Wikipedia talk:Requests for verification

Consensus Points
This proposal has not reached consensus, but is continuing to move forward in addressing a long standing and ongoing debate in en.wikipedia. Please enter points that you beleive we have consensus on so far in this discussion. Jeepday (talk) 02:49, 25 July 2007 (UTC)
 * While WP:V does put the burden to provide reference on the contributing editor, WP:AGF requires that we assume that reliable sources do exist and before an editor nominates a complete article for deletion (regardless of process; CSD, PROD, etc) based on lack of reference the nominating editor should attempt to reference the article themselves. This is in keeping with behavioral guidelines and is in the best interest of the Wikipedia community.  Jeepday (talk) 02:49, 25 July 2007 (UTC)
 * What precisely do you mean by "burden"? &mdash; Carl (CBM · talk) 03:01, 25 July 2007 (UTC)
 * Verifiability Jeepday (talk) 03:57, 25 July 2007 (UTC)
 * Contributing editors (new or old) are not expected be fluent in using Citation templates. When a reference is provided there is no requirement that it be properly formated. Jeepday (talk) 16:48, 25 July 2007 (UTC)
 * Automated or bot tag tagging or challenging of contribution's is NOT appropriate. Jeepday (talk) 16:52, 25 July 2007 (UTC)

Opening discussion for RFV
Opening discussion for 'Requests for verification (RFV} can be found at Wikipedia talk:Verifiability Signed Jeepday (talk) 02:46, 12 July 2007 (UTC)

Simply no
This sort of thing has been proposed many times and never gets consensus. If you want to see more sources in articles, just add them - more process is not the way to resolve these backlogs. Isn't there a perennial proposals page that this can be added to? &mdash; Carl (CBM · talk) 03:00, 12 July 2007 (UTC)
 * I agree with Carl, this will never fly. And I don't think it should.   The unit at which verification-challenges and deletions should be made is individual sentences within an article, not the article itself.  But we don't need this policy to do that, we're already doing this: it's called editing.  semper fictilis 03:36, 12 July 2007 (UTC)
 * I should have included a link to Speedy deletion criterion for unsourced articles, the last incarnation of this proposal. Its talk page and archive are over 512KB together. I encourage everyone not to recreate that half megabyte of discussion here. There are more useful things we can do for the encyclopedia. At the very least, read through that discussion before commenting here. &mdash; Carl (CBM · talk) 03:40, 12 July 2007 (UTC)
 * That's not the last incarnation, the last was Proposed deletion process for unsourced articles Night Gyr (talk/Oy) 18:33, 18 July 2007 (UTC)
 * If such a process were implemented I would never add another fact tag again, and would become quite bold in removing them wherever there was the slightest reason to (I've already stopped using the unreferenced tag after I saw a couple of AfDs using it as justification, though IMO that one's a fairly useless tag in any event). This would not be optimal for improving Wikipedia to say the least. An incomplete article is a work in progress, not a blemish to be expunged. Bryan Derksen 15:25, 12 July 2007 (UTC)

Rebuttal to "Simply no"

 * Past failure to uphold policy is not a valid argument to continue not following policy. No original research is clear that all content must be referenced  - Wikipedia is not the place for original research. Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not presenting original research is to cite reliable sources that provide information directly related to the topic of the article, and to adhere to what those sources say. - Though giving thought to past argument is valid request thank you for providing the reference CBM. Signed Jeepday (talk) 13:12, 12 July 2007 (UTC)


 * Suggesting that editors who challenge content are responsible for adding reference is in direct conflict with Verifiability changing WP:V to place the burden on the challenging editor was recently discussed and rejected at Wikipedia talk:Verifiability Signed Jeepday (talk) 13:12, 12 July 2007 (UTC)


 * Yes, this sounds like a good idea. Contrary to what Semperf says, there are many articles that are entirely unsourced and likely to stay that way even in the face of research. This information needs to be cited and removed, and this is a fine way to give everyone fair notice so they don't cry foul when the unsourced content gets removed. Until(1 == 2) 13:06, 12 July 2007 (UTC)

alternative
"This page has been listed on the requests for verification list. (Add entry to list.) It has been suggested that this entry might not meet Wiktionary's criteria for inclusion without additional documentation. If evidence is not provided, the article may possibly be proposed for deletion." We can say : it will be deleted--Afd decides on deletion, and they do not follow this criterion--they follow verifiable. capable of being verified, and this is one of the principles there. How do we propose to remove articles without going there? How do we propose to convince the people who form the consensus to follow this policy? We cant. If this is adopted, it will be ignored. If it is by chance not ignored, it will shift the fighting to the degree of verification. We should not be trying to find reasons to delete, or a way to delete. We should be trying to find a way to persuade people to try to source material. All material, of any age. The whole point of this project is to improve WP, by upgrading the articles, and only as a very last resort, removing those which cannot be shown to be important by adequate sources. (whatever degree of importance may be the criterion--which is a whole additional story)DGG (talk) 05:13, 12 July 2007 (UTC)

Five premises and a conclusion
When I proposed Requests for verification I was working under the assumptions below that deleting an unreferenced article is supported by Wikipedia policy as indicated in this History entry and we only needed to build the template. Jossi responded that "it sets policy, by stating a 30 days period" here.

New article warning
When an editor opens Editing NEW ARTICLE to begin a new article intermediately above the edit window is the bold statement Articles that do not cite reliable published sources are likely to be deleted.



Challenged
Per WP:V "Any edit lacking a source may be removed", some would argue that content must be challenged before it is removed. I think placing placing RFV with the statement "If evidence is not provided within a month, the disputed information will be removed." would qualify as a challenge.

Burden
Verifiability "The burden of evidence lies with the editor who adds or restores material". Clearly the burden of providing reference lies with the editor who added the article.

Reference requirement
No original research ''Wikipedia is not the place for original research. Citing sources and avoiding original research are inextricably linked: the only way to demonstrate that you are not presenting original research is to cite reliable sources that provide information directly related to the topic of the article, and to adhere to what those sources say.''. Clearly to meet the core content policy NOR you must cite references.

Prohibiting deletion
Last there is no policy or guideline prohibiting the deletion of unreferenced articles. I have looked and I have asked, if there is a policy prohibiting the deletion of unreferenced article someone please point it out.

Conclusion
We have multiple policies and expectations in Wikipedia to remove unreferenced questionable material and articles. I am a believer in There is no deadline, but no deadline does not mean we ignore encyclopedic expectations indefinitely. The is no question if an editor sees unreferenced content they are encouraged to find reference and improve the article, But they are not required to search for references. Any editor may challenge any unreferenced content, there is no criteria that an editor can only challenge specific unreferenced content. The only policy question to decide here what is a fair length of time to allow the contributing editor (or any editor) to add references once an article has been challenged? Jeepday (talk) 03:38, 13 July 2007 (UTC)

How long before delete unreferenced article?
It has been suggested that a speedy delete is to fast and has already been rejected. Others have argued the even the 5 days of prod is not enough time to do a decent job of adding references. It has been proposed that the 30 days used in Wiktionary:Wiktionary:Requests for verification is appropriate. What do others think? Jeepday (talk) 03:38, 13 July 2007 (UTC)


 * this is just a preliminary reply to Jeepday:

I continue to disagree with setting a time. The proposal for 10 days was rightly rejected, and so will be one for 30. Wikitionary has much easier problems--they merely have to find a quote illustrating usage. We have to document the theories, facts, accomplishments, importance. Sometimes I can do this in 5 minutes, in the frequent case that the person placing the tag was sloppy enough to not check google before tagging. Sometimes it can take 1 or more days in a physical library. Sometimes it can take days in an excellent library by one of the WPedans who can deal with the language and who understands the subject, and we may have very few in some cases who will meet all the requirements, be interested and willing and capable. I can do perhaps one or two a month of that difficulty, but could not possibly do more, no matter how many articles might need it. Suggested replacement text: ''It has been suggested that this entry might not meet Wiktionary's criteria for inclusion. If evidence for notability is not provided, the article may be considered for possible removal. " Gentle. no ultimatums. But the necessary stimulus to activity is still there. The next step, of course, if the evidence is not provided, is eventual AfD. But the Afd is for failure to verify notability, not for having unverified statements.
 * and, regardless of the responsibility being primarily with the author, the responsibility is also with all of us. We're supposed to be here to improve the encyclopedia. this often does mean deletion, but it often means rescue. It is in my view uncooperative to place an AfD tag on an article for lack of verified notability unless one is either confident that one knows enough to say accurately that it is unverifiable, or has actually checked. Carelessness makes extra work for everyone, no matter where it occurs. This is not a game, this is not a contest.   DGG (talk) 04:55, 13 July 2007 (UTC)


