Wikipedia talk:Proposed deletion process for unsourced articles

Initial thoughts

This is slightly more palatable than the original, but my same concern applies - why won't this encourage false sourcing? --badlydrawnjeff talk 21:50, 16 February 2007 (UTC)

It also encourages lazy people to use deletion as a tool to force clean-up. If something is unsources, try to source it. Only try to delete if it proves unverifiable.--Doc^g 22:00, 16 February 2007 (UTC)

If someone is routinely falsely sourcing, they'll eventually get caught (especially with Google Print and Amazon's preview features, it's easy to find out if an offline source actually exists and says what they say it says). Knowingly providing false sources should be a block for hoaxing, whether or not we implement this proposal. Seraphimblade ^{Talk to me Please review me!} 23:25, 16 February 2007 (UTC)

If an article is created the way it should be, sourcing should be a trivial matter. Night Gyr (talk/Oy) 23:56, 16 February 2007 (UTC)

Prod already does this?

People can already use {{subst:prod | no sources}}, so what additional value does this add? A category, perhaps, but we already have Category:Articles lacking sources. Friday (talk) 21:54, 16 February 2007 (UTC)

Yeah, but you can remove a prod for that without providing sources. --badlydrawnjeff talk 22:18, 16 February 2007 (UTC)

Sure, but if something's prodded for having no sources, it's unhelpful in the extreme to unprod without sourcing it. No responsible editor would do that. Friday (talk) 22:20, 16 February 2007 (UTC)

Anyone using {{subst:prod | no sources}} should be blocked for vandalism. If it lacks sources - then bloody well try to source it. If you can't then {{subst:prod | appears to be unverifiable}}, will be quite acceptable.--Doc^g 22:21, 16 February 2007 (UTC)

Well, let's not overstate the case. Many people prodding for no sources probably really mean "I think this is unverifiable", they're just not saying what they mean in very exact terms. I speedy delete unsourced stuff all the time for lack of verifiability, and I'm sure many others do too. But, yes, the difference between "this is not verified right now" and "this is not verifiable" is important. Friday (talk) 22:27, 16 February 2007 (UTC)

Yes you are possible right, but the distinction is very important, and I've seen too many debates recently where someone hasn't even done a basic google and simply points to WP:V and says 'the onus is on the one wishing to keep it to source it' - well, sorry, that's not good enough. The distinction between unverifiable and currently unverified is vital. I'm sure all the stuff you've deleted is not simply on the grounds of it not being sourced, there will always have been more to it than that.--Doc^g 22:32, 16 February 2007 (UTC)

Then again, by calling something currently unverified, but not unverifiable, we're just asserting that proper sources exist but haven't yet been found. This is generally fine in the short term, but if enough time goes by and nobody has yet found sources, its not unreasonable to say "this is unverifiable". Friday (talk) 22:45, 16 February 2007 (UTC)

Yes, I don't like things hanging about 'just in case a source turns up'. However, anyone saying "this seems unverifiable", should at least have made a rudimentary attempt to verify it and failed. I'm content that once someone has made a determined stab at verification and not succeeded, then the onus does fall on anyone who thinks the thing should be kept to source it.--Doc^g 23:07, 16 February 2007 (UTC)

Whenever I prod or AfD for lack of sources, I make sure to make clear what I've already tried to find sources (running the name through Google/Google Print/Google Scholar, and if it looks reasonable I'll usually give it a run through Proquest too). I tried to specify in this that it should not just be used because an article lacks sources, but that before tagging someone should do at least a reasonable search. I'll try to clarify that here. Seraphimblade ^{Talk to me Please review me!} 23:23, 16 February 2007 (UTC)

Just keep the distinction between 'currently unverified' and 'apparently unverifiable' clear, the later are quite properly nominated for deletion.--Doc^g 23:28, 16 February 2007 (UTC)

(indent reset) I edited the second section to clarify that, what do you think? Seraphimblade ^{Talk to me Please review me!} 03:15, 17 February 2007 (UTC)

Encourages deletion instead of sourcing

Oppose. In general, if something is unsourced you shouldn't nominate it for deletion, you should try to source it. Only if you honestly believe it can not be sourced (in other words, it's Wikipedia:original research or an outright hoax) you can ask it be deleted, but not otherwise. The only exception is controversial information in Wikipedia:Biographies of living persons. But otherwise, it is terribly discouraging to new editors who are not good at citing sources to have their articles nominated for deletion by someone who didn't care enough to do a basic search for sources. --AnonEMouse ^(squeak) 22:03, 16 February 2007 (UTC)

The notion that unsourced material is removed from Wikipedia, unless the person submitting it sources it is a misunderstanding of WP:V. It does not mean that I can say, 'hey this is unsourced', then sit back, and if you don't source it in 24 hours it gets removed. It means that if I challenge the material either as 1)dubious or untrue, or 2) unverifiable (and I've tried to verify it), then the onus shifts to the one who wishes it kept to source it. We try to source unsourced material, we don't remove it (except by BLP). We remove it only if it is proves unverifiable or is subject to factual challenge. The problem at the moment is that certain people are trying to turn WP:V into a deletionists charter - it isn't. Improvement not deletion is always the goal - deletion is a last resort.--Doc^g 22:18, 16 February 2007 (UTC)

Which this states-I figured on that misinterpretation/misuse, and actually put specifically to look for sources before tagging. I would absolutely agree that one should look for sources before removing information (unless it's BLP, and even then you should remove and then look). However, there's a lot of unsubstantiated and apparently unverifiable stuff going up, and I think this would help with it. Seraphimblade ^{Talk to me Please review me!} 23:21, 16 February 2007 (UTC)

If I can't prove information right or wrong, why shouldn't I remove it? Night Gyr (talk/Oy) 23:57, 16 February 2007 (UTC)

Unverified/unverifiable is a false dichotomy. The easiest way to demonstrate that the content of an article is verifiable is for it to be verified. Any admin that blocks someone for a no-source prod should themselves be blocked for abusing the tools. Chris cheese whine 09:33, 17 February 2007 (UTC)

Because there's a good chance you just don't know where to look for confirmation. Really, that amounts to an "I've never heard of it" kind of argument, but extended to include "and I don't know where to look for it". -86.148.127.202 16:46, 17 February 2007 (UTC)

