Currently, there are over 114,000 articles on Wikipedia that contain no citations or sources, making it one of the largest clean up categories on the site. WP:WikiProject Unreferenced articles has been one of the main WikiProjects attempting to dig through this giant haystack in order to give as many articles proper sources. Unfortunately, a main obstacle to cleanup has been how stringent deletion policy is. If you WP:PROD an article, it takes a week to delete, which is fine, and can be reversed by anyone. The issue is that many of these articles are unsourced stubs with no indicated notability, an article that me and others would agree to be a uncontroversial deletion via WP:PROD. Many of these PROD's are contested and then must go through the possibly month long review process VIA WP:AFD. The conclusion to this process usually is delete, but I believe that a criterion should be proposed that biases an article in favor of deletion, which is not having any sources. If this is written into the WP:DELETE policy, then I believe that editors like me will have a much easier time combing through the massive garbage dump that are unsourced articles. Tooncool64 ( talk) 06:39, 5 January 2024 (UTC) reply

Doesn't our current policy effectively do that? Editors arguing for notability are already required to provide or attest to the existence of sources which support notability. Horse Eye's Back ( talk) 07:09, 5 January 2024 (UTC) reply

True, but this is more directed at solidifying a valid reason for deletion, or a secondary reason, an article lacking sources, such that a PROD could say "Article fails WP:NGEO and WP:NOSOURCE", and be viewed as uncontroversial. Tooncool64 ( talk) 07:12, 5 January 2024 (UTC) reply

I think to some extent PROD will always be controversial. Horse Eye's Back ( talk) 08:15, 5 January 2024 (UTC) reply

Ironic. Folly Mox ( talk) 14:02, 11 January 2024 (UTC) reply

Interesting. I wasn’t aware of that. I proposed deletion for this article [1] but the tag was reverted. The reason was supposedly that other elections later on are notable, but regardless, the problem is many of the earlier articles are unsourced and redundant, and many just redirect to the nominated Emperors' pages. Yr Enw ( talk) 07:46, 5 January 2024 (UTC) reply

Its not a great reason, but its nice that they gave a reason at all (none is actually required to remove a PROD). The next step would be opening a talk page discussion on notability, hopefully the editor who removed the PROD is willing to work with you to find sources and if not will be willing to support a move towards AFD. Horse Eye's Back ( talk) 08:15, 5 January 2024 (UTC) reply

I did talk page them, but they never responded. I get the need for collab, but often it can just become unintentional filibustering Yr Enw ( talk) 09:19, 5 January 2024 (UTC) reply

What if the concepts of WP:BLPPROD were expanded to non-BLPs without any sources? At a minimum, a deprod could be required add one reliable source.— Bagumba ( talk) 08:35, 5 January 2024 (UTC) reply

This might be the best option. I wasn't even aware of the WP:BLPROD policy myself, but having a similar policy apply to unsourced articles would allow for both one, editors to more quickly sift through unsourced articles, and two, editors who want to do specialized research to find obscure sources for articles that are proposed via this hypothetical process. If no sources can truly be found, reliable or otherwise, then it would be an uncontroversial deletion that would be able to avoid the lengthy WP:AFD process. Tooncool64 ( talk) 08:43, 5 January 2024 (UTC) reply

Brilliant. 100% support this.— S Marshall  T/ C 08:51, 5 January 2024 (UTC) reply

Also support this. JoelleJay ( talk) 20:16, 5 January 2024 (UTC) reply

Support this idea myself as well. Let'srun ( talk) 15:06, 6 January 2024 (UTC) reply

Well, I don't. AfD exists for a reason: inclusion criteria are based on whether sources exist, and whether it's possible to write an article on a subject. They aren't based on whether Bill has time tomorrow afternoon to go get an interlibrary loan and then drive out to pick up eighteen books and spend the entire evening going through to frantically reference 53 articles before the guillotine falls. AfD lasts seven days. If an AfD is relisted because of lack of participation, it means that there isn't enough volunteer effort available to properly assess the article. If there isn't enough volunteer effort available to properly assess an article...there isn't enough volunteer effort available to come to a firm conclusion that the topic is non-notable. jp× g 🗯️ 09:39, 6 January 2024 (UTC) reply

If someone wants to re-create the article in the future with sources, then more power to them. It would be a soft-delete, allowing an editor to re-create the page. Tens of thousands of these articles have no reason to exist, no content, no usability for information. Like I said previously, Wikipedia is not meant for collecting items that exist. Tooncool64 ( talk) 10:02, 6 January 2024 (UTC) reply

Well, it's not meant to be a shoot-em-up game either -- the fact that deleting articles causes an enjoyable sensation on the back of the neck isn't a reason to do it. There are plenty of reasons why stubs exist. They're written by someone who had access to some information, or maybe to a lot of information, but who for whatever reason wrote a very short article; for the vast majority of them, it's completely possible to write something longer. If it's not, and the article is such a turd it needs to be wholly extirpated from the project, we have AfD, which sees approximately fifty nominations per day, with a turnover of somewhere around a week. In fact, we also have draftification, PRODs and speedy deletion -- that makes four separate processes by which stuff can be taken out of mainspace if it's bad. Why do we need another? jp× g 🗯️ 10:33, 6 January 2024 (UTC) reply

Bagumba, a proposal to establish the system you describe recently failed at an RfC a few months ago ( Wikipedia:Village pump (proposals)/Archive 207#Request for comment: Unreferenced PROD). Curbon7 ( talk) 08:54, 5 January 2024 (UTC) reply

I can completely understand why many people where against this in the way it was worded. If an unsourced PROD were to exist, it would need to have at the very least a 7 day time limit, like current WP:PROD. The major reason I am in favor of something like this is because I believe, at the very least in 2024, articles on Wikipedia need to have sources, even if it is just one. No article would pass WP:AFC without sources attached. Tooncool64 ( talk) 08:59, 5 January 2024 (UTC) reply

Thanks for pointing that one out. After a quick glance, it seems it involved a new tagging process that people objected to, as opposed to just expanding a known process, PROD. The proposal just waved at a link, and some likely thought TLDR or made some wrong assumptions, and rejected for that reason. Not saying this would necessarily pass, but an improved presentation and concise pitch could go a long way. — Bagumba ( talk) 09:18, 5 January 2024 (UTC) reply

Yep. I tried an RfC on that. Snow-opposed. (Although the wording was really badly done, as I recall, so everyone was at least moderately confused.) 🌺 Cremastra ( talk) 00:42, 7 January 2024 (UTC) reply

This is close to becoming a perennial proposal. Policy is the way it is a foundational principle of this project is that imperfect content is an opportunity for collaboration, not something that needs to be expunged. If you instead choose to look at articles that fellow volunteers have taken the time to write as a "garbage dump" and deletion as the preferred way of dealing with them, then of course you're going to meet friction. –  Joe ( talk) 09:05, 5 January 2024 (UTC) reply

My rhetoric might be harsh, but unfortunately, many of these unsourced articles are tens of thousands of one sentenced geography stubs, that may or may not even meet WP:NGEO, or tens of thousands of unsourced "Topic in Year" articles. If you are looking at these articles as part of a maintenance category, which they are, then you are forced to realize that many of these are not worth keeping, if for the very fact that they are unusable for information. Tooncool64 ( talk) 09:13, 5 January 2024 (UTC) reply

Clearly at least one person disagreed with you about that, or the articles wouldn't exist. –  Joe ( talk) 10:01, 5 January 2024 (UTC) reply

Unfortunately, the standards for creating articles was much lower back in the day. Tooncool64 ( talk) 10:04, 5 January 2024 (UTC) reply

Wikipedia would not have flourished had today's standards eben in place from the beginning. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:25, 29 January 2024 (UTC) reply

Agreed. But this is not 2009. 🌺 Cremastra ( talk) 16:27, 29 January 2024 (UTC) reply

So what? Here are some "one sentenced geography stubs", generated as single-sentence stubs from a database: Chain Island, Tinsley Island, Bull Island (California), Kimball Island, Joice Island, Island No. 2, Russ Island, Atlas Tract, Empire Tract, Brewer Island, Fox Island (Detroit River), Spud Island, Hog Island (San Joaquin County), Fordson Island, Tule Island, Headreach Island, Stony Island (Michigan), Aramburu Island, Bradford Island, Van Sickle Island, Powder House Island. You will notice these are twenty GAs and a Featured Article, all of which were written from said stubs -- the "garbage dump" of which you speak. The issue is that writing things requires effort and skill: the solution is not to spend all day sitting around coming up with new ways to delete stuff. jp× g 🗯️ 09:44, 6 January 2024 (UTC) reply

That's amazing how much hard work and care went into those articles! If an editor in the future wants to re-create an article that was deleted via this hypothetical process, it wouldn't be difficult. We do not need to hoard unsourced articles currently for the possibility in the future that they may be found to be notable. Tooncool64 ( talk) 10:03, 6 January 2024 (UTC) reply

Yes it will: our hypothetical editor will have to notice that something's a redlink (from where?), look through the deletion log, ask the deleting admin for a WP:REFUND, wait on a response, and then get it restored to their userspace or draftspace. This is a rather long and complicated process that, generally, only power users are able to do. What concrete benefit is brought by forcing them to go through this? jp× g 🗯️ 10:33, 6 January 2024 (UTC) reply

Or they can just...create the article themselves without going through REFUND... The difference between expanding and de novo creating a 1-sentence stub is like, the one minute it takes to create a 1-sentence stub... An admin could literally paste the entire REFUND of the text in an edit summary, it's not like we're talking about valuable starting material here. JoelleJay ( talk) 18:24, 6 January 2024 (UTC) reply

Creating a new article on a title that has been deleted before requires one to know that one is encouraged to recreate some, but far from all, deleted articles. The box that comes up for all deleted content is far from encouraging. Espresso Addict ( talk) 11:13, 7 January 2024 (UTC) reply

The problem in that regard is that the way PROD is set up collaboration is "encouraged, but not required." Why not require collaboration as a requirement of challenging a PROD? Horse Eye's Back ( talk) 09:17, 5 January 2024 (UTC) reply

The fallback collaboration option is a formal AfD. PROD offers some rare opportunites for lightweight deletions if nobody is looking or people agree and don't contest, but a WP:REFUND is typically possible. — Bagumba ( talk) 09:24, 5 January 2024 (UTC) reply

TBH I think the ideal collaboration option is actually in between the two... A talk page discussion should be able to settle the issue the vast majority of the time... If the challenger was required to open a talk page section with their rationale (preferably in the form of sources) I think that would go a long way towards facilitating collaboration without the wounded feelings that jumping to AfD can cause. Horse Eye's Back ( talk) 09:35, 5 January 2024 (UTC) reply

Could not agree more. Tooncool64 ( talk) 09:36, 5 January 2024 (UTC) reply

Honestly I consider PROD a failed experiment at this point. The grey zone between CSD and AfD is just too narrow to support an extra process, and the awkward process (add a template, wait a week, keep checking back in case it's removed and you need to turn it into an AfD) makes it useless for anybody who's patrolling articles en masse. –  Joe ( talk) 10:03, 5 January 2024 (UTC) reply

I would love to see some statistics on PROD... What percentage get challenged... What percentage of those go to AfD... What percentage of those survive AfD... Horse Eye's Back ( talk) 10:12, 5 January 2024 (UTC) reply

@ Joe Roe: what do you think of the notability tag? Also in the grey zone? Horse Eye's Back ( talk) 10:25, 6 January 2024 (UTC) reply

The problem is that nominating an article for prod takes a few seconds, and editors often nominate many in a short space of time. Finding a source will often take hours or more, and needs to be specific to the article in question. They are not symmetrical operations. Espresso Addict ( talk) 00:55, 6 January 2024 (UTC) reply

The creator of the article can take as long as they need to find sources, years even. There is no need to create the article in mains space to work on it, it can be done in draft or namespace . Horse Eye's Back ( talk) 10:15, 6 January 2024 (UTC) reply

No problem with requiring sources for new articles, but we're talking about the backlog of old ones here. Espresso Addict ( talk) 11:07, 7 January 2024 (UTC) reply

There has never been a time when that wasn't true, its as true of the old ones as the new ones... If the creator didn't want them judged by mainspace standards they wouldn't have created the article in mainspace. Horse Eye's Back ( talk) 15:52, 7 January 2024 (UTC) reply

These "mainspace standards" you speak of... Where exactly are they documented? Can you find a policy that actually says something remarkably like "Even if whole books have been written on the subject, the article should be deleted unless someone added a little blue clicky number"? I've never seen such a statement in any policy or guideline, and I've been editing Wikipedia for 17 years now (and its policies and guidelines for nearly that long).