 * DGG as is often discussed at Template:Unreferenced There are 2 categories of unreferenced articles; Completely unreferenced (i.e. tagged with Unreferenced and Under referenced (i.e. tagged with refimprove. Nobody is suggesting that articles be deleted because they are under referenced (though moving or removing unreferenced sections of text may be appropriate), template Requests for verification (RFV) should only be on articles without any references.  If an article is partially referenced with reliable sources, then it would not be appropriate to delete it as "unreferenced" (not addressing notability).  In your hypothetical article above - Someone creates an unreferenced article.  Later someone else tags it with RFV and Unreferenced, If you come along and add a single reference then you should remove RFV and Unreferenced and if you feel it is appropriate tag it with refimprove or Underconstruction.  DGG we have interacted often, and I have considered what I have learned from you in drafting this proposal.  I would very much like your support on this proposal.  I just added the usage instructions History to address your concern, which I should have had on there from the start. I think 30 days is more then ample time for articles to meet at least the barest minimum of verifiability, by including at least one reliable published (online or offline) reference.  What other concerns do you have that I have over looked? Jeepday (talk) 14:10, 14 July 2007 (UTC)


 * Considering that unreference articles, especially stubs, are often created by novice editors, 30 days is not necessarily "more than ample time". Frankly, the citation codes aren't all that easy to learn – and the current pages of instruction aren't at all new-user-friendly – and when you're still busy learning a lot of different aspects of Wikipedia (in your spare time), 30 days isn't very much time. If there has to be a deadline, then I wouldn't mind seeing 6 months or even a year. While I certainly see and support the need for encouraging better citation, I really don't see the urgency for deleting such articles. If it comes to pass, I think we may start seeing articles disappear at the same high speed non-free images are currently be tagged and deleted – and seven days doesn't seem to be long enough for most tagged articles to get fixed. Askari Mark (Talk) 02:38, 21 July 2007 (UTC)

Downsides of the proposal
I think the costs of this proposal outweigh the benefits for the following reasons:

Deadlines are evil. Any institution with a formal structure (like Wikipedia) requires deadlines to function. That's why we have deadlines for prods and XfDs. However, these mostly address articles that shouldn't exist at all. We've so far been able to avoid deadlines on articles that should exist but are simply of poor quality. Wikipedia is an evolving project and deadlines on content changes should be avoided when possible.

Bureaucracy is evil. This proposal will create an additional level of deletion bureaucracy based on a deletion criterion that is regularly rejected at AfD (legitimate topic, but poorly written).

Tagging is evil. Given the relatively low threshold for satisfying a request for verification (adding a single source), what is the purpose of the tagging? Instead of adding a tag, why not just add a source? Although the statement that editors "are not required to search for references" is technically true, I do not believe it is one we should embrace or encourage on a mass scale. Editors are under no obligation to cleanup a vanity or spam article, of course, but we should make the effort when the deciding factor is merely the addition of a single reference. We are a collaborative project and a "not my business" attitude undermines the spirit of collaboration and the idea that we all edit for a common purpose.

Until these concerns are somehow addressed or rebutted, I do not think I can support this proposal, despite its goal.

However, if it is adopted, I suggest two modifications:
 * 1) The process should not apply to articles with external links. In many cases, some or all of the content in articles lacking in-line citations and a "References" section is supported by one or more external links. These articles should be handled via the regular improvement and deletion processes.
 * 2) The image of the paper being thrown into the wastebasket absolutely cannot stay. It's the equivalent of saying "this is garbage" when the only issue is lack of sourcing.

Black Falcon (Talk) 19:31, 14 July 2007 (UTC)


 * I agree, at least in part with most of your points here, but disagree with your opening conclusion.
 * Image, Arguable true and as you have raised the point others will likely have similar thoughts. I would support removing it or replacing it maybe [[Image:Diamond-caution.svg|30px]] or [[Image:Purple question mark.png|30px]]?
 * It's is agreed that links to external sites are often valid references. Maybe we should add some text to the  Suggested usage and a link to No original research?
 * Evil is bad, but I hope this policy/tag will decrease the evil.
 * For one it extends the time period from a few days for AFD and PROD to a month for obtaining references. Many contributing editors do not visit Wikipedia daily.
 * For another it does not address the quality or the appropriateness of the article. Both of which are reflected in part by references.  RFV provides a first step towards improving an article.  It is relatively easy to pass RFV so passing RFV will not ensure that an article will survive a AFD or a PROD, it will prepare the article to better fend off the both and will be more likely to survive if nominated.
 * Signed Jeepday (talk)


 * Thanks for your response. After more consideration and several dozen clicks of the "random article" button, I have changed my mind somewhat and would like to see if there's a way to make the proposed process workable (as written now, I don't think it is). To expand on the points I made:
 * This process does not replace any of the three existing article deletion processes and so will create a fourth process. Though extra bureaucracy is undesirable, so is lack of sourcing. If this process succeeds in improving article sourcing, I think the excess bureaucracy generated by it may be worthwhile. However, I believe RfV (if adopted) should be like prod where anyone can contest deletion for any reason.
 * There's nothing inherently wrong with tagging articles as it draws attention to the articles' problems. That said, I think tagging is unproductive in cases where fixing the problem will not take significantly longer than tagging the article (e.g., adding a wikify tag to a one-sentence stub). Given the low threshold for satisfying a "request for verification", I think this continues to be an issue, but perhaps one worth overlooking iff the process can be made workable.
 * My main problem with this proposed process is that it places deadlines on content. Though the philosophical issues involved (immediatism versus eventualism) require consideration, the practical side is more concerning for me. There are currently nearly 80000 articles tagged for having no references. If a group of editors adds RfV tags to even just 10% of pages in Category:All articles lacking sources, then the process will be overwhelmed. If RfV is to result in deletion after a 30-day period, there must be a way of limiting the number of articles that are tagged each day.
 * I would appreciate any thought you may have on the matter. Black Falcon (Talk) 18:24, 15 July 2007 (UTC)

Thank you Black Falcon, those are all extremely good points. I have been waiting for someone to bring up the last one in particular. (saving the best for last). Jeepday (talk) 13:39, 16 July 2007 (UTC)
 * Anyone can contest deletion, Add a reference and remove the tag. I think we can address your concern by adding direction similar to prod.  We could use the prod version with very little change.  Would you like to make a change directly to the draft template or discuss it here first?
 * Tagging - You are correct, you can not dictate usage you can only suggest. Editors will either use it appropriately or not, I lean to Assume good faith for this point.
 * Forcing a time line. There is no deadline has been on my user page since I discovered it. I have added some buffer in already and plan for more
 * First a goodly portion of those 80,000 are old, see Unreferenced articles for progress on cleaning them up.
 * We have placed an age limit on the use of the RfV RFV should not be used on articles with history before January 1, 2007 see third bullet Suggested usage.
 * We can not control how many articles are tagged each day, but we can control how many are deleted. Only an Admin can delete the article, and we need to provide clear direction for that. Example if the article has even a single reference the RfV should be removed. see first bullet Suggested usage
 * If the "history before January 1, 2007" or "30-day period" period need to be adjusted we can do that here. Wikipedia is it is always changing.

Deletion without fixes considered harmful
From WP:NOR:
 * Any material that is challenged or likely to be challenged must be accompanied by a reliable source.