Concur the difference between unverified/unverifiable isn't significant. In reality, at the most, I would perform 4 or 5 separate web searches (google, google news archive, google news, google book, plus amazon books). For the avoidance of doubt, these searches don't demonstrate "apparent unverifiablility". Any admin who was to block a good faith user over a supposed grammatical error in an edit summary should be desyopped. Addhoc 18:38, 17 February 2007 (UTC)

So I'm giving the author 14 days to tell me, before I delete it. Night Gyr (talk/Oy) 18:53, 17 February 2007 (UTC)

oppose the distinction between unverified and unverifiable is very critical, and anyone who doesn't get that doesn't understand WP:V. Much of Wikipedia is currently unverified. Much of it will never need verified. But providing the material isn't unverifiable that's not a problem. The statement 'Mars is a planet in the solar system' does not require verification, but it is certainly verifiable. Statements that are currently unverified may be removed if challenged - the onus being on the one who wants to keep it to verify it. That's fine, but by challenge we mean that the challenger 1) disputes the factuality of the information or 2) asserts that the claim appears unverifiable. Only if challenged on those grounds, is there a burden on the one wishing to keep the statement to source it. I am not allowed to point to a claim and say 'that is unverified and if you don't verify it, I will remove it'. Only if I say 'I believe that to be false' or 'I've tried to verify that and it appears unverifiable' is the claim removed unless sourced. Yes, we must ensure that unverifiable material is removed - but unfortunately as it stands WP:V is being abused by deletionists as a pretext to remove things they don't like. If you see something that is unverified, your initial response is to attempt verification, not deletion.--Doc^g 22:45, 17 February 2007 (UTC)

There should be no exceptions to sourcing requirements for new articles. We simply should not be creating unsourced articles. see User:Uncle G/On sources and content. Night Gyr (talk/Oy) 00:28, 18 February 2007 (UTC)

So source them.--Doc^g 00:43, 18 February 2007 (UTC)

Burden is on the person adding the material. Attempting to source before removing is desirable but not mandatory. Night Gyr (talk/Oy) 00:47, 18 February 2007 (UTC)

Bullshit.--Doc^g 00:49, 18 February 2007 (UTC)

Actually, Doc, he's quite right, swearing aside. From WP:V:

• Any edit lacking a source may be removed, but editors may object if you remove material without giving them a chance to provide references.
• Be careful not to go too far on the side of not upsetting editors by leaving unsourced information in articles for too long, or at all in the case of information about living people. Jimmy Wales has said of this: "I can NOT emphasize this enough. There seems to be a terrible bias among some editors that some sort of random speculative 'I heard it somewhere' pseudo information is to be tagged with a 'needs a cite' tag. Wrong. It should be removed, aggressively, unless it can be sourced. This is true of all information, but it is particularly true of negative information about living persons."

So actually, Night Gyr's dead right. I personally agree with you that it's desirable to do a source search oneself first, but it is unequivocally not required. Seraphimblade ^{Talk to me Please review me!} 00:59, 18 February 2007 (UTC)

No. Of course, material that is controversial, prejudicial, or just plain dubious should be removed if it is unreferenced. Any uncited material that is subject to factual challenge gets immediately removed. I've zero-tolerance in such cases for the notion that we leave it as 'maybe a citation will be found later'. Similarly, if I can't verify something, I will remove the material, it can be replaced if verified later. But we don't remove information simply because it is currently uncited - or we'd literally be removing half the encyclopedia - including statements like 'mars is a planet'. From WP:V "Editors should provide a reliable source for quotations, and for material that is challenged or likely to be challenged." We certainly don't delete articles simply for being unsourced, especially if three minutes on Google would produce 50 reliable ones. Anyone nominating something for deletion simply on the grounds of lack of sources, who hasn't even bothered doing a basic google, is frankly being disruptive. The purpose of WP:V is to make us verify things and remove the currently unverifiable, it is not a deletionist's charter.--Doc^g 01:20, 18 February 2007 (UTC)

Your worries are about undermining the existing content, and I agree that generally good articles should have a request for sources before going straight to deletion, but should people keep creating new articles with zero sources? Night Gyr (talk/Oy) 01:25, 18 February 2007 (UTC)

Ideally, no. But if the articles are actually verifiable, then we should verify them and not delete them. Removing unproblematic verifiable content from wikipedia is unacceptable. They should only 'go for deletion' if the nominator has a reasonable belief that they are unverifi-able.--Doc^g 01:30, 18 February 2007 (UTC)

(edit conflicted reply) Well, anything can be taken to the point of disruption. We allow any editor to remove a prod tag. That's a good policy, it's what makes prod work. But if I go through CAT:PROD and remove every prod tag without any explanation, I'll probably be seen as causing a disruption. Similarly, removing "Mars is a planet" might be a bit pointy. But in almost all cases, that's not what we have. You say "Don't remove non-dubious information!" Dubious to whom? Me? You? A longtime fan of American Idol might know tons of details about that show I'd find dubious and questionable, but that they know and accept as a matter of course. On the other hand, that person might not have the first clue that a half life has anything to do with anything but a popular video game, and find that a very questionable claim, but I'd know that in my sleep. The fact that anyone might find something questionable means it is questionable, and needs to be sourced. And really, how hard is it to find a source saying that yes, Mars is indeed a planet? But WP:V is very clear-err on the side of removal. No, that doesn't mean every last sentence, every last word must be removed or sourced-quite often, a couple of sources support an article's whole content. But just remember that because something is not in the slightest dubious to you, we have a worldwide audience which does not necessarily draw from the same knowledge base we do. Seraphimblade ^{Talk to me Please review me!} 01:34, 18 February 2007 (UTC)

If you have reason to believe that something maybe untrue, and it is unsourced, then challenge it, and it the challenge is not met, then remove it. I'll do the same. Sure we'll have different knowledge bases. But we both agree that mars is a planet - so no need to challenge that. And if we're not sure if Bob won American Idol, then a quick google will tell us whether to challenge it or not. But we don't go about removing information that is probably true or easily conformable without checking. If you're not sure that mars is a planet, then ask the person who's making the claim to provide a source, but if you have no reason to doubt it, then don't.--Doc^g 01:44, 18 February 2007 (UTC)