Fun fact: Do you know how many little blue clicky numbers were used in Wikipedia articles on the date of its four-year anniversary (15 January 2005)? We know the exact number. WhatamIdoing ( talk) 00:49, 12 January 2024 (UTC) reply

Thats because you have created a straw man, I never argued that and we both know that its absurd. Horse Eye's Back ( talk) 12:11, 13 January 2024 (UTC) reply

It looks to me like you're saying that there are "mainspace standards" and that those standards judge articles according to the presence of citations to sources in the article, rather than according to the presence of reliable sources in the real world. Do I understand your view correctly? WhatamIdoing ( talk) 20:20, 13 January 2024 (UTC) reply

Requiring a source be added to dePROD an unsourced article would be ideal. JoelleJay ( talk) 20:14, 5 January 2024 (UTC) reply

+1 Mccapra ( talk) 22:15, 5 January 2024 (UTC) reply

• oppose incredibly strongly. Merge and redirect them if you can't source them rather than run them through afd. Deletion isn't clean up. If the merger is undone, build consensus for the merge. Job done. There's no deadline. We don't need to constantly revise the rules to do the work, we just need to use the rules we've got to make Wikipedia better and coach the editors around us on the way to use the tools. Merge. Redirect. Build consensus that that's the right approach. Hiding T 22:05, 5 January 2024 (UTC) reply

  exactly! Newystats ( talk) 02:04, 16 January 2024 (UTC) reply

• Oppose. Haven't we discussed this recently? The existing procedures (prod, then AfD if challenged) are adequate for removing nonnotable unsourced material. Espresso Addict ( talk) 00:32, 6 January 2024 (UTC) reply
• I think this is a bad idea for the reasons I've explained above. jp× g 🗯️ 09:44, 6 January 2024 (UTC) reply
• Oppose. This proposal is not compatible with WP:NEXIST or WP:ATD. The topic of an unsourced article is often notable. The content of unsourced articles is often accurate and verifiable. In fact, the content of unsourced articles is often WP:BLUE. James500 ( talk) 11:40, 6 January 2024 (UTC) reply

  “The content is often WP:Blue”… really? prove it! 😉 Blueboar ( talk) 15:13, 6 January 2024 (UTC) reply

  The article is so obvious, we don't need sourcing! Lee Vilenski ^{( talk • contribs)} 00:48, 7 January 2024 (UTC) reply

• So your suggestion is to merge unsourced material elsewhere...? JoelleJay ( talk) 22:07, 6 January 2024 (UTC) reply

If the info is unsourced, then we shouldn't be merging it anywhere. Lee Vilenski ^{( talk • contribs)} 00:47, 7 January 2024 (UTC) reply

Nah, part of the merge process is either sourcing what is unsourced or discarding it and improper for merging. This discussion is about entire erstwhile articles with no sources, not about snippets of text without sources in articles that otherwise are sourced.  —  SMcCandlish ☏ ¢ 😼  03:14, 7 January 2024 (UTC) reply

Merge and redirect them if you can't source them is directing us to merge the unsourced content of an unsourced article into another article. JoelleJay ( talk) 18:44, 7 January 2024 (UTC) reply

• Oppose. I agree with Hiding, above. We aleady have the policies we need; what we lack is editorial focus to get the job done.  —  SMcCandlish ☏ ¢ 😼  03:14, 7 January 2024 (UTC) reply

My idea would be to increase the “unsourced article deletion” time to 60 days. Then I would probably accept it. 71.239.86.150 ( talk) 20:08, 7 January 2024 (UTC) reply

RfC: Should a special PROD category, similar to WP:BLPROD, be created for unreferenced tagged articles?

Should a special PROD category, similar to WP:BLPROD, be created for unreferenced tagged articles? Tooncool64 ( talk) 06:33, 8 January 2024 (UTC) reply

This category for deletion would have four caveats.

1. Articles could be removed from this process by having at least one sourced attached to it, removing the unreferenced tag.

2. Articles held within this category would not be deleted until 30 days have passed, hopefully allowing editors ample time to go through these articles and potentially find sources.

3. This would not be an automatic process. Unsourced articles would optimally be only tagged for this special PROD after editors have looked for a source and have failed to find one.

4. This proposed PROD policy would not supersede WP:AFD or WP:CSD.

Survey (RFC for an unreferenced PROD procedure)

• Oppose per Expresso Addict above and Phil Bridger below. This proposal is not compatible with WP:NEXIST or WP:ATD or WP:BEFORE. The topic of an unsourced article is often notable. The content of unsourced articles is often accurate and verifiable. This proposal was overwhelmingly rejected just three months ago. The word "optimally" in caveat 3 appears to mean that tagging without a search for sources would not actually be forbidden. Caveat 3 does not specify that the search for sources must comply with the "minimum search" specified by criteria D1 of WP:BEFORE, let alone require a search for sources on sites that are not indexed, or not properly indexed, by Google, and which can only be searched, or properly searched, with their own internal search engine. The proposal assumes that if at least one source exists, then someone will search for, find and add that source within 30 days. In view of WP:NOTCOMPULSORY, there is no reason to assume that anyone will do that, not even the closing admin. James500 ( talk) 18:28, 8 January 2024 (UTC) reply
• Oppose Per my comment in the section below, which was made before the title of this section changed. Phil Bridger ( talk) 21:02, 8 January 2024 (UTC) reply
• Oppose - purging articles without references after a 30-day waiting period will not contribute to building an encyclopaedia. What is more, as written this proposal will inevitably lead to editors stripping citations from articles that have references (whether due to concerns about source quality and relevance, or for quite different motives), then PRODding them under the new criterion. Of course similar things already happen prior to some AfD noms, but even our largely broken AfD process calls in more eyeballs than this new "unreferenced PROD" process would do - I imagine that if the initial tagging is turned over to bots and carried out at a mass scale, then the judgement of only one human editor would typically be involved in each deletion. Not in the interests of encyclopaedia-building, as I say. Newimpartial ( talk) 21:24, 8 January 2024 (UTC) reply

  editors stripping citations from articles that have references [...] then PRODding them under the new criterion; yes, I have seen similar many times with BLPPRODs. Some IPs will vandalize a BLP by removing the citations and then an editor drive-by tags the article for BLPPROD without checking the page history. Curbon7 ( talk) 22:18, 8 January 2024 (UTC) reply

• Oppose - As far as I can see, no examples have been given of unsourced articles that these proposals intend to target, so I looked in WP:WikiProject Unreferenced articles and cherry-picked a few: Echo chamber, Agricultural aircraft, Armoured companion. I don't see any value to the encyclopedia by speedily deleting articles like these. If they must go (and I hope they don't), then they should at least go through the AfD process. Toughpigs ( talk) 22:17, 8 January 2024 (UTC) reply
• Support This is clearly going to fail, but my comment the previous time this was proposed remains valid: The current situation, through no fault of anyone else, amounts grandfather clause (old unreferenced articles survive, new ones usually get deleted), which is exactly what should not be happening * Pppery * _{it has begun...} 06:07, 9 January 2024 (UTC) reply
• Support, per WP:V, which says any material whose verifiability has been challenged or is likely to be challenged, must include an inline citation to a reliable source that directly supports the material. This will provide a practical way to challenge material in articles where there are no sources.
  This change would also bring us into stronger alignment with WP:5P2, which says All articles must strive for verifiable accuracy, citing reliable, authoritative sources, especially when the topic is controversial or is about a living person.
  Beyond policy and principles, we want to build an accurate encyclopedia that readers can trust, and we want to demonstrate through practice rather than instruction that new material, including articles, should be referenced (" Do as I say, not as I do"); this change will help towards both ends. BilledMammal ( talk) 06:26, 9 January 2024 (UTC) reply

  I believe this proposal is primarily about articles whose contents (see Wikipedia:Glossary#verifiable) have not been challenged and whose contents are WP:UNLIKELY to be challenged – you know, the sort that say things like " Christmas candy is candy associated with Christmas", not articles that people really wonder whether it'd be possible to find a source for the contents. WhatamIdoing ( talk) 00:08, 10 January 2024 (UTC) reply

  I think you misunderstood my !vote; the prod would itself be the challenge, and in doing so would address the current issue in that there is no practical way to challenge material on WP:V grounds in unreferenced articles. BilledMammal ( talk) 07:25, 10 January 2024 (UTC) reply

  Wikipedia:Proposed deletion is a process for claiming that you don't believe the subject qualifies for an article. It is not a process for claiming that you don't believe the contents of the article could be supported by reliable sources. Even if the contents are already cited to reliable sources, that doesn't mean that we should have an article on the subject. Only the belief that it's impossible to source the material justifies a WP:CHALLENGE.

  For example, even though "WhatamIdoing is a Wikipedia editor", followed by a link to Special:Contributions/WhatamIdoing, is verifiable, if anyone ever creates such an article, you should feel free to add {{ subst:PROD}} to the top – but not to add {{ fact}} to the end of the sentence. WhatamIdoing ( talk) 01:14, 11 January 2024 (UTC) reply

  To be fair I think this is the strongest argument in favour of this and similar past proposals. There is an imbalance in how policy treats unreferenced content within articles compared to wholly-unreferenced articles.

  However, there is one key difference between "challenging" unreferenced content by removing it from an article and by deleting the entire article: the former can be undone by anyone, at any time, when they find a source, because it remains in the page history; the latter can only be undone by an administrator, because the material is removed from public view. This is an important difference because this is supposed to be the encyclopaedia that anyone can edit, and it's why the deletion of articles is strongly regulated by the administrator right, a dedicated policy and multiple well-defined processes, whereas removing content (even all the content) is basically treated like any other edit. In other words, the imbalance is intentional and (in my view) necessary.