Clearly, the policy does not require EVERYTHING to be cited top to bottom, only controversial material. We have to have accurate general sources, but not for every single trivial thing in an article. The extremist view that this section is not significant is not useful. Time after time we have seen people run around tagging wildly, including some clearly not controverisal information which really does not require any sort of citation or reference.

It is not acceptable to just go around removing stuff. If it's probably true but not referenced, your job is to fix it, not delete it. Georgewilliamherbert 23:47, 15 July 2007 (UTC)


 * You may be interested in these two essays: You don't need to cite that the sky is blue and You do need to cite that the sky is blue. Cheers, Black Falcon (Talk) 00:04, 16 July 2007 (UTC)
 * Now I'm not sure if I should thank you for giving me more reading material, or complain that I've now clicked away from the topic at hand and may never remember to get back to this page to finish reading and reach an opinion one way or the other. I guess I'll go with a quick "thanks for the new info", and start reading. Spazure 05:27, 20 July 2007 (UTC)
 * It is not the intent of this policy to dictate what can or can not be challenged, only to provide a tool for doing so. As I presented in Five premises and a conclusion there is only one policy consideration here, how long before challenged and unreferenced articles get deleted, other policy and guidelines are preexisting.  Jeepday (talk) 12:57, 16 July 2007 (UTC)


 * As someone put it recently in the mailing list: If something is garbage, it should be prod'ed for being garbage not because it isn't sourced. Sourced garbage is still garbage. Put up or shut up is bureaucracy which is a violation of WP:NOT. Articles should never EVER be tagged for deletion for the sake of a process. If an article has trivial flaws (such as lack of sources), correcting is much easier, faster, and productive than involving any kind of process. Which summarizes it pretty well. ≈ jossi ≈ (talk) 12:30, 18 July 2007 (UTC)


 * The argument in the mailing above is so fragmented it is impossible to address "it should be prod'ed" followed by "put up or shut up is bureaucracy which is a violation of WP:NOT"'' completely counter arguments in immediately following sentences. That being said here is a rebuttal.
 * Lack of sources is not a trivial flaw, it is counter to 2 of 3 core content policies WP:V and WP:OR. There is already an internet that is full of wild unvalidated statements.  Per WP:NOT Wikipedia is an online encyclopedia and, as a means to that end,  an online community of people interested in building a high-quality encyclopedia in a spirit of mutual respect.  As said in WP:NOT Wikipedia is not a place to publish your own thoughts and analyses or to publish new information not heretofore published. and in WP:NOR That is, any facts, opinions, interpretations, definitions, and arguments published by Wikipedia must already have been published by a reliable publication...The only way to show that your work is not original research is to produce a reliable published source.   Jeepday (talk) 01:49, 19 July 2007 (UTC)

Updating to address concerns
I have updated the main page to address many of the concerns above. Please take a look and provide feed back. Jeepday (talk) 13:56, 17 July 2007 (UTC)
 * I actually like this proposal (it's being discussed on the mailing list). Seems to me like a good way of enforcing citation, which will ultimately improve the wiki (if it forces people to cite).  Giggy  UCP 04:22, 18 July 2007 (UTC)

This proposal is a huge mistake
My opinion: this proposal is a huge mistake. Rather than tag and delete articles without sources, do some research and add sources. We are here to build an encyclopedia and we welcome contributions of people that do not necessarily read or know our policies. We should welcome their contributions, and source material that we find unsourced. ≈ jossi ≈ (talk) 04:21, 18 July 2007 (UTC)


 * See Article_Rescue_Squadron. A project worth joining, IMO. ≈ jossi ≈ (talk) 04:37, 18 July 2007 (UTC)


 * Jossi, in your knee-jerk reaction to any mention of deleting articles you seem to have completely missed the point. ARS and this project are not opposites, they are partners. If this is concentrated effort to bring attention and force the verification of articles, then it's doing a favor for those who don't want to see articles deleted for a lack of sources by making a handy list of articles for ARS to work on. VanTucky  (talk) 04:50, 18 July 2007 (UTC)


 * If an article is not verifiable, meaning that when you encounter such an article and cannot find even one source for it, you place prod on it. The proposed process does not entice people to look for sources, but places the burden on others, mnost likely the ocassional editor that will never see the warning. Rather than tagging with RVF, do 5 minutes of research and find a source. ≈ jossi ≈ (talk) 05:24, 18 July 2007 (UTC)


 * The difference between RfV and prod are significant and those differences work to improve the likelihood that good articles will not be lost. For one RfV only addressed lack of references so projects like ARS can focus on articles that have more potentially to be saved.  Second it provides 30 days instead of 5 to address the lack of reference.  In my work at Requests for verification I have found that you can't always find a reference in 5 minutes, if the article is clearly not notable then by all means place a prod-nn, but if there is any chance of the article being a valuable article in Wikipedia put a RfV on the article and a uw-unsor1 on the authors page. Jeepday (talk) 13:01, 18 July 2007 (UTC)

I've gone through Category:Articles lacking sources and it's embarrassing how articles just sit there literally for years without anyone doing anything. Maybe this isn't the best solution, but the current system of asking nicely and hoping people aren't hoaxing us doesn't seem to be yielding results. --W.marsh 14:21, 18 July 2007 (UTC)
 * Oh yeah, If 10 people opposed to this process will promise to reference 10 unreferenced articles a week for the next year, or delete them if there aren't references to be found, I'll oppose this proposal :-) --W.marsh 14:23, 18 July 2007 (UTC)

tag template
This seems like a duplicate to or a less-effective version of AFD, as it provides no space for extensive discussion. And is going to be applied to any unreferenced article? VanTucky (talk) 04:46, 18 July 2007 (UTC)
 * See response above at Jeepday (talk) 13:04, 18 July 2007 (UTC)
 * Not really... people can remove PROD for any reason, they don't even have to give a reason (some people in fact revel in not giving a reason). This process requires an actual reference to be added before the template can be removed. That is a huge difference. --W.marsh 14:18, 18 July 2007 (UTC)

"Unsourced" does not mean "deleteme". It just means, "unsourced".
When we decorate an article (perhaps with one of the handy templates) to warn that it "does not adequately cite its sources", the implication is not that "you, the feckless editor who wrote it, had better supply some references pronto, 'cause the rest of us are just itching to delete it."

Rather, the much more important intent of those templates is to warn the reader, and the message is simply, "so sorry, we're less sure of this fact than we'd like to be, so please take it with as many additional grains of salt as you deem necessary."

If an article (or a sentence or section within an article) is not merely unsourced but also POV, nonnotable, slanderous, or wrong, we've already got plenty of mechanisms (including, of course, the lowly Edit button) for correcting the problem. But we also have scads of facts which, though unsourced, are reasonably accurate and eminently useful and absolutely not in need of speedy deletion.

We absolutely do not want to give deletionists, policy wonks, and POV-pushers another handy tool which will make it that much easier for them to finesse the deletion of information they don't like. —Steve Summit (talk) 05:00, 18 July 2007 (UTC)


 * As mentioned in above the deletion of unreferenced content is addressed in other policy pages.  RFV only addresses the time line for deleting completely unreferenced articles.  Extending the 5 days of prod to 30 days for RfV and providing 6 times longer to save the article. Jeepday (talk) 13:25, 18 July 2007 (UTC)


 * Under what circumstances would you want to give an unsourced article more than 5 days, but less than forever to "save"? --Steve Summit (talk) 19:49, 18 July 2007 (UTC)


 * This is the first I've heard of the policy but IMHO it's a good thing. And basically you have answered your own question. Editors have 5 days in which to contest a proposed deletion. If they do, then they have 30 days to find at least one reference for the article. Failing that, then yes, the article should be deleted. If an editor is fighting to save an article, they should at least be willing and able to find one reference for it. If they're not, then wikipedia is almost definitely a better place without the article and really I don't see why their views that the article should stay count for much Nil Einne 15:22, 19 July 2007 (UTC)

Steve Summit has saved me a lot of effort; I agree completely. Don't delete ureferenced articles, just flag 'em. Chris the speller 15:30, 18 July 2007 (UTC)
 * How about actually referencing them? Google Scholar, Google Books, News archive, JSTOR if you have it... it only takes a few minutes for most articles. --W.marsh 15:43, 18 July 2007 (UTC)