And in this case, the challenge/request to add a source is slapping a {{sourcedelete}} tag on it, and giving them, and anyone else who cares, a 14 day deadline before you remove it. If things are so trivially sourceable, 14 days should be plenty of time to add at least one. Night Gyr (talk/Oy) 04:25, 18 February 2007 (UTC)

Also, counterexample: I have no reason to doubt that the GNAA is a trolling organization, but it got deleted anyway because it couldn't be sourced. Night Gyr (talk/Oy) 04:39, 18 February 2007 (UTC)

Many things are this simple, but not all. The ones that still require printed books can take much longer, & much commercial or web material is v. difficult to source conventionally. What is needed is for everyone concerned to try to source material, rather than for people to set out to delete. Many things I've sourced on a challenge to some article at AfD, where the nominator could just as well have done so & saved everyone's time. (And, as User:Seraphimblade says, one should usually avoid challenging on a topic one doesn't understand. ) Writing WP articles shouldn't be a contest--the bias should be that everything valid can be sourced, & we should use longer periods & consultation with experts when needed. DGG 05:34, 18 February 2007 (UTC)

Uh...I did not say that, I actually agree with WP:V-err on the side of removal. If someone needs a month to find a source, that's alright! They can then create the article in a month. What I am looking to change is the "create from memory, source/remove inaccurate crap later" that seems to be pervasive. How about "source first, create second?" Seraphimblade ^{Talk to me Please review me!} 05:38, 18 February 2007 (UTC)

But if you're creating a new article, shouldn't you already have the source at hand? If you don't, we can bring it back when you find the info. If things are that obscure, you could be adding wrong information as easily as right information and no one could know, even the person mistakenly adding incorrect information. Night Gyr (talk/Oy) 06:07, 18 February 2007 (UTC)

Not necessarily, especially when you're creating stubs. --badlydrawnjeff talk 06:16, 18 February 2007 (UTC)

Jeff, do you think we should allow stubs that are based on original research, or are not neutral? I wouldn't imagine you do-but if so, what makes unverified material any different? Seraphimblade ^{Talk to me Please review me!} 06:29, 18 February 2007 (UTC)

Unverified or unverifiable? The latter, of course not. The former can be fixed with some effort. --badlydrawnjeff talk 02:41, 19 February 2007 (UTC)--badlydrawnjeff talk 02:41, 19 February 2007 (UTC)

Lack of sourcing for articles is a severe problem. My opinion is that anyone who creates an article with no sources is either ignorant of our policies, or is simply being rude to every other editor and reader. If the editor is unfamiliar with the verifiability policy, then PRODUS is a good way to educate them while giving them plenty of opportunity to supply some sources. If the editor is aware of the verifiability policy, and creates a new article (or adds to an existing article) without providing sources, it says to me that they don't agree with the verifiability policy, or don't care, or are too lazy to be bothered. Saving an unsourced article that someone else created (or expanded) by finding the sources they couldn't be bothered to provide is enabling a culture that sources don't matter. We need to be building a culture that everything needs to be sourced from the start. -- Donald Albury 11:36, 18 February 2007 (UTC)

...is a good way to educate them..., in the same way that public executions work so well. Deletion is easier than sourcing. That doesn't make it more useful and more like building an encyclopedia. It is entirely possible that a new, helpful, and enthusiastic editor is ignorant of our policies (we have one or two, after all); this does not mean we should set fire to everything they do. You seem to finish by going further: that finding sources for someone else is a bad thing, and that we should instead be building a culture of deleting things as an educational process! This has been the problem with the proposal from the start - that those who would see this done know their beans and wish to actively punish those who haven't got the hand of it yet, mainly because they might not have heard of the obscure topic the other party is writing about. The bar to contributing to Wikipedia was low when you joined, and it should remain low in the future. That's how this whole thing comes to exist in the first place. 86.148.127.202 13:33, 18 February 2007 (UTC)

Having to source what you write isn't some kind of hoop that we're trying to make newbies jump through, it's an essential part of creating a reliable reference work. Night Gyr (talk/Oy) 19:14, 18 February 2007 (UTC)

How does deletion aid the creation you mention? 86.148.127.202 22:33, 18 February 2007 (UTC)

In the same way that speedying vanity bios and spam aids with the creation of non-vanity bio, non-spam articles. It acts as a filter. It keeps the signal-noise ratio higher. That's why we have deletion procedures at all-there are some things that just don't belong here. Unverified material is one of those things. Those who create an unsourced article are, basically, expecting everyone else to do what they should have done in the first place. And deleting spam and vanity bios is not either primarily intended to teach or to punish-it's intended to remove material which does not belong here. The same is true of this proposal-its primary aim is to remove unverified content, any other benefits would be incidental. Seraphimblade ^{Talk to me Please review me!} 23:17, 18 February 2007 (UTC)

Oppose: Conflicts with WP:BOLD. Responsible editors put a "citation needed" tag first before removing contentious material. This policy would just encourage the sort of editor who tends to delete without offering justification, on the grounds that the "burden of proof" lies with contributors. There are many areas where having some material is better than having none. This proposal just encourages edit wars and newbie-stomping, neither of which is good for Wikipedia. Avt tor 20:21, 23 February 2007 (UTC)

If the article is unsourced, how can we tell whether or not it belongs in Wikipedia? By the way, the 'burden of evidence' lying on the contributor is policy. Anyone who adds unsourced material is doing Wikipedia a disservice. It is not newbie-biting to help new users learn about our attribution policy. -- Donald Albury 03:01, 25 February 2007 (UTC)

Oppose, for same reasons as the no source, no article essay. First off, there is no deadline on completing articles. Secondly, the whole verifiable/verified thing. Lastly, applying this would allow hundreds of thousands of articles to be put up for deletion. I don't think anyone could keep up with that.--Rayc 03:44, 25 February 2007 (UTC)

First, do you think that we should have articles that have no source? That would make them original research. Sedcond, how do we know that an article is verifiable unless source are provided. Finally, your last reason is spurious; this proposal would not apply to existing articles, but only to new articles. -- Donald Albury 00:41, 26 February 2007 (UTC)