  Unreferenced articles can still be "challenged" by stubbing, blank-and-redirecting, draftifying (for new articles), or with {{ unsourced}}, {{ citation needed}}, and related templates. –  Joe ( talk) 09:21, 10 January 2024 (UTC) reply

  And, in that latter case (e.g., stubbing, blanking-and-redirecting, etc.), you aren't challenging "the article's existence" (which is what all forms of deletion do); you are challenging "the current contents". WhatamIdoing ( talk) 01:16, 11 January 2024 (UTC) reply

• Oppose. As I've said countless times the majority of old unsourced articles are valid topics for which sources exist (often in related articles or other language wikis) on uncontroversial topics which would never be outright deleted at AfD. I have spent time going through the unsourced categories in the past and can't recall encountering a single article that was a hoax or completely inaccurate. I would also estimate that a significant fraction of articles marked as unsourced (perhaps as much as a third) actually already have primary sources and/or external links present. I fail to see how deleting valid articles helps us to build an encyclopedia. (Also noting did not receive ping.) Espresso Addict ( talk) 08:05, 9 January 2024 (UTC) reply
• Oppose - We're already struggling with not seeing WP:BEFORE being done at AFD. This just seems like a recipe for chaos. - jc37 05:07, 10 January 2024 (UTC) reply

  Definitely a good point. WP:BEFORE should be explicitly required, and a speedy keep if not. SmokeyJoe ( talk) 01:43, 11 January 2024 (UTC) reply

• Support, and what kind of "editor" would deprod a totally unsourced article anyway?— S Marshall  T/ C 08:02, 10 January 2024 (UTC) reply

  Maybe someone like you, three days ago? And not for the first time, either. Sometimes de-prodding an unsourced article is actually the right thing to do (e.g., to send it to AFD, to redirect it, or to merge it to another article). WhatamIdoing ( talk) 01:24, 11 January 2024 (UTC) reply

  Yes, three days ago I prodded an article, then changed my mind, deprodded, and redirected it instead. As you rightly say, sometimes I advocate redirecting instead of prodding. I rephrase my question.

  What kind of "editor" would deprod a totally unsourced article and try to leave it in mainspace?

  We have a problem, and it's that there is no deadline. That rule made sense in Wikipedia's early days when it was struggling to attract a critical mass of editors and fill out the core articles any encyclopaedia should have. It does not make sense now. Unfortunately there were early editors who sought to "scoop" other editors, rushing to start sometimes very large numbers of articles very quickly, and other times writing content without referring to sources. It's true that a lot of this content is fixable. But fixing it might well take a solid hour of a good editor's time. That's disproportionate when the content was created in a few minutes by a scooper.

  My view is that WP:TIND needs to be rescinded and now, in the 2020s, we need to enable practiced, competent editors to cleanse the Augean stables of the scoopers with less bureaucracy and obstructionism than currently obtains. I think it's a lot easier and more fun to start from scratch than to fix someone else's defective content.— S Marshall  T/ C 09:22, 11 January 2024 (UTC) reply

  Me. I would. If, for example, I found a prod'ed stub in a subject area that I know very well, but I was stuck editing on mobile and away from both my home and office libraries, I might de-prod it to avoid it being deleted unnecessarily and then come back and add references later (or alert a relevant WikiProject about it, etc.). XOR'easter ( talk) 21:34, 12 January 2024 (UTC) reply

• Oppose. Lack of references is a surmountable problem and a long-standing, core policy of Wikipedia is that surmountable problems are an opportunity for collaboration and not an appropriate reason for deletion. The vast majority of unreferenced articles consist of verifiable, uncontroversial content that merely lacks citations, usually because they were created back when we weren't so strict about such things. We can afford to let them be while we slowly work on filling in those references – there's no hurry. –  Joe ( talk) 09:08, 10 January 2024 (UTC) reply
• Support. I don't see a connection between this proposed deletion process and how it would delete valid articles. Keeping an article that has no sources available to verify its content is worse than not having the article at all. This will probably fail, but I trust that anyone using this process would try to look for sources, including administrators who will act upon the tag and users who put the tag. Problematic users who are causing disruption by mass-tagging can be dealt with separately. 0x Deadbeef→∞ ( talk to me) 09:29, 10 January 2024 (UTC) reply

  Could you elaborate on how it's worse than not having an article? If I put myself in the shoes of a reader wondering what Kegnæs was, for example, I would currently find that:

  Kegnæs ( German: Kekenis) is a peninsula on the southern coast of Als in Denmark.

  Which I might take with a grain of salt because there's no citation, but on the other hand there are coordinates which I could look up on a map to verify most of the content, and links to articles on Kegnæs in five other languages, including a 1000-word dawiki article with sources. But if we deleted this unreferenced article, I'd find:

  Wikipedia does not have an article with this exact name. Please search for Kegnæs in Wikipedia to check for alternative titles or spellings.

  ...not much use there. –  Joe ( talk) 10:12, 10 January 2024 (UTC) reply

  This is a very specific example, but broadly speaking unsourced articles have a greater chance of having hoaxes and/or incorrect information that could mislead readers. If someone cannot be sure that the article subject exists at all, then we're doing a bad job. In this case incorporating a source from dawiki or maybe proposing that this deletion process does not apply to articles where subject on other Wikipedias have sources would probably fix it. 0x Deadbeef→∞ ( talk to me) 10:34, 10 January 2024 (UTC) reply

  To add on: We're here to build an encyclopaedia, so our priorities should be around improving the readers' experience. If a reader somehow trusts Wikipedia's validity and does not consider the existence of sources much, (it might be bias that we don't even talk about them, since everyone who's participating here should care about sources) unsourced articles actively harm them. We really need to consider the stakeholders here: Do we care more about readers who want Wikipedia to be reliable or editors who can't be bothered to find even one source to support an article they have created?

  The alternative beneath this proposal isn't really an alternative. The issue we're addressing here isn't about articles where sources can be found, but articles where it isn't the case. Isn't it common sense that any one using the tag would only apply it on articles where it is unclear whether any sources can be found? Why are we picking out examples where sources can easily be found (just that no one really bothered to find sources) as a reason to oppose this deletion procedure? Can't we oppose AfDs altogether if we're to accept that logic? Since there's apparently nothing that prevents someone from opening AfDs without conducting a WP:BEFORE, and nothing that prevents someone from nominating an article where sources are obviously just there, heck, I've even got one myself.

  Some of the Oppose !votes above just translate to !voting Delete at the MfD for our deletion policy for me. There's nothing in the deletion policy that prevents us from deleting an article per the policy that could actually be a valid article, after all. 0x Deadbeef→∞ ( talk to me) 15:30, 10 January 2024 (UTC) reply

  I've encountered a lot of articles where I truly thought that there is no RS avaliable and when I nominated it to AfD/PROD, somebody suggested a few sources that are just enough to cite the article. The alternative I've suggested here is exactly like what is being proposed right now, except that no new policies or processes are needed. Just plain old cite, PROD, or AFD. Again, I want to repeat, the drive does not require editors to completely cite the article. Just one is enough to pull the article from the {{ Unreferenced}} backlog (and push it to another backlog called {{ More citations needed}}, but that's a mess for another time). CactiStaccingCrane ( talk) 15:40, 10 January 2024 (UTC) reply

  But a source that is verifiable is a much lower bar than RS, or even notability.

  Given the size of the unreferenced backlog, it would make sense for us to create a more prioritized backlog, where one editor has thought that sources are not available. Just one source is enough to pull it from the prioritized backlog too, and it would have less articles than 158,034, so we can reasonably work on them. 0x Deadbeef→∞ ( talk to me) 15:44, 10 January 2024 (UTC) reply

  One reliable source. That's it. I'm currently working on January 2024 and considered January 1st 2024 as the day where all articles MUST be sourced to at least one RS. CactiStaccingCrane ( talk) 15:51, 10 January 2024 (UTC) reply

  One week after, we seem to have reduced about 410 articles (it is now 157,024 articles) in a week, and that is given the suggestion here to simply clear the backlog. Even if we're on an optimistic estimate based on this data point (~500 articles per week), it would still take another six years (${\displaystyle 157024({\frac {1}{500}})({\frac {1}{52}})\approx 6}$) to completely clear the backlog, and that is assuming that we don't get new unreferenced articles at all in the backlog.

  I appreciate your work here @ CactiStaccingCrane, and I think a backlog drive can certainly be helpful in reducing the sheer amount we have here, but I just don't think we can realistically address the problem quickly enough to have massive burnout because of the volume of work. 0x Deadbeef→∞ ( talk to me) 12:42, 17 January 2024 (UTC) reply

  @ 0xDeadbeef, what does "quickly enough" mean to you? You seem to be operating on some sort of mental deadline.

  I do think there is value in prioritizing or categorizing the backlog. It could be useful to split it by page views (so the most popular pages are at the top of the list) or WikiProject (so that each group can see only the ones that might interest them, e.g., .gw, 28-bit computing, 2M (DOS), 3D Wayfinder, AMG LASSO, ANI (file format), ANSA pre-processor, ANSI T1.413 Issue 2, AOL Active Virus Shield, etc. for WikiProject Computing but Aeromedical Center, Dübendorf, African League of Associations for Rheumatology, American Anti Drug Council, Antibiotics and Chemotherapy, Antonio Ramon Horta AG7, etc. for WPMED). It's easier for people to tackle the most important or relevant ~20 pages than to see an unsorted list of hundreds or thousands. WhatamIdoing ( talk) 16:06, 17 January 2024 (UTC) reply

  It was perhaps more about the burnout. It's hard for someone to keep doing the same thing for six years, and it's hard to know whether the effort we currently have will remain the same for six years. 0x Deadbeef→∞ ( talk to me) 22:46, 17 January 2024 (UTC) reply

  It's been different people during those six years, so I think that it won't be all the same someones for the next six years. I am personally a bit daunted by the size of the Category:Articles lacking sources from December 2009 (which was mostly populated and dated by a pair of bots), but I think the current system can handle the rest of the months okay-ish, and that it would do better if we added some additional supports (e.g., trickling out small, obviously manageable lists of higher priority articles). WhatamIdoing ( talk) 21:16, 18 January 2024 (UTC) reply

  Isn't it common sense that any one using the tag would only apply it on articles where it is unclear whether any sources can be found? Not really, no. It seems likely to me that some people would take a content area they dislike and start looking through it for unsourced articles to delete, and that some would do the same indiscriminately because they think it gives them wiki-points of some sort. Anomie ⚔ 16:19, 10 January 2024 (UTC) reply

  Damn, some people on Wikipedia are not sane. CactiStaccingCrane ( talk) 16:21, 10 January 2024 (UTC) reply

  Well, that's true, but it's not relevant.

  What is relevant is that some people have different ideas about Wikipedia's purpose and the value it brings to the world. For example, a person who feels strongly about government transparency might think it's highly desirable for Wikipedia to have at least a stub about every government agency, even if all it says is something like "The Whoville Water Agency is the government agency that is responsible for providing water to the city of Whoville." Another person, who feels strongly that only "important" subjects should be covered will see such an article and be unhappy: it's some tiny little agency for some small town! Worse, it says almost nothing, and it cites nothing! Wikipedia:Wikipedia is serious business, and we just can't have this kind of pointless article lying around.

  People who get a bee in their bonnet about these things already buzz around trying to find excuses to delete the articles they dislike. This proposal would just make it easier for them to do that without as the oversight that AFD provides. WhatamIdoing ( talk) 01:36, 11 January 2024 (UTC) reply

• Oppose. The topic of the unsourced article might be notable. WP:NPP flowchart specified that if the topic is notable, even though it is unsourced, it is still "marked as reviewed" though being tagged as unreferenced. Even if notability is borderline, as long as there is useful prose the move is to draftify the article. PROD is seen as the last resort - where there is no useful prose or the article is obviously not notable. This change will change how WP:NPP deals with new unreferenced articles. If the topic is not notable and it's not referenced we can deal with it through WP:AFD. Drive-by PRODs would be a problem, other editors shouldn't be forced to find references. ✠ SunDawn ✠ (contact) 09:54, 10 January 2024 (UTC) reply

  Even if notability is borderline, as long as there is useful prose the move is to draftify the article

  This definitely sounds like using draftification as pseudo deletion.