 * Hey Chris, what do you think happens after you flag 'em? Good volunteers at Unreferenced articles (and similar) help us reference them, but the unreferenced content is adding up faster then the references are getting added. Jeepday (talk) 02:04, 19 July 2007 (UTC)


 * Steve the honest answer to your question is that a good number of unreferenced articles maybe 5% or 10% are not easy to reference but are good candidates for encyclopedic content. Only the original editor or someone with access to similar sources can provide the references that lead to a quality encyclopedic article.  By providing the extra time you increase the likelihood that the article will be referenced, if it can't be referenced in 30 days, that it would seem to pretty clearly fail WP:NOR and be deletable by that policy.  Does that answer your question? Jeepday (talk) 02:04, 19 July 2007 (UTC)


 * Thanks for the answer. I was with you for the first two sentences.
 * Where you lost me was with the "providing the extra time" part. Are you saying that, today, one of these "good [but unreferenced] candidates for encyclopedic content" is a candidate for  ?  I'm no expert on the fine nuances of every Wikipedia policy, but I didn't think "insufficient references" was one of the suitable determiners for either a   or a speedy deletion.
 * If an article does, in fact, "clearly fail WP:NOR and be deletable by that policy", then we can obviously just delete it by that policy; we don't need a new one. —Steve Summit (talk) 03:16, 19 July 2007 (UTC)
 * When I search for references and don't find any, it is potentially a prod-nn, as I am mostly searching online, the availability of references is limited, many quality references are only available in libraries of far away places.  Using WP:AGF I assume that the authoring editor used a reference that is not online (this is a good thing!) but neglected to list it.  If the reference can't be found then WP:NOR says "The only way to show that your work is not original research is to produce a reliable published source that advances the same claims or makes the same argument as you." It might be a 3 day hike to the library, I want to give the authoring editor every opportunity to meet Policies and guidelines.  This is where the timeline comes in, how long is long enough to hike to the library?  Jeepday (talk) 03:42, 19 July 2007 (UTC)

Mailing list thread on this proposal
WikiEN-l: Proposed "prod" for articles with no sources ≈ jossi ≈ (talk) 05:07, 18 July 2007 (UTC)


 * The conversation at the mailing list thread sort by thread is slowing down. Jeepday (talk) 12:49, 20 July 2007 (UTC)

Variation
I placed the variation of the thread on WP:V and WP:OR as well as all the active project pages I am aware of.

Please see: Requests for verification

A proposal designed as a process similar to prod to delete articles without sources if no sources are provided in 30 days.

It reads:

Some editors see this as necessary to improve Wikipedia as a whole and assert that this idea is supported by policy, and others see this as a negative thing for the project with the potential of loss of articles that could be easily sourced.

I would encourage your comments in that page's talk or Mailing list thread on this proposal WikiEN-l: Proposed "prod" for articles with no sources

Signed Jeepday (talk) 14:15, 18 July 2007 (UTC)

Another suggestion
I propose that the editor placing an RfV tag on an article should have to do a Google search to find obvious sources first. In my experience, 99% of the time something usable turns up from just cutting and pasting the headwords into Google. If someone else later finds an appropriate reference from a simple Google search strategy, the first editor should be liable to a punitive block.

Also certain types of article should be exempt as self referencing, including: Espresso Addict 05:47, 18 July 2007 (UTC)
 * Books/journals with an ISBN/ISSN
 * Physical locations with a grid reference or lat/longitude


 * I think that everyone agrees that all editors should make attempts to improve articles. But as discussed in  above the requirement and burden of sourcing is addressed in other policy pages.  RFV only address the addresses the time line for deleting completely unreferenced articles.  There are multiple projects working to verify and reference articles some are discussed on this talk page. Jeepday (talk) 13:28, 18 July 2007 (UTC)

Time limit and automated deletion are not appropriate.
See WP:TIND for reasons why time limits should be avoided.

Deletion without community oversight is potentially problematic. WP:PROD was a very contentious proposal, and only succeeded because it solved a serious problem: the volume of uncontested deletions was becoming too high to manage. Deletion of unsourced articles is a more contentious subject than uncontested deletions (obviously). Unsourced articles are sometimes kept when they are discussed at AFD (although it is less likely for them to survive now than it was six months ago, I note). This oversight is necessary. JulesH 14:16, 18 July 2007 (UTC)
 * The solution to unverified articles isn't to defend their right to sit around in limbo waiting to be our next press embarrassment... it's to actually go through the year+ backlog we have and reference these articles. It's nothing that 15-20 experienced editors spending a few hours a week couldn't do. Yet no one really seems to be interested in working on that backlog. If nothing else this proposal is an amusing jolt to the status quo. --W.marsh 14:29, 18 July 2007 (UTC)
 * If we don't go through the backlog within 30 days, what's the point of deleting the ones we didn't get to yet? Bryan Derksen 16:17, 18 July 2007 (UTC)
 * I don't think you understood my point. We need to work on the problem and oppose proposals like this, not just oppose proposals like this and let the backlog grow, assuming someone else will fix it. --W.marsh 16:27, 18 July 2007 (UTC)
 * Of course working on adding references is a good thing, I'm all for it. But I'm not obligated to provide a "better" solution in order to oppose a bad one. IMO we're doing okay with our current processes; having articles "wait around in limbo" until they get references added is not a terrible thing. Wikipedia is a work in progress. Bryan Derksen 01:22, 19 July 2007 (UTC)


 * I am working on Unreferenced articles and making this proposal. Jeepday (talk) 02:10, 19 July 2007 (UTC)

Let's write an encyclopedia
Let's write an encyclopedia. People who are here to delete good articles on any pretext they can find should be shown the door. People who think tagging articles they haven't read (as in "there is no ref section so it has no refs") is a good idea should told to stop. Forcing people to add cites with a tag will merely result in thirteen year olds going around tagging and refs being added that don't actually support the claims made. People have got to get back to the idea that this is an encyclopedia writing project. Thoughtful contributions welcome. Mindlessness robotic behavior not welcome. WAS 4.250 14:29, 18 July 2007 (UTC)
 * I think you're probably right. But unreferenced, unverified articles are still something we should be working on. Myself included. That there's a year+ backlog is an embarrassment. --W.marsh 14:32, 18 July 2007 (UTC)
 * "Embarrassment" is a bad reason for deletion. It doesn't result in more referenced articles, it just hides the fact that there were a bunch of articles we didn't have references for (and throws away what work has been done on them to that point). Bryan Derksen 16:20, 18 July 2007 (UTC)
 * I'm not saying we should delete them, I'm saying we should acknowledge the problem and work on fixing them. --W.marsh 16:24, 18 July 2007 (UTC)
 * I hope I'm not "defending the right to sit around in limbo" when I say this, but: we really shouldn't get too worked up about the possibility of "embarrassment". Wikipedia is something the world has never seen before, and somehow, despite all its obvious flaws and all the obvious reasons why it can't possibly work, it's nevertheless working very, very well.
 * Anything as big and visible as Wikipedia will always have its detractors; anything as big and complicated as Wikipedia will always make the occasional mistake. When we make mistakes we'll fix them, and when someone reports an "embarrassing" gaffe we'll acknowledge or ignore it (as appropriate) and move on.  But we can't possibly try to eliminate the possibility of every conceivable mistake or embarrassment before it happens -- that would be completely contrary to the incremental, perfection-by-successive-approximation approach which is one of the several foundations of Wikipedia's enormous success. --Steve Summit (talk) 17:08, 18 July 2007 (UTC)
 * I'm not saying we should aim for perfection, it's just that it seems like only a microscopic fraction of editors ever go through backlogs like the "articles needing sources" one and try to fix problems, rather people either want to axe them all or ignore them as if actually working on it is not an option. Working on it should be the main option.- -W.marsh 17:12, 18 July 2007 (UTC)
 * Wholehearted agreement. I've been working on the WikiProject Biography assessment drive of late, and found it's almost always trivially easy to find at least one or two sources for the unreferenced articles I've encountered. Espresso Addict 21:57, 18 July 2007 (UTC)