...and anyway this is already rejected

...because it's a simple rehash of whichever version of the old proposal. That's already been rejected (principally for being too draconian); I don't see what is different about this proposal. Add a tag, delete in 14 days. No different. This has already been rejected. 86.148.127.202 13:35, 18 February 2007 (UTC)

The difference is that the old proposal was about speedy deleting unsourced articles, whereas this proposal isn't.Hut 8.5 13:55, 18 February 2007 (UTC)

No, it was always 14 days. It just used the word 'speedy' as that is the terminology that comes easily to mind. 86.148.127.202 15:34, 18 February 2007 (UTC)

Speedy deletion is completely different from proposed deletion, and a lot of the objections on the old proposal were because of the speedy deletion component, so the old decision does not apply. Hut 8.5 20:16, 18 February 2007 (UTC)

Speedy is very different to prod, yes, but in the opening lines of the old proposal, it refers to 14 days. The old proposal was never speedy is the usual sense of WP:CSD, and at least one of the mechanisms suggested there (i.e. the first one!) is essentially identical to this proposal. In any case, the principal objections to that proposal were that deletion is not the broadly-desired solution, and that instead of burning our way through articles (because that's decidedly easier), we should be trying to source them. The same criticisms apply here; and there is clearly no consensual agreement that the actual basis of either the old or new proposals is a basis that is to be proceeded upon. 86.148.127.202 22:31, 18 February 2007 (UTC)

The old proposal went through many revisions and evolutions. At its start, it did say speedy-deletion. The modification to 14 days came much later. While the proposal remained contentious, this new proposal was intended to give the idea a fresh start - clean from the early objections which were based on the original speedy-language. Rossami (talk) 02:05, 19 February 2007 (UTC)

It was actually suggested at the old proposal that a new one should be made, to give a "fresh start" to the discussion. By the end of the discussion, the name of the proposal ("speedy deletion") didn't even match what was being proposed, and the whole thing was a confused mess. Basically, there wasn't a single proposal that got rejected there-the whole thing basically turned into a train wreck, as the thing got changed and old comments didn't even apply to what was there anymore. This seemed the most reasonable and least bureaucratic of the various things that got suggested there. And yes, of course, it's taken from one idea there that did seem to have pretty widespread support-the whole point of leaving rejected proposals around is so that we can learn from them! Seraphimblade ^{Talk to me Please review me!} 02:35, 19 February 2007 (UTC)

Alternate route?

Would it be possible to achieve this result by altering the existing WP:PROD policy? This would require two principle changes

• the introduction of templates giving standard reasons, in the same way we have {{db-bio}}, we could have {{prod-us}},
• modification of WP:PROD, such that removal of the prod template is only valid if the stated reason has been complied with.

The first stage, creating {{prod-us}}, could possibly happen now. Addhoc 14:09, 18 February 2007 (UTC)

Not on the back of a rejected proposal, no. If a broadly-agreed-to proposal can be formulated, then of course that's entirely different. But those proposals for which people have gone out and stuck their template on articles before it has reached such a state are those that generally cause the most upset and get dismantled the fastest. (Plus, what you suggest is identical to the main idea here, so I don't see it as an alternative route). 86.148.127.202 15:34, 18 February 2007 (UTC)

It's different in that it can be an extension of the existing WP:PROD policy rather than a new policy. Hut 8.5 15:48, 18 February 2007 (UTC)

Same difference. New part of existing policy = new policy. 86.148.127.202 15:54, 18 February 2007 (UTC)

Not sure I've explained this properly. I've created the {{prod-us}} tag, which doesn't really involve any policy modification. After the template is substituted, its completely identical to a normal prod. Addhoc 16:58, 18 February 2007 (UTC)

I like that new tag idea. Though, as of right now, anyone can remove it for any reason, still and yet-I think this could still be problematic. I've seen many times where people will deprod for no reason when an article is prodded for being unverifiable, and leave no reason. If asked, they'll either give WP:ILIKEIT, an "Include all X", or "There's got to be a source out there" (despite descriptions of the search efforts in the prod!). Of course, in no such cases does the person actually add a source. It would sure be nice if that could be made non-optional. Seraphimblade ^{Talk to me Please review me!} 23:25, 18 February 2007 (UTC)

cusotmers who bought this, might also be interested in....

m:User:Doc glasgow/seconding. A more comprehensive solution to ensure all new articles are checked by at least one experienced user.--Doc^g 15:37, 18 February 2007 (UTC)

Interesting. That's a bit like m:Patrolled edit. 86.148.127.202 15:54, 18 February 2007 (UTC)

Concerns about cultural impact

I remain deeply concerned about the potential impact of this proposal on the culture of Wikipedia and on the volunteers that we get. Some people have extensive academic experience with sourcing, footnoting and indexing and are very comfortable with the practice of explicit citations. Most of our editors, however, are not from that background. They are normal folks trying to improve the encyclopedia by adding what they know. More than that, we have to remember that they are all volunteers. As volunteers, we all have different skills and interests. Some start stubs, some copy-edit existing articles, some add categorization tags, some patrol for vandalism, some wonk on policy pages and some add citations. All of us have a place in the project. Forcing us to do parts of the project that we don't enjoy is going to frustrate users and drive otherwise good contributors away from the project.

Sourcing is important but using the threat of deletion as a sword over the head of every article (and by extension, every contributor) seems to me to be an over-reaction. Some have said that a 14 day window is a reasonable grace period to ensure that articles are sourced. I disagree. Not every editor is as deeply addicted to the project as are the people commenting on this page. Most of our editors read for a while, add an article in good faith and then get on with their lives, perhaps for months before coming back to see what's become of their contribution.

I also have concerns that if this policy is implemented aggressively, it will result in a pattern of fictitious citations. Yes, doing so is an act of vandalism. But so was adding the false content that drove the need for citation in the first place. Bad editors are going to game the system no matter what rules you put in place. Having a false citation in the article is far worse than having an article with no sources. With no sources, the reader can make up his/her own mind about what and how much to trust the content. Citations, especially print citations which can not be easily cross-checked, add the appearance of validity with no more confidence than we actually had in the original content.