  You are supposed to be sure that it would be deleted at AfD.

  There’s an old suggestion that all draftifications should reference WP:DRAFTOBJECT in the edit summary. SmokeyJoe ( talk) 01:51, 11 January 2024 (UTC) reply

  I refer to this flowchart. It is clear that AFD is the way to take if notability is not borderline AND PROD is likely to be contested. The flowchart is clear that if I suspect some information is unavailable to me or that notability is borderline, the correct way to take is Draftify them. If the article is obviously notable it didn't even go to deletion route, it will still be marked as reviewed. ✠ SunDawn ✠ (contact) 04:21, 12 January 2024 (UTC) reply

• Oppose per what I've said on this, above, already. jp× g 🗯️ 10:05, 10 January 2024 (UTC) reply
• Support as an extension of WP:BURDEN. I've written extensively on this before, and am too lazy to write new arguments. 🌺 Cremastra ( talk) 12:59, 10 January 2024 (UTC) reply
• Oppose: If this is implemented, what's stopping people from tagging hundreds of articles at once, far outstripping attempts by other editors to add a source? I'm also concerned that this process doesn't even require a search for references beforehand. The backlog of unreferenced articles is slowly but steadily decreasing, with over a thousand articles being cleared a month. There's no rush, and from my experience, most tagged articles are largely accurate, even when unsourced. ARandomName123 ( talk)^{Ping me!} 13:46, 10 January 2024 (UTC) reply
• Alternative solution per below. CactiStaccingCrane ( talk) 13:47, 10 January 2024 (UTC) reply
• Support. If it doesn't summarize sources, it's not worth saving and it's irresponsible to publish. The most important thing is that Wikipedia be accurate and not (as it currently does and always has so far) spread misinformation. An encyclopedia that is "mostly true" isn't good enough, it has to be 100% true. An unsourced article has no value because we can't be sure if it's true. Believing that articles without sources have value is vanity -- an unsourced article is just a blog post that someone wrote at some point. Who cares about saving it? If it is deleted, anyone can recreate it with a source, but the risk of spreading misinformation to the reader while we wait for someone to source an unsourced article is not worth the benefit of continuing to publish someone's blog post from yesteryear. It's irresponsible of Wikipedia to allow unsourced information to be presented to the reader for years. Levivich ( talk) 14:27, 10 January 2024 (UTC) reply

  To add: there is hypocrisy in banning LLMs because they hallucinate sometimes but allowing unsourced articles to persist because they're right most of the time. "Most of the time" ain't good enough for truth. Levivich ( talk) 15:01, 10 January 2024 (UTC) reply

  An unsourced article has no value because we can't be sure if it's true: @ Levivich, are you sure about that? Take a look at User:WhatamIdoing/Christmas candy. It's 100% unsourced. Are you genuinely uncertain about whether the contents are true?

  Earlier this week, I responded to a request for help with some sourced content. Guess what? Half of the sourced content was completely untrue. Little blue clicky numbers don't prevent people from putting garbage in articles. WhatamIdoing ( talk) 01:42, 11 January 2024 (UTC) reply

  I will grant, however, that when you're unfamiliar with the content area, it's usually faster to find out whether the material aligns with an already cited source than to find a new source. WhatamIdoing ( talk) 01:43, 11 January 2024 (UTC) reply

  User:WhatamIdoing. It’s not 100% unsourced. The image is a source, of a kind, but is definitely a source. SmokeyJoe ( talk) 01:47, 11 January 2024 (UTC) reply

  A photo of a candy cane is not normally considered a source of any kind, much less a source for the claim that candy canes are associated with Christmas. WhatamIdoing ( talk) 02:07, 11 January 2024 (UTC) reply

  Normally? It is a source for the existence and appearance of a candy cane. A pure primary source. SmokeyJoe ( talk) 08:17, 11 January 2024 (UTC) reply

  Are you unaware of deepfake candy canes, then? I don't think a photo without any publisher or author attached qualifies as a source. Newimpartial ( talk) 14:06, 11 January 2024 (UTC) reply

  It's a tangent ffs. CactiStaccingCrane ( talk) 14:12, 11 January 2024 (UTC) reply

  @ WAID: the thing about 100,000 articles is that I can't take a look at them to see if I'm uncertain about whether the contents are true. But will say that whenever I look through a Wikipedia article fact-by-fact -- and I mean the sourced ones -- I almost always find at least one thing that's wrong and needs to be fixed. In some cases, hugely wrong. I have no reason to believe the 100,000 unsourced articles are going to be any more accurate than the 6 million sourced ones -- they're as accurate or less. To get a decent sample size of the 100k, we'd need to review thousands of them to see if they're accurate; we haven't and will never do so. Mind, I'd also remove unsourced content from articles, not just PROD unsourced articles themselves. I'd have a bot remove anything that had a {{ cn}} for a year. I'd cull unverified content from the encyclopedia, and it would shrink, and its reputation for reliability would grow, and that would attract new (and better) editors, and then it would grow again, with higher quality than before. *sigh* If I ruled the wiki... Levivich ( talk) 02:45, 11 January 2024 (UTC) reply

  Amen Tooncool64 ( talk) 02:46, 11 January 2024 (UTC) reply

  • We have no evidence that blanking uncited information would cause Wikipedia's reputation for reliability to grow. That's pure speculation and IMO {{ dubious}}. We do have evidence that 299 out of 300 page views do not result in anyone clicking on the refs per doi: 10.1145/3366423.3380300, so that suggests that 299 out of 300 readers don't care about this issue (and the rest mostly care when the article is short and they want more information).
  • We have at least anecdotal evidence that having less information leads to fewer readers and less satisfied ones, which does not seem to be a good way to improve our reputation. "Wikipedia, the website that has a few brilliant jewels but never the thing I want to know" is not IMO an improved reputation.
  • We have good reason to believe that incorrect information is a good way to attract new (and better) editors. See Cunningham's Law and https://xkcd.com/386/. Also, the contributions of every good editor who began editing because something was wrong on wiki. I mean, you didn't start editing because the articles were so perfect, right? Neither did I.

  WhatamIdoing ( talk) 03:44, 11 January 2024 (UTC) reply

• Oppose - we want articles to be written, there is no deadline, and this is a lazy substitute for the content checks normally expected before deletion. Deletion discussions about articles lacking sources very often result in sources being found and vigorously reviewed, and the article improved; this proposed new process would eliminate that safeguard and certainly would result in good information on notable topics being discarded. AfD is a better process for this. Ivanvector ( ^Talk/ _Edits) 14:56, 10 January 2024 (UTC) reply

  Why are we not calling the creators of an article lazy for not finding the sources themselves? Why should we apply a double standard when it comes to challenging unverified material? This is equivalent to "if you see an unsourced section, take it to the talk page or the WikiProject first instead of removing since sources can be found and people who remove those sections are just lazy". If there is a reason that removing unsourced sections and deleting unsourced articles are different, it must be WikiPolitics i.e. Inclusionism vs. Deletionism, right? 0x Deadbeef→∞ ( talk to me) 15:10, 10 January 2024 (UTC) reply

  If I blank uncited material (which is not the same as unverifiable), then anyone can check my work, see if I made a mistake, and even revert me and add sources. If you delete the article, the average editor is unable to do any of that. WhatamIdoing ( talk) 01:47, 11 January 2024 (UTC) reply

  You would be hard pressed to demonstrate to me that out of the 114,000 unreferenced articles, by deleting them, we would be losing anything substantial. I do agree that there are some large articles on that list, but the overwhelming majority of them would be of no loss to Wikipedia. Tooncool64 ( talk) 02:30, 11 January 2024 (UTC) reply

  I've been reading some older (1920s through 1940s) books recently, and I've looked up some unfamiliar things on Wikipedia. Most of the things I'm looking up are pop culture from the time. I never checked for the presence of absence of refs. I rarely read past the first paragraph. Mostly I just wanted to know things like whether the car that was mentioned was supposed to be a normal one or a luxury model, or what that name means in American English, or what kind of fabric just got mentioned. When the subject wasn't covered, then I was sorry not to find the information I wanted here. If the existing articles were deleted, I would feel like I had lost something.

  This is apparently a common pattern for Wikipedia's users. Editors get all excited about long-form FAs that take half an hour to read, but most of our readers spend significantly less than one minute on the page. "Large articles" aren't more important than tiny ones, and even if we write a large article, people mostly don't read much of it. What matters to the reader is that we have the piece of information that they actually want. It's great that someone wrote five thousand words about Silk, but that's not helpful if the thing you want to know about is Lisle (fabric). WhatamIdoing ( talk) 06:40, 11 January 2024 (UTC) reply

  No longer a red link. With sources too. CactiStaccingCrane ( talk) 06:49, 11 January 2024 (UTC) reply

  Here's my first article – completely unreferenced. It's like that because it was one of my first edits, I didn't know that sources were expected (indeed, they weren't really, in 2005), I didn't know how to cite sources (I was still in school), and I'd read many other articles without sources. If somebody had immediately slapped template on it saying it was going to be deleted unless I did extra work, and called me lazy for not doing so without being asked, I'm quite sure that would have been the end of my time on Wikipedia. Instead, it was left alone and eight years later someone added the first citation. In the mean time, I was encouraged enough to go onto become an editor who knew how to cite sources, so the project gained both an article and editor, and precisely nobody was worried about the lack of references on the plot summary of a mediocre novel assigned to high school students in England. –  Joe ( talk) 08:00, 11 January 2024 (UTC) reply

• Oppose as a long-standing member of Wikipedia:WikiProject Unreferenced articles. We have heard proposals like this for years, with variations on how long the deletion waiting period will be, draftifying all unreferenced articles after 6 months, etc. I believe the most recent one was in October 2023, but there are discussions about this type of proposal on the URA talk page going back to 2020. In my opinion, these proposals would only lead to a huge swath of articles that are in fact notable and worth keeping getting deleted due to insufficient attention and a finite amount of editor hours and resources. Furthermore, there is a serious issue with recency bias and topics that more obscure when it comes to what gets referenced and what doesn't; this would lead to a significant skew in the subjects/types of articles that are more likely to be deleted under proposals such as these. If we lost these articles, I'm not so sure we would get them back.
  The way to get the number of unreferenced articles down is to do the frustrating and often lengthy work of actually finding citations for them, or putting them through deletion procedures if that is not possible due to lack of notability/reliable sources. What the URA needs is editors doing regular work on this maintenance category, or big, one-time events like a referencing drive that will clear out a big chunk of articles and make the backlog more manageable.
  Speaking of this backlog - the statement 'there are over 114,000 articles on Wikipedia that contain no citations or sources' lacks pertinent context, which is this: on 4 November 2022, the number of unreferenced articles was 135,240. In just over a year, that is a decrease of over 21,000 articles. We are getting through this backlog. It is happening, but there are no shortcuts to this work. Please help us by doing this work of finding and adding sources rather than trying to find a deletion policy that will, in my opinion, harm the project more than it will help. Thank you, Kazamzam ( talk) 16:47, 10 January 2024 (UTC) reply
• Support Any editor can remove unsourced content from any article, but, anomalously, they cannot remove an article that is entirely unreferenced. This inconsistent state of affairs can be rectified by the proposal. ( t · c) buidhe 01:06, 11 January 2024 (UTC) reply
• Oppose, not yet, due to too little evidence. Whether the delay is seven days or sixty days, with a low ratio of active editors watching articles to article count, this is a pseudoCSD that doesn’t pass WP:NEWCSD. Instead: WP:PROD unsourced articles. If the PROD is removed, AfD it. Repeat a couple of dozen times. Bring the list of AfDs back and show that they resulted in “Delete” everytime. — SmokeyJoe ( talk) 01:41, 11 January 2024 (UTC) reply
• Procedural close and Oppose. Procedurally, the discussion prior to opening this RfC was only open for 3 days, and this same proposal was rejected three months ago. While consensus can change, this should've been discussed much more before this RfC and the proposal should've been workshopped. On substantive grounds, I also oppose. If an article topic is notable, it should be improved, not deleted. If it's a new article, it can (and usually will be) draftified by NPP. I also have some issues with the proposal as framed. Why one month? Seems like quite a long time, particularly when a regular PROD/AfD usually last one week. Why one source? N requires multiple sources, so why should one source defeat a PROD for lacking adequate referencing? Finally, I have the same concerns as other editors who note that this would result in the deletion of a lot of old articles that were created when PAGs were looser. I would only be able to support a proposal that has some sort of restriction, like the not older than 90-day rule for draftify-ing. voorts ( talk/ contributions) 01:55, 11 January 2024 (UTC) reply

  one source defeat a PROD for lacking adequate referencing - maybe because PRODs are supposed to be uncontroversial? When an article has a source evaluating its notability becomes less uncontroversial. 0x Deadbeef→∞ ( talk to me) 02:01, 11 January 2024 (UTC) reply

