weeks, this proposal allows (and was created for the specific purpose of allowing) the deletion of an article irrespective of any consensus reached in the resulting discussion. Even if there were a unanimous consensus amongst 20 established editors that the article should be kept, the article could still be deleted if a source were not provided within two weeks. This sort of deletion is not merely a possibility of abuse -- it is the specific purpose for which the criterion was created[4], and the specific reason why the criterion includes no consideration of community consensus before an article is deleted. John254 15:33, 24 November 2006 (UTC)

I still disagree that there's any potential for "abuse" here. This tag would be almost as trivial to remove as the prod tag-all one would have to do is cite one source that verifies the thing exists. Not is notable, not is encyclopedic, not even necessarily that the source is reliable, just that the thing exists somewhere. A person could cite a blog post that mentioned the thing in passing-that's not likely to get it to survive an AfD if that's all there is out there, but it would stop this process (just as "Removing prod-I think this article's cool!" will stop the prod process). Therefore, I think the comparison to prod is appropriate. If no source, even so trivial as the one I mentioned, can be found, the article should be deleted. I also assert that this (just like prod) is decidedly not unilateral-the discussion would simply center around finding sources, and the location of any sourcing whatsoever stops here. Any discussion of whether that source is reliable or shows notability would then be reserved for AfD, if that still proved to be necessary.

WP:V is policy, and this is the logical extension of it-"The burden of evidence lies with the editor who adds or restores material. Any material that is challenged or likely to be challenged needs a reliable source, which should be cited in the article. If an article topic has no reliable, third-party sources, Wikipedia should not have an article on it." So, one could just go around blanking every article which includes no sources, but this would (probably correctly) be seen as disruption to make a point, and would also drive AntiVandalBot nuts. One could AfD every such article, but this wastes AfD time-especially if sources are out there and no one's gotten around to citing them yet. Instead of creating those issues, this creates a process by which a major problem (unsourced articles) can be dealt with in a way specifically tailored to that problem, instead of being "shoehorned" into other processes or the problem simply ignored. And with strong time limits, there's no more "potential for abuse" here then with WP:PROD-an admin that deleted a prodded article after two days (unless it also met a CSD criterion) would easily and correctly be seen as out of line. The same would apply here. Also, this idea, much like prod, provides that an article be easily and unquestionably undeleted if a source is found after time expires. Seraphimblade 20:51, 24 November 2006 (UTC)

This proposed criterion, like all existing criteria for speedy deletion, is indeed unilateral, in the sense that consensus is ignored by the closing administrator. While provision of a reference would, theoretically, result in the article being kept, if no references are provided in a "UfD" discussion, the closing administrator is directed to delete the article, even if a large number of established editors supported keeping the article, and none besides the nominator supported deletion. Of course, as I have explained previously, this proposed criterion was created for the specific purpose of circumventing the outcome of AFD discussions, and deleting articles against consensus.[5] While Wikipedia:Verifiability is indeed foundation-level policy that cannot be overridden by local consensus, Wikipedia:Verifiability does not require that every article lacking references be deleted 14 days after it is nominated for deletion. The actual provision of this policy concerning article deletions is that "If an article topic has no reliable, third-party sources, Wikipedia should not have an article on it." That a Wikipedia article does not presently have any sources certainly doesn't establish a strong probability that none exist. Furthermore, the inability to produce a source for the article after 14 days in the "UfD" process doesn't establish the probable non-existence of sources, especially given differences between AFD discussions and "unverifiable for deletion" discussions. In the AFD process, the default outcome is to keep the article; only if there is a consensus for deletion will the article be deleted. Consequently, if large numbers of articles are nominated for deletion, in such a manner as to overwhelm the ability of users to respond to the nominations, and to prevent any discussion of most nominations from occurring, the articles will be kept with an outcome of "no consensus." In the "unverifiable for deletion" process, the default outcome is deletion. Thus, if the number of articles that are nominated for deletion through this process is too large to allow users to make an effort to find references for each of them in the prescribed period of time, the articles will simply be deleted, even if references for any one of the articles could be found with a modicum of effort. While it would be possible to prevent a single user from engaging in deliberate "deletion trolling" by personally nominating such large numbers of articles for deletion through this process as to exceed the ability of users to attempt to find references for them, the more likely disruption of this process would be created accidentally by large numbers of users nominating relatively small numbers of articles for deletion, so that the aggregate number of articles being considered for deletion at any one time would exceed the ability of users to search for references for them. Moreover, since deletion of articles under "Proposed text #4" is apparently not restricted to articles created after its enactment, and since it appears that at least half of Wikipedia articles are presently unreferenced, "Proposed text #4" could potentially be used to rapidly delete hundreds of thousands of articles, almost all of which could be referenced if there were sufficient time available to do so.

Wikipedia:Verifiability is policy. However, this policy needs to be enforced in a manner that provides some deference to community consensus with respect to particular applications. Articles can, and have, been deleted to enforce Wikipedia:Verifiability through the AFD process, sometimes as a result of my nominations (see Wikipedia:Articles_for_deletion/Dualist_nihilism, for example). However, the proposal to always delete any unreferenced article if sources cannot be found within fourteen days after the nomination for deletion, consensus be darned, is excessive, and is likely to result in the rapid deletion of large numbers of perfectly salvageable articles, and a large number of wheel wars over the deletions. John254 22:46, 24 November 2006 (UTC)

• On the other hand, consensus differs from vote counting. The above argument seems to be directed against current policy. existing policy already directs administrators to ignore mere numbers when looking for consensus, and especially so when the issue is verifiability. A single "keep" argument, in the form of a good source, should win the day over any number of "delete" unvotes. Likewise, a host of "keep, there must be source somewhere" unvotes should be ignored if there are neither sources nor a reasonably precise idea of where to look for sources. The UfD proposal (as opposed to the CSD proposal) does nothing new in this regard, except to make it more obvious when administrators ignore the policy. Robert A.West (Talk) 18:54, 26 November 2006 (UTC)

On the contrary, this proposed criterion was created for the specific purpose of circumventing the outcome of AFD discussions, and deleting articles against consensus, as the creator of this proposal admits here. Moreover, the "Deletion guidelines for administrators" actually state that

Where it is impossible that an article on any topic can exist without breaching these three policies[ WP:V, WP:NPOV, and WP:NOR ], such policies must again be respected above other opinions.

An article's present lack of references certainly doesn't necessarily imply the impossibility of finding references. Rather, whether an article is so inherently unverifiable as to justify deletion against the wishes of most established users is a matter which must be decided at the closing administrator's discretion in an AFD discussion. Ultimately, though, the extent to which the administrator closing an AFD discussion actually can act against community consensus is rather limited; if other administrators are sufficiently dissatisfied with a deletion, the article will be undeleted. The closing administrator can't re-delete the article; a claim that one is enforcing Wikipedia:Verifiability through a deletion is not a license for wheel warring. Moreover, if, say, ten established users, all administrators themselves, support keeping an article, and no one besides the nominator supports deletion, why should the closing administrator substitute his/her own judgment for the judgment of the ten other administrators? How can the closing administrator actually be sure that he/she evaluated the situation correctly, and everyone else evaluated it incorrectly?

The "Deletion guidelines for administrators" also state that the closing administrator should "respect the judgment and feelings of Wikipedia participants..." and clearly state that "When in doubt, don't delete." (quoted in the original boldface type). This proposed CSD is quite inconsistent with existing guidelines, because instead of allowing a presumption against deletion, it creates a presumption in favor of deletion, by assuming that any article that has remained unreferenced for 14 days is automatically unverifiable, absent any particularized consideration. Articles may be deleted to enforce Wikipedia:Verifiability; however, it is prudent to employ AFD discussions as a safeguard against the unilateral deletion of large numbers of salvageable articles. John254 23:43, 26 November 2006 (UTC)

• So now I am puzzled. You specifically cited UfD by name: that is a 14-day discussion-permitted alternative to CSD (#4 on the project page). You then oppose it in large measure for lacking an opportunity for discussion. I don't get it. The UfD alternative demands due diligence on the part of the nominator, which AfD does not: 60 days of tagging, or a source that makes the topic appear unlikely, or enough of a search that one would expect to find something if there is something to find. After that, there is 14 days to overcome the presumption of no sources. That's a total of seven weeks -- half a semester. I can hardly think of that as a rush to judgment. The only "anti-consensus" measure contained in UfD is a prohibition of WP:ILIKEIT and its close cousin, the Morton's fork: if you haven't found sources, you haven't looked hard enough. Robert A.West (Talk) 01:20, 27 November 2006 (UTC)

"UfD" proceedings, as described in the proposal, wouldn't actually require that the article be tagged for 60 days prior to nomination for deletion. They would merely require "that the nominator asserts a good-faith belief either that there are no valid, reliable sources for the topic" (which will not be challenged, unless the nominator is personally nominating high volumes of articles for deletion through this process). That "the article has been tagged as unsourced for at least 60 days and is not being actively rewritten" is not actually a prerequisite for nomination, but merely a suggestion as to one of many reasons why an article might be nominated for deletion through this process. I do not "oppose... [the UfD process] in large measure for lacking an opportunity for discussion" -- rather, the difficulty with the UfD process is that consensus is ignored. The claim that Wikipedia:Verifiability somehow mandates the outcome of the UfD process, and forces decisions against consensus, is incorrect. WP:V does state that "Editors should provide a reliable source for material that is challenged or likely to be challenged, or it may be removed." -- however, what "removed" means, in this context, is "removed from the current version of the article", not "removed from the page history". Thus, if an article has no sources, it is perfectly consistent with WP:V to blank and redirect the article to an article on a related, more general topic. Note that the creation of a functional redirect will not be reverted by AntiVandalBot or similar bots. However, since the content of an article blanked and redirected in this manner remains accessible through the page history, editors may examine the article, find sources for it, and restore it. Administrative deletion of an article, however, is a far more permanent measure, because it prevents editors without administrative privileges from subsequently examining the content of the article, and restoring it once sources are found. While Wikipedia:Verifiability also states that "If an article topic has no reliable, third-party sources, Wikipedia should not have an article on it", the failure of an article to have reliable, third-party sources for fourteen days doesn't necessarily imply that no reliable, third-party sources may be found for the article topic -- whether it is probable that no reliable sources may be found for an article topic is a matter of judgment with respect to the particular article in question. Even if an article appears to concern a topic with respect to which "Wikipedia should not have an article", this doesn't imply that administrative deletion is the proper course of action. A redirect is not an article; consequently, blanking and redirecting the article to a more general topic satisfies the requirements of Wikipedia:Verifiability, while allowing editors to continue to view the page history, and attempt to find sources for the article. Thus, it has been shown that

(1) Wikipedia:Verifiability never requires that an article be administratively deleted (though this policy might require that an article be blanked and redirected until reliable sources are provided, if it is not administratively deleted)

(2) As a corollary to (1), whether any article should be administratively deleted to enforce Wikipedia:Verifiability is a matter to be decided purely by community consensus.

(3) By (2), this proposal, which was created for the specific purpose of circumventing the outcome of AFD discussions, and deleting articles against consensus [6], should be rejected. QED. John254 03:31, 27 November 2006 (UTC)

The foregoing assumes, counterfactually, that there is always a sensible redirect target for any topic. If that were true, we would never need deletion at all. This means that assertion (1) is true only in the trivial sense that Verifiability does not prohibit blank articles, verifiable articles on topics that do not match the title or random redirects. Other policies do. If there is no sensible redirect, then we have no valid alternative to deletion. Since (1) is false, its corollary fails, and the conclusion cannot be logically supported. Robert A.West (Talk) 03:50, 27 November 2006 (UTC)

Redirects are not blank articles. If the name of a page has any meaning at all, it can almost always be redirected sensibly to a more general topic of which the subject treated on the page is a subset. For instance, if the subject of an unreferenced article is a school in Fargo, ND, the article can be redirected to Fargo, North Dakota. The article on "Dualist Nihilism" that was deleted as a result of my nomination here could have been redirected to Nihilism. An unreferenced article concerning a chemist could be redirected to Chemistry. There is really no difficulty in formulating sensible redirects unless an article is about nothing, in which case it is already subject to speedy deletion under the existing criteria (CSD G1, CSD G2, CSD G3, or CSD A1). This is not to say that blanking and redirection is always preferable to deletion -- sometimes deletion is preferable; however, whether a page should be redirected or administratively deleted (or, for that matter, whether a sensible redirect for a page exists) is a matter to be decided by community consensus. Note that the UfD proposal essentially excludes the prospect of blanking and redirection; if references for an article are not forthcoming within the prescribed period of time, the page is automatically deleted. (Though, interestingly, if an article is blanked and redirected before a UfD nomination, it cannot be considered for deletion through this process.) Administrative deletion is a serious measure that should not be taken against pages unless

(1) The pages are of a nature that would almost certainly be deleted if nominated for deletion through the AFD process; almost all such pages are described in the existing CSD.

(2) There is a consensus to delete the pages through the AFD process.

(3) The pages are libelous, copyright infringements, contain sensitive private information, or otherwise really need to be deleted. (Though, even here, there are difficulties. If ten administrators claim that a page isn't a copyright infringement, but one believes that it is, can that one administrator really be sure that his/her assessment of the page is correct? Perhaps the most we could claim here is that the deletion of a page that "really needs" to be deleted couldn't be overridden by a non-administrative consensus, since an administrative consensus may provide a better indication of the truth in this matter than a unilateral assessment.)

or

(4) The pages are deleted by the Wikimedia Foundation.

Community consensus on Wikipedia is entitled to substantial deference, and should not be bypassed without a good and sufficient reason. John254 04:34, 27 November 2006 (UTC)

• I never said that redirects are the same as blank pages -- please let us disagree reasonably and about what each other actually said. I said that meaningless redirects, blank pages and so on are prohibited by policies other than Verifiability.

• In many cases decent redirects won't exist. Where does one redirect an article on Slobovia? Or its prime minister and his cabinet, or Her Serene Imperial Majesty, the Empress of that vaunted nation, all made up in the common room of a game store one day? Does one redirect P-P-P-Powerbook to internet hoax or to scam baiting? Actually, if we had a single reliable source to tell us which way to point, we would still have the article.

• Your Point (3) brings me back to my original proposal, which was limited to unsource BLP's, which can be harmful even if not obviously libelous. I really see no argument for keeping any unsourced BLP's.

• Question: an article has no reliable sources cited. An editor states that he has done a pretty standard pre-AFD search: Google, Lexis/Nexis, Google Scholar and Google Books, and come up with nothing remotely reliable. How much more do you really need before you conclude that there are no sources to find? Sure, there may be something on D-floor of some huge library, or an obscure newspaper, but why isn't 14 days a reasonable waiting time before we conclude that the article is probably bull? Why shouldn't ILIKEIT and "Look harder" arguments be ignored? Robert A.West (Talk) 05:03, 27 November 2006 (UTC)

There might not be sensible redirects for all articles. However, if there is a community consensus for a certain redirect, that would tend to indicate that the redirect is sensible. One generally expects the Wikipedia community to act in a reasonable manner. 14 days might a reasonable waiting time before we conclude that the article is "probably bull" in the vast majority of cases; however, the question here is whether the 14 days is a reasonable time period where there is a consensus for a different period of time. Indeed, what this question really boils down to is: how is the truth with respect to an article ascertained? Do we place a greater reliance on an assessment by a single individual, or the considered opinions of many members of the Wikipedia community? Perhaps we could place a greater reliance on the opinions of editors with greater qualifications; administrative opinions could weigh more heavily than non-administrative opinions. However, where other administrators comment in an AFD discussion, is the closing administrator in a better position to determine the correct outcome than anyone else? John254 05:23, 27 November 2006 (UTC)

• No, he's just the guy with the job. The administrator should look at the arguments, and should throw out the garbage. IDONTLIKEIT is just as bad as ILIKEIT: neither is a considered opinion. The nominator says, "I did A through E and found no sources. I think none exist." The only possible refutations are, (1) the nominator is obviously untrustworthy and should be ignored; (2) another editor finds a source; (3) another editor makes a good, reasonable argument why a source exists, and why more time is needed to locate it. What else could possibly qualify as a considered opinion? Robert A.West (Talk) 05:44, 27 November 2006 (UTC)

List of unsourced articles

I wish to see here an uncommented list of articles that any editor believes are entirely unsourced. I don't say such do not exist; I just want to see a list. Thank you. John Reid ° 11:11, 24 November 2006 (UTC)

• Wikipedia talk:Speedy deletion criterion for unsourced articles#Sample group
• Lennox Sebe
• Virgem da Lapa
• Kaffeeklatsch
• Carolyn Watkinson

Notes

My additions to the list above were found using Special:Random; I've permlinked them in case they get sourced between writing and reading. Per the request above, I won't comment further on them unless asked. --ais523 13:28, 24 November 2006 (UTC)

Comment by MacGyverMagic

I come across unsourced articles on a regular basis and often it's quite easy to fix it. I would only support an AFD-like process where people discuss if sources exist (and add them if it's the case). Articles tagged for speedy deletion, no matter if it is 5 days or 14 days generally get ignored. Of course, another project that works to source articles before they are deleted is acceptable too. It would be a shame to lose good articles just because a newbie creator didn't source it. - Mgm|^(talk) 11:30, 24 November 2006 (UTC)

• Extra note: If implemented, this should be subject to a site-wide strawpoll before any decisions are taken. - Mgm|^(talk) 11:35, 24 November 2006 (UTC)

• See option #4. By splitting off verifiability issues from AfD into a separate UfD we can give more time to find sources (14 days) and make out-of-process "keep" arguments (and "delete" arguments, for that matter) stand out more. Unlike the CSD proposal, it would also permit explicit consideration of the quality of the sources, which will lessen the moral hazard of fraudulent sources. Overall, we will make better decisions, and that will improve Wikipedia.
• That proposal is consistent with supporting the CSD for bios of living persons (including bios that lack enough context to know whether the person is living), because of the great harm that such articles can cause. Robert A.West (Talk) 12:06, 24 November 2006 (UTC)
• Yes, I've read option 4 in more detail and while it may not be the best option for AfD according to some, I think it would be an excellent way to handle unverified info, provided there's enough people working to get sources. - Mgm|^(talk) 09:07, 27 November 2006 (UTC)

Enforcement of Wikipedia:Verifiability

An alternative approach is to move to a more active enforcement of existing policy - Wikipedia:Verifiability.