 * IMO, Wikipedia worries too much about embarrassment and not enough about quality. It's not an embarrassment to be unfinished if what you've put up is labelled as a work in process. Matthew Brown (Morven) (T:C) 18:33, 18 July 2007 (UTC)
 * Standards are still important... if we write off every problem as "Well we're a work in progress" that eventually just seems like we're content with being unreliable. I want to reduce the number of unverified articles (not through deletion, unless they're truly unverifiable), it's hard to believe that could be spun as a bad thing. --W.marsh 18:44, 18 July 2007 (UTC)
 * I don't write off every problem with "we're a work in progress." I do, however, write off this particular "problem" that way. And especially with respect to this particular solution. This isn't just some generic proposal that it really would be super to have more references in articles somehow. It's specifically a deletion mechanism. If that line about deletion in 30 days is removed from it then there's no significant difference between this proposal and the current mechanisms already in place for marking articles in need of additional referencing. Bryan Derksen 01:32, 19 July 2007 (UTC)

Tagging is sometimes appropriate
I've seen a number of people here state that editors should always find references instead of tagging with,  or. (Emphasis mine). I invite any of you who feel that way to work on the Dead End Pages project for a while. Every couple of months we get several thousand articles to slog through, trying to evaluate which should be proposed for deletion and which just need a little improvement. Yes, it's ideal to completely wikify each article, properly categorized it, and find quality references for each article, but the reality is that we'd only get through about 1% of the articles that way. (Even fewer, because we'd lose about half the people currently working on the project.)--Fabrictramp 16:59, 18 July 2007 (UTC)
 * My proposal above was just a teensy bit tongue in cheek, and wasn't meant to attack those brave souls who attempt to improve the really poor articles. I do, however, think the current proposal to delete tagged articles is ill founded, and would lead to the destruction of valuable articles that could easily be sourced if the right people could be induced to look at them. Espresso Addict 22:07, 18 July 2007 (UTC)
 * Sorry that my reply was a teensy bit frustrated. *grin* You'll notice that I didn't comment on the proposal itself, because I'm not quite sure yet how I feel about it. I certainly don't want potentially good articles deleted. But I do find a lot of articles that I feel conflicted about -- I can't find good sources in the first several pages of ghits, but a 5 day prod seems unnecessarily harsh.--Fabrictramp 00:03, 19 July 2007 (UTC)
 * A note of context, when judging this proposal consider that is coming from an editor with thousands of edits, most of which are adding references or other clean up work. The same editor who was a founding team member of Unreferenced articles and contributer to many previous clean up projects.  You both have good points, I invite you to support this proposal and join Unreferenced articles or a similar project. Jeepday (talk) 02:22, 19 July 2007 (UTC)
 * I'll consider joining when the current assessment drive on WikiProject Biography is over. I just don't have any spare time right now. Espresso Addict 02:49, 19 July 2007 (UTC)
 * Great! I look forward to seeing you there. Feel free to stop in anytime and just do one or stay for a while and try to complete a letter of the alphabet. Jeepday (talk) 02:57, 19 July 2007 (UTC)

Not sure what I think about this proposal
Normally I would be strongly against adding a new deletion process. However there is one problem where I feel is that the existing processes completely fail us on. This problem is an article being made which is completely incorrect, maybe a hoax, maybe a rumor, who really knows. This article has no sources and it sits around long enough for other places to pickup on it and copy this information. After a year or so when someone tries to source it, the WP article predates all the google hits. Are they really possible sources, or is this self-propagating circle of bad information? Who can tell. This problem is rare, but I think it serious enough consider making a new effort to see that articles have at least one source before they get very old. I don't know if this proposal is the answer to that, but I think it is the most problematic sourcing issue out there. I have two separate suggestions which might answer some of the objections raised against this proposal. I am not sure either of those suggestions would make this proposal workable, but I do think something needs to be done to at least ensure new articles coming in are quickly meeting some basic standards. We can eventually get through the existing sourcing backlogs, as long as we can stop more newly created articles from joining the pile every day.-- Birgitte SB  19:39, 18 July 2007 (UTC)
 * 1) One way I would like this proposal better if it focused on new articles, say created in the past week. Not only will such a restriction focus this on a manageable group of articles, but it will have a much better chance of catching the attention of the original contributor.  It would also make this a poor tool for people who just want to delete articles on certain subjects.  This would  definitely prevent more of problems I outlined above from developing.
 * 2) Another thing change that would make this proposal more acceptable to me is if rather than deleting articles after 30 days they would be put in line for on AFD. This would still allow this tag to create a useful list of article needing sources as VanTucky suggests (I can assure you unreferenced does NOT already provide this).  I would imagine it is possible to have a bot add these articles to AfD after the 30 days expire making it as painless as possible.  If someone had to tell the bot how many to add each day; it wold prevent AfD from ever getting overloaded because alot were tagged at once or there it already being a busy day at AfD.  Then every article would get (a) a fire lit under the original contributer and anyone watchlisting it to source it, (b) put in convenient list for sourcing projects, and (c) the individual evaluation it deserves at AfD.  At the same time we should make a template to leave on talkpages of people who tag articles that can be sourced in five minutes with google, a tailored version of sofixit.  This would not be the right tag for easily sourced articles.


 * Birgitte we are of similar minds. I agree that this would be used to best effect on newer article, but with the current back log it would be challenging.  The proposal already is only focused on newer articles (after January 1, 2007).  When this proposal gets approved we can work on the backlog from opposite ends of the unreferenced timeline.  On the forward to AfD, I hear where you are coming from by adding even more committee involvement.  But I think that would be counter to the rationale for prod to decrease the impact to AfD. Jeepday (talk) 02:37, 19 July 2007 (UTC)


 * Re suggestion #1. I understand the proposed focus, but there are some valid objections to applying this to all articles since Jan 1 2007.   If people creating articles without sources would get a message in the first week about this tag, we may be able to train them to start declaring sources from the beginning.  Yes the backlog would still be a separate problem, however we will be much better off a year from now and the smaller number of articles going through this system will mean they get closer attention.-- Birgitte  SB  15:44, 19 July 2007 (UTC)


 * Are you suggesting that it should only be used on articles less then a week old?Jeepday (talk) 12:29, 20 July 2007 (UTC)


 * Yes-- Birgitte SB  13:04, 20 July 2007 (UTC)

I can see where that would address many concerns of those opposed. It clarifies that this is tool for encouraging referencing and limits (severely) the potential for misuse. If we make the appropriate changes to the delete instructions and make other relevant changes like the third bullet in suggested usage to I am going to ask Jossi and DGG to come take a look at this suggestion and get there feedback. Jeepday (talk) 13:27, 20 July 2007 (UTC)
 * RFV may only be used on articles articles less then a week old.
 * I will mull this over are reply later. ≈ jossi ≈ (talk)