  The notability guidelines all require multiple independent sources, not just one source, so I would expect that an UNREFED-PROD would require at least two references. I probably should have stated that more clearly instead of in the form of a rhetorical question. voorts ( talk/ contributions) 01:29, 13 January 2024 (UTC) reply

  @ Voorts, that's not actually true. First, WP:N only requires that sources exist in the real world (e.g., at your library or on a website); it does not, and never has, required that any source be cited in the article itself (e.g., for the convenience of New Page Patrollers, who would very much appreciate it). Second, I think you've forgotten about WP:NPROF, which still does not require any independent sources whatsoever – not merely that they not be cited, but that it is unimportant to them whether any independent sources have ever existed. WhatamIdoing ( talk) 02:06, 13 January 2024 (UTC) reply

  I agree that N only requires that sources exist, not that they be cited. My point about requiring two sources is in response to the logic of how this proposal would work. In particular, as I take it, one of the arguments being advanced in favor of this proposal is that it would result in the easier deletion of non-notable articles that are unreferenced. However, as proposed, only one citation would be needed to defeat the PROD. In my view, just one citation does not make sense by the logic of the argument that this would result in the deletion of non-notable articles, because under almost all notability guidelines (with the exception of NPROF, as you point out), multiple (i.e., more than one) sources must exist to establish notability. Thus, adding one source to defeat the PROD would not plausibly indicate notability, undermining that argument in favor of this proposal. voorts ( talk/ contributions) 02:19, 13 January 2024 (UTC) reply

  Also, this proposal wouldn't require independent sources, which is another way that it would fail to fully demonstrate notability. It might push people to stop doing their current activities (which is maybe bad) and rush to empty the unref cats (which is desirable), but I think that the real goal here is just to eliminate the visible backlog, not to properly demonstrate notability. WhatamIdoing ( talk) 02:27, 13 January 2024 (UTC) reply

• Oppose. I'm getting the sense that people are becoming WP:IMPATIENT with having to deal with a perpetual backlog despite the fact that there is no deadline. Deleting articles like this would only frustrate efforts to find and improve these articles, particularly when they're doing no harm just sitting there while saying only the obvious ("Town is a place in State"). - BRAINULATOR9 ( TALK) 02:00, 11 January 2024 (UTC) reply

  How is "Town is a place in State" any obvious? 0x Deadbeef→∞ ( talk to me) 02:15, 11 January 2024 (UTC) reply

  The longer and more detailed the article is, the more likelihood of getting something wrong. When the article says nothing more than "Town is a place in State", a quick trip to your favorite online map can determine whether it's correct. If, on the other hand, it said something like "best known for the mayor killing his wife in 1892" or "best known for the school being built out of locally mined basalt", it's not always obvious how to verify that content. WhatamIdoing ( talk) 02:29, 11 January 2024 (UTC) reply

  To be fair, 113,000 articles, or about 1.8~% of all articles on the English Wikipedia are tagged as unsourced, some in a better state than others. If one of the five pillars of Wikipedia is reliability, it is insane to me to think that we are allowing all of these unverifiable subjects on the wiki. Tooncool64 ( talk) 02:28, 11 January 2024 (UTC) reply

  They're not "unverifiable", just currently unverified. There's a big difference. Toughpigs ( talk) 02:31, 11 January 2024 (UTC) reply

  Something cannot be unverified and not unverifiable, or when these articles were created, sources would have been attached, or at least references. Tooncool64 ( talk) 02:35, 11 January 2024 (UTC) reply

  That's an odd thing to say. It is possible for an article to have information that is currently not sourced, and then you can go and find a citation for that information. We do it all the time. Toughpigs ( talk) 02:38, 11 January 2024 (UTC) reply

  My point is that it is a silly distinction to make. Verifiability means sources, and sources means notability, per Wikipedia:Notability and Wikipedia:Verifiability. Tooncool64 ( talk) 02:40, 11 January 2024 (UTC) reply

  @ Tooncool64, they're tagged because they're uncited, not because they're unverifiable. We require all material to be "verifiable", which means "able to verify", not "already verified". Note "able to verify" does not have restrictive clauses like "able to verify through the sole and exact method of clicking on a little blue clicky number to see what source someone else claimed contained this information". If you can verify that candy canes are a type of Christmas candy by asking your favorite web search engine about it, then it's verifiABLE even if the claim is not cited. WhatamIdoing ( talk) 02:32, 11 January 2024 (UTC) reply

  If these articles were as "verifiable" as you say, then they should have sources, should they not? Unless of course every article states "The sky is blue". Tooncool64 ( talk) 02:33, 11 January 2024 (UTC) reply

  No. @ Tooncool64, please go read the definitions for those words: cited, uncited, verifiable, and unverifiable. They are not synonyms. WhatamIdoing ( talk) 02:38, 11 January 2024 (UTC) reply

  Verifiability means sources, and sources mean notability per WP:N and WP:V. Tooncool64 ( talk) 02:41, 11 January 2024 (UTC) reply

  Sure, but it means sources that exist in the real world, not sources that are cited in the Wikipedia article. I know it's common for editors to cite shortcuts without actually reading them, but there are big, bold-faced words in WP:N that say Notability is based on the existence of suitable sources, not on the state of sourcing in an article. Verifiability does not mean sources have already been cited. It means editors would be able to cite reliable sources. WhatamIdoing ( talk) 03:00, 11 January 2024 (UTC) reply

• Oppose In my experience the gap between the probability that referenced vs unreferenced articles are actually accurate is fairly small. The same goes for cn-tagged sentences. Johnbod ( talk) 02:43, 11 January 2024 (UTC) reply
• Oppose Though this is a legitimate issue, mass-deletion is almost never a good solution. That is incongruous with WP:BEFORE, WP:NEXIST, and WP:DINC. That the backlog is shrinking at a substantial rate is proof that the current modus operandi is working. Curbon7 ( talk) 09:48, 11 January 2024 (UTC) reply
• Support — Fourthords | =Λ= | 14:05, 11 January 2024 (UTC) reply
• Weak oppose. While we should either find a source for unreferenced content or just remove it, I agree with numerous commenters that mass-deletion would contravene WP:NEXIST. Furthermore, deleting a wholly unreferenced article is not exactly the same as deleting an unreferenced section from a page that has references elsewhere. In the latter situation, the deleted content is still visible in the page's history should a user choose to find sources for the content that was challenged. Conversely, deleting an unreferenced article makes the content inaccessible to the vast majority of editors, meaning that anyone seeking to find and add sources would have to either request a WP:REFUND or rewrite the article from scratch. – Epicgenius ( talk) 17:31, 11 January 2024 (UTC) reply
• Support per WP:5P2. and WP:V. Let'srun ( talk) 21:44, 11 January 2024 (UTC) reply
• Oppose this is special pleading. There's no practical difference in usefulness between an unsourced article, an article only sourced to an unreliable source, an long article with one tangential fact source etc. Moreover, the existing PROD system works. See the page history for User:DumbBOT/ProdSummary from 7+ days back and look at all the redlinks. Mach61 ( talk) 02:59, 12 January 2024 (UTC) reply
• Oppose as another longstanding (admittedly not very active) member of Wikipedia:WikiProject Unreferenced articles. If the current mountain of unsourced articles looks hopeless, consider that back in 2010, there were—hold onto your hats—280,000 of them. Worse, new unreferenced articles were being created at nearly the same pace the old ones were being fixed. Today, there are around 114,000 unreferenced articles left, and we are making steady progress as ever clearing out the backlog. I fail to see good reason for short-circuiting the process now. Altamel ( talk) 05:01, 12 January 2024 (UTC) reply

  I find the "steady progress" argument convinces me to support this even more, not less. The fact that the backlog went from 280k to 114k in fourteen years is, to me, proof that Wikipedia does not have the resources to clear this backlog manually, despite the awesome efforts of a number of dedicated volunteers over the years. The way I see it, I'd like to give the "PROD tool" to the people going through the backlog so that they have a chance of actually clearing it -- having to either source articles or AFD them has proven to be unsustainable ... after all, we've only gotten through half the backlog in 14 years. Ironically, the people who I want to give this tool to don't seem to want it. Levivich ( talk) 04:06, 14 January 2024 (UTC) reply

  I think your frame scale is off. Yes, it is correct that it went from 280,000 to 114,000 in 14 years, but it is also correct that it went from 135,000 to 114,000 in this past year. Keeping this pace (21,000 per year), the backlog would be fully cleared in just over 5 years. Curbon7 ( talk) 08:24, 14 January 2024 (UTC) reply

  Yeah that makes me support it more. Over five years, if Wikipedia keeps the current pace (which is the fastest pace in history I believe). And then Wikipedia will have to keep up some pace forever to stay ahead of it. No thanks, these unreferenced articles aren't worth all that effort. Just have a rule that in order to be published on Wikipedia, a statement must be cited. Then everyone working on referencing other people's unreferenced blog posts can go work on something else. Levivich ( talk) 14:37, 14 January 2024 (UTC) reply

  Why is it more important to include a blue clicky number (which may or may not support, or even bear any relation to, the article content) than WP:V? Where is your evidence that even a significant minority of unreferenced articles are not verifiable? Where is your evidence that "blog posts" are a significant proportion of unreferenced articles? Thryduulf ( talk) 14:53, 14 January 2024 (UTC) reply

  I think that the gist of the proposal is very good but IMO a new PROD process is just not the right tool for the job and is addressing the wrong problem. This proposal hinges on the fact that an editor can remove a normal PROD by asserting "this topic might have a reliable source", but in reality such a removal would be quickly reverted, plus the normal PROD process has a shorter timeframe for closing (7 days) than the new UNCITEDPROD is proposed here (30 days), so arguably the normal PROD process is more effective than what is proposed here. Most of my PROD in uncited has been people suggesting RS or source finding strategies or suggest putting the article in AfD for further discussion. What we need now is not another deletion process but a way to channel editor's efforts towards slaying the gigantic backlog. CactiStaccingCrane ( talk) 08:35, 14 January 2024 (UTC) reply

• Support. The unreferenced article backlog is growing and this may focus minds on the need to add references rather than spew out new content. Stifle ( talk) 09:23, 12 January 2024 (UTC) reply