Without any source of any description, an article does not demonstrate that the subject matter actually exists. It would be entirely consistent with WP:V to ask for a source. This can be done politely - eg "thanks for your article, I'd be grateful if you'd add to it how you know that information by adding your source to it" - and in most circumstances, this will provide a positive response. And indeed, why shouldn't it? It shows someone has read their work and wants to know a little bit more.

If the author refuses to provide a source (or no longer edits), then two options are available. An editor may choose to find a source himself (even though, as WP:V makes clear, he is under no obligation to do so). Alternatively, he may blank the article as it contains information that has been challenged. Once blanked, of course, the article is eligible for deletion under another criterion.

As noted above, this approach is entirely consistent with existing policy. And as long as it is done politely, and couched in positive terms, encouraging people to improve articles and the quality of their edits, fears that inappropriate decisions will be made or people will be put off contributing will be allayed. jguk 12:49, 24 November 2006 (UTC)

• While an editor is under no obligation to source an unsourced article, doing so is a good way to improve Wikipedia, so it should be encouraged. I am a little doubtful about blanking the article as a process, since it is hard to distinguish from simple vandalism. It also doesn't get more eyes on the matter. I don't think anyone looks forward to deleting accurate articles about notable subjects for which sources exist. No matter what we do to improve compliance with Verifiability, that will happen sometimes, but we want to minimize the incidence of that result. Having more eyeballs is important to that effort. Robert A.West (Talk) 13:24, 24 November 2006 (UTC)

I've written {{db|Contains nothing but unsourced statements about a BLP, which could all be deleted under current policy (CSD A3)}} before, and had the article deleted. I'm not sure if this is entirely within the letter of CSD or BLP, though. If the BLP rules were changed to any unsourced statement rather than just negative ones, the trial run failing to reach consensus above could be changed to an A3 deletion - on existing articles with no lag time! --ais523 13:39, 24 November 2006 (UTC)

Agreed with Robert A West-while jguk's method would technically be correct, getting the article in front of a bunch of people who are specifically out to find sources would be highly beneficial, and two weeks would be a reasonable period of time after which to presume that if no sources have been found, none are to be found and the article should be deleted. Also, tons of blanking would drive AntiVandalBot nuts. Seraphimblade 13:28, 24 November 2006 (UTC)

It would be easy to set up categories by date to highlight where a request for sources has been made. Anyone would then be able to see what has been requested and when - and anyone would be free to add sources for such articles if they so wished. A key point, however, would be to encourage those authors who are not currently providing references to their new articles to do so. So I wouldn't help them immediately by adding them. After a week of inactivity, however, I agree that getting more eyes in to help to see if the article can be sourced is a good idea. jguk 13:35, 24 November 2006 (UTC)

• 90% of the time when I ask the editor of an unsourced article for sources, the result is no response. Blanking is no good, as SeraphimBlade points out the VandalBot issue. I prefer instilling sourcing into WP culture, though that is very tough. ColourBurst 16:15, 24 November 2006 (UTC)
  • When I've asked for sources for edits (rather than new articles), I've tended to get over 90% positive response. So it's interesting to hear that ColourBurst gets an opposite result for new articles. I think the way forward is to give it a bash (up to possible blanking/deletion of articles) and see how things progress. If it looks as if gentle prodding just isn't going to work, we can reconsider before making any mass blankings (especially as the whole idea is that there shouldn't be mass blankings or deletions if the approach is working). jguk 16:26, 24 November 2006 (UTC)

• • (edit conflict) I have found the threat of AfD an effective goad in some cases where I suspected that an article could be sourced, but the sources were going to be hard-to-find, and the editor was uncooperative. I've been called six kinds of bastard, but the eventual result was a well-sourced article in each case. Robert A.West (Talk) 16:29, 24 November 2006 (UTC)

Comments by Rick Block

This proposal is fine as far as it goes, but given the experience of the great image purge (was it a year ago?) my guess is if we ever extend it beyond "newly created" articles we'll end up with thousands of pissed off editors. Tagging an article someone created two years ago and then giving them 14 days to provide a source after which the article is summarily deleted is really not that different from bot-deleting articles (as user:Elaragirl suggests in jest above). We have at least tens (if not hundreds) of thousands of articles written by users who are no longer active. I'd be OK with this policy (which is more similar to PROD than CSD) with the additional proviso that tagging an article (even a "new" article) as unsourced when a source can be readily found should be treated as vandalism and the procedure for deleting an article so tagged should include at least a perfunctory attempt to find a source. If you're not willing to look for sources, don't go around tagging articles as unsourced. If you're not willing to look for sources, don't go around deleting articles for being unsourced. IMO, making "sourceless" a PROD-style criteria for deletion of newly created articles is defensible, but allowing editors to apply such a tag without requiring them to try to fix the issue is a horrible idea. -- Rick Block (talk) 23:58, 27 November 2006 (UTC)

Moving forward

Since this criterion seems to be losing steam/going nowhere fast/choose your own analogy, please allow me to advertise a way to extract some good from the ideas generated here. {{unreferenced}} now includes a date parameter for sorting purposes, maintaining backwards compatability (both {{unreferenced|Foo|date=Bar 2006}} and {{unreferenced|date=Bar 2006|Foo}} will have the same result).

Thank Rich Farmbrough for this new functionality and have fun; usage of the new parameter sorts the article into subcategories of Category:Articles lacking sources by month and out of the parent category, allowing the oldest problems to be identified and dealt with first. Personally, I'm trying to backtrack when the tag was added so we don't end up with 33K as of November 2006 articles "sorted", but maybe we can get a bot to run that gamut eventually. -- nae'blis 06:33, 28 November 2006 (UTC)

Triage

It may be worthwhile to step back a second and consider this in more generality. I have now looked through all the sample group; and I have the following modified triage to propose, depending on what should be done with the article, independently of this proposal.

• Class I: Articles we would keep, sourced or unsourced. This includes Wikipedia navigation aids, like 263 (the year) or List of crossings of the Saint Lawrence River. For some of us, at least, simply connected space and Koelreuteria_elegans are also Class I.
• Class II: Articles we would keep if sourced, delete unsourced. I don't see very many of these.
• Class III: ASrticles that should be deleted, sourced or unsourced. There are some articles, like The Alien Within, which we should be able to agree are Class III; one of the sample has already been prodded.
• Class IV: Unverifiable articles, which either are hoaxes, or are so obscure that they have no verifiable sources. These should be deleted, but are really very rare.

The chief virtue of these proposals are that they deal with Classes II and IV; deleting some Class I articles, and keeping some Class III's, are costs, which the proponents see as minor. But I doubt any article in this sample belongs in Class IV, and I don't see any unambiguous Class II's.

We will differ on the line between Class I and Class III. For example, some people will say that all album lists should be deleted on sight, and I could be persuaded to go along; but, if an article is an album list and it is sourced (and almost all of them can be, from Amazon, or Schwann's, or some such), why should this save it?

Comments?Septentrionalis 17:44, 28 November 2006 (UTC)

I fail to see how implementing this would result in more class III articles not being deleted, things with refs can still be deleted under other criteria. Classes I and III are more or less immune from this policy --T-rex 19:07, 28 November 2006 (UTC)

We may be agreeing at length. But this is the wrong tool for class III articles, like The Alien Within. Making that a sourced bit of cruft, instead of an unsourced, is no help to the encyclopedia at all. I think it can be dealt with by prodding (and AfD as backup) on grounds of notability. If it can't, we need a tool to deal with cruft, sourced or unsourced. Septentrionalis 22:57, 28 November 2006 (UTC)

I think at least some of the proponents have a different primary advantage in mind. They want to move the culture so that we don't have new articles created without sources. (They really don't care about good deletions, though that would be an extra benefit.) While I happen to agree with this culture shifting goal, I don't think this is the right path to that goal. GRBerry 23:21, 28 November 2006 (UTC)

What would be? Certainly just putting WP:V and WP:RS in place haven't done it-and a lot of new editors I've spoken to didn't even know you're supposed to source things. This will make it very clear that we're as serious about sourcing as we are about notability and other policies and guidelines, while simultaneously offering an ample amount of time for sources to be found or a novice editor to request help. I've had the opportunity to help several new editors learn to contribute better-but generally the only time they care that we have guidelines to what should and should not be done at all, is when the prod/speedy/AfD tag goes up. All of a sudden, there's an incentive to learn! This would be no different-it would encourage sourcing without deleting articles in an unreasonable time frame. If one can't find something that mentions a given subject within a couple weeks, then chances are none are to be found. Seraphimblade 23:28, 28 November 2006 (UTC)

I've advocated, once or twice in the swamp of text above, starting with a sourcing based pseudo-AFD process. (The original proposal, although labeled as "speedy deletion criteria" is in fact a pseudo-PROD.) I think the proposed text #4 would meet this concern. But I don't think that it is better than AFD however, so why add it? We are getting AFDs closed solely on the basis of sourcing. (To take an example that is under deletion review right now, see Wikipedia:Articles for deletion/Gay Nigger Association of America (18th nomination), which was a purely sourcing discussion.) GRBerry 02:26, 29 November 2006 (UTC)

I think that goal is profoundly unrealistic. Changing cultures requires engineers of the human soul, and we haven't got them. Even if we did somehow change the culture of regular editors, the creators of many of these articles are newbies, who are gone and beyond the reach of our culture-changing. Septentrionalis 06:08, 29 November 2006 (UTC)

Comparison with CSD I4 (unsourced images)

The last section brings up an interesting point: if this CSD becomes policy, will it change new users' attitude to sourcing? It may be interesting to compare this with no-source deletion for images, which is enforced by bot, and has a shorter lag time. Although many new users find image upload confusing (and OrphanBot gets a lot of hate mail), it has lead to a big improvement in the sourcing status of images; it might be interesting to check the change that this has had on the new-image sourcing rate (although I can't think of a way to check that right now). The lag time for images is also shorter (CSD I4 has a lag time of seven days). The main difference is that image sourcing is primarily a copyright matter, and that article sourcing is a verifiability matter, but verifiability (and verifiedness, for that matter) is still one of the core Wikipedia policies and could do with being enforced. Any opinions? --ais523 10:47, 29 November 2006 (UTC)

I think the comparison is interesting but of somewhat limited use. As you say, the primary driver for image sourcing was copyright compliance. It was an urgent legal problem recognized by the community as a whole as a clear and present threat to the project. The image sourcing standard was driven from the top and had wide support that such draconian measures were appropriate. Verifiability and sourcing, while important to an encyclopedia, do not represent an immediate legal threat. The community members (even the newbies) recognize the difference and will react differently. Rossami (talk) 14:25, 29 November 2006 (UTC)

If we're talking about information which is potentially libelous or slanderous, we sure are talking about an imminent and very real legal threat. Of course, not all unsourced information is potentially libelous or slanderous-but then again, many untagged images probably aren't actually copyright threats either. However, we certainly should err on the side of caution in either case. Seraphimblade 22:19, 1 December 2006 (UTC)

WP:BLP is already policy; we don't strengthen it by adding a guideline. Indeed, we may weaken it; if a completely unsourced article goes through any of these processes and is saved by a source verifying the existence of the subject, it will be defended more strongly against removing possible derogatory statements.

For the record, none of the unsourced biographies in the sample is derogatory; this is a small problem, which we already have multiple solutions for. Septentrionalis 17:49, 2 December 2006 (UTC)

Primary sourced articles not covered?

Maybe I missed something here, but I think this proposed policy should take into account the large number of articles that are about novels, television episodes, etc. where the subject at hand happens to be the only source available. For example, take the article Saint Errant which is a stub for a short story collection (which I'll be expanding in the near future). Would this article be deleted because there was no third-party source writing about it? Of course if I were to add something to it such as "This book is believed to have sparked World War II" then that's different, but on the surface this policy would seem to be discriminatory against articles on works for which there are no additional sources available. Same goes to the TV episodes; articles that basically refer to the episode itself as the source shouldn't be penalized because someone hasn't written a book yet on the episode of Jericho that aired 20 hours ago. I understand the rationale behind the policy, and I appreciate the inclusion of the phrase "The term reference is to be used broadly" (this will help me in a dispute I have over a couple of articles of mine that have been marked unsourced despite the presence of a bibliography), but without some notwithstanding clauses this policy could end up with many deserving articles being deleted or disrupted. (PS. I'm aware that the term "primary source" is often used when addressing Original Research issues; this is not related to that.) 23skidoo 21:35, 30 November 2006 (UTC)

If there are no third-party sources to be had on a subject, then it is not a worthy article. —Centrx→talk • 04:31, 1 December 2006 (UTC)

That statement is beyond absurd, Centrx. I have the novel The Immaculate Conception right here on my desk. The only third-party references I've seen do nothing but give the title and a couple of sentences of summary (there may be further resources in French of which I am unaware). As a finalist for at least three major literary awards in two languages, it is worthy to have an article by the Novel Project's notability standards. Does Wikipedia have to wait three or five or ten years until some university professor writes an article about it? How useful and current is an online encyclopedia that has to wait for several years for third-party sources to appear?--Ibis3 16:59, 1 December 2006 (UTC)

The statement's not absurd, it's official policy: Wikipedia:Verifiability: "If an article topic has no reliable, third-party sources, Wikipedia should not have an article on it."

In terms of the book to which you refer, if it was a finalist for literary awards in two languages, that information had to come from somewhere, and certainly belongs in an article about the book.—Chidom ^talk  17:37, 1 December 2006 (UTC)

You mean to tell me that any sort of fanfiction or self-published book can be on Wikipedia? That's essentially what you're saying when you say that books/films/etc can source themselves. ColourBurst 18:50, 1 December 2006 (UTC)

No, they should be AfD'd for notability, as should at least one of the books in the sample. But this proposal doesn't addres that problem. Any source includes the book as a source on its own content. Septentrionalis 20:25, 1 December 2006 (UTC)

Books in general and in our case "Novels" in particular are their own best reference and statements of fact about content should be deemed verifiable from the book's content, i.e. they are intrinsically self referencing. However I agree any statement that is POV should be sourced from 3rd parties and verifiable. Also references of all types should be strongly encouraged. If this set of policies go through as they currently are I can see a deletion mayhem taking place and "many many" editors being so hacked off that they leave Wikipedia entirely. I can see that happending to me. There are just too many "barrack room lawyers" that editor by "policy" and treat a "used broadly" as a license to delete. Is there anyway to mitigate these risks and still manage to increase the level of referencing that editors use. :: Kevinalewis : ^{(Talk Page)}/_(Desk) 09:14, 1 December 2006 (UTC)

In the case of Books, the primary source should be the book itself. Original research in this case would refer to unsourced interpretation or literary criticism. I am currently drafting a proposal for Importance Assessment for the Project in which I am also stressing third-party sources and verifiable evidence in the assessment process, but to demand that the source of an article on a book be something other than the book goes against the whole scientific rationale of Wikipedia's policy on sourcing. It is a well-established scholarly principle that the primary source document is always superior to secondary sources. The first source for an article about a book should be the book, the first source for an article about a movie or television episode should be that movie or episode. To have it any other way is to diminish not increase the credibility of Wikipedia.--Ibis3 16:59, 1 December 2006 (UTC)

Wikipedia:Verifiability is official policy; it states that "If an article topic has no reliable, third-party sources, Wikipedia should not have an article on it."

In this case, what's available from the primary source—the book or novel itself—would be a plot summary; not commentary on or critique of the book/novel itself. If the article's only content is a plot summary, it's not encyclopedic. If the book is notable, someone must have written a review of it or something about it somewhere. If there is information about the book/novel from non-trivial, published sources, that information should be discussed in the article and its sources given. Bear in mind NPOV; unless the work is universally held to be either good or bad—I doubt there are many detractors of Les Liaisons dangereuses—a balanced representation of the opposing viewpoints should be given.

There is usually some form of professional review on a book included with its listing at Amazon.com. (However, the reader reviews at Amazon.com would most likely qualify as "trivial".) Since the Wikipedia:Notability guideline is to provide multiple non-trivial published works, additional material would need to be found and referenced.

Once the article contains more critical information or commentary, the addition of a plot summary could be an option. However, unless the plot summary was crucial to understanding the topic of the book and what is being referenced in the reviews/etc., it might be best not to give a detailed plot summary; a brief description of the book, more along the lines of what is found on the inside jacket flap or back cover, might be more appropriate. If a detailed summary is included, the appropriate spoiler template should be added.

I could probably make a case for considering plot summary written entirely by an editor or editors familiar with the work, instead of being drawn from a published synopsis, to be original research. In my mind, an argument could also be made for deeming "statements of fact" drawn from the work by an editor or editors original research. While that may seem too extreme, it's not really up to us to read a book and include our evaluation of it or commentary on it in an article; we are to read what others have to say about it and compose an article based on that information.