DGG (talk) 19:44, 20 July 2007 (UTC)
 * DGG The general idea is certainly acceptable. But there are some major considerations involving the interaction with other processes, and also with the original goals of the project.
 * 1) First, the original goal of the project was, at least in part to improve existing articles, and the need to do so remains. This should continue, perhaps as a separate project. The key element of the process would be that it does not lead inexorably to deletion if nothing is done, but rather to successive rounds of improvement--together with the merging or deletion of whatever unacceptable articles may be found. I see this as dealing with completely unsourced article as well as articles containing grossly inadequate sources.
 * 2) The grossly inadequate sources include material taken from older encyclopedias with no updating, or with the updating being unsourced; some of these articles  will be among the most difficult to deal with adequately for those without convenient access to major print collections.
 * 3) They also include articles taken  from other language WPedias that do not have consistent sourcing--most notable the German WP, where material that would reasonably be expected to be available from common German-language reference sources has not in general been specifically sourced---but where most of these articles will be about notable subjects & the contents will be accurate.  Sourcing some of these will require a knowledge of German, an awareness of German-language sources--especially those likely to have been relied on in writing the articles, access to the material, and often subject knowledge. The same problem is present to a lesser extent in material from the French and some other language WPs.
 * 4) More important, there are also a considerable number of older articles dealing with individuals or containing material on individuals, that are in gross violation of current BLP standards. In many of these cases it is a question not just of sourcing, but of NPOV, accuracy, and notability.
 * 5) It is also the case that a certain percentage of the unsourced articles will be quite clearly non-notable, and best dealt with by merging or deletion. Probably the least obtrusive way to handle them is by merging when possible. There is a separate notability project, fairly well organized and efficient, which has now dealt with all notability tags placed through the end of October 2006, and will soon reach the beginning of 2007. I'd suggest those primarily interested in this aspect work within that project. I help out there a little myself.
 * 6) The problem with newer articles is first the existing  policy on stubs, and then the coordination with New Page patrol and the existing deletion processes, and with the
 * 7) . Stubs are presently not handled in a satisfactory manner. There are two sorts of stubs--what are informally called permastubs, or very short articles, and the stub intended to be expanded. Permastubs are, for example, articles on such people as 19th century major league baseball players, or small villages, where it has been decided that the entire class is individually notable but there are insufficient secondary sources to write a full article without doing Original Research in Primary Sources. There is precedent for this even in print encyclopedias--Brittanica for example has always had one-paragraph articles on less important people and towns. (Approaches to changing this policy should be done very cautiously; personally, I'd leave the established classes of this sort alone, but resist additional ones.)
 * 8) But the sort of stub to establish a place for an new article and encourage someone to eventually expand this has always been considered acceptable. There are minimum requirements for one--see WP:STUB. However, this requirement does not include a source! The assumption has apparently been that the stub tag will result in a further check to see that it gets expanded properly.
 * 9) Perhaps that is not a good idea. But the first step is to discuss it of WP:STUB. I would support changing the policy to require at least one source, whether a reference or an informative external link, on new stubs. I'm not sure I would support having a 3rd part reference at first. It is in my opinion reasonable to enter a page about a clearly notable organization based ont their home page, as long as it does not sound like advertising.
 * 10) As for New Page Patrol, most people who do it look for articles immediately after creation. Due to system limitation, it may not be possible to go back one week without some programming. There is no point in looking at the immediately created pages with this only in mind--a great many of them will be deletable  on other grounds; but if one is doing NPP anyway, this could well be something additional to look for. (another problem is that people often make multiple edits in an article in the first few minutes, and this way you'll get it before they have a chance to add references. A more experienced admin told me he checks by looking a few thousand items back, and I have been doing likewise: the method is to  go to special:New pages, and click "next 600" In the browser address box, you'll se the parameters "limit" and "offset" Change offset to 6000. This currently take you 3 days back, after most of the junk has been deleted. I just looked at 20 of them.   4 articles that were almost certainly about notable subjects and that  clearly should have had refs. did not have them. (and I found one A7 to speedy-delete). There were also 2 articles with refs referred to in the text, but without a ref. section--these of course just need moving the information.  So the yield might be worthwhile.
 * 11) The question is then whether you would want to try to   source at least from Google rather than notifying immediately.  In favor of trying, It can be just as easy, & it immediately solves the problem. In favor of notifying, is that it educates people. In any case, if nothing is done,  I would very strongly advise continuing your present procedure of trying seriously to source it yourselves rather than nominating them for deletion.
 * 12) Next question is how to handle them. I would try very hard to merge instead, and I would do it by a separate proposed merge tag, according to WP:MERGE, which requires another five day wait.  If its disputed, then its a matter of convincing the people there that merge would be better.
 * 13) It would be highly inadvisable to add to the work at Afd. there are already between 100 to 150 items a day, and to follow it carefully takes several hours a day. I follow only about one in 10 & it still takes an hour a day. Further, it is the established practice at afd that an article will not be deleted just for being unsourced if it seems notable and sourceable. So if you do bring things there, you should be prepared to document where you had tried to source it.

I could not possibly oppose this idea any more strongly
To begin with, the image of throwing the article into the trash is offensive. Secondly, although ideally everything should be sourced, to throw things away just because they haven't been is, almost literally, throwing the baby out with the bath water. Many excellent articles have taken a long time to develop. Tag them, list them for cleanup, ask for third opinions, whatever, but to just delete something because it isn't sourced (unless it's a BLP violation) is a horrible idea. Corvus cornix 01:43, 19 July 2007 (UTC)


 * The image of the trash is used on the Wiktionary version of Requests for verification there is no image on this proposed template. Second I for one and every contributer to the conversation I have seen on this talk page (opposed or in favor) seems to believe that when RFV becomes an accepted policy it will be a last choice after we have tried all the "whatevers" you listed.  This is just the last of the "whatever". And lastly as addressed in  the policy for deleting unreferenced is addressed in other areas of Wikipedia, all we are addressing here is the timeline for deleting it.  What part of this (RFV) idea you are opposed to?  Jeepday (talk) 02:52, 19 July 2007 (UTC)


 * I can't speak for Corvus, but (a) IMO the proper timeline for sourcing an otherwise good article is "as long as it takes", and (b) the only parts of the proposed RFV I'm opposed to are the 30-day deadline, and the template. —Steve Summit (talk) 03:03, 19 July 2007 (UTC)
 * It is easy to have excellent policies, and harder to implement them. The merest glance at AfD or any other discussion page will show that there is a great diversity in what WPedians think reasonable; adding another policy with a challenge period will result in some reasonable challenges, and many not so reasonable--no matter what point of view. Every process in WP acts to push articles down hill to deletion--it is so much less work than improving them. The last thing we need is encouragement for this.
 * How about this: anyone who can provide two articles with sources gets to nominate one for being unsourced? Jeepday, who understands very well how to source articles, will still be able to do his nominations, and those who do not yet know can learn, before they start in on projects to make more work for other people. DGG (talk) 03:17, 19 July 2007 (UTC)
 * DGG if you can figure out how to make that 2 for one work, sign me up. I would raise it to 5 references for 1 nomination and have extra left over. Jeepday (talk) 03:25, 19 July 2007 (UTC)
 * We obviously thing almost exactly alike--I had originally put 5 and then changed it to 2, so as not to expect too much :). see below. DGG (talk) 04:25, 21 July 2007 (UTC)
 * My response above is probably not completely appropriate, but it does illustrate that many commenter's on this proposed policy are out there editing articles, we interact, we question each others actions, and we work to make wikipedia better. Actually we all know that deletion is not the goal of any editor who has commented here so far, the goal is to reference the article and improve Wikipedia.  Sometimes you just can't reference it for a many reasons, You really don't want to delete the article but at the same time you don't want to leave the job undone.  What do you do?  I think RFV is a good compromise between differing perspectives.  Look back at this talk page, there is everything from Speedy delete, to prod, to AFD to never delete mentioned here; 30 days is a good compromise time.  Could it be misused? sure anything can and will be misused.  Remember WP:AGF most editors will use whatever tools you give them well, to improve Wikipedia. Jeepday (talk) 05:24, 19 July 2007 (UTC)
 * It is easier to destroy than to build, a manifestation of the second law of thermodynamics. It takes work to overcome this, and we therefore need greater care to get articles improved, than to get them removed. The fast way, the tempting way, will always be to remove them, so this is what we must guard against. Anyone productive here over a long period must learn to avoid attachment to individual articles. This means more than not trying to own articles. it also means not trying to be responsible for them all. The article that i can not imp[rove today, can be improved by someone else tomorrow. the article that I cannot delete now, can be deleted eventually. It is all any one of us can do to take a few articles, and improve them--to find some that need deleting, and delete them. If one can do more, one can teach others how to do the same thing properly--I hope to help a few people, I know I can not teach them all. I'm not the only person here who can help people write better articles. DGG (talk) 04:25, 21 July 2007 (UTC)

This proposal has the right heart, but is implemented really badly
I've long suggested something along these lines (and I'm an inclusionist). The problem with this particular suggestion - as with the abuse of PROD and AfD for similar ends - is that it's completely arbitrary. The article gets nominated when some random who wants the article deleted nominates it, and anyone who wants the article kept has to drop what they're doing and fight it. It breeds nastiness, and it wastes time.