  The backlog is not growing; as Kazamzam states above, it has decreased by 21,000 in the past year. Curbon7 ( talk) 09:47, 12 January 2024 (UTC) reply

  @ Stifle, what makes you think that this backlog is growing? WhatamIdoing ( talk) 16:57, 12 January 2024 (UTC) reply

• Oppose. The proviso in point #3 is doing a lot of work here. Regardless of good intent, I just don't trust people to be sensible with this 100% of the time (even if they always had noble intentions); and looking through the Wikipedia:WikiProject Unreferenced articles backlog there's definitely quite a few that shouldn't be deleted. Yes, there's a lot of geography/music/person stubs but there's also a decent amount on topics of reasonable import. ― novov (t c) 09:54, 12 January 2024 (UTC) reply
• Support - Unreferenced articles are bad. Many of them can be straight-deleted as WP:NOT, WP:V, or WP:N fails. The fact that this may be similar to a perennial proposal does not say anything really about the merits of it. A PROD is still just a PROD and can still be removed by literally anyone for any reason. FOARP ( talk) 10:08, 12 January 2024 (UTC) reply

  @ FOARP, this particular proposal is to prohibit the removal of this particular PROD "by literally anyone for any reason". The only acceptable reason for removing this PROD would be that you already added sources to the article. WhatamIdoing ( talk) 17:00, 12 January 2024 (UTC) reply

• Oppose per Curbon7, Altamel, and Kazamzam, as well as a general distate for creating yet another procedure with yet another track and yet another pile of acronyms. XOR'easter ( talk) 15:53, 12 January 2024 (UTC) reply
• Support. I would expand this to add PROD to anything which solely sources a database. InvadingInvader ( userpage, talk) 18:46, 12 January 2024 (UTC) reply

  Alt proposal: Codify an instant right to send uncited or single-source articles to draftspace. InvadingInvader ( userpage, talk) 18:50, 12 January 2024 (UTC) reply

  Why single-source articles? WhatamIdoing ( talk) 18:53, 12 January 2024 (UTC) reply

  @ InvadingInvader how does an article being in draftspace or mainspace bear any relation to whether the content is verifiable? How does moving something to draftspace increase the chances of citations being added? Thryduulf ( talk) 14:55, 14 January 2024 (UTC) reply

  What's wrong with databases? Does the format somehow make them automatically unreliable? XOR'easter ( talk) 19:12, 12 January 2024 (UTC) reply

  Oh, editors who say that don't actually mean databases. They mean cheap websites that contain a few statistics and no (or very few) sentences written by actual humans. No editor has ever said that https://omim.org/entry/609423#text is bad merely because they use a database to store the content. They generally also assume that it's difficult to write more than a couple of sentences from a database, though User:WhatamIdoing/Database article proves them wrong on that point (at least for some database records). WhatamIdoing ( talk) 19:35, 12 January 2024 (UTC) reply

  Well, if that's what they mean, then that's what they should say! Otherwise, we'll end up with guidelines full of statements that we don't actually mean, which would be a sorry state of affairs indeed. XOR'easter ( talk) 21:30, 12 January 2024 (UTC) reply

  Such a proposal would certainly have to define what sorts of databases were meant, and so on. Almost every modern website is technically a database, even the simplest WordPress ones.  —  SMcCandlish ☏ ¢ 😼  22:52, 12 January 2024 (UTC) reply

  Personally I'm specifically targeting the LUGSTUB articles, which only reference Olympic statistic websites and no evidence of passing GNG. InvadingInvader ( userpage, talk) 00:29, 13 January 2024 (UTC) reply

  Then say so in the first place rather than assuming everyone reading it is familiar enough with this years-long vendetta against a past-user to pick up on the dogwhistle. Gnomingstuff ( talk) 23:29, 27 January 2024 (UTC) reply

• Support an alternative/modified version: This "super-PROD" could not be applied without first complying with WP:BEFORE. I think that would solve every issue. E.g., it could not be abused to delete obviously sourceable/notable topics just because no one had gotten around to injecting the citations yet, it wouldn't be possible to mass-tag a zillion articles, it would be consonant with WP:BURDEN, it would mirror AfD requirements and comply with WP:ATD and WP:NEXIST just as AfD does, it would enable us to gradually get rid of junk pages. It would simply be a way of getting rid of something that should be AfDed without wasting multiple editors' time at AfD. The "editors stripping citations from articles that have references ... then PRODding them under the new criterion" would also basically be invalidated, though it is already invalid under WP:FAITACCOMPLI and other provisions anyway (which is why it's not happening all the time already).  —  SMcCandlish ☏ ¢ 😼  23:01, 12 January 2024 (UTC) reply

  What consequence for someone who skips the WP:BEFORE part? WP:UfD? 😀 Anomie ⚔ 23:10, 12 January 2024 (UTC) reply

  We already see people bringing pages to AfD without having done a cursory WP:BEFORE. Why wouldn't the same thing happen here with YAPDM (Yet Another Page Deletion Mechanism)? XOR'easter ( talk) 16:41, 13 January 2024 (UTC) reply

  I am sure that we would see, if this proposal is passed, exactly the same sanction for not following WP:BEFORE as we currently get at AfD and PROD. We would also get serial offenders claiming that they looked for sources when they either didn't or were grossly incompetent at it, and being believed. Phil Bridger ( talk) 17:41, 13 January 2024 (UTC) reply

  Neither normal PROD nor AfD have the "must be deleted if no source is added" requirement that this proposal does, though. Since SMcCandlish suggested that WP:BEFORE be a requirement before this new super-PROD gets added, I was wondering what would happen if someone fails to do WP:BEFORE properly. Anomie ⚔ 20:47, 13 January 2024 (UTC) reply

  I believe SMcCandlish stated that a failure to do a BEFORE would result in the nomination being speedily kept. However I don't recall any mention of what would happen in the case of an inadequate BEFORE (just because the fist page of Google results doesn't turn up in-depth sources about a 19th century railway station in India or a Soviet cross-country skier active in the 1970s is not evidence that those sources do not exist), nor whether there would be any comeback for the editor concerned. It also doesn't address the BURDEN being placed (rightly or wrongly) on other editors to determine whether the BEFORE was performed or not (which takes a lot more time and effort than nominating an article without doing a BEFORE does).

  Personally, I think an appropriate response to a missing or inadequate BEFORE would be to speedy keep the article, prohibit renomination of that article for a period of time (perhaps 2 weeks or a month, excluding newly discovered copyright violations) and automatically restrict the nominator to a maximum of one deletion nomination per 24 hours until they demonstrate they can consistently do an adequate BEFORE (with any nomination in excess of this limit being speedily kept with the same moratorium on renomination). Thryduulf ( talk) 15:07, 14 January 2024 (UTC) reply

  Honestly, I never have "enforcement" in mind (one of the reasons I never ran for AfD in the last 14 years). It would be entirely up to the community to decide what to do for people who abuse the process without any evidence of doing BEFORE (like if someone else can cough up appropriate sources in a minute of effort just with an Internet search? I dunno). How to specify and determine something like that sounds like a big discussion, and to be applied first to AfD itself. One idea might be a very narrow T-ban, from nominating stuff for deletion again for X amount of time. But this is really not my forte.  —  SMcCandlish ☏ ¢ 😼  02:18, 18 January 2024 (UTC) reply

• Oppose good faith nom but I don't think this solves the issue presented (actually think it creates completely new issues instead). As mentioned multiple times above, many if not most unsourced articles pass WP:GNG and are inherently valuable. The issue of hoax articles is already mostly handled by the existence of WP:PROD imo. If someone thinks an unsourced article has no potential of having sources/is blatantly wrong/is a hoax, they could just PROD it yk. Sure, it's slower and oftentimes there'll be an editor who'll remove the tag from the article arguing that there's sources. Personally I'd be in favor of amending WP:BURDEN to include a clause saying "Editors who remove a proposed deletion tag must provide sources within the article that demonstrate verifiability" (also amending WP:PROD to reflect this as well). If sources are not provided, the PROD tag is reinstated and stays until suitable sources are provided. Rather than allow unsourced articles to be deleted en-masse, let them be challenged with PRODs as is typical, and the editors who contest those PRODs (who also tend to have greater familiarity with the articles' subjects) address the issue of the sourcing, thereby avoiding losing the valuable encyclopedic coverage and at the same time improving the sourcing. This approach is fundamentally in the best interest of Wikipedia imo. Cheers, Dan the Animator 02:22, 13 January 2024 (UTC) reply

  Please use {{ db-hoax}} if you run across an actual hoax article. WhatamIdoing ( talk) 02:39, 13 January 2024 (UTC) reply

  Yup, thanks WhatamIdoing! :) Also for the other WP:CSD criteria too. Dan the Animator 04:55, 13 January 2024 (UTC) reply

• Oppose, yet again, and for a second time per my comments above. Deleting articles does not improve them. It "improves the ratio" of bad articles to good articles -- but why should anybody give a rat's ass about the "ratio of bad articles to good articles"? This is a great example of Goodhart's Law -- the measurement of success is not the same thing as success. Similarly, we would not consider "dispatching hitmen to the cancer ward" to be an effective method of "curing cancer", even if it did improve the ratio of cancer patients to the rest of the population. jp× g 🗯️ 03:10, 13 January 2024 (UTC) reply
• Oppose. Category:Articles lacking sources is not a "garbage dump". There are lots of actually pretty good articles in there. For example, Focusing screen and Genetic analysis. These are useful articles. They just don't have any sources yet. Nosferattus ( talk) 18:09, 13 January 2024 (UTC) reply
• Support Wikipedia has minimal quality standards. Having a routine non-debatable expectation of a minimum of one citation is reasonable.
  The citation of sources is the foundation of our fact-checking process. Articles with zero citations are also articles with zero documentation of fact-checking. I support raising our expectations from zero evidence of fact-checking to one citation. I even support going much further, and eventually planning to have citations for nearly every sentence in every article, but this is a great step for our current position. Bluerasberry (talk) 21:51, 13 January 2024 (UTC) reply
• Oppose per every time similar proposals have been rejected previously. The effort required to nominate an article for deletion is already very significantly less than that required to evaluate whether sources exist and add them to the article. Unless and until WP:BEFORE is mandatory and includes a requirement for the nominator to look for sources in the place they are most likely to exist (which is far from always places Google knows about, even if asked in the local language) it is completely contrary to the project's fundemntal goals to make deletion even easier. Thryduulf ( talk) 10:21, 14 January 2024 (UTC) reply
• Oppose Honestly this should've been closed early as relitigating a discussion from 3 months ago, but there are many useful unreferenced articles, and I don't see any reason to shortcut the work of going through them and either referencing them or deleting them as needed. Galobtter ( talk) 22:27, 14 January 2024 (UTC) reply
• Oppose The assumption behind the proposal seems to be that going through AFD takes too much time. However, users have scripts to tag and list articles for AFD relatively fast, and the deletion process is straightforward. Improving and adding even just one ref to an article is a much more time-consuming process. -- PeaceNT ( talk) 05:01, 16 January 2024 (UTC) reply
• Oppose I have a bold idea. Why not we try to improve these articles. Implementing this would mean decreasing the overall quality of the site. NW1223< Howl at me• My hunts> 00:32, 17 January 2024 (UTC) reply

  @@ NightWolf1223, I tend to agree with you, but from the other POV, I believe the feeling is that having only 100,000 excellent articles is higher overall quality than 100,000 excellent articles plus 6,000,000 not-so-excellent articles. They're not wrong about the average ( mean) being higher in such a case, but I personally think that the value is in the total ( sum) rather than the average. WhatamIdoing ( talk) 00:52, 17 January 2024 (UTC) reply