If an article is unsourced and you have reason to be suspicious of the claims made in it, by all means, challenge them and if necessary pull them out. But if the claims make sense, Wikipedia:assume good faith.

Wikipedia has been so successful in part because it has been able to successful capitalize on our diversity of skills and interests. Locking us into a single mold - expecting every user to be as comfortable with footnoting as you are - is unreasonable and not a good thing for the project. Rossami (talk) 06:22, 19 February 2007 (UTC)

I'm more concerned, realistically, with the effect not implementing something like this would have on the culture here. Many newer editors may not be familiar with many of our policies. They may want to write an article about their viewpoint on something or based on personal experience. But we don't let them do that, nor do we accord them "months" to come back and see if they fix it-AfD is only 5 days, and speedy can be under a minute. Yes, many people have many different areas of expertise, but there are some things we require of everyone. Verification is, and well should be, one of them. What we're saying right now is "We care about NPOV, but you know, we really don't care that much if you're putting up your own interpretations or garbage information. We're not going to ask that people show what they're saying actually is verifiable." WP:AGF doesn't apply. Someone might make a POV fork or vanity bio in good faith. That's fine. It's not getting deleted for bad faith. It's getting deleted because, good faith, bad faith, or ugly faith, it doesn't belong here. Neither do sourceless articles. Seraphimblade ^{Talk to me Please review me!} 06:35, 19 February 2007 (UTC)

Moved from the village pump

Absolutely not. Terrible idea. We remove unverifiaABLE material. We don't remove something simply for being currently unsourced/unverifiED (except by WP:BLP), unless someone is challenging the veracity of the claim. If something is unsourced - don't nominate it for deletion - try to source it. If you can't then by all means nominate it as unverifiABLE. I'm sick of lazy people using deletion as a tool for clean up. Stop it.--Doc^g 21:54, 16 February 2007 (UTC)

And if I decide to level a blanket challenge? Per WP:V, unsourced information may be removed by any editor at any time. Night Gyr (talk/Oy) 23:49, 16 February 2007 (UTC)

Questioning the veracity of a statement or an article for any other reason than that you think it might be untrue, with the aim to have the article deleted, is a blockable offence (WP:POINT). -- Eugène van der Pijll 10:38, 17 February 2007 (UTC)

Agree with the Doc. AnonEMouse ^(squeak) 21:58, 16 February 2007 (UTC)

No, no, no, bad idea. —Doug Bell ^talk 14:30, 18 February 2007 (UTC)

I also agree with Doc. We really, really can't do this. This process will be horribly backlogged, and many worthwhile articles will be deleted because no online sources are available (or worse, all online sources are paid access that editors can't get to). Most unsourced content is either good or obviously false; this is not necessary. Just {{prod|No sources, questionable: possible hoax}} and get over it. --N Shar 07:07, 20 February 2007 (UTC)

Perhaps an unrelated question: Sources that are not available online are considered to have equal weight as those found online. I have read somewhere in either guidelines or policies that sources that can be found online are better (and I am not referring to scholarly works or rare books and the like). Is there such a preference stated in WP?Arcayne 22:42, 16 February 2007 (UTC)

I haven't heard of such a guideline and I'd strongly argue against it. We should use the most reliable sources possible, regardless of how they are available. For most articles, the best sources are not available online. --Cherry blossom tree 22:55, 16 February 2007 (UTC)

I absolutely agree. Online sources should never be considered to carry more weight then offline ones, nor should the reverse be true. If the source is reliable, it should count the same, period. Seraphimblade ^{Talk to me Please review me!} 03:32, 17 February 2007 (UTC)

The -ed/-able debate is utterly pointless. The easiest way to show something is verifiable is to actually verify it. Wikipedia:Cleanup has a backlog of around 18 months. Even with our slowest mirrors, that's not an acceptable timeframe. It should be made clear at this point that the onus is, and always has been, on those seeking to add information to justify its existence (which is why in turn we require consensus for controversial deletions). Chris cheese whine 09:37, 17 February 2007 (UTC)

Not at all. What matters is whether something is true or not, not ticking boxes on form. Most of the good material on Wikipedia was added without sources. I doubt that Wikipedia would have succeeded if it has all been deleted on sight. CalJW 00:21, 18 February 2007 (UTC)

Actually, WP:V says the precise opposite - it matters not whether or not something is true, but whether it can be verified. We can happily include information that is verifiably false, and point out that it is so. Chris cheese whine 10:58, 21 February 2007 (UTC)

Agree 100% with CalJW. Some people seem to treat the addition of unsourced material as a heinous crime, which is ridiculous. When I first started editing Wikipedia almost nothing was sourced and few people added references. If all that material is now deleted as unacceptable we are going to lose a hell of a lot of good material. I'm not saying that I don't agree with sourcing, incidentally, but I think there can now be too much of a mania for sourcing every little bit of information and too much of a mania for deleting period. -- Necrothesp 14:24, 18 February 2007 (UTC)

On the whole offline sources are better, they are just less accessible. An ideal Wikipedia would use mainly offline sources. CalJW 00:21, 18 February 2007 (UTC)

And again, I agree with CalJW. I've seen people claim an article should be deleted because it uses print sources and should only use online sources, on the basis that print sorces "can't be verified" by most users. There is absolutely no basis for this claim in Wikipedia policy, and it is frequently true that print sources are more detailed and more reliable. -- Necrothesp 14:29, 18 February 2007 (UTC)

The WPV1.0/Project assessment scale prefers "hard" offline sources for articles of GA class and above where possible, also - a lot of online refs die or will die with page moves etc and not all are accessible via the Internet Wayback Machine; whereas the majority of print sources such as books, journals and newspapers have physical archives in large libraries such as the British Library; their life and usefulness for inclusion into an encyclopaedia is arguably longer. - Foxhill 14:51, 18 February 2007 (UTC)