In theory, a good editor and writer should be able to research and write an article on any notable topic—even one the editor knew nothing about beforehand (including a book they had not read). This is why the topic needs to be notable to start with—so that there are already materials pertaining to it to be used as references. (This isn't so far-fetched—it's basically what I was often asked to do in high school.)—Chidom ^talk  17:31, 1 December 2006 (UTC)

I think it's a mistake in what is written in WP:V, which I have corrected. WP:Verifiability is, and should be, about verifiability. Notability issues are, and should be, discussed elsewhere. If a source is not a third-party source, there may be question marks over its reliability, and we will, in practice, be more rigorous in assessing its reliability before using it. But there is, in practice, no bar to using non-3P sources provided that they are reliable for the claims being made. jguk 17:49, 1 December 2006 (UTC)

It was no mistake -- there was significant discussion and broad consensus for that exact wording before it was added. The debate is available in the archives. The only signficiant debate was whether the requirement of sources independent of the subject of the article was so obvious that it was not necessary to state.

Obviously, consensus can change, so feel free to propose removing it, but this was not a change made lightly.

In any case, this is irrelevant to the CSD proposal, since any sources that are not obvious garbage would save an article from speedy deletion. Robert A.West (Talk) 18:23, 1 December 2006 (UTC)

WP:V is moronically contradictory on this note, but in the event of a book/movie/piece of music, there's not a single one that doesn't have some sort of third=party verification, so the whole thing is rather moot. Besides, this criteria is not supposed to take into effect the type or quality of the source, so it's moot on two levels. --badlydrawnjeff talk 19:06, 1 December 2006 (UTC)

There is no contradiction, but that discussion belongs under WT:V, not here. Robert A.West (Talk) 19:20, 1 December 2006 (UTC)

For this particular proposal, any sourcing which is even marginally relevant would prevent automatic deletion by this process, just as any dispute whatsoever to a prod (even if it's just "I like this article, please keep it!") prevents deletion under the prod process. Realistically, though, if no reviews or anything of the book are to be found, the book might be verifiable, but it is probably not notable. So, citing the book as a source on itself would prevent this procedure from taking place-but if that's all there is, it's highly unlikely to survive AfD. Also, if the article makes no assertion whatsoever of the book's notability, the article can be speedied anyway as a non-notable product. Seraphimblade 22:26, 1 December 2006 (UTC)

If the book has been written about elsewhere in published works, it qualifies as notable; that's also the requirement that has to be met for Wikipedia:Verifiability—that it has been the topic of multiple, reliable, published sources. I drafted an article: The Immaculate Conception based on the references I could find online. I found a fair number of references; I don't have access to the book. It needs work, but I think it's a good stub, at least.—Chidom ^talk  22:50, 1 December 2006 (UTC)

UfD Un-nessecary

UfD to me reeks of m:Instruction creep. Why would we need a whole separate section when AfD is already handling such existing cases. Yes, we may receive some extra workload, but in the past we always have been and are able to handle them. - Mailer Diablo 05:48, 2 December 2006 (UTC)

While I wholeheartedly agree with the intent and philosophy behind this proposal, I agree with Mailer Diablo that a new deletion process is superfluous. Especially now that nae'blis and Rich Farmbrough have added a date parameter to Template:Unreferenced, our existing tools should be sufficient to deal with this problem. -- Satori Son 16:27, 2 December 2006 (UTC)

The single advantage of UfD is that it provides a longer comment period. This is important because we want to encourage not just regular 25hr/day users, but the once a week users to participate, because they qare a key source of new content. If they find their stuff deleted the next time they come, we will never see them again.DGG 17:21, 3 December 2006 (UTC)

If that is truly a serious problem, then the best solution would be lengthening the period of debate for AfD, not creating an entirely new, mostly redundant deletion process. Please note that I am only referring to WP:CSDUA#Proposed text #4, or "UfD", not the entire proposal. -- Satori Son 18:18, 3 December 2006 (UTC)

Comment by Roninbk: (How is a 14 day waiting period a speedy deletion criterion?)

Apologies for skimming the debate, since it is FREAKING HUGE, but here's how I'd do it.

• Any editor can tag an article for UfD. 14 day clock should start from tagging, not article creation.
• Any unsourced article should be fair game. No grandfathering. If you're gonna do this, do it right, and flush some of the existing crap off the Wiki.
• No additional comment should be needed, we know why you're tagging it.
• The article template should be clearly worded so that the author isn't scared off from Wikipedia.
• The article should be put into a category for patrollers to find and help out if they can. (see WP:WPPDP)

So basically, it's a specialized PROD.

Only thing I am unsure of, is what to do with removed tags. Precedent from PROD would kick it over to AfD, but I think we want to make AfD less cluttered. Kicking over to Speedy would be a bit drastic. Maybe replace/revert the tag once with a vandalism warning, and speedy delete + warn2 on the second offense?

In any case, that's my $0.02. --RoninBK_ETC 16:26, 3 December 2006 (UTC)

Improve not delete

I agree sources are needed for controversial topics but they should not be OBLIGATORY for the run-of-the-mill "Issum is a town in Germany 20 km from X; population Y.".

Those people who think sources are absolutely necessary for every article would IMO be better employed just getting on with finding sources for those articles that lack them rather than creating yet another rule - which wont deter the malicious few (who will create a false, foreign-language source that will probably go undetected) but will catch the many, non-controversial, stubs by well-meaning amateurs who are the backbone of wikipedia.

Demanding a source for every article is raising the bar way too high for a project like this which should not REQUIRE the standards expected in professional publications.

If you can encourage people to add sources as a matter of routine that's great, but if they don't, they don't. No reason to delete non-controversial stubs. If it offends you in a particular case, then please find a source and add it yourself, rather than invoking a new procedure like this. Jameswilson 05:04, 7 December 2006 (UTC)

Couldn't agree more. I've even seen people asking for sources for the principal page of a composite article where all the sources were in the subpages. I've seen AfD debates asking for sources of each event in a fictional story or show for the article about that story or show. I confess I have once written a posting in a major listserv to serve as a source to keep an article from deletion. The word used in WP's definition is "verifiable" not "verified," meaning verifiable if necessary. Out of laziness, we are using "sourced" as if it meant "accurate." Articles vary: it is probably necessary to ask for at least one outside source for a contemporary bio. to prove existence. Products are difficult either way: it is necessary to show they exist in reality, but there may be no source other than the producer. DGG 05:22, 7 December 2006 (UTC)

Absent a source, how did the Wikipedian who added that information know that Issum is 20km (not 19km) from X and has population Y (not Z)? If he/she had a source, what is the conceivable objection to naming it in a references section? Writer's cramp? As for "creating another rule," the rule exists, and has existed for a long time: it is called, "Verifiability." Every article, even a stub, should contain at least one fact significant enough to deserve an explicit source, or why bother? If an editor creates a decent stub and does not disclose the source, he/she is foisting the job of looking it up and adding the source onto someone else, and that is unfair and unhelpful.

As for articles with no possible source other than an interested party (producer of a show, subject of a bio, maker of a product), how are we supposed to write an NPOV article based on self-published and undoubtedly self-serving information? The answer is, we cannot. The neutral point of view is the irreducable, non-negotiable requirement for any Wikipedia article: Topics that are not well enough documented to permit that have to be out of scope. (BTW, creating a listserv posting to justify an article or edit was creative, but astute Wikipedians should have noted that the source was self-published, and therefore not reliable.)

As for academic fraud, it is the second-worst offense against the project that I can imagine. The only things I can think of that would be worse are crimes or torts motivated by edits on Wikipedia, or using Wikipedia as a means. Is there any reason that an editor committing academic fraud should not be permabanned? Robert A.West (Talk) 08:27, 7 December 2006 (UTC)

I don't think a lot of people understand what plagiarism is. I think someone who, upon being corrected, is willing to take the advice and ensure to do better in the future should be forgiven a beginner's mistake, even if that is blatant plagiarism. On the other hand, those who should know better and do it deliberately should most certainly meet some type of consequences. I think, however, that "We absolutely require sources, or the information gets removed or deleted" is an important policy. Further, if no one has seen fit to study the events of a fictional series, why do we have it here? If one cannot cite a source, a plot synthesis based on the editor's own perception is original research. Same with some type of product-I agree, if no one has written about it besides those with an interest in promoting it, how could we possibly write something objective using that source? And why would it be notable anyway? Seraphimblade 09:59, 7 December 2006 (UTC)

I should have been clearer -- I meant fraudulent citations of non-existent or irrelevant sources, as proposed above. I was not speaking of dire consequences for someone who quotes without citation or something of the sort. I apologize for being unclear. Robert A.West (Talk) 10:38, 7 December 2006 (UTC)

Robert, its a matter of trusting people who take the trouble to create an article. If a resident of Issum ( or anybody else) bothers to create an uncontroversial stub on Issum, we should start from the assumption that they know what they are talking about. Innocent until proven guilty. There is no reason to get all paranoid and demand proof for everything. If a stub makes assertions that seem odd, then put a fact tag or whatever on it. This is not a legally-binding document about Issum, its just basic info.

Anything else would be seen as insulting by many contributors who are not accustomed to the rules of the academic world where people accept having to justify everything. 99% of articles are created by non-academics who dont live in that world. Granted it doesnt take long to find a source but the point is that most people "expect to be believed" and will rightly (in their terms) get very annoyed at being told their word is so suspect that the whole article (not just one fact) should be deleted unless they back everything up.

I've perhaps put the case in a rather extreme way, but my basic point is please dont seek to impose the standards of academia on wikipedia contributors. For a wiki user-friendliness is paramount, even if that is sometimes at the expanse of other concepts such as comprehensive verification. Genuine factual errors which slip in as a result (maybe Issum is 25km from X not 20) can and do get corrected later. This proposal would compromise the project by deterring "ordinary people" rather than strengthening it. Jameswilson 00:39, 8 December 2006 (UTC)

(Indent reset) The trouble is, no one is getting accustomed to it either. The current "Cite sources, we'd really like you to, pretty please?" is just not getting the point across-and citing sources is critical to correcting factual errors later! I think the "speedy deletion" bit in the header is really deceptive-in fact, I'd like to get rid of the "speedy" bit altogether. This deletion process, in fact, would give a far longer time even then the very generous proposed deletion process, and would only require a single marginally relevant source cite to prevent automatic deletion after the 14 days. The finder of a source need not even be the one who originally wrote the article-I would imagine, should this go into effect, that a "source patrol", similar to the prod patrol, will form quite quickly to go find sources. 14 days is plenty of time, especially since as proposed this will only apply to new articles-articles written before this goes into effect would be immune, at least while this process is being tried out.

As to "not hurting someone's feelings", WP:V is pretty clear on that-remove unreliable information whether it does or not. This is a pretty logical extension of that. Have I had someone irritated and upset because an unref tag went right on a new article, or because an unsourced and potentially problematic assertion got removed? You bet. Am I sorry I did it? Not a bit-the lack of sourcing was a real problem which needed to be addressed. Some people even get mad if you put a stub tag. This does not mean that we shouldn't place them where needed-and the fact that requiring verification might hurt someone's feelings shouldn't stop us doing it. In fact, we already do-it would be perfectly justifiable to flat out blank any article with no sources, "unsourced information may be challenged or removed by any editor", and to subsequently nominate the article for a real speedy as blank. I don't think that would be a good solution, however, and I don't think our current asking-but-not-enforcing is working either. This seems an excellent compromise. Seraphimblade 01:07, 8 December 2006 (UTC)

I find the "improve, don't delete" argument objectionable. It is all too often used as an excuse to not do anything about unsourced (and often unsourceable) articles. I provide sources for every article I start or expand, and have searched for sources to save articles others wanted to delete, but I really resent being told that I am also responsible for finding sources for all those poor-quality unsourced articles that I think should be deleted. If everyone would just note the sources for their edits, we wouldn't have any problem. Wikipedia:Verifiability is quite clear on this; "The burden of evidence lies with the editor who adds or restores material." -- Donald Albury 01:54, 8 December 2006 (UTC)

But as I see it there's no problem with using it as an argument to not do anything about unsourced articles immediately. A major problem with speedy deletion/AfD in my opinion is the strict deadlines they impose; I simply don't see the need in most cases for an article's problems to be resolved within a matter of minutes or days. When I find statements in articles which I consider dubious but not obviously wrong, I flag them as needing citations and then I sometimes wait for months before I get to the point where I simply remove them. What's the rush? There are lots of other things to do in the meantime. Bryan 02:45, 8 December 2006 (UTC)

Here's the problem, though-we're not talking about articles which simply have problems. No one's suggesting attaching deletion to the cleanup tags, or allowing "Not written very well" as a deletion criterion. What we are talking about is articles which violate a core policy (WP:V). We're not talking about all the article's problems being resolved within days-simply the problem of having no sources whatsoever. Even inadequate or questionable sourcing would prevent this automatic deletion, and the process would indeed be a slow one-two weeks to find one source. I don't think that's placing a "rush" at all-if a subject is so poorly covered that a single source cannot be found in two weeks, it likely doesn't belong anyway. Seraphimblade 03:01, 8 December 2006 (UTC)

Seraphimblade, you;'re assumming experienced editors who know how to play the game. It is not obvious that it is much more desirable for an unfinished article to have totally wrong, made-up, or irrelevant sources, than to have none at all. People do not realize that a useless link may never be followed by anyone who cares to go back to the article or talk page and comment. People don't know the policy here, that V really means plausible, rather than accurate..

I know the above sounds a little sarcastic, but it is meant to get attention for my real proposal:

That nobody be allowed to place a V tag or a speedy delete or a prod or an AfD unless they actually documented the sources they tried and failed to find information in, or the places for checking notabilty or whatever, or the evidence that something was commercial spam instead of the bare assertion.

We should return (or start) being an editing community. I'd liketo start seeing people list on their pages the AfD candidates that they had successfully improved, or the prod tags had been able to remove because they themselves had found what was missing. We are--after all--editors not censors. And the admins are administrating the editing--not setting up as master censors. DGG 05:27, 8 December 2006 (UTC)

I am inclined to agree with DGG. At the moment it feels a great deal like systems are being set up to criticise and censor what is being added to Wikipedia, rather than systems being put in place to improve Wikipedia. The phrase "improve, don't delete" has been criticised, but surely as editors that is what we're all trying to do? Improve the encyclopedia as a whole. Most reasons for deletion do already seem to be covered by existing protocols. This one seems entirely unnecessary. Silverthorn 14:49, 8 December 2006 (UTC)

I agree with "improve, don't delete!" This is why this proposal allows ample time to improve (by adding a source), before concluding after two weeks that such improvement is either impossible or not going to happen. At this point, such an article is unlikely to survive an AfD in any case. This would simply streamline the process and take some strain off AfD. There has to be some "censorship" as to what comes in-and WP:V is clear that unsourced information is one of those things which should be subject to challenge and removal, just like spam or attack pages. This would simply make a logical and coherent process for doing so. Seraphimblade 19:38, 8 December 2006 (UTC)

Comment by Errabee

Hmm, while some topics should definitely be referenced, I feel this proposal is too much. Will we next delete teapot as well because it is unreferenced? Errabee 00:55, 9 December 2006 (UTC)

Everyone has different examples. This is one of the cases where there could be references (anything collectable usually has them)--there are certainly books about tea-making and about pottery--so I just looked-and Amazon showed a few specific ones. There certainly are websites--and even a museum. I just added the refs. This is what we mean by improve --it is just as easy and better and more fun to do this than to write a AfD or the like. (I did not, however, source the individual statements in the article). DGG 02:57, 9 December 2006 (UTC)

There are statements in teapot that are disputed on the talk page and could really use references. There's an argument that it could collect sources if pressed, however. ColourBurst 00:28, 10 December 2006 (UTC)

Of course there's information out there about teapots. This is an example of something it would be ridiculously easy to source (and has been done). If this proposed template had gone on the article, sources would have been readily and easily found, and no deletion would occur. The idea of this is only to delete stuff which is not only unsourced, but unsourceable, and for the rest to simply give the message "Sourcing is not optional." We already have a rule stating that sourcing is not optional-why not make a logical extension of that so it means something, instead of having it amount to "Hey cite a source? Please? Pretty please?" Seraphimblade 21:21, 11 December 2006 (UTC)

Would you please let me know which rule that is? I have never found a WP policy that says "a source must be provided when information is added to an article." In particular, I just checked WP:V and it doesn't include that sort of requirement. CMummert 11:35, 12 December 2006 (UTC)

"The burden of evidence lies with the editor who adds or restores material. Any material that is challenged or likely to be challenged needs a reliable source, which should be cited in the article. If an article topic has no reliable, third-party sources, Wikipedia should not have an article on it." WP:V#Burden of evidence. One could, in theory, have an article consisting entirely of statements that no rational person would challenge, but I doubt it would be worthwhile. Robert A.West (Talk) 12:11, 12 December 2006 (UTC)

Yes, but that is not the same as "Every new fact must be provided with a source when it is added to the article." Although information must be verifiable, it is not required to be sourced unless challenged or disputed. This distinction is present in both WP:V and the proposed WP:ATTRIBUTE. I believe that several editors on this page make implicit claims that WP:V requires citations when it does not. This is not to say that I am against well-referenced articles; I am just pointing out that references are not required by policy unless the material is challenged. CMummert 12:23, 12 December 2006 (UTC)

Comments by BirgitteSB

I think this policy would lead to people grabbing any ISBN off of Amazon and saying it is referenced from there. It will be much harder to root out spoofed references than it is to work on the unreferenced articles the way they are now. Even though referencing these seems overwhelming I think this policy would create a much bigger problem. At least now we can have some faith that if people went to the trouble of referencing something then they must have had it at hand. I think the policy sounds like a good idea, but is completely impractical.--Birgitte§β ʈ Talk 21:43, 10 December 2006 (UTC)

I am actually horrified at the assertion that Wikipedians would commit fraud so lightly. Isn't anyone who would be capable of such a thing worse than useless to the project? Falsus in unum, falsum in omnibus. Specifically, if I cannot trust someone's citations, then it would be madness for me to trust his or her unsourced assertions. Robert A.West (Talk) 05:36, 11 December 2006 (UTC)

I am actually horrified at the assertion that Wikipedians would commit fraud so lightly You do realize this is a project anyone can anomynously without necessity of learning the rules first. I don't imagine any well-established Wikipedians would use such tactics but generally editors don't like to see their work deleted. I can easily see someone thinking "I am an expert on Foo. I know what I am talking about and according the reveiws to The Complete Guide to Foo on Amazon that book has all the basic info on Foo. Certainly it have the common information from my edits, which are certainly correct. I will use that as my reference." Seriously you should not underestimate the ability of people to delude themselves that doing the lazy thing is not actually wrong. I doubt anyone that would use these tactics would consider themselves as "commiting fraud". While they may realize they are "technically" wrong, they will believe they are helping the project and will expect to be proved that the book does not actually reference anything from the article before they can believe it was truly a wrong thing to do. That is my prediction in any event, sorry if it horrifies you. I believe it is based on good understanding, but you are welcome to disagree; it is certainly highly hypothectical. Honestly a cousin of this idea was discussed on foundation-l and it was a common conclusion that editors would "game" the system. There are certainly editors trying to game the others rule out there, I don't know why this would be different. As to not trusting unsourced assertions by these editors who would spoof citations; that is hardly a reason for this policy as you can at least identify the untrustworthy unsourced assertions easily. Spoofed citations will sit there being falsely trusted until actual fact-checking is accomplished and this project is far a away from real systamatic fact-checking IMHO.--Birgitte§β ʈ Talk 13:07, 11 December 2006 (UTC)

If the majority of Wikipedians, or even a substantial minority, will create fraudulent offline citations to defeat a CSD, what makes you think they will not do so to defeat a verifiability-based AFD? Or to prevent a particular passage from being stricken? Should we avoid nominating unverifiable articles for deletion or demanding citations altogether? That problem is not created (nor will it be fixed) by a CSD criterion.