• Oppose Either you a) leave the page alone or b) do the work and research sources -> use normal PROD if you are sure that no one will be able to find sources, and normal AFD if you are pretty sure that no one will be able to find sources. If there are questionable unsourced statements, remove those. Lazily slapping deletion tags on Category:Unsourced articles en masse dosen't improve the encyclopedia. -- TheImaCow ( talk) 16:10, 17 January 2024 (UTC) reply
• Oppose I remain unconvinced that the existing PROD process is broken. And if it ain't broke, ya don't fix it. CaptainEek ^{Edits Ho Cap'n!} ⚓ 23:14, 18 January 2024 (UTC) reply
• Oppose without very strong enforcement of a requirement that no editor may tag such articles without performing a proper WP:BEFORE search for sources themself, and adding sources in preference to prodding. — David Eppstein ( talk) 02:31, 22 January 2024 (UTC) reply
• Oppose for essentially the same reasons I opposed the proposal three months ago. Lots of the unreferenced articles in the backlog are both useful and not all that hard to find sources for, and the existing PROD/AfD process is sufficient for dealing with the ones that can't be referenced. TheCatalyst31 ^{Reaction• Creation} 05:27, 22 January 2024 (UTC) reply
• Support. I don't have faith this will pass but adding my voice here regardless. I agree with Pppery here. We're basically allowing old unsourced articles to be grandfathered in. These articles, might seem helpful, but if they aren't sourced, they really aren't. This might seem harsh, but if one of the deleted articles is really notable, someone will come back and recreate it, and this time with references. I personally don't see any loss here. -- Gonnym ( talk) 17:17, 22 January 2024 (UTC) reply

  @ Gonnym, what does "helpful" mean to you? I'm sure you don't mean that you personally know what's helpful to every single reader under ever single circumstance – such that if I say an uncited article actually was helpful to me, you'd feel like I didn't know what I was talking about, because you know better than I do what's actually helpful to me – but I'm not sure what you do mean. WhatamIdoing ( talk) 20:01, 22 January 2024 (UTC) reply

  I've seen your other comments here so I'm not going to engage with you other than this single reply. If an article has no sources to it, anything written there can be incorrect, falsehoods or plain lies. So even if you wrote it and you are sure of its validity, it means nothing if there is no sources backing it. Sending readers (and editors) to do the job of finding if that is true is just not a valid solution. Articles with no sources (which again, can be incorrect) damage the creditability of the project as a whole. So no information is always better than bad information. And again, if you, or someone else, wants to restore a deleted article with sources no one is stopping you. Gonnym ( talk) 20:45, 22 January 2024 (UTC) reply

  I agree with Gonnym here. And, WhatamIdoing, the things you derisively call "clicky blue numbers" are part of WP:V and one of the foundations of Wikipedia. 🌺 Cremastra ( talk) 20:50, 22 January 2024 (UTC) reply

  WP:V requires inline citations only for four specified types of content:

  • direct quotations,
  • material that has already been challenged (e.g., by the addition of Template:Citation needed or Template:Dubious),
  • material that you personally believe is actually WP:LIKELY to be challenged (e.g., "Cooking over an open wood fire causes lung cancer", not "Red is a color"), and
  • contentious matter about a living person.

  If an article contains none of that type of content (e.g., the article's entire contents are "Italian renaissance sculpture is sculpture from Italy during the Renaissance"), then WP:V does not actually require any inline citations. Editors might agree or disagree about whether this is a desirable state for WP:V, but this is the current state of WP:V. WhatamIdoing ( talk) 21:50, 22 January 2024 (UTC) reply

• Oppose. There is no way to enforce looking for sources as we have seen at AfD. Many of our technical articles came in very early before sources were required, especially for standard textbook information. Textbooks are almost impossible to access online these days, so finding reliable sources could be time-consuming and would be skipped over. Editors looking at backlogs often do not have expertise in the subjects of articles they look at. They neglect to look at the article history, which may have had then-considered reliable sources such as professors' lecture notes. This would lead to flooding project to-do lists with too many prods to handle and, since only some projects track prods, under-the-radar deletions of articles that can currently be adequately handled by existing processes. StarryGrandma ( talk) 21:42, 22 January 2024 (UTC) reply

• Oppose - this is unrealistic. We don't have the editors and admins to deal with a surge of tens of thousands of new PRODS. Proposing deletion is so easy compared to reviewing and salvaging content.

Every day, I try to review most of the articles recently proposed for deletion. I'd say 1/3 are likely notable but I don't have time to dig up references for all of them - this might take on average 30 minutes per article. By contrast, it may take 5 minutes at most to PROD an article: tag the article, tag the talk page and then notify the creator.

Increase the daily list three-fold or ten-fold and editors will be overwhelmed reviewing PRODs. Admins will get overhwhelmed, too; they don't just delete articles but they cleanup red links in related articles, etc.

The result will be a chaotic, mindless purge of content, probably a third of it useful. -- A. B. ^{( talk • contribs • global count)} 01:45, 23 January 2024 (UTC) reply

• Oppose Wait 30 days for a PROD? Just using AFD would be faster than this process and would probably yield better results. On the underlying philosophical issue, I find myself agreeing with jpxg. Deletion doesn't fix sourcing problems, but it does make it harder for people to write about things we know are notable (newbies need to wait for AFC, others need autoconfirmed account to create in main, non-admins need to ask an admin for a refund). Also, using deletion as a threat (add a source or else!) is probably not the way we want to encourage volunteers to continue contributing. — Wug· a·po·des 23:26, 24 January 2024 (UTC) reply
• Oppose as contradicting WP:NEXIST. I'd consider myself quite liberal in WP:TNTing unreferenced material, but I do not think we should be outright deleting unsourced articles, barring any bigger issues. ^Queen of _Hearts 14:45, 26 January 2024 (UTC) reply
• Oppose based on many excellent points above, particularly those raised by Mach61 and A. B. In my own words: references in most cases are irrelevant to user value (although crucially important to our work for other reasons), and further reducing barriers to deletion would only lead to deleting even more encyclopedic content. I would add that these perennial proposals put the cart before the horse: they are motivated by a focus on product over process, but that focus is precisely why we are in this pickle to begin with. The continuing substantial number of unreferenced articles is a result of Wikipedia's largely frozen growth and development, which has left so much of our coverage fossilized in the early 21st century. These unreferenced articles are just some of the more visually apparent fossils. This fossilization is itself the result of exclusionary get-it-right-first approaches that took hold as Wikipedians became more concerned with looking presentable to the world than with getting on with the messy work of building and perfecting a comprehensive encyclopedia. An unreferenced article is, in the vast majority of cases, a suboptimal article -- but a hundred suboptimal articles do less harm to the project than a single missing article. Accordingly, we should avoid shifting our practices in a way that makes deletion of suboptimal articles easier. -- Visviva ( talk) 22:48, 26 January 2024 (UTC) reply
• Oppose. Do we not have enough deletion processes already? -- œ ^™ 00:37, 27 January 2024 (UTC) reply
• Strongest possible oppose. I figured there must have been a reason for there being an unreferenced articles drive. I spend most of my time on Commons these days (in no large part because the way the winds are blowing is beyond demoralizing) but I used to spend quite a bit of time referencing unreferenced articles. These include, among others, unquestionably notable literature by Italo Calvino, Virginia Woolf, Margaret Atwood, Stephen Crane, and Stephen King, where the challenge was not finding sources but pulling together the sheer huge scope of scholarship on them. Any proposal that views indiscriminately nuking articles like these as an acceptable loss has completely lost the plot on what Wikipedia is supposed to be doing. That time could be better spent doing the work and sourcing articles. Gnomingstuff ( talk) 22:34, 27 January 2024 (UTC) reply
• Support per WP:V TarnishedPath ^talk 04:21, 11 February 2024 (UTC) reply

  WP:V requires that content be verifiable in reliable sources. It doesn't require those sources be present on the article. Thryduulf ( talk) 11:47, 11 February 2024 (UTC) reply

Alternative solution: a massive backlog drive

As an editor who are working on Category:Articles lacking sources from January 2024 and a member of Wikipedia:WikiProject Unreferenced articles, I found that just simply PRODing every article that does not have a source is not effective towards addressing the backlog. Like a lot of people above have said and consistent with my experience, most of these articles are notable though sources for it might be hard to find. It would be much more helpful and exciting if everyone here can chip in to the cause! If enough people are participating, we can set up a large backlog drive and kill Category:Articles lacking sources once and for all.

The drive would work in a similar manner to the suggested PROD, except that there would be no need for new hard-coded policies or processes. Just one reliable source attached to the article is enough to move the article out of the backlog. If the source cannot be found in a reasonable timeframe, they can PROD it or AfD it, as it has always been the case. CactiStaccingCrane ( talk) 12:56, 10 January 2024 (UTC) reply

In the long run we might need a process to delete articles more efficiently, but such a proposal should only being made if PROD is overloaded with these requests. CactiStaccingCrane ( talk) 12:57, 10 January 2024 (UTC) reply

Pinging @ Kazamzam to the discussion as a long-time member of the WikiProject, who can explain about this much better than I do. CactiStaccingCrane ( talk) 12:59, 10 January 2024 (UTC) reply

I don't begrudge anyone who wants to do this -- and some people already are and have been for a long time, and that's great -- but I don't really think editors should stop what they're currently doing and do this instead, because whatever they're already working on currently is probably more important. Or to put it another way: if an article has been unsourced for years, it's probably because nobody cares, which is probably because it's one of the least important articles on Wikipedia. Whatever editors are doing instead of sourcing these articles is probably more important, up to and including writing new articles. (There are exceptions of course.) But for example I wouldn't want people to stop patrolling NPP or patrolling vandalism or making new articles (about topics that have more readers), etc., in order to source old unsourced articles. New projects and drives sound great but resources are limited, so it's a matter of balancing priorities and I think prodding these is a better use of time than a backlog drive (although no reason we can't have both). Levivich ( talk) 14:42, 10 January 2024 (UTC) reply

Ultimately it is about balance right? You don't want to spend hours finding sources for an article about a random sport seasons, and at some point the editor must decide that this is enough and PROD/AfD the articles (for me my limit is 10 minutes of searching on Google news/books/scholar). Based on my experience, when you PROD/AfD these articles, it's also a way to request others finding more sources, so it is an additional process to check again whether a topic is truly notable or not.

I truly do think that such a drive, going through every single article and cite at least once source to them, is 100% compatible with the policy proposal that we have here. In fact, you could say that this drive is essentially identical with the proposal at hand here, with much less controversy. Both the drive and the policy proposal's end result would be that all articles on Wikipedia will have at least one reliable source attached and all other articles that cannot be sourced in a reasonable amount time will have been PRODed or AFDed. CactiStaccingCrane ( talk) 15:00, 10 January 2024 (UTC) reply

And to address others who are opposing on the basis of Wikipedia:There is no deadline, no, in this case there is a deadline. The reliability of Wikipedia is at stake here. If we are at least making a centralized effort to deal with this issue then we are effectively saying that Wikipedia tolerates original research. That's hypocrisy. CactiStaccingCrane ( talk) 15:04, 10 January 2024 (UTC) reply

Can you provide a few specific examples of currently unsourced articles that you believe are undermining the public's general sense of Wikipedia's reliability? Toughpigs ( talk) 02:03, 11 January 2024 (UTC) reply

I would like to be a doer rather than a talker. CactiStaccingCrane ( talk) 02:34, 11 January 2024 (UTC) reply

In other words, you are either unable or unwilling to back up your argument. The closer of this discussion will take that into account when determining how much (if any) weight to give it. Thryduulf ( talk) 02:24, 28 January 2024 (UTC) reply

I don't really care about this RfC and it is a drain of everyone's time. CactiStaccingCrane ( talk) 02:29, 28 January 2024 (UTC) reply

@ CactiStaccingCrane, I think you've confused "no citations in the current version of the article" (the complaint here) with "information never published anywhere until a Wikipedia editor posted it on wiki" (=original research). WhatamIdoing ( talk) 02:14, 11 January 2024 (UTC) reply

That's on me. But my point still stands: we cannot procrastinate and kick the tin can down the road to future editors any longer. CactiStaccingCrane ( talk) 14:38, 12 January 2024 (UTC) reply

This argument comes up time and time again, and is self-contradictory every time:

- We must deal with uncited articles now as they are a huge problem and "the reliability of Wikipedia is at stake." If so, then this problem is by default more important than most other things that editors could be doing instead.