What would be a much more effective process is to treat unsourced articles similarly to the way we used to treat images back when copyright was just beginning to be enforced properly. We could have a bot tag a certain number of articles a month, and give people say, a three month window to actually get them sourced. You'd still have the time window before deletion to put pressure on people to actually source their work, but you'd be giving them more than enough time to do it - actually giving people time to get to libraries if necessary, and without forcing people to immediately drop any other projects on someone else's whim. Rebecca 06:34, 19 July 2007 (UTC)


 * The suggest for AfD at one end is one persons suggestion the same as your suggestion of an increase to 3 months prior to delete is one persons suggestion. Jeepday (talk) 13:27, 19 July 2007 (UTC)

Time to tag as rejected?
There is clearly no consensus for this to become a guideline. Perhaps we should tag it as rejected and move on? JulesH 07:38, 19 July 2007 (UTC)


 * What definition of "clearly no consensus" are you using to base this suggestion on? Per Consensus requires serious treatment of every group member's considered opinion Per Consensus decision-making a decision-making process that not only seeks the agreement of most participants, but also to resolve or mitigate the objections of the minority to achieve the most agreeable decision. I don't beleive that 6 days qualifies as serious treatment of this proposed Policy. Many of the objections to this proposed Policy are unrelated to this policy, they are complaints about other policies, while valuable to this discussion they are not relevant to forming consensus for a time period to implement a standing policy.  It is interesting you offer to reject this guideline, when you have not logged any opinion on the proposed policy talk page.  If you have objects, please state them. Jeepday (talk) 13:46, 19 July 2007 (UTC)

Very good proposal
I'm leaving wikipedia soon, so this will be my only comment on the proposal. It seems to me this is a very good proposal. Sadly, as with a number of other good proposals, it looks like it's going to die because the existing viewpoint is too entrenched. Perhaps when things get messy enough that people will see the merits it will come back, but I'm not holding my breath. Sad to say, far too many people seem to think that utterly crap content is better then no content (IMHO a lot of the time this is not true). And many also seem to think that we shouldn't encourage and expect editors to contribute content that meets policy. But even though it's a good thing if an editor is willing to contribute content in the form of an article, if said editor is unwilling or unable to provide at least one reference for said article in 30 days then as hard as this may be for others to accept, we are in fact nearly always better off without their contributed articles. This may sound harsh but it's the truth. Verifiability and NOR are cornerstone policies and it's simply not acceptable to encourage editors to contribute without also encouraging and expecting them to abide by these policies. Also, contrary to what some seem to be suggested on this talk page, I don't think many supporters of the policy are saying that we should jump on newcomers the moment they write something unreferenced. Instead what is being said is that it is vitally important that all editors (newcomers or old) reference what they write and if they don't then there is nothing wrong with telling them that they should. Also, I think all this talk of wikilawyering and POV pushers is missing the point. Yes, wikilawyering and POV pushers are a problem and can be incredibly frustrating. But rather then shunning good policy because of the fear it'll be misused by wikilawyering and POV pushers, we need to fight them by not giving them room. Proper referencing and not allowing useful to be baited will make it a lot more difficult for these people to operate. But far to often they are almost encouraged to operate by the poor editing practice and behaviour of well meaning editors Nil Einne 15:48, 19 July 2007 (UTC)

Possible merit
I see some posisble merit in this idea, but it needs refinement. I think that many of DGG's commetns above are well taken. I also think that an editor placing thsi tag whould be required to certicy that s/he has made a reasonable search for references, and has been unable to find any relevant ones. It was said above that current policy palces the burden of findign refs on the contributor, not the challanger. Thnat is true, but if this propoisal is to change policy by proposing a new deletion mechanism, it is IMO reasonable to also change it that anyone choosing to tag via this mechanism be required to do at least a basic search for references, since no one wants to delete reasoanble articels merly because redily available refs have not yet been added to the article. DES (talk) 21:45, 20 July 2007 (UTC)

After further consideration...
... I still do not see anything in this proposal that cannot be addressed by prod, and existing processes. ≈ jossi ≈ (talk) 23:50, 20 July 2007 (UTC)


 * The existing processes were never meant to deal with the issue of unreferenced but notable articles, and they deal with it remarkably badly, generating minimum results for maximum nastiness. This process was at least an attempt, albeit a flawed one, at coming to a rational solution to the problem. Rebecca 03:36, 21 July 2007 (UTC)
 * I don't think so... Use  Unreferenced article since August 2006 , for example. ≈ jossi ≈ (talk) 15:51, 21 July 2007 (UTC)
 * Though I don't support this proposal as suggested, I have to agree with Rebecca. It's very tempting to simply remove a prod tag with a rationale of "Unreferenced article since August 2006". That rationale does not suggest that any sort of attempt was made to find referenced before slapping on the prod tag. Black Falcon (Talk) 20:05, 21 July 2007 (UTC)
 * It's also completely not what PROD was intended for, and I'd consider an appropriate response to such a tagging to be to remove it and tell the tagger to get stuffed. We need to be sourcing material, not deleting notable material through incompetence. Rebecca 02:33, 23 July 2007 (UTC)
 * I also agree that  Unreferenced article since August 2006  is not appropriate use of PROD. Jeepday (talk) 13:21, 24 July 2007 (UTC)

the fundamental question
Back up in Five premises and a conclusion, Jeepday quoted Verifiability which says, "Any edit lacking a source may be removed". The key point, for me, is that the policy does not say, "Any edit lacking a source must be removed".

If you come across a dubious statement or one you suspect of having been entered in less than good faith, by all means, tag it with   or   or whatever and then, suitably later, if no citation is forthcoming, delete it. That's all doable under existing policy. The objection many of us have with the proposed RfV policy is that it seems to steer us ever closer to a more draconian Wikipedia in which any and every unsourced statement will be removed, no matter what. —Steve Summit (talk) 15:19, 21 July 2007 (UTC)


 * RFV is discussing the deletion of complete articles that are unreferenced. Jeepday (talk) 13:24, 24 July 2007 (UTC)

A suggestion for encouraging and better enabling proper sourcing
BirgitteSB made a good suggestion I would like to build upon as at least one part of dealing with the issues under discussion – and one which does not in of itself lead to deletion. I think we’re all agreed that we need to strongly encourage new editors – actually, all editors of new articles – to begin sourcing their work as early as possible. (If nothing else, it’s easier to do so when the sources are at hand, than for the creator or someone else to try to rediscover them afterwards.) As I pointed out earlier, the citation codes really aren't all that easy to learn – and the current pages of instruction aren't at all new-user-friendly – and when you're still busy learning a lot of different aspects of Wikipedia (in your spare time), which can be rather daunting and quite overwhelming, 7 or even 30 days isn't very much time.

In a related discussion, I suggested that we create a "sample" stub page – perhaps along the lines of Annotated article – which could serve as a general template (in the informal sense of the word) for new editors to refer to as a guide to the minimum requirements for starting a new article A short explanation of how each part satisfies the basic requirements would clarify what needs to "be there" and why. Furthermore, a quick look at the coding would show basic formats for citations, categorizing, etc.

I would like to recommend that, say, 7 days after a new article is created, if no citations have been added, that a bot send a “pleasant, respectful” notice to the creating author’s talk page (and perhaps as well to the article’s talk page) reminding them that the article they created still lacks sources, briefly summarizes their importance, and directs them to a “Annotated stub” page for an example of our “minimum article requirements”. This would be helpful regardless of what eventually gets decided about a tagging/deletion process. Askari Mark (Talk) 18:42, 21 July 2007 (UTC)


 * Seems as though this doesn't need proposing. I like the annotated stub idea, and if that comes to exist then I can't see anyone objecting to pointing it out to editors along with the offer of helpful information. That's the kind of non-controversial thing that doesn't really need proposing—just go ahead and do it. As far as having a bot detect which articles haven't been sourced, there's a can of worms worthy of its own topic with wider applications than what is proposed here. At any rate, with so many bots already the amount of willing programmers likely exceeds the need for new bots. I wouldn't expect it to be a problem to find someone willing to attempt this relatively complicated parser design task.  Big Nate 37 (T) 19:12, 21 July 2007 (UTC)


 * This still runs into the problem that WP:STUB does not require a reference. We would need to first change that--and I do not think that we either could or should. People probably should be able to start an article about something that will clearly be notable, but without having any reliable source specifically in mind, with the intent that others will complete it--that was originally the purpose of stubs and it remains a purpose. It's not the best way to do a stub, but it does serve the purpose of getting non controversial content into WP for later development.