Don't be sorry that I am horrified -- I am sorry that you are not. Robert A.West (Talk) 18:49, 11 December 2006 (UTC)

Why would I be horrifed at my own assertion? Horrified at the fraud, yes; horrified that I would suggest this policy will lead to an increase in fraud, no. I find this proposal very dissimilar from the other processes you mention. AfD involves several people looking at the article and debating it, the same is usually true of talk pages when removing passages. This policy sets a very low bar to overcome deletion with no forum for discussion. I think the interaction with other editors is significant factor here in how people will react. I do not think this CSD criterion will create an entirely new problem, but it will make a problem significantly worse.--Birgitte§β ʈ Talk 20:23, 11 December 2006 (UTC)

How would you feel if the CSD were limited to biographies of living persons? I see no advantage to tolerating any unsourced BLPs, even if they appear not to be defamatory, but the moral hazard would still exist. Robert A.West (Talk) 21:03, 11 December 2006 (UTC)

Such possibilities of fraud will always unfortunately exist. Any editor may remove unsourced information and demand that it be sourced prior to reinsertion, this could lead to fraud as well. It is already "gaming the system", to a good degree, to write unsourced and unverifiable material and expecting others to figure out how you know what you wrote is true. I do imagine some fraud would be committed as a result of this (or any) system, but I think the good impacts would greatly outweigh the bad. Most editors who are already to the point of creating decent articles on genuinely encyclopedic subjects do want to follow the rules, and simply may not see the need to cite sources. Oftentimes, a polite "Where did you get that information?" request goes unanswered, or is answered with a vague "I'll source it sometime." This generally ends up with the article landing on AfD, and then "I like it/It should stay because it's a.../Don't ever delete anything" come into play. This would make clear that sourcing is a requirement in and of itself, and that no matter how "good" the article may be, it must, 100% of the time, source its information.

Of course, anyone caught committing fraud should be swiftly and harshly penalized-whether this passes or not. Seraphimblade 21:17, 11 December 2006 (UTC)

we are not equipped to determine if fraud has or has not occurred, except in the most obvious of cases. Fraud requires an intent to defraud, not making an error that one believes justified. What we are equipped to do, is for the multidue of editors here to look for themselves and report what they see. Of course this leaves opprtunity for both fraud and--much more likely--innocent error, especially in the less edited articles. This occurs in much more rigoursly edited publications as well. False accustation of fraud are not unknown, and to do so is a libel. WP removes libels when detected, so in a contested case it is necessary to properly decide the matter--and the AfD and similar processes here do not meet even the most rudimentary requirements.

One of the most rudimentary requirement is that the judges of fact can see the evidence, and it is the practice here to remove material in some cases before an AfD discussion.

We are concerned about keeping error out, but we are also concerned about keep good material in. We are in the end dependent upon the good faith of the author or photographer--and their knowledge that anything that justifies true legal action is liable to it. Just like in the RW.

In order to have good material, it is necessary that it be sourced but anyone can try to source it & people should be strongly encouraged to do so. No one should be allowed to remove material without being required to attempt to source it, and in many Afd cases the proposers do in fact say where they have looked.

It is also hypcritical to insist that new material be sourced, while accepting the material already here.DGG 04:54, 12 December 2006 (UTC)

Can DGG please clarify why he/she believes AfD has anything to do with deciding charges of misconduct? I thought it obvious that any accusation of fraud would require an arbitration. I don't see why Arbcom should not be up to the task. Obviously, a frivolous accusation of fraud would itself be a policy violation, and Arbcom is fully empowered to deal with that as well.

The hypocrisy charge makes no sense to me. Policies change to meet new circumstances, and grandfather clauses are routine methods of dealing with the practical problems of those changes. New houses must meet new code requirements, but older houses can wait until someone complains. Wikipedia can tolerate older unsourced articles without condoning them, but anyone is free to bring an older article before AfD. Many articles were created by anons. We decided as of a certain date that this was no longer a practical policy, but the older articles remain. None of this is hypocritical in the least. Robert A.West (Talk)

I had sort of forgot about this page, sorry. I have no problem with this being enacted for biographies of living people. There is an active task force on that subject that I would expect to be vigilent for problems in a way that cannot be expected for the entire encylopedia. I would not want to see such a policy adopted wider than that until we have a sucessful systematic method of fact-checking in place.--Birgitte§β ʈ Talk 14:50, 5 January 2007 (UTC)

Comment by RAN

I see the process of warning the creator, then deleting unreferenced articles as bad policy. Most of the people that worked on articles early on. are long, long gone from Wikipedia. They wrote the articles before referencing was required. I like to think when I am dead, that my contributions to Wikipedia will still be around, even though the rules are going to change many more times over the upcoming years. The traditional AFD process has people getting together to reference the articles, lets stick to that. I have already had several articles of mine on religion threatened with deletion by User:Wikipediatrix who has been using the unsourced rule as a cudgel. She said external links didn't count and demanded that every sentence have a source without her having to show me a counter fact that showed my fact as wrong. We already have several tags to show the level of authentication of articles. --Richard Arthur Norton (1958- ) 06:56, 12 December 2006 (UTC)

This proposal has always applied specifically and only to articles created after its adoption. I don't see how this possibly endangers any existing content. There is a reason to draw a line in cyberspace. At one time, new articles had a high probability of being general information on a topic well-known to the stereotypical Wikipedian. Such articles have a sizeable base of knowledgable Wikipedians who can evaluate for NPOV out of their own knowledge, and those that have stood the test of time are probably reasonably good on the facts. We should source those. Now that we have reached 1.5 megaarticles, the situation is different. More and more, new articles are likely to be subtopical expansions of older articles, to concern specialized topics that few Wikipedians can evaluate readily, or topics that expand our horizons beyond our systemic biasses, and that will have correspondingly fewer editors who can confirm them from personal expertise. This makes it harder to source them after the fact, so insisting on sources up front becomes more important. Robert A.West (Talk) 10:32, 12 December 2006 (UTC)

There are several comments higher on this page (search for "grandfather") that advocate applying the new policy to all articles, possibly after a trial period. So the opinion that old articles will be exempt is not unanimous. CMummert 12:27, 12 December 2006 (UTC)

At the moment, the only proposal on the project page that does not include a grandfather clause is #4, which follows the XfD model, not the CSD model nor the prod model. When discussing a proposal, I prefer to discuss the text(s) on the table. If the procedure is adopted, any extension would not be automatic, but would require discussion and consensus. Robert A.West (Talk) 13:07, 12 December 2006 (UTC)

Comment by Dalbury

I've voiced my support for this general proposal before, but I haven't had time to participate at length in the discussion. I would just point out that I have recently (in the last month) sent several articles to AfD (Wikipedia:Articles for deletion/Second Round, Wikipedia:Articles for deletion/Point of No Return (3LW album), Wikipedia:Articles for deletion/Solange: The Beginning(album), Wikipedia:Articles for deletion/No Turning Back (Christina Milian Album), Wikipedia:Articles for deletion/Structure and Evolution of Chinese Social Stratification, Wikipedia:Articles for deletion/Words whose meanings changed when people misunderstood them in context, Wikipedia:Articles for deletion/The Kelly Rowland Story, Wikipedia:Articles for deletion/Ms. Kelly and Wikipedia:Articles for deletion/I'm Back Again) because they were unreferenced and appeared to be unverifiable. They have all been deleted (or will be in a day or two). I don't see anything controversial about deleting such articles. However, AfD is crowded and messy, and it would help if AfD could be reserved for more substantial issues, and newly created articles with no sources, and for which no one can come up with a source within two weeks, should be deleted without going through AfD. Oh, and note that these articles got deleted faster than they would have un this proposal. -- Donald Albury 22:16, 12 December 2006 (UTC)

Comment by Derex

Laziness. If there are no sources provided, spend a few seconds googling one up. The vast majority of Wikipedia is poorly sourced; there's no call to dump it. The solution is to provide those sources, not to delete the imperfect. If no sources turn up, and requests for sourcing are unanswered, then off to AFD it goes, as has always been the case. Quite often the array of eyeballs at AFD turns up sources. It's just as quick to place a source-request template on a page as a speedy template, and far more constructive. Derex 21:04, 14 December 2006 (UTC)

If you google, you'll get pages relevant to the subject, but quite possibly they won't have been used to write the article and so aren't sources. Sources are used to indicate where the information in the article comes from (which the person who writes the article must surely know), and the CSDUA rule would 'encourage' the person who wrote the article to provide them in the first place. Googling for marginally connected pages and adding them to OR doesn't stop the page being OR. --ais523 12:07, 15 December 2006 (UTC)

If you want to cite 'laziness', then it is laziness on the part of contributors to not cite sources when they add material, thus creating work for the rest of us. Moreover, many of the things that get added are not easily googled. I've spent a considerable amount of time trying to determine whether an edit is reasonable based on google hits. How many google hits am I supposed to open and read before I can conclude that a 'fact' can't be verified from Google? It is far easier for the original contributor to provide a source than it is for a later editor to find an appropriate one. Therefore the onus of 'laziness' belongs on the contributor who could not be bothered to supply a source, not on an editor who questions the material. -- Donald Albury 12:54, 15 December 2006 (UTC)

Comment by Grace Note

No way. This is just a means to have more and more articles deleted by people who do not actually contribute anything much. Let's write a policy saying "if you find an article without sources, have a go at finding some sources", not yet another policy that few people will read but some will use to destroy articles that they personally do not like.Grace Note 10:08, 16 December 2006 (UTC)

I find your comments quite offensive. I wholeheartedly support this proposal, and I have started 169 articles, all of which are sourced (as well as adding sources to other articles). Please do not be imputing motives to other editors. -- Donald Albury 01:24, 20 December 2006 (UTC)

Although Grace Note's message might be offensive to you, the bit about "if you find an article without sources, have a go at finding some sources" is perfectly reasonable. I have the same issue with people blindly deleting images from articles saying "free alternatives are available" and then moving onto the next article without making any apparent effort to actually provide said free alternatives. The more I read about some of the hoops that article creators like myself (and my tally of created articles is probably in the 500-600 range and that's a low estimate) are being expected to go through, the more I think I should start sending Jimbo Wales an invoice for my time. I charge $40/hr, plus GST. 23skidoo 00:49, 22 January 2007 (UTC)

Please remember that the policy is that it is the responsibility of whoever wants to keep information in Wikipedia to source it. To try to shift the burden of finding sources to someone who requests them is, IMHO, quite rude. It is the equivalent of saying, "I'm an idea man, but I'm too busy to do the scut work of finding sources. If you have questions about the material, you have to find the sources yourself, and you're not allowed to remove it if you haven't wasted a lot of time searching for sources." -- Donald Albury 13:24, 22 January 2007 (UTC)

I would agree, and also-there's no one in the world who knows the source of added material better than the person who added it. If the "source" was "I just know that" or "That's how I see it", that person needs to review WP:NOR. Otherwise, why not just ask the person who knows the source, to cite the source? Far easier then anyone else trying to read minds! Seraphimblade 13:36, 22 January 2007 (UTC)

WP:V only states that facts must be "verifiable" meaning they have appeared in at least one reliable source. Only when a specific fact is "challenged or likely to be challenged" does the existing policy require a source to be explicitly provided. This would be the first policy to say that an article with unchallenged, obviously verifiable information must have sources provided - it is not an application of WP:V. The entire point of this proposal is that a lack of sources is not currently a valid criteria for article deletion, so a new policy would be required to make it a valid criteria.

Adding verifiable material that you "just know" does not violate WP:OR, even if no source is provided, so long as the material has appeared in a reliable source. WP:OR only covers material that does not appear in any reliable source, not material whose source is not provided in the article. CMummert · talk 13:49, 22 January 2007 (UTC)

Well the application of this policy and the actual tagging of the article would be challenging the verifiability of the article. In regards to OR, if the material is so readily available in reliable sources then it should be readily available to cite.Agne 19:02, 22 January 2007 (UTC)

I've made that same point all along. If you need to cite the assertion that the Earth is round and orbits a star commonly known as "the Sun" or "Sol", it's easy as can be to cite all of those assertions. I would imagine there are literally thousands, if not millions, of highly reliable sources which back all of those. Try to find me one thing that most people would consider "obviously true", that would not also be trivially easy to source. As to adding what one "just knows"-let us say it tends toward violating WP:NOR, because, as we all know, firstly, memory can be unreliable, and secondly, our heads are not hard drives in which information passively sits, awaiting retrieval in its exact form. We analyze information, we crunch it, we look at it in the light of other things we know, we draw conclusions. And well that we should-as humans, that process is exceptionally useful to us! Unfortunately, when working on this project, working from memory is working with information we are almost certain to have performed some kind of "original analysis or synthesis" on, that's our nature as humans. So, while working on this project, we should work right from the source, and leave analysis and the drawing of conclusions to the source. Our reader, of course, should be left free to draw h(is|er) own conclusions, as (s)he will. Seraphimblade 21:36, 22 January 2007 (UTC)

I would like to point out to Agne that whether or not it is "easy" to explicitly source things is irrelevant - the current policy is that they don't need to be sourced unless challenged, and the point of a challenge is that you honestly suspect the material might be false. Challenging facts that are obviously verifiable (in principle) merely because they don't have a source attached is not acting in good faith, and is generally a waste of everyone's time.

In the context of article deletion, the current deletion policy says that the question is not whether the current material is verifiable or not, but whether any verifiable article at all could be written on the subject. So deleting an article just because no sources are provided would be a great leap from the current policy.

I would respond to Seraphimblade's comments by pointing out that writing articles does not mean just copying facts from various sources with footnotes attached - it means gathering appropriate facts, giving them due weight, explaining what is going on, giving examples when helpful, etc. The difference of opinion we have about working closely from sources may be because we work on different areas of the encyclopedia.