- We should not spend time sourcing unreferenced articles, because they're not important and nobody cares or looks at them. If so, then their existence poses little threat to the encyclopedia's reputation.

(Plus, well, there's the fact that if you're reading this then what you are currently doing is doomscrolling through a gargantuan procedural RfP. Almost anything you can do, on Wikipedia or in life, is more important than that.) Gnomingstuff ( talk) 22:48, 27 January 2024 (UTC) reply

The problem with backlog drives is the math. Someone said above that the backlog is declining by about 1K articles per month. Good – but that means that it'll take another ten years to get it down to zero. We ran the numbers during the last big discussion, and it was something like needing 300 (three hundred) editors to commit to adding a source to one unref'd article every single day of the year (all 365, without exception) to get everything handled during the next 12 months. In practice, that means needing ten times that many editors to sign up for this, because anyone could get sick, and most people won't stick with it for a whole month, much less a whole year.

We only have about 10,000 high-volume editors at any given point, and relatively few of them are engaged in sourcing articles. Getting it done "right now" is just not realistic. WhatamIdoing ( talk) 02:20, 11 January 2024 (UTC) reply

Its amazing the Wikipedia ever allowed unsourced articles to be created in the first place, or we wouldn't have to deal with this mess. Tooncool64 ( talk) 02:24, 11 January 2024 (UTC) reply

Check out nostalgia: to see what Wikipedia looked like in the early days. Click on nostalgia:Special:Random to open a few pages in tabs. WhatamIdoing ( talk) 02:36, 11 January 2024 (UTC) reply

I've made a test backlog drive to get some operational experience before launching a much bigger backlog drive (expect >1000 articles): Wikipedia:WikiProject Unreferenced articles/Test backlog drive. CactiStaccingCrane ( talk) 14:40, 12 January 2024 (UTC) reply

A suggestion that sounds like a joke but isn't: get a grant from the WMF to hire people to add references to uncited Wikipedia articles. If volunteer effort isn't making up the gap, make it somebody's day job. XOR'easter ( talk) 19:17, 12 January 2024 (UTC) reply

I doubt that they would fund that. You could perhaps get a grant to fund an event (e.g., rent a meeting room) that would do this kind of work, but AFAIK they don't offer grants for work that would replace efforts by volunteers. They try to stay out of paying people to create content, even for something as relatively uncreative as adding a citation to existing articles.

However, it's possible that some other organizations would be interested in funding such work for a given subset of articles (e.g., "Society to Promote Geography" might be willing to hire a Wikipedia:Wikipedian in Residence to add citations to geography-related articles). WhatamIdoing ( talk) 22:45, 12 January 2024 (UTC) reply

@ XOR'easter, I think it's much better if there is a small monetary effort (i.e. a subscription to a paid journal or $50 amazon gift card) to the best participant. This sort of thing has been done before in the WP:The Core Contest and WP:Reward board. CactiStaccingCrane ( talk) 02:25, 13 January 2024 (UTC) reply

We tried that for improving physics articles and got a grand total of zero takers. XOR'easter ( talk) 20:41, 13 January 2024 (UTC) reply

XOR'easter, is it a part of a drive or is it just an entry in the reward board? CactiStaccingCrane ( talk) 10:49, 14 January 2024 (UTC) reply

If you are interested about the backlog drive, sign up at the newsletter: Wikipedia:WikiProject Unreferenced articles/Mailing list CactiStaccingCrane ( talk) 05:42, 16 January 2024 (UTC) reply

The drive is now online at Wikipedia:WikiProject Unreferenced articles/Backlog drives/February 2024. Please join in the effort and clear out 6 massive backlogs (July–December 2023)! CactiStaccingCrane ( talk) 13:43, 20 January 2024 (UTC) reply

Discussion for proposed PROD

Tagging Horse Eye's Back, Yr Enw, Bagumba, S Marshall, JoelleJay, Let'srun, JPxG, Curbon7, Cremastra, Espresso Addict, Mccapra, Hiding, James500, Lee Vilenski, SMcCandlish, and Joe Roe per previous discussion. Tooncool64 ( talk) 06:57, 8 January 2024 (UTC) reply

@ Tooncool64: I don't understand your fourth point. Anybody can recreate any non-protected page now, without going through WP:REFUND or anywhere else. WP:REFUND is for restoring a page (i.e. recreating it with its former content and history) and there is no way for a non-admin to do that without help. –  Joe ( talk) 07:13, 8 January 2024 (UTC) reply

I see. I was under a different impression and confused about what can and cannot be re-created, will edit that now, thank you. Tooncool64 ( talk) 07:15, 8 January 2024 (UTC) reply

• Can you clarify how this is different from Wikipedia:Village pump (proposals)/Archive 207#Request for comment: Unreferenced PROD? Additionally, is this to apply retroactively or only for new articles? Curbon7 ( talk) 08:58, 8 January 2024 (UTC) reply

  Yes, this would apply to all unsourced articles. No unsourced articles are being created today anyways, they wouldn't pass new page patrol. Tooncool64 ( talk) 17:07, 8 January 2024 (UTC) reply

  @ Tooncool64 - not a comment on anything else above, just a note that plenty of unsourced articles are being created today and get past NPP because they are overwhelmed with volume and lack sufficient editors to keep up. I have PROD'd/draftified a number of these articles myself for the exact reason of being a BLP without sources. Kazamzam ( talk) 16:22, 10 January 2024 (UTC) reply

• No. This proposal was overwhelmingly rejected just three months ago. Let's spend our time looking for sources and using the many existing procedures rather than looking for new ways to delete articles. I note that there is a section for votes in support of this but none for votes against. Hardly neutral. Phil Bridger ( talk) 11:15, 8 January 2024 (UTC) reply

  Votes "for" does not mean "votes in favor of", but "votes regarding". It is not meant for only supporting votes, obviously. Tooncool64 ( talk) 17:05, 8 January 2024 (UTC) reply

  If we get more of these proposals, we might need to put it in WP:PEREN. Or we could set procedural rules, like requiring editors to provide an inline citation in at least 10 of the articles they want to delete before they can propose deleting them. WhatamIdoing ( talk) 00:11, 10 January 2024 (UTC) reply

  BTW, if you all want to look at the articles that this proposal recommends deleting, for your own favorite subject area(s), then I suggest going to https://bambots.brucemyers.com/cwb/index.html and looking down the list for your favorite subjects. There are 779 options at the moment. Click on it (warning: some of them are v-e-r-y big pages) and look for the heading "Cites no sources".

  Once there, consider whether you think Wikipedia would be best served by deleting those articles, having you take time from other things to cite them right now, or maybe just to let them alone for a while. I've found that there's nothing quite so informative as looking through a list of articles for a subject you know well to figure out what the result of this proposal would be. WhatamIdoing ( talk) 00:33, 10 January 2024 (UTC) reply

  It'd also be interesting to mandate that, before you can propose the mass deletion of thousands of articles, you have to try and write at least ten of your own... –  Joe ( talk) 09:27, 10 January 2024 (UTC) reply

  Or to be Wikipedia:Autopatrolled. If you're trying to mass-delete thousands of articles, we should have some confidence that you know what a decent article requires. WhatamIdoing ( talk) 19:32, 12 January 2024 (UTC) reply

• Personally, I don't think the bar for an article existing being a single source is strong enough, but that being said, adding a way to get articles that have never been sourced closer to either getting sourcing, or being removed is a positive in my book. Lee Vilenski ^{( talk • contribs)} 14:10, 8 January 2024 (UTC) reply

  @ Lee Vilenski, I see that Royal Dutch Billiards Federation, Saluc, and The Baltimore Bullet are all tagged as being uncited, and all tagged by WP:CUE. Do you think that Wikipedia would be better off if these were all deleted?

  Do you think that a threat of deletion would help you decide to find and add some sources yourself? WhatamIdoing ( talk) 00:26, 10 January 2024 (UTC) reply

  I do actually, I can't say I'm familiar with any of those articles, but I shall look for some sources, so they aren't part of this issue. I'm a firm believer in Wikipedia being a summation of what reliable sources say. If the article cites nothing, it can't be a summation. Lee Vilenski ^{( talk • contribs)} 07:26, 12 January 2024 (UTC) reply

  Are you sure that If the article cites nothing, it can't be a summation?

  Imagine that I:

  • read a book, and
  • write what I learned from the book in an article, but
  • can't figure out how to cite the book, so I give up.

  Does my addition really stop being a summation of the book, merely because I couldn't figure out how to cite it? This happens fairly often with new editors (especially in the wikitext editor). WhatamIdoing ( talk) 18:52, 12 January 2024 (UTC) reply

  To you, it's a summation of sources, but to anyone else reading there's no information as to what is being summarised from. We aren't really talking about new editors with the above, as newly unsourced articles rarely stay in mainspace for long. These are generally ones from quite a while ago. Lee Vilenski ^{( talk • contribs)} 20:53, 12 January 2024 (UTC) reply

  Even if you don't know what book it's from, it still actually is a summation of the book.

  Adding inline citations helps other editors find out whether the cited source contains that information, which is certainly valuable, but it doesn't change the nature of the content itself. It's not like if you type two sentences from a source and cite that source, then it's a summary of the source, but if you type two sentences from a source and don't cite that source, then you made it all up out of your head. Citing the source may demonstrate that it's from a source, but it's actually from that source even if you don't WP:PROVEIT for some reason. WhatamIdoing ( talk) 21:10, 12 January 2024 (UTC) reply

  • For what it's worth, most of the things that come to AFD for having no sources are not the result of people reading books. Oftentimes, reading books turns up very different information, and it appears that the original author is just putting vague personal knowledge of the subject into the wiki. This hypothetical side-discussion is rather a red herring. Uncle G ( talk) 03:07, 14 January 2024 (UTC) reply

  These three are now reasonably well cited. Not even really enough to satisfy GNG, but I think this should have been the baselimit for an article. Lee Vilenski ^{( talk • contribs)} 12:35, 12 January 2024 (UTC) reply

• Added to WP:CENT. Curbon7 ( talk) 04:50, 10 January 2024 (UTC) reply
• See also Wikipedia:Village pump (policy)#Utterly badly sourced business articles where I suggested several possible courses of action for dealing with one specific class of unreferenced articles (articles for businesses that are basically a vehicle for a clone of the business's own blurb and an external hyperlink to its WWW site), ranging from changing policy to using Proposed Deletion more, and asked for discussion, and some people foolishly tried to turn it into a vote. Uncle G ( talk) 03:07, 14 January 2024 (UTC) reply