Burden of evidence

Doc changed "It is a logical extension of this core policy [Wikipedia:Verifiability] that unsourced articles may also be challenged and, if no sources whatsoever are provided, removed" to "It is a logical extension of this core policy that unsourced articles may also be challenged and, if no sources whatsoever can be provided, removed." In my opinion, the first version is the logical extension of Wikipedia:Verifiability#Burden of evidence, not the second. I think two weeks is ample time for at least one source to be provided. If not even one source is provided in those two weeks, it indicates to me that either reliable sources don't exist or that no one is interested enough in keeping the article to bother finding sources. Deleting a new, unsourced article does no harm to Wikipedia. If the subject deserves an article, there should be reliable sources available, and anyone who wants to keep the article needs to be responsible for citing those sources. I strongly disagree with the proposition that an editor proposing deletion is obligated to perform the search for sources that the creator and other editors who want to keep the article have not done. New editors who do not know about or understand our attribution requirements need to be educated about them. Editors who do know about the sourcing requirements, and continue to add unsourced information to Wikipedia are being lazy and rude to the rest of us. I already spend too damn much of my time trying to chase down sources for unsourced material. I thoroughly resent the implication that I am in the wrong when I nominate an article for deletion for lack of sources without having engaged in a major search for sources. The burden for finding sources is not on the editor who proposes deletion, it is one editors who want to keep the article. -- Donald Albury 14:19, 20 February 2007 (UTC)

I would say that the burden for finding sources for an edit is on the user who makes that edit, and on the creation of an article the burden is on the user who creates it. Only the person who adds the edit can know their sources; other people can only guess. If the person who makes the edit isn't using sources, then it's original research. I've already pointed out on WT:CSDUA how easy it is to find sources for 'obvious' facts; so there isn't any good reason to make unsourced edits. We have to warn editors that their edits are unsourced and may be removed, and remove them if the sources aren't forthcoming, because nobody else will ever know for certain what the sources were, and the article will remain unverified and unverifiable for ever. For an edit to be verifiable, it has to be verified. --ais523 17:59, 20 February 2007 (UTC)

Now, I often do search for sources to support an unsourced statement in an article, and I will edit the article if necessary to reflect the sources I found. I have helped save some article up for deletion this way. However, my objection is to the claim that I am obligated to search for sources before I can propose an article for deletion. -- Donald Albury 18:24, 20 February 2007 (UTC)

What about new pages?

A lot of new pages come in at Special:Newpages that are neither speedy candidates nor acceptable stubs. They're basic prose, without sources or wiki format, but often on noteworthy subjects. This policy would basically turn those into secondary speedy candidates, and would substantially inhibit the growth of the encyclopedia's information base. YechielMan 19:48, 20 February 2007 (UTC)

This proposal has nothing in it about speedies. Normally, those would see the creator asked for sources, and if none were forthcoming (or readily found on Google), the article would be prodded. This gives a much longer period then prod does, and makes clear that lack of sourcing is the problem. Once again-why is the burden not on the article writer to show that the subject is noteworthy, by showing who has made note? In this case, I fully agree with Jimbo Wales-better no information then garbage information. Unsourced information is "growth" we don't need. Seraphimblade ^{Talk to me Please review me!} 20:34, 20 February 2007 (UTC)

• • Often a user creates an article, then never looks at it again, but the information in the article may very well be good, even though it is unsourced. From doing newpage patrol, I have noticed that most unsourced articles are either obvious garbage or obviously valid. The proposed process will not help to delete the obvious trash; that can be prodded or speedied. The obviously valid articles need not be deleted; they need to be cleaned up and sourced. Why, therefore, are we proposing a deletion process? It makes no sense. --N Shar 20:41, 20 February 2007 (UTC)

Note that we already enforce this with anonymous users wanting to create articles at articles for creation. Why should registered users be able to get away with not providing sources when anons can't? Hut 8.5 20:40, 20 February 2007 (UTC)

At AfD debates when the issue is lack of evidence for Notability, it happens that during the debate someone --occasionally the "author", sometime the subject, sometimes an interested bystander finds good sources and demonstrates notability, and the nomination is either withdrawn or closed with a decisive keep. The article may have sat there for months or years, the sources may have been easily available from the first, but nobody does the work until the need becomes urgent. As it now stands, the nominator is strongly advised to do this first, before nominating the article for deletion--and I am sure many people do--they obviously aren't visible at AfD.

I think the true justification of this policy is to produce challenges that will succeed in getting articles properly sourced--not to provide a rationale for discarding good articles. But I wonder if the energetic implementation of this policy might provide too much work for the editors to handle.Most weeks I find I have time to source only one article, if that. The others who do it seem to have about the same capability. If ten times as many articles wer challenged, the editors who are willing to do this work could not keep up. DGG 21:02, 20 February 2007 (UTC)

Which is why it should be for new articles; I doubt anyone is creating quality articles too quickly to source them properly. Night Gyr (talk/Oy) 00:05, 21 February 2007 (UTC)

• • If the article creator doesn't give their sources, but sources are easy to find, is there really much harm done if the article is deleted? Seriously, sourcing an unsourced article is equivalent to rewriting it from scratch, as you have to change the article to match the sources anyway (if you don't they aren't sources), so deleting the article doesn't lose any information (either the information's in the sources, in which case it isn't lost, or it isn't in the sources, in which case it's original research and has to be deleted anyway). Information without sources is much less useful than a list of sources with no information given, but at the moment the first often hangs about for ages until it has to be deleted or completely rewritten, whilst the second is a CSD A3. --ais523 10:30, 21 February 2007 (UTC)

    • If the information is in sources and the article is deleted, then whilst the universe does not lose the information, Wikipedia obviously does! I don't see the point in deleting information that someone knows to be true just because I/you happen to be ill-educated in their particular area. Deletion isn't some tool for masking away the average ignorance of mankind. 137.222.189.198 11:03, 23 February 2007 (UTC)