I do not think we should change that without a general community discussion. That the proposal might not succeed in one is not an an argument for trying to change the policy here. DGG (talk) 19:26, 22 July 2007 (UTC)
 * I have just found "The perfect Stub Article" in Meta--written by Jimbo in 2001. It explains things very clearly--a much better version similar to what I was trying to say--and it does not mention references. of course, this is six years later, and probably now he would say that at least one reference would be desirable.
 * I therefore see two courses: one is to change the stub guidelines to require references--the other is to rework entirely what we think we were going to do here, for any move to remove unreferenced articles is simply against longstanding WP guidelines, and that's the end of it. DGG (talk) 19:39, 22 July 2007 (UTC)
 * Er, the stub guidelines simply outline what a stub ideally should consist of, which should of course include references. I'm not sure what the point would be of "changing the stub guidelines", apart from stating the obvious. Rebecca 02:35, 23 July 2007 (UTC)

The matter of the example having citations where they are not necessary to avoid CSD (but I argue they are necessary in the final version) is easily solved: show citations and in the annotation explain they are not required to avoid speedy deletion but recommended for the good of the article.  Big Nate 37 (T) 08:30, 23 July 2007 (UTC)


 * Reading this section I am wondering if people are confusing references and citations. When we are talking about unreferenced articles, we mean an article without a single useful external link or off-line reference given.  It has nothing to do with if the reference is in a properly labeled section or in citation style.  We are only talking about articles which have nothing to back up the information.  These are where we will find hoaxes and misinformation (remember Porchesia).  The truth here is no one is wanting to delete valid information, but sometimes threatening deletion gets things done.  I am not supporting this proposal as it stands.  However if we were going to use the threat of deletion to improve the reliability of articles, these articles (the ones without a single bit of information to back them up) would be the appropriate focus.  This proposal may be entirely unworkable, but I want to be sure that everyone is clear what sort of articles are under discussion: ACME Animation Factory, ACGT, ABCO Foods.  And also remember that everyone has agreed that anyone placing a tag should do their to source the article first.  As I said I do not think it is very workable, a version with enough checks and balances to get an acceptable result would be too bureaucratic.  I did like the idea of restricting it one week from creation so mostly because it would keep the number of actively tagged article small and I think everything that could be sourced would be saved (Examples of qualifing articles: Mr. del, Liga Nacional de Fútbol de Honduras 1983/84, List of shipwrecks in 1998).  But that is not easily done as DGG pointed out.  Another idea, which I admit is very radical, is to blank the article when placing a "request for verification".  There would be no deletion and the template could link to the last version before tagging so editors could easily restore the article when adding a reference.  At en.WS our copyright violation template somehow automatically blanks any page it is displayed on.  The idea would be that the article is hidden until verified.  As I said it is pretty radical.-- Birgitte  SB  14:19, 24 July 2007 (UTC)
 * "Hidden" radical but interesting concept. Jeepday (talk) 04:04, 25 July 2007 (UTC)

[Unindent] I certainly wasn’t proposing sending bot messages that threatened editors with deletion of their stub articles if they don’t attend to identify their sources immediately, nor do I believe the stub guidelines need to be changed to enable this (however worthy such a change might be). I was intentionally using the terms “referencing” and “citing” colloquially; there are, after all, three main ways to do it and what is important here is encouraging editors to get into sourcing early and easily. Right now, though, the “Help” pages aren’t very helpful as they are anything but clear and easy to understand for the novice editor to understand. I started on Wikipedia a year ago and while I am quite computer-literate and familiar with academic citation, it took me a while to figure out how to do it. Those of you who’ve been around a few years probably don’t realize how intimidating Wikipedia can be to those who are just starting out – and I cannot say that sourcing has become any easier to learn since I joined.

I was unaware that it would be difficult to detect a lack of any sort of citation, but that is easily worked around. One possibility is an automatic “reminder” message generated a week after an article is created advising the creator along the lines of


 * “If you haven’t already added references to the article you recently created, XYZ, please take some time to do so soon. It’s best to do so while you still remember them, and it’s very difficult for another editor to determine what they might have been.  Don’t worry about getting them perfect, but please ensure that sufficient information is added so that they can be improved later.  If you’re unfamiliar with how to add references, a simple example of a stub article with references can be found at Annotated stub.”

I was thinking that such an example might subsume “Good stub” and perhaps work Larry Sanger’s “[The Perfect Stub Article]” into a single piece, but possibly somewhat more along the format of Annotated article. Askari Mark (Talk) 03:28, 25 July 2007 (UTC)
 * Sorry I didn't mean to imply you meant to send out bot messages like that. Somehow my comments on your idea grew into comments on the range of things on this talk page.  I don't see why not to work on a help page like you suggest.  However I think citation-style references are a little complicated for newbies.  I still have problems getting them to work at times (basically I copy and modify a citation someone else did without a complete understanding of how it all works).  I would suggest you just focus on encouraging people to list their sources at the stub stage.  I wouldn't want anyone to decide it is to complicated they will just leave it out.  Once you have something together, I would suggest making a talk page template to direct newbies there in a friendly way and ask the people who do new page patrol to start leaving that on the talk pages of new editors whose stubs are lacking something.  I don't feel comfortable sending a bot message to everyone making a new page,-- Birgitte  SB  13:14, 25 July 2007 (UTC)


 * Just to add my experience with citations, I use Citation templates. I copy the code, paste it to the article, fill out the fields I know and delete the ones I don't, and remove the line breaks. Finding this from following links from Template messages seemed easier than learning from help pages would have been.  Big Nate 37 (T) 14:37, 25 July 2007 (UTC)

[Unindent] I have been following all discussions for a while, and I also like Birgitte's suggestions, as well as providing a link to "This is how a start-class article/perfect stub should look like" in the message template. I am not sure however if the 30-days limit for all articles created after Jan 1 is really appropriate. If articles without references (and if it's just an external link) have survived for half a year now and no-one has prod-ed them, I think we shouldn't put RfV on them now. What I think would work however is if mainly Newpages patrol uses the proposed RfV for newly created articles that assert notability but just happen to be unreferenced. Usually, the article creator sees them immediately and can address the issues with the helpful perfect-stub link without much outside help if he is really willing to create a lasting article for wikipedia, not just be another lazy spammer ("But I didn't know how" won't work no more). 7 days before deletion would be sufficient I think to provide one book name or one external link that addresses the main claims made in the article. In this case RfV is a special case of prod, yes, but at the moment, Newpages patrolers are more concerned with keeping spam out by using speedy tags, not prods, because prods need some kind of well-sourced justification, which the patrolers can't "make up" in such a short time. This proposed RfV would be easiest for Newpages patrol, newbie contributers (if the template contains a link to "the perfect stub") and everyone else. (And I am aware that this means knowing even one more tag for NP patrolers, but since most suspicous new pages are checked anyway, adding RfV to where there is no Reference section is no more work than deciding whether a page needs a speedy tag.) – sgeureka t•c 15:11, 25 July 2007 (UTC)

Next Steps
Conversation seems to have stopped on this page. The last entires are ones I made at. Does anyone have thoughts on consensus that has been reached or next steps for this proposal? Jeepday (talk) 13:47, 4 August 2007 (UTC)
 * there was no consensus. The next step, in my opinion, would be to mark this as rejected. DGG (talk) 23:38, 14 August 2007 (UTC)
 * I think I am going to have to agree with you. We did make some headed way maybe in a couple years it can be readdressed.  I see Radiant! has marked it appropriately, so... Jeepday (talk) 02:23, 16 August 2007 (UTC)