The overall summary of my position is: the fastest way to improve bad articles is to work on editing them, not to work on deleting them. If it is so easy to find sources then why not just add the sources instead of nominating the article for deletion? CMummert · talk 22:53, 22 January 2007 (UTC)

If it's easy to add sources, generally I will, same as I will if it's a trivially easy fact to find a source for (such as the earlier bit regarding the Earth and Sun). However, explicitly citing sources provides many benefits. When I became convinced of its utility was based on my own experience-someone challenged one of my additions, stating it was an urban legend, though a widely believed one. I was quite embarrassed to find out that they were exactly right! If I'd looked at sources first, rather than relying on my own knowledge, that wouldn't have happened-and if that astute person hadn't been monitoring the article, it would be serving to perpetuate that falsehood. Sometimes, something that "everyone knows" is still wrong, or is only half-right but imprecise and distorted. For example, I would venture a guess that most people "know" that the Earth is a sphere, if one were to ask them. The only problem with that is that what they know is wrong-it's ovoid. "It's common knowledge" is the classic example of the argumentum ad populum fallacy. Working from sources also offers the advantage that our readers have a good, complete list of where we got our information. Since, as you stated, we must paraphrase and tie together our sources, that allows future editors to ensure that we have done so correctly. Also, since our objective is to provide an overview of topics rather than in-depth coverage, source citations allow readers to easily follow up with deeper research, if they so desire-whether to further improve the article or for any other purpose. And finally, as I discovered, checking for sources helps us clear up our own misperceptions, inaccuracies, and distortions too! Seraphimblade 03:38, 23 January 2007 (UTC)

(lowering indent level) As I said, it is probably true that we edit different types of articles and that is why we have different opinons here. An article I recently worked on is Uncountable set (which is not great but isn't bad). I added two general references, when the article had none, and edited it. First I made sure the definitions matched the correct ones, then I convinced myself to my own satisfaction that the remaining claims were not only logically correct (by proving them for myself) but were also unoriginal. I left a comment on the talk page for one claim that I did not think was correct. Several other editors were watching the page, and they subsequently edited the article and commented on the talk page. If any fact in this article were challenged, apart from the section that is under discussion on the talk page, I could search out a reference. Therefore, I have no motivation to go through and add inline citations for each fact "just in case" - it's a waste of my time. In less objective articles, it might be more helpful to have more detailed citations, but in this case anyone who pulled one of the general references would be able to verify the content of the article. Arguments that I have heard others make, such as "it might still be incorrect" or "but some readers just aren't able to look at a reference without a page number," aren't compelling to me. In the end, I think that it is the wiki process rather than detailed citations that we have to rely on to produce high quality articles. CMummert · talk 03:58, 23 January 2007 (UTC)

CMummert, please read Wikipedia:Common knowledge. Much of what people 'just know' is not accurate, or sometimes it is just plain outright wrong. One example that tickled me a few months ago was when someone stated in one of these debates that, "of course, we don't need to cite a source for the fact that George Washington had wooden false teeth", as I was immediately able to cite sources that explained that his false teeth had no wood in them. If you can't point to a published source for a statement, you should not be putting it into Wikipedia. -- Donald Albury 02:55, 23 January 2007 (UTC)

The more common complaint I hear about technical articles is that "common knowledge" is too specialized, not that it is incorrect. The point is, if I know that I can point to a reference if pressed, there is no policy that suggests I have to go find a reference before editing, just in case some editor might someday challenge what I write. The goal of WP is to produce verifiable articles, meaning that the material is theoretically attributable to reliable sources, not to produce articles that list a source for each fact they contain. CMummert · talk 03:27, 23 January 2007 (UTC)

But what happens when you are no longer on Wikipedia? What happens when someone comes across an article and becomes curious at something that you originally edited many months (or even years) ago. Maybe they tag, maybe they ask on the talk page requesting some source or follow up. We will have to hope that some other editor would respond and be able to readily verify and source this material that at the time you knew that you could source (but happen not to). Now for some subjects that tend to draw many editors from several realms of knowledge and familarity, this might not be an issue. But for more obscure subjects and areas that require more depth and expertize, this becomes a more pressing issue in that "experts" may not be perpetually available. Agne 08:07, 23 January 2007 (UTC)

The premise of WP is that (1) no article is never "finished" and (2) there is always a supply of editors willing to work on every article. Without those premises, the entire enterprise falls apart. So we already assume that there will always be someone around to work on every article, answer questions, etc. If future editors decide they don't like what I wrote they can change it to their liking. If all knowledgable editors dissappear, lack of sources will not be the most pressing issue.

The level of inline citation I use when writing on WP is quite similar to the level of inline citation I use when writing in real life - and since I think this level is appropriate for papers that will be archived in perpetuity, I think it is also resonable for an article that can be easily changed as time goes by.

You might also be assuming that the number of requests for further citation will be high, but my experience does not bear this out. The requests for citation that I see in practice are for facts that I could already predict were likely to be challenged; I try to add sources for these as I write. I also make an effort to list general references for each article, so that the reader can go read about the subject from an expert if they really want to learn about it. CMummert · talk 12:19, 23 January 2007 (UTC)

I had someone request a citation for a specific fact I added to an article, and I had to reply that it was one of three references I had listed at the end of the article, and I couldn't say which one it was until I could get back to the library to look at the books. As a result, I now try to use in-line citations for everything I reference. As for the volume of requests for citations, I don't make all the ones I would like to, because I want to keep track of and follow up on them, and that takes time. IMHO, there are nowhere near enough requests for citations being made, and Wikipedia is filling up with unsourced 'facts', which may or may not be verifiable. The only way we can be sure that something is verifiable is to have references to check. Every thing that does not cite reliable sources is, as far as I am concerned, rumor. -- Donald Albury 13:35, 23 January 2007 (UTC)

If you knew that the fact was in one of three books, the you knew it wasn't rumor; you're being hyperbolic. It isn't my experience that the science and math articles are filling up with false statements; and a sample taken by Nature magazine showed WP and Britannica to be very comparable in terms of quality. If there are some specific other areas where the articles have problems, it seems to me that the appropriate wikiprojects should set up stricter criteria for those articles rather than trying to make a one-size-fits-all solution. It is good editors, not good references, that make good articles. CMummert · talk 13:52, 23 January 2007 (UTC)

It was 4 inaccuracies for WP versus 3 for Britannica, right? I don't like the idea of Wikiprojects WP:OWNing their domain articles any more than they already are. Maybe it's not a problem for the Wikiprojects you reside in, but there are a few Wikiprojects that get annoyed if you even touch an article without telling them. In addition, that assumes that everything can be classified into a Wikiproject, or an active Wikiproject for that matter (ever see the WP:CSB page? That's a Wikiproject but the material is too diverse to get a modicum of people who are interested and research the material in detail.) ColourBurst 16:21, 23 January 2007 (UTC)

(lowering indent) The Nature article is free (to me at least) online [7]. The fact that the two encyclopedias came out so close indicates to me that the wiki process has been successful at creating articles of value, at least in the physical sciences.

I agree that a wikiproject shouldn't try to prevent outsiders from editing, but participants in a wikiproject are in the best position to ensure that the articles are correct, rather than random editors who happen to come by an article. I can't believe that a large number of articles are maintained by editors completely unfamiliar with their subject material. CMummert · talk 17:01, 23 January 2007 (UTC)

The Nature comparison was very limited. (Oh, and you're lucky, I don't have access to the Nature site.) They only looked at articles on subjects that were covered in both encyclopedias, and that were of comparable size. Try looking at our many articles on popular culture, living persons and places. I've also watched and participated as articles on historical figures (such as Ponce de Leon) have slowly had the crap weeded out and replaced with sourced material. The great majority of our articles are still unsourced or very poorly sourced, and contain a great many unverifiable, or even falsifiable, statements. We are not citing sources in articles for readers who already know the subject, we are citing sources for readers who are not familiar with the subject, and may want to be reassured of the validity of the article or learn more about the subject. -- Donald Albury 22:53, 23 January 2007 (UTC)

I agree that listing references at the bottom is very helfpul for readers who want to learn more. CMummert · talk 00:06, 24 January 2007 (UTC)

Comment by Wwwwolf

Here's a small thought from someone who has been somewhat dismayed by the eagerness of people to delete stuff based on lack of sources.

My big problem is this: Why is it bad that article histories have unsourced facts? We allow embarrassing glitches in thoughts to exist in the histories. We allow redundant information to exist in the histories. We allow vandalism to exist in the histories, for crying out loud.

What makes unsourced material so evil that they cannot exist in histories either?

I have a small beef with people who take the "It's unsourced, therefore it's EVIL and should be nuked to hell" approach. There are degrees of lack of sources. People add facts that are correct, it's just that they don't have sources for that right now. People forget that correct facts can usually be sourced. People who go "well, if it's so correct, why isn't it sourced now? DELETED!" are just being mean and aren't productive at all. (Perhaps we should amend "Assume good faith" with "Assume the unsourced information is correct and can be sourced, and make good effort at sourcing it before declaring it incorrect and unsourceable.")

Now, I do understand that people have the "Zero information is better than misleading information" thing. However, from the point of view of an encyclopaedia user, zero information is annoying, especially if the topic can be demonstrated to be notable.

My favourite unreferenced piece of crap right now is Expansion pack. Clearly, a notable concept in computer gaming, and anyone who claims otherwise is just delirious. Good luck finding a good source for that article, though, every scientific source I've seen seems to talk of them like they're already part of the landscape and need little explanation. But heck, the information on the article right now is good.

So there's zero sources at hand. If anyone has a good source for fundamental definitions in computer gaming, I'm all ears.

The big question is, What are we to do if the topic is notable, but there's no sources? We shouldn't frustrate users by telling "sorry, the thing doesn't exist."

I'm definitely in favour of userfying or developing the "Unsourced for Deletion" process - with the caveat that if an editor promises to look at the sourcing issues and move the article to their userspace, UfD can be withdrawn. Userspace is the right place for articles that are still in development... and that includes articles that someone believes can honestly be sourced if given enough research time. I'm also extremely hesitant to give any deadlines, because finding a book source can be really tricky in certain fields. (If someone mails me a copy of The Official Book of Ultima, which isn't in any of my local libraries last I checked, for free, I may be willing to change this position =)

That said, I really like the clauses along the lines of "The nominator is expected to have made a reasonable effort to find sources". I dislike AfD nominations that say "there's no sources" and, if you peek at the nominator's monitor, you don't see a Google page that says there's 6,910,500 hits for this term. People who say "okay, but those are all dupes" should at least be able to say so with a straight face. --wwwwolf (barks/growls) 12:28, 16 December 2006 (UTC)

And sorry for rambling. --wwwwolf (barks/growls) 12:29, 16 December 2006 (UTC)

Actually, "Fundamental definitions in computer gaming" is something that really really needs sources right now, because without it there's a lot of original research flying around. We've got probably tens of thousands of people who think they're video game experts editing wikipedia, so everyone thinks they know the right answer, and what we end up with is things like Real-time tactics with a single source and then a bunch of OR to point games that manufacturers sold as RTS are really RTT. If you look at Expansion pack, I see a bunch of things in there that sound plausible but could be completely wrong. " the first video game expansion was most likely " has all the hallmarks of OR bullshit. We don't want people writing what they think they know, because then we're no better than asking anonymous strangers on the internet. We want people writing what they can source, and nothing past that, because otherwise we're going to end up with speculation claimed as fact. As the policy says, if you can't find reliable secondary sources, don't write the article. Night Gyr (talk/Oy) 14:07, 16 December 2006 (UTC)

OK, this goes a little bit off topic to the tangent, but anyway: "The first game expansion" may be OR guessing and can and the Fact of that matter should be tracked down by someone competent. I'm absolutely not objecting to someone removing that claim for the article, unless there's at least a plausible external source that tries to do that job. However, that was not my point - it's the definition of expansion packs that needs the big fat source. Where's the scholarly/industry/journalistic source that says what an expansion pack is? What I'm thinking of is the Finnish article on welding: "According to the standard SFS 3052, welding is defined as 'connecting parts to each other using heat and/or pressing, so that the parts form a continuous surface'." That's a good example of a sourced definition. Where's the Game Glossary with an Authority? --wwwwolf (barks/growls) 15:26, 16 December 2006 (UTC)

By the definition of Notability, notable things have been written about in secondary sources, so by definition there should be one out there. Notablility isn't your subjective opinion, it's the question of whether they've been written about enough for us to write about them. I'm sure some gaming magazine has written on the subject somewhere; someone just needs to dig that up. Writing an article shouldn't be a matter of writing down what's in your head, you should be finding sources and including what they say. Night Gyr (talk/Oy) 15:41, 16 December 2006 (UTC)

Okay. So expansion pack isn't notable because we can't find sources for it. I sure can't think of any sources right now, so let's delete it right now, okay?

I rest my case. --wwwwolf (barks/growls) 18:31, 16 December 2006 (UTC)

When it was originally written, the information had to come from A. Sources or B. the writer's head. We only want to include information from A. That's why this proposal is only for new articles. Night Gyr (talk/Oy) 18:38, 16 December 2006 (UTC)

And I'm overreacting (as usual - so what else is new). Once again, I wish to apologise for my senseless paranoia about "Lawful Evil" types.

Anyway, one more thing springs in mind: If we're particularly concerned about the sourcing of new articles, where's the Utter Newbie's Guide To These Cite Templates? Or templates in general? I'd wager for most people, finer points of template syntax are Week 2 material. Why isn't article sourcing yet part of Tutorial? (It's barely discussed at all!) If this is implemented, then that has also be reflected in the introduction material. (Heck, while we're at it, someone ought to write a really good summary of image Fair Use. It's confusing the hell out of newbies. Sorry, I forgot I'm working on this.)

And also: We need extremely soft gloves. The deletion processes as they are are scaring newbies and "random" contributors. It's apparently not fun for them to get editing and then run into this giant wall of rules we have, and this one proposal here may make it even taller wall. It's hard enough to calm down people who go "They deleted the article about my favourite garage band"; it's more heavy-hearted to explain what went wrong to people "They deleted the article of my favourite garage band, even when they were really big in Pakistan in 1968."

Will we see the death of the stub article, or witness the miracle of consistently well-sourced stub articles? Okay, that idiotic sentence confirms it: I'm an idiot. --wwwwolf (barks/growls) 00:45, 17 December 2006 (UTC)

PS. Sorry if this was already discussed before, but how about a completely new namespace for unsourced material? Move completely unsourced material about that band to Unsourced:Supposedly Good Band.

And yet again, sorry for these delirious ramblings. I will assume the entire responsibility for it and not blame coffee. --wwwwolf (barks/growls) 00:51, 17 December 2006 (UTC)

People can keep articles not up to standards in their userspace until ready. Night Gyr (talk/Oy) 13:31, 17 December 2006 (UTC)

Yes, and people keep talk pages archived in Talk: space, just like they can use their fridge to store DVDs - The system arguably works, I'm not questioning that, and is logical enough if you learn the (somewhat simple) secrets, but it's not elegant. =) --wwwwolf (barks/growls) 17:19, 17 December 2006 (UTC)

Comment by Jll

What proportion of articles are without sources? If it is only 1% then that is 15,000 articles. I suspect it is more.

Because it is easy to do, most of these would articles be tagged very quickly after the adoption of this proposal. But references would be found for them very much more slowly. So the effect will be to swiftly delete the majority of these articles without having adequate time for references to be found.

There would be thousands of otherwise perfectly good articles deleted and have to be rewritten from scratch rather than merely having to have references for them found. I would be happy with the proposal if initially the period before speedy deletion was, say, six months, to be gradually reduced to 14 days over time for the enormous backlog that will be almost instantly created to be cleared. Jll 13:13, 17 December 2006 (UTC)

• ONLY FOR NEW ARTICLES sheesh. Night Gyr (talk/Oy) 13:30, 17 December 2006 (UTC)

• Only one of the four proposed texts talk about this applying solely to new articles. The other three do not. Jll 15:43, 17 December 2006 (UTC)

It was the original proposal and the one that 90% of the discussion here concerns. The worries about backlogs of old articles are really missing the point, because we can gradually work into them, or make a requirement only for new ones. Night Gyr (talk/Oy) 20:37, 17 December 2006 (UTC)

Comment by Grandmasterka

This seems like MEGA instruction creep to me. If you think an article should be sourced, then source it. It would also create a lot of extra hoops to jump through for new users, and it would be abused frequently... I just can't imagine making this work. Please prove me wrong. Grandmasterka 05:56, 18 December 2006 (UTC)

Every article should be sourced, and WP:V places the burden of sourcing on editors who add material, not those who think it needs a source. Night Gyr (talk/Oy) 17:23, 18 December 2006 (UTC)

Well, first, I can prove you wrong on the "extra hoop" bit just by pointing to WP:V and WP:RS. Citing sources is not an "extra hoop" or a nice thing to do, it is a requirement, on one's first article or one's hundredth. This policy creates nothing extra, it simply enforces what already is policy and has been so forever. And how would it be "abused"? An article is sourced or it isn't-the addition of any source to an article would prevent deletion under this procedure. If that's abused on the other end, by someone citing a very dubious source, the article can be sent for a full AfD at that point, preventing abuse of that sort. But there's no potential for abuse here-if someone places this tag on a well-sourced article, as for example a form of vandalism, it'll just get quickly removed with a "What were you thinking?"

But it puts a very strong sanction on that material, by speedy deleting it. It's essentially a shoot on sight policy, a deletionists dream.

Mostlyharmless 00:31, 2 January 2007 (UTC)

Fourteen days to come up with a source, any source, is hardly 'shoot on sight'. -- Donald Albury 02:21, 2 January 2007 (UTC)

Comment by Agne

Strong support, especially with the 14 day period. I think that is very fair and it help prevents biting newbies. What this proposal does it put WP:V back to the forefront of this project's focus. If an editor has time to create an article, then they'll certainly find time within 14 days to put in at least one source. Agne 09:47, 19 December 2006 (UTC)

Comment by msh210

"Reference" here is to be interpreted broadly. References may be in-line citations, works in a reference section, external links, or mentions of a source in the text; questions of whether the references are adequate must be settled by discussion, normally at V, AfD or PROD. —from several of the proposals

I think that the list of possible places for references should include <--HTML comments-->.—msh210℠ 19:01, 20 December 2006 (UTC)

And also edit summaries.—msh210℠ 01:10, 28 December 2006 (UTC)

Comment by Fyre2387

The 14 day rule just doesn't make sense to me. PROD and AfD are both faster than that, so calling it speedy deletion really doesn't make sense. I'd prefer to either have this be true speedy deletion (ie, articles with no sources at all are deleted on sight) or just leave it to AfD.--Fyre2387 ^{(talk • contribs)} 03:05, 22 December 2006 (UTC)

The purpose of this is similar to prod-there are tons of unsourced articles, and mass AfD's would clog the system horrendously. This system can also be closed/stopped immediately once a source is found, whereas the AfD would continue to run for the full 5 days. This also would allow enough time for editors who wish to check for offline sources on a subject, or similar. Finally, this would make clear that unsourced articles are to be deleted if they don't get sourced, period-on AfD, a lot of people bring in "I like it!" or "It'll get sourced eventually, give it time to grow!" (How it's supposed to grow without any idea where the information is from is beyond me, aside from adding more speculation/unsourced info/"I think I remember something about this".) This problem really is big enough to require its own procedure, and this is the perfect thing for it. Seraphimblade 06:03, 22 December 2006 (UTC)

Comment by Kizor

Since we still don't appear to have a working definition on what constitutes a reliable, acceptable source and what doesn't - beyond such things as "well-known," which some take to mean "known to many who are interested in this particular field" and others take to mean "the New York Times" - I look forward to desperate efforts to save hundreds of articles nominated by those whose personal take on this is different than mine. --Kizor 13:13, 24 December 2006 (UTC)

Huh! What is Wikipedia:Reliable sources? If you don't think that is specific enough, you can propose changes there. -- Donald Albury 13:20, 24 December 2006 (UTC)

that's why the main proposal has no questions about reliability of sources--any source will do. Night Gyr (talk/Oy) 06:10, 26 December 2006 (UTC)

Exactly, and it should be-this would only apply if no sources whatsoever are cited. If someone does cite an unreliable or questionable source, the matter would be solved through AfD as normal. Seraphimblade 06:20, 26 December 2006 (UTC)

Comment by RockMFR

My biggest concern: how does this proposal help the project? I've thought over every step of how this would work, and every step involves wasting the time of thousands of editors.