• • • • We don't work from our own knowledge. Seraphimblade ^{Talk to me Please review me!} 11:08, 23 February 2007 (UTC)
        • That's a complete misinterpretation. The main reason everybody decides to write an article is because they know stuff about it. Our best articles are written not by the ignorant but by the specialist. NOR does not say "you must not use your own knowledge", it says that Wikipedia does not publish original thought. The two are totally and completely different. 137.222.189.198 16:17, 23 February 2007 (UTC)
          • See the related policy WP:V. Whether you 'know' something or not isn't the question, it's whether you can show you aren't making it up. Hut 8.5 16:29, 23 February 2007 (UTC)
            • And you would argue that, if 1) I forget a source and 2) you know nothing about my specialist area you will 3) delete the article out of, essentially, ignorance. Tell me directly: if you too knew something about that area, and knew the article to be correct in fact, would you still advocate deleting it? And, as Doc glasgow began the thread by saying, WP:V does not urge the deletion of unverified statements, only those that are challenged as likely to be incorrect and/or unverifiable. Setting aside Night Gyr's suggestion that he will simply walk around saying that all 1.6million articles are likely to be incorrect in toto, this proposal is neither a logical extension of WP:V, nor a re-statement of WP:NOR. It is, at its core, though not deliberately, a proposal to delete anything you know nothing about. Seraphimblade argues somewhere that it is to increase the signal-to-noise ratio; but one does not achieve that globally by deleting parts of the signal. One may achieve it locally to oneself, because now one doesn't have to worry about all that confusing mathematics/sociology/history/science etc but that is quite a different prospect to having increased the total quality of the encyclopedia by deleting factually correct information. 137.222.189.198 18:04, 23 February 2007 (UTC)

Alright, now I'm just going to leave this proposal to die the death it needs. 137.222.189.198 18:04, 23 February 2007 (UTC)

• We can categorize articles submitted to Wikipedia as follows:
  1. Obvious junk -- these already get deleted speedily
  2. Unsourced articles about non-notable subjects -- these already get prodded/AfD'd per WP:V and WP:N
  3. Unsourced articles about notable subjects -- subdivided as follows:
     1. Factually correct and neutral -- we don't want to delete these; since they are on notable subjects, they will eventually become useful articles
     2. POV or incorrect -- these are dangerous and should be deleted or corrected
  4. Sourced articles about non-notable subjects -- already go through prod/AfD per WP:N
  5. Sourced articles about notable subjects -- what we really want

This proposal, if implemented, would result in the deletion of many articles in categories 3.1 and 3.2. The essential question is: Is the benefit of deleting articles from category 3.2 greater than the cost of losing articles from category 3.1? --N Shar 21:02, 23 February 2007 (UTC)

Well, the 3.1 and 3.2 are not black and white categories - most unsourced articles are probably a mix of 3.1 or 3.2, with probably a tiny minority in 3.1 or 3.2. Unravelling what content is in 3.1 and what's in 3.2 is the nasty part, and probably takes longer than rewriting the article from a source. That being said, the proposed deletion option is probably my least favourite one - Userficiation or "draftspace" seem a lot better in comparison. ColourBurst 05:30, 24 February 2007 (UTC)

As you say, 1 (vanity bios, spam, etc.) are already speedied, and still would be. This actually would help with 2 though-there's no way to know whether or not something's notable without seeing sources, since notability is reliable secondary sources. Rather then having any fights over what's "notable" or not, this totally eliminates ILIKEIT/IDONTLIKEIT and puts sourcing as the priority. (Of course, prod or AfD still might be necessary due to inadequate or unreliable sourcing being all that's available, that's outside this proposal's scope. This is only for totally unsourced stuff.) On 3, again, how does anyone know if it's a 3.1 or a 3.2 without sources? That's the whole point of having sources. I can't read someone's mind to tell where they drew their information from, if it's a fuzzy thing they half-remember or if they carefully reviewed a highly respected peer-reviewed journal. If they cite that source, that gives several benefits-I know they got the information there, and if I can get hold of it (and likely I can, either personally or through the library), I can verify for myself that it really says what they claim it does. Of course, once I do that, I can also add to the article using that source! (By the way, what's this "draftspace" bit, and what do you mean by userfication? That sounds rather interesting.) Seraphimblade ^{Talk to me Please review me!} 05:40, 24 February 2007 (UTC)

Seraphim, I'm not the one who was talking about the 1-3 categories - that was User:N Shar (The response you gave seems to conflate the responses both of us made. I just want to clarify.) I agree with most of your points here, being that non-experts need to know what's correct and what's not correct as well, and 137 above hasn't really come up with an alternative mechanism, other than actual sourcing, for seperating the wheat from the chaff (because "experts" saying that articles are "correct and neutral" isn't it). Userfication is something that commonly gets brought up in DRV (and I think the old SDUA talk page as well) - the article is moved to userspace and the editor gets to work on sourcing the article. Draftspace was also one of the SDUA talk page ideas thrown around - a space that all articles would go and if articles satisfied a certain criteria set (sources, etc) they'd be moved to article space. This is probably the less likely proposal as it involves a huge amount of changes in the current encyclopedia structure. ColourBurst 06:18, 24 February 2007 (UTC)

I agree that we can't tell whether something's from category 3.1 or 3.2. That's why we don't suggest using Wikipedia as a reliable source, especially when sources are not provided. This is what we have {{unreferenced}} -- to warn readers that pages not citing sources may be unreliable. The point is that this criterion would affect both categories equally. On the subject of "draftspace," I've often thought that this would be a Good Thing for Wikipedia. Perhaps users should be able to view a "stable version" of the encyclopedia if they set it in their preferences or something. From the perspective of those users, the unsourced articles would not exist. Meanwhile, they are still available to editors and users who want to see everything. Your point about notability being secondary sources is well taken, though. --N Shar 03:10, 25 February 2007 (UTC)

If there are no sources and we are warning people not to rely on the article, then what is the point of having the article? -- Donald Albury 00:43, 26 February 2007 (UTC)

Because it's a starting point. --N Shar (talk • contribs) 02:04, 26 February 2007 (UTC)

Not all unref articles necessarily get tagged. And you know what? I use Wikipedia a lot for quite a few different purposes-I didn't stumble upon it by accident, I'd been using and reading it for years before I started editing! Certainly, it's not my final stop for research, but it's an excellent starting point and way to get an overview of a topic, and quite realistically, I find as few errors in well-sourced, mature articles as I find in any more "conventional" source. However, for that to remain workable, we've got to keep the signal-to-noise ratio high. Every article requires vandal patrolling, maintenance, etc., etc. There's a certain threshold below which that's just not really worth it. (That being said, the userfication proposal presented earlier is intriguing-I think that may be a viable alternative to outright deletion.) Seraphimblade ^{Talk to me Please review me!} 02:12, 26 February 2007 (UTC)

Userification could work. One concern I would have is the possibility that userfied articles might never be touched again. In that case userfication would be worse that outright deletion. Another concern would be cross-space links. However, I think both of those concerns are solvable. I think there would be a definite advantage to encouraging editors to create and develop articles in user space, and only moving them to main space when they were reasonably well sourced and formatted. I see one problem, though, which is that copyvios might go undetected longer in user space than in main space. All in all, though, it is worth discussing. -- Donald Albury 22:32, 26 February 2007 (UTC)

Userfication

I think this is well worth discussion. It could easily ease the concerns of both sides-the information is not lost, but isn't presented as an encyclopedia article either.