1. Unsourced article is created by (newbie) user. If deleted, obviously this is a huge waste of their time.
2. Editors watch new pages for articles without sources (which essentially is watching to make sure article has external links to a page about the subject, as hardly anybody ever cites books when creating articles). Watching pages for vandalism and other speedy criteria is already annoying enough.
3. Editors tag the article (either with some sort of unsourced tag or with a UfD tag). Time consuming.
4. If UfD tag, tons of time wasted through UfD bureaucracy.
5. Ultimately every UfD would end in two results: speedy through another criteria (for articles that are complete crap), or someone would google and find some shitty source. Bam, passes UfD (reliability of source does not matter under this proposal). Again, waste of time. Either the article should have been deleted immediately, or someone just wasted their time finding some shitty link so the article gets past UfD.
6. Rinse and repeat. --- RockMFR 00:16, 30 December 2006 (UTC)

What about a third and forth potential result under your #5? That editor finds a legit source to add to the article, improving both it and the project or that maybe a newbie editor is made aware of Wikipedia's WP:V, WP:OR and WP:CITE policies and not only adds sources to the initial article but is more likely to do so in future edits, making that newbie a better editor in a shorter period of time? I don't doubt that your two "end results" will happen, and probably happen often, but they are not the exclusive outcome and there will be some benefit to the project. Agne 01:28, 30 December 2006 (UTC)

I believe the main benefit is not in the case of any specific article-just like blocking for vandalism doesn't benefit any specific editor who receives a block. Rather, it establishes a culture that vandalism is unacceptable. Of course, this proposal does not involve blocks, but it does similarly have some "teeth". It establishes that WP:V is policy, not simply a nice thought, that citing one's sources (and having sources rather than putting up OR) is required, not just something to do before an FA nomination, and that we as a community value these things enough to enforce them. I hope that not one article is ever actually deleted under this proposal if it goes into place-rather, I hope that lots of articles are sourced, and a good number of editors learn how. (Also, if sourcing is what's -specifically- at issue, I believe that those editors concerned with it are more likely to go check out a source added to prevent deletion, and to question it and AfD if necessary if the cited "source" is unreliable or does not support what's in the article). Seraphimblade 14:51, 30 December 2006 (UTC)

Let me correct your point 1:

1. Unsourced article is created by (new) user. They see the UfD tag, or a big warning on the screen before they create it, and add their sources.

At the moment, we have:

1. Unsourced article is created by (new) user.
2. {{unref}} is added. The new user ignores it, because it isn't exactly a pressing matter.
3. Weeks go by. The new user has left Wikipedia.
4. The article goes on AfD. There's no longer any way to find out what the sources were, so there's now no way to tell that the article is correct.
5. Either the article is deleted, or someone finds a website that contains similar information to the article. However, there's no way to know that that was the source so the article has to be rewritten to match the source (it could just have been written from scratch at this point). --ais523 17:09, 30 December 2006 (UTC)

Which of these scenarios looks better? --ais523 17:09, 30 December 2006 (UTC)

Ideally, new users would actually add sources to articles. In reality, it just won't happen. From my personal observation at Afd, it has become clear that even with a scary deletion tag on the top of an article, editors generally won't bother trying to find sources. For many editors, they just don't want to do that kind of stuff. They want to make new articles and add the stuff they know (which may or may not be sourceable), and other users come along and add sources. A vast majority of users will never cite their sources. This is something that the project is going to have to live with. --- RockMFR 21:56, 30 December 2006 (UTC)

Why does the project have to live with editors who refuse to add sources to their contribution? Shouldn't it be the other way around? That editors have to live with Wikipedia's policy on WP:V and if they don't, then their content is simply deleted. I'm an optimist and I like to think the best of folks. I think there is motivation for editors to become better contributors through the project with a gentle but firm nudge to cite their sources. If there are some editors who steadfastedly refuse to add sources, then they will simply have to leave with the outcome. The beauty of this proposal is that the ball is squarely in their court. They dictate the next move. Agne 00:22, 31 December 2006 (UTC)

The beauty of Wikipedia is that it currently has hundreds of thousands of contributors. Every "source or delete" proposal like this one only pushes us closer and closer to a system where outsiders cannot edit the encyclopedia. The barriers to entry of Wikipedia are getting bigger and bigger. --- RockMFR 04:53, 31 December 2006 (UTC)

To some extent-the barriers to entry should be that high! "The encyclopedia anyone can edit" does not mean "The message board anyone can come along and put whatever they want on." WP:V is policy, not a nicety, not a stylistic element that should happen before the article goes to FA, and not something you only have to do if you don't "remember" stuff about the subject! While we should treat editors who fail to cite sources much more gently then, say, vandals or those who make attacks, we should not forget that, while inadvertent rather than intentional, their behavior violates policy and makes tons of work for other editors (who have to find all kinds of sources or rewrite the article properly, and that may not even have been the one(s) used originally.) We should, then, have this deletion policy-it gently but firmly states "Look, when we say required, we really do mean required."— Preceding unsigned comment added by Seraphimblade (talk • contribs)

I think Seraphimblade has stated the case very well. -- Donald Albury 19:58, 31 December 2006 (UTC)

This can be seen a little more constructively. Those who have not yet learned how to write sourced articles should be helped. The worst sort of "help" is to have their work deleted--which benefits neither the WP nor them.

Warnings do not work very well for this because these are not the users who come regularly enough to see them, or who are experienced enough to pay the proper attention. This is especially the case when speedy procedures are used. It is seen at its very worst when those adding the speedy tag do not bother to warn the user on his talk page, even though there is a very easy to use template for he purpose. Such behavior is disruptive to the intended profess. We should of course have this policy, but we should implement it in a more friendly way. The minimum is to soften the wording on the template, extend the time periods, and remove all tags where the first and recent editors were not warned. (Since it takes a little thought to see who arethe significant recent editors, I do not immediately see how it could be done by a bot.) DGG 02:59, 2 January 2007 (UTC)

Comments from Lankybugger on Proposed Text #4

After reading this policy, I have to agree with it. Currently a large amount of articles in the AfD areas could be made better with proper sourcing but might not get the chance due to the 5 day time period, during which a seemingly overwhelming amount of resistance could be levelled against it. Likewise an AfD can be percieved as a hostile attack which the user must defend against (especially to new users), while this policy could gently ease new users into the idea of citing sources and neatly maintains itself strict with regards to WP:Bite.

More importantly, a notice could be added to the unsourced tag which explicitly states that if the article doesn't recieve sourcing within sixty days it might go to an UfD. This would prompt the new user to read and discover what a UfD is and the UfD policy page would, I'm assuming, lead to the appropriate policies.

It also breaks down the size of the AfD page, which makes it easier to browse for users with slow connections. I would recommend text for Speedy to UfD be drafted so that part of the policy would allow the transfer from an AfD TO an UfD, but that's just me.

Cheers, Lankybugger 00:32, 3 January 2007 (UTC)

Comments from CyberAnth

I totally support the idea of this policy proposal. In fact, if it develops out enough, it just might help keep me away from Citizendium.

About unsourced material

Allowing 14 days for an unsourced article to remain as an article is TOO MUCH time. Users should indeed be drafting articles on their UserSpace.

The criteria of "until it has at least one reference" is far too lax. Good gracious, that is a standard for fourth graders!

While one reference may be fine for stub-creation or very short articles, larger articles should be well-sourced before becoming an article. I prefer a criteria of, "until all major assertions are adequately referenced," or "until the majority of the article is adequately referenced".

About drafting articles on userspace

Articles should indeed be drafted on userspace. At User:CyberAnth/Religious views on masturbation, I began drafting an article in this vein. To garner contributors, I visited various article talk pages and invited others. The article is shaping up very nicely. Some would say it is past time for it to be posted on article space. I don't think so.

About Wikipedia culture change

Without this policy, Wikipedia will only encourage the amateurism that is has become too well-known for. Let's take steps away from that ASAP. There is no good reason why Wikipedia should not utterly require articles to be well-sourced, and speedily so, except as an argument for amateurism.

Granted, this new policy, plus the ones I urge below, will change the culture of Wikipedia. Great!

And to do that, standards for contributing will have to be raised.

Right now, newbies come in and are encouraged to start editing basically right away. That is patently irresponsible, unless the primary goal underneath of this whole Project is to actually to raise the number of Wikipedia "hits" as as tactic in increasing the flow of grant funds.

New users should be required to know the standards for articles before attempting to edit. The "Welcome message" new users receive should be much more explicit about the fact that editors need to be responsible to train themselves in content policies before editing. Otherwise, they are wasting their time and, most particularly, quality editor's time. In the end, this drives quality editors away while encouraging the retention of amateurs, which leads to an increasing downward trend in Wikipedia quality.

Concomitant policy proposals

Premise: If an author does not cite while they write (e.g., Bonny Hicks I wrote recently), they simply should not be contributing in the first place. Period.

• Wikipedia: Cite-while-you-write needs to be created.

This policy in a nutshell: Wikipedia editors should cite their sources while they write material. Inadequately cited articles will be speedily removed to user space as drafts. All non-sourced additions to articles will be removed on sight.

NOTE: The above policy should contain, "Non-sourced additions should be placed on the article's talk page, where other editors can help find sources."

• For articles removed from article status to user space, a template should be placed where the article was:

This article has not yet reached Wikipedia standards for verifiability. You can help Wikipedia by contributing to this article's draft at User:Anyname/Article name.

Alternately, and perhpas even better, create "draft space" Draft:Article name, while placing whatever it is that text file is that hides pages from being crawled by search engines.

Also, I think it is just insane that Wikipedia allows I.P contributions.

Conclusion

What I have said may not be popular. But like I said, there is always Citizendium just down the road.

CyberAnth 03:51, 3 January 2007 (UTC)

Comments by DGG

Not that you are wrong, but I think many of our contributors would have difficulty working the way you suggest.

DGG's Point #1

On many topics, there are more than 1 person working, and it would not be practical to work by withholding an article until finished. #Furthermore, on many topics, more than one person is needed. I recently defended the article on cock tease at an AfD debate, and found myself collecting the references that should have been there at the first. (It required considerably more patience than most people here have--I had to go through all the non-useful ones on Google, just as a first step.) But after I finished, and had enough to justify the article and it was OK'd, two of the other people began also contributing appropriate examples and links, and finding better sources for some of mine.

DGG's Point #2

Even in the academic world, when I write a paper, I use what I know, and leave space for the references, and then put them in at the end, along with whatever I had forgotten. I would not actually teach students this method, but it works, especially on familiar topics.If I start with the sources, I find it difficult to follow my own theme. (I of course have collected and categorized sources as I find them or as they are published.)

DGG's Point #3

Some people at any level work best in small increments. Many of the times when 20 or 30 successive small edits are made to a page, the result is very satisfactory.

DGG's Point #4

Even in the sciences, some of the best contributors at WP are amateurs, often of high school age--especially in biology, but there are math and even chem amateurs as well making useful contributions. They do not usually have the library facilities some of us have, and put in what they do have, and others update the editions & find additional.

DGG's Point #5

Writers learn by writing. More precisely, they learn by guided writing. The multi-editorial process here--if done in a careful and friendly way (which is of course not always the case) --can teach them. You can look at edit histories, and see the development of sophistication. As I have time, I try to guide a beginner or two. It would work if others did likewise. You learn most about writing, or about a subject, by teaching others. Good teachers do not look down on beginners.

DGG's Point #6

The level of contributions in 06, at least in some fields, is much higher than it was at the beginning. To really see the worst of unsourced articles, look at the ones that haven't been worked on in the last two years.

DGG's Point #7

Specifics--the 2 week period is necessary to accommodate those who do not spend their whole life here.--articles in different areas require different degrees of sourcing, and any review of controversy here will show the lack of agreement on standards even within a field. It is difficult to be prescriptive when it is not clear what to prescribe--and, many WP articles are remarkably over-sourced--e.g. in the field of philosophy. Not everything needs a primary source.

As for Citizendium, the editorial process seems similar to here. People contribute edits, people contribute sources, people ask about their outlines. The main difference is that the working copy will be free of vandalism. You are, usefully, trying to imitate this with your user space page, but if the word gets out, the vandals will find it. And you have been deprived on the benefits of those who didn't know it was there. Now that I do, I can see if I have anything to add.

DGG 04:57, 3 January 2007 (UTC)

CyberAnth's Reply to User:DGG

Thanks for your thoughtful comments, DGG.

Making it a bit more difficult via higher standards to be a Wikipedia editor is part of the idea. Greater difficulty tabulates to more serious editors and an higher quality product.

Reply to DGG's point #1

Editors are not excluded by the creation of Draft:Article name in place of Article name. The world is, from thinking that what they are reading is a real encyclopedia article. When anyone, including editors, searche Wikipedia with a term, they find a page as I said:

This article has not yet reached Wikipedia standards for verifiability, thus it is not posted as an article. You can help Wikipedia by contributing to this article's draft at Draft:Article name.

Reply to DGG's point #2

The writing process you explain is actually pretty common ~chuckle with big friendly grin~. But if you do not hand in uncited academic papers read by one person, your instructor, then you most certainly should not place an uncited article on Wikipedia namespace to be read by the world.

In the interim, editors can plod away on their article at Draft:Article name. For existing articles, the place to work out uncited additions is the article's talk page, or perhaps at draftspace from which articles at userspace are updated periodically after consensus.

Reply to DGG's point #3

Working in increments best happens on additions proposed on talkpages, or the individual's userspace or basic text editor. Afterward, you post a substantive verifiable change...rather than 20 increments that degrade the article during the process.

Reply to DGG's point #4

Kudos to the bright high school students! None of what I am proposing excludes them. Also, they always do have their school textbooks, and school and/or public libraries who have collections and inter-library loan capabilities.

Reply to DGG's point #5

You stated, "Writers learn by writing. More precisely, they learn by guided writing" (emphasis added). Really good stuff, and as a college instructor I most heartily agree! But responsible learning-by-writing should best occurs at Draft:Article name, hidden from search engines and not in real-time to the world. Your argument provides excellent support for: 1) the Citizendium system where authors are automatically placed under the resident Ph.D. for that article's domain; and/or, 2) the enormous expansion of WP:ADOPT (which you sound like you might enjoy joining, BTW).

Reply to DGG's point #6

It is immoral to host encyclopedia articles that seek notoriety as reputable while being irresponsible. Moreover, Wikipedia now faces potentially serious competition from Citizendium.

Reply to DGG's point #7

A 2 week-length period before moving an article to draft space is not at all necessary to accommodate those who do not spend their whole life here. The length is too large. Their sub-par contributions are viewable to the world, after all. When casual editors view their watchlist after visiting Wikipedia after an absence, it will contain their article contributions with a notation indicating a template has been placed in said contribution, pointing them to Draft:Article name. There, they can continue their work as before, but in not in view of the world where it does not belong. Users of the world will still find article pages with articles still in process of verifiability indicated by their templates indicating such; and, they can read the draft if they choose. However, the template will declare to them, in essense, "Wikipedia says this content is not yet worthy of being an actual article."

A few counterpoints to User:CyberAnth from User:Lankybugger

User:Lankybugger's point #1

Regarding unsourced material, I believe fourteen days is fair. Most wikipedia editors don't check the site every day. I consider myself a decently contributing editor and there have been days or even weeks at a time where I find myself with little to contribute. The fourteen day time period would take this into account, allowing those who might not have the time to check every day to still contribute without fear of returning to find their in-progress article (which might require sourcing which is available, but has yet to be added) gone. Likewise, drafting the article itself in the user's page is not necessarily a good idea as it might limit people from contributing. Lord knows I've wandered into articles which I've improved via linking from other articles and I imagine it's the same for others.

User:Lankybugger's point #2

As far as Wikipedia being limited to registered users is concerned, I fail to see the reason to do so. A major portion of Wikipedia's success has been the ability to attract new users by allowing them to edit and then introducing them to proper forms of writing. A plumber might not be computer literate, for example, but I'm willing to bet he knows his job and can certainly provide better information than an academic. So it is for a great variety of subjects. There's a reason Wikipedia has so many articles (and a great deal of them good ones!) and it is NOT because it discourages new users and ideas because they aren't aware of every policy we may have. The editors who genuinely want to contribute can be taught to write properly and it's a matter of finding the right balance between academic responsibility and being an encouraging, open teacher.

User:Lankybugger's point #3

Demanding that everything be cited during creation is academic standard. While some articles may well be considered worthy of academic praise (such as some Featured Articles), to demand that every article confirm to academic standard at creation would be a large breach of WP:BITE.

Cheers, Lankybugger 05:44, 3 January 2007 (UTC)

User:CyberAnth's replies to User:Lankybugger

Thanks for your thoughtful comments. I appreciate very much the opportunity to engage in civil discussion of ideas with other Wikipedians. Thanks again. :-)

Reply to User:Lankybugger's point #1

See Reply to DGG's point #1.