I see two ways we could work this:

• After 14 days tagged without a source, a user may move the unsourced article to the userspace of the principal contributor. The resulting redirect may be tagged for speedy deletion as a cross-namespace redirect. That contributor, and any other major contributors, will be informed of the article's new location on their talk pages, as well as what must be done (adding a source) before the article is moved back.

The advantages I see there are that the article is still out of mainspace, and if its editor(s) can source it, it can easily be moved back. This would remove it from the attention of others, but I believe we could solve that with a category such as Category:Userfied unsourced articles, which anyone who wanted to help with sourcing could go through, and the cat could easily be added by the template placed on these userfied articles. (This would also allow defunct articles which no one has been able to source and the author has lost interest in to be periodically prodded.) The old page being deleted would also allow a direct move back from userspace once the article is sourced, preserving all article history. The disadvantage is that it would cause redlinks in any article that's been linked to it, and would require an administrator to delete the redirect page (though not to userfy the page, as any established editor may perform a page move).

• Same as above, but a soft redirect explaining the situation is left in place of the article text, and the old mainspace location is not deleted entirely.

The advantages here are that this solution wouldn't require any additional administrative attention (users could perform the move and place the redirect template), and would leave blue links from incoming articles. However, this might be confusing to readers which follow those links, and wouldn't allow the moved article to be directly moved back, resulting in a GFDL issue from cut-and-paste moves back into mainspace (or eventually requiring admin attention anyway to delete the old page and perform it as a requested move). Also, these pages would have to be pruned pretty frequently so that mainspace wouldn't get crapped up with a lot of them.

Personally, I would prefer the first solution, but what does anyone else think? Seraphimblade ^{Talk to me Please review me!} 07:57, 27 February 2007 (UTC)

• That is not a bad idea at all. I wonder if this is something that a bot could be set up to do. I prefer the first solution as well. I'm not too concern about the redlink because it may actually inspire someone to write a better, sourced article to fulfill it. Agne^Cheese/^Wine 08:23, 27 February 2007 (UTC)

• You've addressed the concerns I mentioned. So, would it be a good idea to fork a new page on this, and if so, how soon? -- Donald Albury 20:54, 27 February 2007 (UTC)

• Well, this in its current form seems highly unlikely to achieve consensus, and the userfication seemed to get a much warmer reception. The first solution seems preferable to me, especially as it doesn't raise GFDL issues with copy-paste moves, so I'll start one based on that if no one else cares to. I'm going to tag this one rejected, it seems very unlikely to gain consensus. Seraphimblade ^{Talk to me Please review me!} 01:17, 28 February 2007 (UTC)

Fair. With a good summary, it may usefully serve as a starting point in a year or so. At least a year (sigh).DGG 06:33, 28 February 2007 (UTC)

Time to revive this?

Given some of the recent shift in policy and how it is carried out, especially around unsourced articles that are under the purview of WP:BLP, it may be time to consider reviving and revising this policy. A rejected policy doesn't necessarily mean that it's rejected forever. ++Lar: t/c 00:10, 11 June 2007 (UTC)

Yup, probably worthwhile. Also, in the case of articles that have been tagged as lacking sources for over a year, I would prefer if the prod tag could be reinstated unless the article is sourced. Addhoc 09:40, 11 June 2007 (UTC)

This page is still on my watchlist. If someone wants to revive it, I'd be glad to help work out the details (I'm in support at the moment, but as it's quite vague so far I'm open to being persuaded otherwise). --ais523 11:06, 11 June 2007 (UTC)

I suspect that a policy of this sort that is restricted to bios of living persons will receive stronger support. I'll play with a new version, see what I can come up with. -- Donald Albury 11:12, 11 June 2007 (UTC)

Was rather surprised to see a comment here. There seemed to be some support toward userfication rather than outright deletion, I'll see what I can do with that. Unsourced BLPs can already be speedied, generally speaking. Seraphimblade ^{Talk to me} 11:16, 11 June 2007 (UTC)

While unsourced BLPs are being speedied, being unsourced is only one factor, generally not the primary factor, being used to justify those deletions. Being unsourced has not been added to the explicit criteria for speedy deletions at Wikipedia:Criteria for speedy deletion. Due to the heightened risk of harming living persons due to unsourced or poorly sourced material about them in Wikipedia (per Wikipedia:Biographies of living persons), I think we need to make it explicit that unsourced and poorly sourced bios of living persons can be speedy deleted, allowing recreation when and if notability (per WP standards) has been established and proper reliable sources are cited. -- Donald Albury 11:57, 11 June 2007 (UTC)

Just so. ++Lar: t/c 19:30, 12 June 2007 (UTC)

Seraphimblade, I do not think that userification is the right approach for biographies, given how understanding of the issues behind BLP has improved. It may be for non biographical things (heck, I have some partly done things in my userspace that are poorly sourced, at best) but not for biographies due to the potential harm factor. ++Lar: t/c 19:32, 12 June 2007 (UTC)

Good point. Perhaps a split approach (place the tag, after X days, it's deleted if it's a BLP and userfied if it's not?) It also could be narrowed to apply only to BLPs, but I'm not sure that would make a tremendous difference. After 14 days, Google Cache and mirrors will have picked it up anyway. Perhaps a better approach would be to widen G10 to include any unsourced BLP or one which solely cites unreliable sources, and have this focus on non-BLP issues? Seraphimblade ^{Talk to me} 20:01, 12 June 2007 (UTC)

Yes. 14 days is a lot of time, unfortunately, when trying to keep things out of cache and out of mirrors. Widen G10, or a new G13... either would work for me. ++Lar: t/c 20:33, 12 June 2007 (UTC)