Moreover, Wikilinking from articles is not in any way inhibited by my proposal. Editors can wikilink into an article just as before; only, what they will find if it is unverifiable is my proposed template:

This article has not yet reached Wikipedia standards for verifiability, thus it is not posted as an article. You can help Wikipedia by contributing to this article's draft at Draft:Article name.

Currently, editors regularly link into This article does not yet exist but you can create it, and then create the article. Since this is so, they can all the more link into an article for which they receive the above proposed template. The former This article does not yet exist but you can create it leads to a blank editing template. The latter Draft:Article name leads to a link they can follow to an article already in process.

Reply to User:Lankybugger's point #2

The reason for limiting editors to registered users is, among other reasons, because it will greatly diminish vandalism and the number of articles that are daily AfD'ed - both an utterly enormous waste of editor energy, which rapidly drives many of the best ones away but keeps many amateurs and the unemployed around to stimulate their own sense of self-esteem and importance. I am sorry to have to say this so forthrightly.

Also, if all created accounts required approval from a sysop before editing were allowed, with the sysop checking the new account for sockpuppetry, it would very greatly cut down on that systemic energy-wasting malady endemic to this project as currently constituted. My proposal step will greatly quell the exodus of high quality editors who do not wish their time wasted dealing with such things. Spend a few hours going in detail through WP:MW to see what I mean.

Reply to User:Lankybugger's point #3

How can Wikipedia even so much as have the gall to claim it is an encyclopedia if its articles do not meet the basic academic standard of verifiability? Apart from such standard, the result is little more than a bastard, sterile cross-breed between an encyclopedia and an Internet forum. Unless the primary goal underneath this whole Project really has all along been ultimately about money, e.g., raising the number of Wikipedia "hits" as as tactic in increasing the flow of grant funds, etc., - in that case, please just allow me to walk down the road to other projects not afflicted so.

Regarding WP:Bite, it is past time to revisit the policy. The policy was developed while Wikipedia was in its infant stage and needed as many new contributors as possible. It was useful for its time, and wise. However, that condition is no longer relevant. While no one would wish to overly inhibit new good contributors, the reason why Wikipedia has developed so many systemic problems now that it has evolved - problems that have have also sparked serious competing forks - is largely because of certain elements of WP:Bite. Now that Wikipedia is in no way at a loss for contributors, it is time for the policy to be re-evaluated.

Times have changed and so must Wikipedia. Yet currently, from the top-level down, its culture is too greatly rooted in its soil of the past. As with all other organizations who cling thusly, it will be superseded by those willing to change, even radically so if need be. Wikipedia right now has a limited window in which to change radically with the times. How will you help it do so?

CyberAnth 07:01, 3 January 2007 (UTC)

Comments on User:CyberAnth's proposals by Agne

There is quite a bit that I agree with CyberAnth on especially relating to the vision of Wikipedia improving through higher standards and motivating editors to better themselves and their contributions. I am an ardent believer that the credibility of the project is vitally linked to our commitment to the core policies of WP:V, WP:NPOV, and WP:OR and this proposal is a positive step forward. But where I break with CyberAnth's sentiments is the sense that more needs to be done right now and in the context of this proposal. The difference between the current incarnation of this proposal and the drastic change to the degree that CyberAnth proposal is, in my mind, the difference between a nudge and a shove.

When you think of human nature, what is a person's instinct when they get "shoved" in any direction? They put their foot down in reflex to stop their forward momenteum. However, if you continue to firmly nudge them along you can sometime find that alot of ground gets covered in a reasonably short amount of time because you don't have that reflex and instinct to put your foot down and halt. While I love the idealism and hope in CyberAnth's proposal, I just don't think the Wikipedia community can move that far so fast. I think there is a lot of merit to what this current proposal can do in terms of nudging the community into the right direction and to a better encyclopedia. Agne 08:21, 3 January 2007 (UTC)

CyberAnth's reply to Agne

Your points are very well taken.

I think the better approach is neither a nudge nor a shove, but a year-long or so plan consisting of a series of incremental "nudges" that tabulate to substantive change. Plans do not always equal actualization, true, but that should not negate a concerted, time-lined effort.

Thoughts?

CyberAnth 11:16, 3 January 2007 (UTC)

Response to CyberAnth's response

First, I'd like to thank you for taking the time to have this discourse, CyberAnth. Though I can already tell we no doubt disagree on many topics, it's nice to see mutual respect in a discussion which is as sensitive as this.

Regarding Point #1

Mulling it over further, I can see the bright points to your suggested Draft system. Perhaps a policy suggestion might be that articles created are automatically noted as drafts, and must go above a certain standard before the template is removed?

For example, let's say I create an article called The Application of Personal Lubricant to Firearms. It creates the article exactly as I've written it, however above the article we are shown this:

This article has not yet reached Wikipedia standards for verifiability and thus is considered to be a working draft only. You can help Wikipedia by contributing to this article.

This may just be a tangent, but personally I believe there should be a policy in place which puts a template on articles warning about the possibility of unverified content in ANY article under a certain standard. Perhaps it's just that cover-your-ass thought process working for me, but I do agree that articles should be seperated by quality in a manner which is immediately noticable on the page as soon as it's loaded. The passive user might not think to look at the article's Talk page to discover what the article has been rated.

This would also make a more robust use out of an already existing system and would keep vandals from creating the article directly under the article name itself by overwriting the draft notice and would encourage the sort of open editing which has helped to foster Wikipedia's growth.

Regarding Point #2

I'm going to have to disagree that pre-screening editors would be a work-saving measure. I would guess that most editors will be clean. In the rare instances of dedicated vandals or sockpuppets, it's probably better to be reactionary. Disruptive users rapidly make a spectacle of themselves and it's far easier to allow the average editor to screen newer users for possible vandals through the observation of their actions than by prescreening them on the off-chance that they're a sockpuppet.

Missing Wikipedians is precisely WHY I would encourage continued open invitation for editors. While some of those editors left due to a hostile climate created by the excesses of vandals, others have left because of personal confrontations with other users in good standing, vanished because of real life issues which might prevent them from editing, or even just plain boredom and a desire to move on to something different. For every good editor that leaves, there's probably going to be another to take his or her place.

Regarding Point #3

Perhaps I should have clarified. I do find myself agreeing with you in theory but not in the manner of execution. You're absolutely correct in stating that it's vital to Wikipedia's credibility to demand that articles be held to an academic standard. I just believe that there's a different way to do it.

WP:BITE needs work in the details but is a good policy in theory. Less emphasis should be placed on fixing a new user's mistakes and more emphasis should be placed on guiding the user (through talk pages and edit summaries) to correct their own mistakes in a gentle but firm manner.

Cheers, Lankybugger 16:33, 3 January 2007 (UTC)

DGG's overall comments

I do not see that WP and CZ are in competition. There is more than one way to make an encyclopedia, and I would urge those who feel that CZ's way is more suitable for their purposes to go to CZ and give it a try. I think we have more to gain from being distinct--not just gain for ourselves as Wikipedians who may want informality even if it brings a certain amount of confusion, but gain for those who may find that an encyclopedia based on the current way of editing fills certain needs better than one made by a more formal process. For those needs where the CZ method meets better, we can leave them to CZ. There is no need for us to imitate CZ, or to adopt cleverly paced changes to get people used to an idea they do not want.

And, I remind CyberAnth that he may not be the only college instructor in this discussion. So, with all our combined experience, what reasonably undisruptve changes can we add to the current structure? DGG 02:57, 4 January 2007 (UTC)

I think the change proposed is reasonably non-disruptive. It's not like we're proposing a genuine "shoot-on-sight" speedy criteria for articles here. Rather, we're simply proposing that editors be given notice that articles really must be sourced, or they don't belong here. WP:V already establishes the principle that unsourced statements don't belong in Wikipedia, this policy is a very logical extension of that. Seraphimblade 03:08, 4 January 2007 (UTC)

I more or less agree with Seraphimblade, so long that it is the first proposal, which would apply to articles created after it went into force. Rebecca 03:22, 4 January 2007 (UTC)

That's the one I'm referring to-massive nominations of articles created previously would be extremely disruptive. Seraphimblade 03:38, 4 January 2007 (UTC)

Ah, good. I see very little harm in enacting such a change; not much content would be lost, as anything that needed to be could be easily undeleted, and it would be a good means of starting to enact a cultural change. It's only the use of lack of references to delete existing articles which I vehemently oppose, mainly because the selection of articles to be deleted is so arbitrary. Rebecca 04:03, 4 January 2007 (UTC)

Regarding Point #2 (and other matters)

I've created a decent number of articles, a very large number of them I've created with things I know (but verifiable, and often backed up by the external links) references will come later, as the articles expand past short stubs. This represents a real step backwards, and a slap in the face to people like myself who don't have hours to create the perfect article, but want to see a stub where nothing previously existed. If I create something and have a speedy tag attached, what do you think I'm going to do? Mostlyharmless 06:29, 4 January 2007 (UTC)

One hopes you would source it, as (per WP:V) you were supposed to have done in the first place. This would prevent the deletion, and you'd have 14 days in order to do so. This criterion doesn't require you to create the "perfect article"-it requires you to cite at least one source. One. Seraphimblade 06:52, 4 January 2007 (UTC)

To add to previous-for purposes of this policy, external links which support the text in the article do count as a form of sourcing, and would prevent deletion under this criterion. It's only totally unsourced articles which would be subject to deletion-it's not necessary that inline sourcing be used for this proposal, only that at least one source is provided. Seraphimblade 06:54, 4 January 2007 (UTC)

How hard is it to add a single source in a reference/external links section? How good is your memory about "things you know"? How susceptible are you to things like urban legends? It's possible (and very easy in fact) to create a stub with a reference attached to it (I'll accept that if the references are readily accessible creating it from your memory is "okay" if the material backs up the subject), so I'm not understanding your objections here. It's not really all that hard to find a source for most things, and the only reason people seem to detest it is because they don't want to do it for some reason. ColourBurst 16:20, 4 January 2007 (UTC)

Well-referenced should be the criteria, not the fourth-grade ONE source. One sources does not even establish notability, which requires MULTIPLE quality sources. Mostlyharmless is free to spend all the time he or she needs in userspace to draft poorly sourced articles. If he or she wants to write things for the pleasure of having them appear immediately on the Internet, do it on Internet Forums, where what is posted is not packaged in an aura of authority. CyberAnth 01:25, 6 January 2007 (UTC)

CyberAnth, arguing merit of sources provided is what article discussion pages are for. AfD is for articles which are found to be non notable. This process has a defined goal of ensuring that every article has at least one cited reference, that is all. Beyond such black and white objectives it is impossible to use a speedy criteria, and currently every speedy criteria can be challenged by a single contributor (usually, the contributor who creates the article).Garrie 21:36, 7 January 2007 (UTC)

'Speedy' delete, and template appearance

For reference, here is the current appearance of {{subst:nsdel}} (thankyou, Special:ExpandTemplates):

(By the way, {{subst:nsdel}} isn't active yet; this is a version for experimentation with the tag while the proposal is being discussed.)

There are issues, for instance raised in the section above, about 'speedy' deletion (which is what is implied by the title of the proposal) putting users off, but this policy proposal is more like prod (for the first proposal, which I support and which this section, and debate for that matter, is mostly about) or AfD (for the 'UfD' proposals). I have two points to make:

• Being slower than prod, this isn't speedy. We shouldn't call it a speedy-delete criteria (and should keep it a separate policy from WP:CSD) because that would just cause confusion.
• The warning template should encourage people to source the article rather than forget about it. (Mostlyharmless in the section above says "If I create something and have a speedy tag attached, what do you think I'm going to do?".) It's important that the correct attitude is for the user seeing it to immediately add sources rather than to abandon their article as hopeless.

Any comments? --ais523 09:10, 4 January 2007 (UTC)

I definitely agree that we should take the "speedy" out of the policy name. This is not a "shoot-on-sight" criterion as speedy deletions are, if it's similar to anything it would be similar to the prod process. Why not just call it "unsourced article deletion"? I like the tag as is, although maybe to clarify-"You may remove this tag if, but only if, a source is added to this article." Might that work a bit better, and make clear to people that deletion is not a foregone conclusion, and they just need to source? Seraphimblade 17:28, 4 January 2007 (UTC)

I agree, so I've made the change on {{subst:nosourcedel}} and {{dated nosourcedel}}. --ais523 17:55, 4 January 2007 (UTC)

As calling it 'speedy' seems to be a problem, then we should change the name. "Unsourced article deletion" is descriptive, but sounds sort of flat to me. "Source it or loose it" has more zip, but also has too much bite. "Saving articles by sourcing them" captures the idea, but sounds, uh, dorky. Sorry, I'm having trouble coming up with a happy medium here. Maybe someone else will be inspired. -- Donald Albury 01:57, 5 January 2007 (UTC)

A proposal without real substance or teeth

The template states, "You may remove this message only if you add a source to the article."

Here is how that will work, I have been around here long enough to know it all too well.

Someone will write tons of OR or other BS, search amazon.com for a book of the general topic, and place it in. Wa-la, all better!

For example, I am right now trying to AfD Wanker. The references support the creation of a dictionary definition (see here) but the article is utterly rife with Original Research.

So during the AfD, someone came along and placed Jenny Cheshire, English Around the World: sociolinguistic perspectives, Cambridge University Press, 1991, into the article in an attempt to make all the OR appear "referenced".

So I looked at the book, actually searched through its entire digitized text. The following showed up and nothing more:

A number of the formulae ["you wanker" and "fuck off, wanker" (from page 205)] use as a form of address the word wanker. Literally, the word wanker in New Zealand English denotes a male who masturbates. But it is also used as a general term of abuse among New Zealand males. Masturbation is popularly regarded as a sign of sexual inadequacy. Therefore a wanker is one is sexually inadequate (page 206).

In other words, support only for a dictionary definition was cited.

Requiring ONE reference is a standard for fourth-graders, not people held to a responsible standard of sound encyclopedic article creation.

I urge You may remove this message only if you add a source to the article You may remove this message only after all major assertions in the article have been reliably sourced.

As written now, it exhibits only the continued paradigm of the past when Wikipedia needed editors, editors, editors, any editors just give us editors. Now, that condition is far past gone. We need only high quality responsible editors and policies that strongly enforce that which will drive anything less to the playground of Internet Forums where they belong.

CyberAnth 02:13, 5 January 2007 (UTC)

If people add a poor source, that can be decided at AfD. Nothing in this would prevent an article from going to AfD if only inadequate sourcing can be found. This, however, is intended to be more like WP:PROD-and if nothing else, it will draw attention from other editors to the fact that the article is in need of sourcing. Seraphimblade 02:36, 5 January 2007 (UTC)

How in the world would such an article get put in AfD for poor sourcing? When the poor sourcing is added the deletion template is removed. That will generally be the end of the story except for the rare case where someone is being extra vigilent. The concerns of CyberAnth will happen because the few editors willing to actually evaluate the quality of the sourcing in an article that have had this template added and then removed will be completely overwhelmed. This proposal will only benifit WP if we have a systematic method for fact checking in place. Currently we do not even have fact checking as a requirement for Featured Articles, our highest quality marker. Going forward with this proposal in the current enviroment will do a great deal of harm to this project. Set-up fact checking first then return to this proposal.--Birgitte§β ʈ Talk 15:04, 5 January 2007 (UTC)

And the editors that try to help articles by adding sources after the fact are not currently overwhelmed with the amount of new articles being created without anytype of sourcing? Backtracking to add sources is certainly a more difficult and time consuming task then fact checking provided sources. Agne 00:24, 6 January 2007 (UTC)

We may be currently overwhelmed with unsourced articles, but the redeeming factor is that they are obviously unsourced articles. Many of them even have a nice purple template on them warning readers they are unsourced. If this proposal goes forward with the provision allowing editors to simply "add a source" and remove the deletion template, I believe we will end up much worse off. We will have many of the same quality of articles as we do now but they will appear to be sourced by the first book on the subject that was located through Amazon. I do not think we should push in this direction unless we have some means of keeping people honest. I do think this proposal has good intentions and it may be a good idea in the future but I believe if this were implented right now the project would put in a worse position than it is now. --Birgitte§β ʈ Talk 01:10, 6 January 2007 (UTC)

You have a good idea! I'm going to begin drafting a proposal for fact-checking. It's definitely something we can never have enough of! Seraphimblade 15:09, 5 January 2007 (UTC)

Wikipedia would be greatly improved if 50% or so of its articles disappeared over the next year. Folks, gobs of elementary schools (!) do not allow Wikipedia to be used as a source, and no high school teacher or college professor alive worth his or her office space allows it. What we have here is a monster cross-breed between an encyclopedia and an Internet Forum and some things need changing starting now.

Speak for yourself. I teach library school students how to evaluate, use and teach it, and the usual academic policy in most universities is to use it with discretion, just like a paper encyclopedia, except for the areas of technology where its the best source. Obviously subjects will differ. DGG 03:53, 23 January 2007 (UTC)

Having to add references to someone else's mish-mash is an utter nightmare. And then when an article gets AfD'ed for OR and non-verifiability, what happens is a mob of partisan "editors" come along and reference-pad. It is an exhausting thing to have to call them on this with proof. And BirgitteSB is completely correct: all the one source rule will do is make people go snatch something on the general subject at amazon.com, then the article stands and whoever posted it can sit and look at their self-pat-on-back "work" and feel like a big boy.

The premise we have to start with is that a great many editors simply do not belong here, and the only way to separate them is through policies that require high standards. Those here for less than serious reasonings will either come to par or go--exactly what needs to happen.

Any article not well-cited needs to be Speedy AfD'ed on sight and any mish-mash removed to someone's name space.

A proposal for fact-checking is a great idea and will strengthen both this currently far too weak proposal and WP:V.

The very last thing any of this will do is damage Wikipedia, though it may cause some "down-sizing" and "re-organizing"!

CyberAnth 01:12, 6 January 2007 (UTC)

You continue to ignore the damage this will do to the culture. This project succeeds because of the social controls and voluntary participation, not because of top-down hierarchies or traditional techniques of collecting and distilling knowledge. Jimbo already tried the traditional approach. It failed. (See Wikipedia:replies for a start on that story.) The power of the diversity of our volunteers should not be underestimated. And it should not be jeopardized by short-sighted attempts to impose traditional approaches on this new enterprise. Rossami (talk) 05:41, 6 January 2007 (UTC)

CyberAnth, while I am somewhat in support of what you are saying, and I'm all for this proposal for new articles, I think that to apply this sort of proposal to existing articles, or even to go even further as you have suggested will really go against the spirit of Wikipedia and its goal, "to give every single person free access to the sum of all human knowledge" and will detract from the information that we do have here, and the work that editors have done here already. Whilst JW may have said it's better to have no article at all than one with sources, and while there are some rubbish articles out there that should be deleted (and we do have processes for that and I'm all for getting rid of them), a lot of the unreferenced articles on Wikipedia at present are quite good, even though they don't have references (probably due to the education of editors - it's only in the last few months on here that I've seen a greater emphasis on keeping your articles sourced, something I didn't really get taught when I first came on here a year or so ago). Yes, it's a shame that this sort of thing wasn't put in place when WP was first started, but I think it's more favourable that this apply to new articles, and that editors put a bit of effort into getting existing articles into the sort of shape that you have suggested above rather than just deleting them willy-nilly - as that will be what it will be, completely arbitrary, with some deletionist editors picking and choosing the articles that THEY don't think should be in Wikipedia. But until the time when most articles are up to scratch, I think it's better that we have something where the article is good, and slowly improve it, rather than confine everything to a userpage and miss out on a lot of the articles that we do have. If this proposal is going to be adopted for existing articles, I would urge that we make the time period a lot longer than 2 weeks - in many cases the sources needed to make an article better are not available on the Web but in printed material, and it takes time to go and access them from your city's major library and research them properly; this would stop the laziness of drawing sources from Amazon and other places on the Web just to stop an article getting deleted— and for those of us who want a life outside of WP, it's a bit more reasonable than to expect someone to start writing a large referenced article in a couple of days. Anyway, just my opinion... I hope that a sensible policy on this is adopted, and not one that will seek to wipe out most of Wikipedia at the expense of the (good) work that has been done on the articles in question. JROBBO 06:00, 6 January 2007 (UTC)

For many of the existing articles without sources (or adequate sources) getting something respectable in there would not be very hard, and the way to do it is to do it ourselves. If it looks like a major library will be required, we could leave a note on the talk p. saying so, & someone with access will see it eventually. There are so many of the old ones to do that we needn't worry about getting them all. Another way is for those of us with access to a partic. good & recent source to try to add it to the articles relevant--there is usually more than one. I suggest that getting a useful reference on the ref. list is more important at first than putting in exact inline references. This could be organized, but perhaps we should try a little before making elaborate plans.

If we needed a project, we could start with, say, all the 1911 EB refs and see if there is something later to add--particularly a book likely to be in many libraries, or a good PD original text from Project Gutenberg. One cannot re-edit an encyclopedia in one sitting. To do what has been suggested for a print encyclopedia takes about a decade, & we have a much wider variety.

And it is meant as a serious suggestion that those who want to work in a more controlled setting give CZ a try. Their requirements for what they term "editors" was a PHD degree, but only knowledge is wanted from authors. All it takes is to sign up and choose an article. DGG 06:16, 6 January 2007 (UTC)

A proposal

This discussion seems to be moving in a constructive direction. While I strongly oppose any attempt to CSD unref'd articles, creating a culture of referencing is something I strongly support. So here's my serious proposal - we get a bot to put an unreferenced template onto any article without ref tags or a references section. Mostlyharmless 08:50, 6 January 2007 (UTC)

I LOVE that idea. Agne 09:09, 6 January 2007 (UTC)

Don't forget the {{cite}} template and its kin, which predate the ref tag. . -- nae'blis 20:45, 8 January 2007 (UTC)

Or plain external links, or parenthetical citations, .... Christopher Parham (talk) 01:44, 13 January 2007 (UTC)

RefBot has recognized all those for several versions, so it can be done. I'll add {{citation}} when I finish its new core. Tagging unreferenced articles is trivial, but if nobody cared enough to supply citations in the first place then why tag it? (SEWilco 03:49, 20 February 2007 (UTC))

Some time ago, I wrote a script to make a list of unreferenced math articles, and while it is not impossible, it is harder than you expect because of natural language issues. The section headings that are used for references in articles are quite varied. Not all referenced articles use any sort of citation template. You should expect several percent false positives the first time you search for unreferenced articles (which I estimate at around a hundred thousand errors if you scan the entire article space and exclude redirects and disambiguation pages). CMummert · talk 04:26, 13 January 2007 (UTC)

Thanks for the response! I don't imagine it would be easy, and whoever did create such a bot would deserve a few barnstars! I'd think that the text of any template of this kind should include a statement asking for refs and inviting users to remove it where references are provided. Mostlyharmless 10:50, 13 January 2007 (UTC)

• Actually, mass-tagging those pages won't really help unless you've got a cadre of volunteers to do the actual referencing. We have a plethora of cleanup processes and nearly all of them are terminally backlogged. >Radiant< 12:34, 16 January 2007 (UTC

I agree. Cleanup lists aren't working all that well at the moment. This is more about giving a clear reminder to users that they have a duty to reference their articles, and getting them to do so voluntarily. I certainly don't think that most editors would do so, but a sizable minority would, particularly on anything that isn't an unwatched and unattended stub. And of course the template could go onto every new article. Mostlyharmless 02:11, 24 January 2007 (UTC)

Use prod instead

Rather than creating yet one more process for people to learn, let's just scrap this proposal and send unreferenced articles to Prod instead. Johntex\^talk 20:17, 6 January 2007 (UTC)

The trouble there is, a prod can be removed for any or even no reason. A well-meaning but inexperienced editor may remove the prod tag from an article without adding a source, and that action cannot be reversed even if they have not addressed the prod concern, in which case the only options are to AfD (which is a drain on the community's time, and may bring WP:ILIKEIT into play), or allow the unreferenced article to remain. This criterion makes clear that sourcing is the problem, and that sourcing must be performed in order to stop the deletion. It also allows more time in which to do so, in case someone needs to seek an inter-library loan or the like, that can't always be done in five days. Also, this makes it clear, by adding a specific procedure to deal with lack of sourcing, that sourcing is a big deal. That's a point I'd like to see gotten across very clearly. Seraphimblade 21:19, 6 January 2007 (UTC)

I think that point is very clear already from WP:V and WP:RS, both of which are prominently displayed and frequently cited. There are also a number of great temlates to use in calling for references. The article Talk page is a great resource to discuss which statements are most in need of sources. There are barnstars to be given to people who do a good job of referencing. There is a growing number of strong WikiProjects that are a great resource to find people willing to help out on any given subject matter. There is even my favorite WikiProject: Wikipedia:WikiProject Fact and Reference Check which has a goal of multiple in-line sources for every statement. Let's use all these channels more effectively, instead of creating yet one more process for people to try to learn about, follow, and enforce. Johntex\^talk 00:44, 7 January 2007 (UTC)

I don't think there could be any harm done by trying to prod these sorts of articles. Why not try it and see what happens and then re-evaluate? I would suggeset trying it with a few different "reason" messages and see if any get a better response then the others.--Birgitte§β ʈ Talk 13:45, 8 January 2007 (UTC)

In many cases, prodding an unreferenced article would only have a tiny chance of getting the article deleted, which means there is no reason to do it. The deletion policy suggests that a mere lack of sources is not a reason to nominate for deletion. Unless someone has a good resaon to think that the article is not verifiable in principle or not notable in principle, there is no reason to nominate the article for deletion. There is no policy saying that material that is verifiable in principle must have sources to stay in wikipedia. CMummert 14:07, 8 January 2007 (UTC)

Maybe I am behind on this disscusion, but how can we expect to get consensus for a speedy deletion for unsourced articles if it is not even a valid reason for an AfD nomination? I am really surprised to learn this. I had thought this discussion was about articles which were a pain to take through AfD, not articles which were invalid for AfD nominations. Shouldn't the regular deletion policy be changed before making this some kind of speedy criteria? I don't think it is a good idea to make an end-around on existing policies in this way. But maybe there is something behind this that I don't realize.--Birgitte§β ʈ Talk 18:41, 8 January 2007 (UTC)

That's not quite correct-the deletion policy does allow deletion for lack of sources, it just encourages you to check and request sources from the article's creators first. There's actually an ongoing discussion/situation we can use for a case study there: Wikipedia:Articles for deletion/List of minor Babylon 5 characters. This is a perfect example of how people don't understand how important sourcing is, especially secondary sourcing. A lot of the stuff in that debate is "Well it references that it happens in the show!" Firstly, it should at minimum reference a specific episode to use a primary source that way. The use of primary sources should be restricted only to totally obvious and uncontroversial stuff-and sometimes it's not very obvious how easy it is to go past facts to interpretation. "Jane shoots Ruth in episode 5 of season 3 of The Dumb and the Dumber." Alright, so long as that actually happens, fair enough. "Jane shoots Ruth in episode 5, season 3 of The Dumb and the Dumber because Ruth slept with her boyfriend in episode 3." Oops! Now we're into interpretation. Even if Jane later says that's why she did it, she could be lying, or there could be other motives-it's not for us to decide that for our readers. Now, of course, that's solvable-we can either directly quote Jane state she did it for that reason (Right before the shooting, Jane says "I'll get you back for stealing him you @#$@^*&!"), or we can quote or paraphrase a secondary source (Bad Shows Weekly states "Jane had always held a grudge against Ruth, and catching her sleeping with John was the final straw"[cite to BSW here]-ah, but now we've got more and better material, this adds that there was always bad blood there-an unacceptable interpretation if we offered it, but perfectly fine to quote a secondary source on.) This is why it's always best to quote secondary sources, and that's why we've got to get across to people that no matter how obvious it seems to you, if anyone might not know it or might challenge it, it needs a source. Seraphimblade 19:09, 8 January 2007 (UTC)

(outdent) Anyone can nominate an article for deletion for any reason; there are no hard and fast critera before you can make an AfD nomination. The question is whether the article has any chance of actually being deleted. Neither WP:V nor WP:ATT requires that every article have sources; they only require that the material is verifiable in principle. Moreover, WP:DEL lists a lack of sources as a reason for which deletion "may not be needed". So an article in which the material is clearly verifiable in principle about a topic that is clearly notable (example: History of Maine) is unlikely to be deleted even if it has no sources. I have raised this point before. CMummert 19:32, 8 January 2007 (UTC)

Right! The word is verifiable not verified. Verifiable means can be verified if the notability or the fact is challenged. I'd think this should normally be done in the talk page of the article. Then if it were a fact in the article, and consensus cannot be reached, there are other directions for settling the point. If it affected the entire article, then it would be appropriate for an AfD, with the nominator referring to the prior discussion. If everyone there agreed it couldnt be verified, then and only then a prod. DGG 22:47, 8 January 2007 (UTC)

Not too sure about that; the definition in WP:V is '"Verifiable" in this context means that any reader should be able to check that material added to Wikipedia has already been published by a reliable source', so to me the policy means that it should be obvious from the article what sources have been used (the 'has already been published' is suggestive that this is actually a sourcing requirement rather than some sort of 'truth' requirement). Verifiable 'in principle' would presumably mean that 'any reader should be able to check in principle that material added to Wikipedia has already been published by a reliable source', which doesn't make a lot of sense. I think the word 'verifiable' is misleading in the context of WP:V, but unfortunately there isn't much chance of changing it now. --ais523 10:32, 9 January 2007 (UTC)

That is not the way that I (or others I have learned from) interpret WP:V in the context of AFD. (There are other contexts, such as biographies, where the sourcing requirements are tighter.) Look at the article Bail: it has no sources provided at all right now and has been tagged since June 2006. However, it is obvious that a verifiable aricle could be written on this topic, and the topic is obviously notable, so an AFD would not succeed in removing the article, no matter how poor the current state is. So putting a prod on this article would be inappropriate. Similarly, History of Maine is a newish article, quite long now, with no sources listed. But it should not be given a prod because there is no chance that it would actually be deleted - the topic is clearly notable and a verifiable article could be written. CMummert 12:44, 9 January 2007 (UTC)

For articles where finding sources might be doubtful, it is of course sensible--and prudent--to add them at the first, and we are certainly justified in asking for them in AfD, and we do ask, and we do delete if not forthcoming. But many people with specialized interests and little experience here write such articles without them, and we need a gentle method to give those new editors a chance for the sake of their subject, and for the sake of retaining them as editors. I think the proposed time delay is a goo gentle way, much better than the current speedy. For some thing in speedy, there is real reason to get them out immediately, but not for ones just unsourced. DGG 19:17, 9 January 2007 (UTC)

Note, though, that WP:V#Burden of evidence makes it clear that as soon as someone challenges an item in an article, it must be referenced from a cited source, or it can be removed. It is clear that the standard is that reliable published sources must already exist, and that if any 'fact' cannot be cited from published reliable sources, it does not belong in Wikipedia. That means that I could go to History of Maine and request citations for every bit of information in the article, and any information that was not supported by citations to reliable published sources could be removed from the article. Personally, unless it is obvious vandalism or potentially negative material about living persons, I wait at least two weeks after requesting citations for items before removing them, and I do not aim to remove all unreferenced items from an article in one swoop, but that is my style, and I still support other editors' rights to be more aggressive in removing unsourced material. -- Donald Albury 22:27, 10 January 2007 (UTC)

If all editors were as careful as you, we would need fewer rules. Anything less than10 days risks discouraging the occasional editor, who will then never become a regular. No rule will protect against a truly irresponsible editor, but a standard at least provides evidence of the irresponsibility, which is useful in encouraging them to follow procedure. It is almost never the case that a V issue has the potential for immediate harm, except of course libel, which is covered elsewhere.DGG 22:38, 10 January 2007 (UTC)

This raises the question of why a speedy deletion criterion for unsourced articles is necessary. CMummert · talk 04:34, 13 January 2007 (UTC)

It isn't really a speedy (that was a pretty unfortunate name). The proposal is actually to allow 14 days after tagging before deletion if the article is still unsourced, which is actually quite a bit longer then prod or AfD allows. Seraphimblade 05:33, 13 January 2007 (UTC)

Userfy is a good option

I think proposal 1, allowing speedy deletion for anything unsourced, is likely to lead to deletion of false positives - articles that could be sourced, but (1) nobody wants to bother, or (2) there is hardly anything on the Internet (e.g. for historical people). I think UfD might work, but as a participant in AfD discussions, I like having every kind of deletion problem discussed in one forum. (I don't read PROD.) I think 2 and 3, which userfy content without deleting it irreversibly, are the best of the options given. —The preceding unsigned comment was added by YechielMan ([[User talk:YechielMan|talk]] • contribs) 05:34, 15 January 2007 (UTC).

Proposal one is only for new articles. The point of the whole proposal (in my opinion) is to make sure that the author of the article does bother to provide the sources used to write the article, as they are the only person who can do so! If the author of the article wasn't using a source, then it's original research and therefore ought to be deleted anyway. --ais523 17:09, 15 January 2007 (UTC)

The "author" of the article is not the only one who can find sources on the subject. WP is cooperatively edited, and--in fact--there is no author of any article. There is merely the editor who begins the article, or adds substantially to it. Very few articles here stand in good form except after many people's work. DGG 22:58, 23 January 2007 (UTC)

Anyone who adds material to an article, but cannot be bothered to cite any sources, is being discourteous to the other editors who later have to try to find reliable sources. We need to actively discourage editors from adding unsourced material to articles. We want editors to collaborate on articles, but we should also be asking them to back up what they add with reliable sources. Editing collaboratively does not absolve any individual editor from the responsibility to provide sources for material that the editor adds to an article. -- Donald Albury 11:42, 24 January 2007 (UTC)

To add to this, I would state that collaboration is far easier when sources are already cited-future editors can easily look to the already-cited sources for more information and ideas for future research. Seraphimblade 12:04, 24 January 2007 (UTC)

In case I was unclear, I'll rephrase it: the editor of a revision is the only person who knows the sources for the edit made in that revision. In some cases (such as vandalism reverts), I can understand why a revision wouldn't need sources (although when adding content to an article I try to make sure that I give a source for my edit, even if it's just 'see reference 1' referring to a reference that's already there); however, the edit that creates an article always needs sources, and it's in that sense that I was referring to the article's "author". --ais523 12:08, 24 January 2007 (UTC)

Unparseable

This is a non-sentence (in the policypage intro): "Even if the information is those things, since it is unverifiable, we (and our readers) have no more reason for trusting the claim than for trusting any of the other bad unsourced claims out there." — SMcCandlish [talk] [contrib] ツ 08:59, 31 January 2007 (UTC)

The sentence is parseable, but relies a lot on context. I can add the context into the sentence like this: "Even if the information is [reliable, accurate, and neutral], since [the article] is unverifiable, [Wikipedia] (and [its] readers) [has] no more reason for trusting the claim[s] [made in the article] than for trusting any of the other bad unsourced claims [in Wikipedia].". --ais523 11:00, 31 January 2007 (UTC)