The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.

---

Regarding the section "Settlements and administrative regions" of the notability guide for geographical features (WP:NGEO), which of the following should we do?:

• Option 1: Status quo. Leave it as it is
• Option 2: Replace the "Settlements and administrative regions" section with:

Settlements and administrative regions are typically presumed notable, provided information about them beyond statistics, region, and coordinates is known to exist. The extent of coverage in secondary sources should be considered to ensure there is enough verifiable content for an encyclopedic article. Editors should avoid creating very short articles on these topics, e.g., "Village X is in County Y, Province Z of Country A, at coordinates C, with a population of P." is not enough content for a stand-alone article.

If a Wikipedia article on a settlement or administrative region cannot be developed using known sources, information on it should instead be included in a more general article on a higher-level administrative region, or in a list article collecting information on similar settlements or regions (per the list notability criteria and common list selection criteria #2).

This criterion applies to settlements and administrative regions including states, provinces, municipalities, districts, parishes, counties, cities, villages, towns, and their equivalents, that provide general-purpose administration for their region, regardless of whether they presently exist or are abandoned, even if their population is low. It does not apply to special-purpose districts such as census tracts, irrigation districts, library districts, etc. which should instead be assessed on a case-by-case basis according to the general notability guideline.

This change to be made in combination with the amonishment that editors should avoid flooding AFD with articles (e.g., more than 10 nominations per day as with minor planets) that fail the amended guide. FOARP (talk) 16:22, 15 September 2023 (UTC)

Background

The proposed amendment is the result of a discussion on the NGEO talk-page involving the editors @Davidstewartharvey, Bookku, Certes, North8000, Horse Eye's Back, Selfstudier, Crouch, Swale, Visviva, BilledMammal, Hut 8.5, Bkonrad, Newimpartial, Hike395, Donald Albury, Viridiscalculus, Mxn, Avilich, Dlthewave, Choess, Joe Roe, Ezhiki, Rschen7754, Atlantic306, XOR'easter, and Imzadi1979:. This change was prompted by discussions involving sets of mass-created articles about Geographical entities based primarily or solely on database sources including:

• Wikipedia:Arbitration/Requests/Case/Carlossuarez46
• Wikipedia:Administrators'_noticeboard/Archive332#Mass_deletion_of_Iranian_abadis
• Wikipedia:Administrators'_noticeboard/IncidentArchive1064#Mass-creating_articles_based_on_one_unreliable_source
• Wikipedia:Administrators'_noticeboard/IncidentArchive691#User:Dr._Blofeld
• Wikipedia:Reliable_sources/Noticeboard/Archive_357#RfC:_GNIS
• Wikipedia:Reliable_sources/Noticeboard/Archive_362#GEOnet_Names_Server_(GNS)

Survey (Changes to GEOLAND)

• Option 2 - This change is needed, particularly in an era of LLMs, to avoid the further flooding of Wikipedia with articles about geographical entities based solely on statistical data from a database and containing no "human" data on the entity in question. The policy basis for this is plain since Wikipedia is not a directory, and Wikipedia is not a database. In as much as Wikipedia has a gazetteer function, it is fulfilled by articles that extend beyond a single sentence and an infobox, and as such the proposed standard, which is a long way short of full WP:GNG, is a reasonable minimum standard guaranteeing that, going forward, we will have articles that at least tell you something about the place beyond simple statistics. It also avoids the mass-creation of articles about geographical entities that are purely bureaucratic such as irrigation districts and the like, which presently is allowed under the GEOLAND standard.
  The present situation, where any geographical entity of any kind that has a population is granted notability pretty much automatically, has repeatedly led us into bad situations. Far from raising the profile of poorer countries, the present GEOLAND standard serves to elevate countries that have easily-accessible electronic databases of geographical entities that the poorest states typically lack but, for example, richer countries tend to have. Given the problems that mass-created articles on EN Wiki about Iranian "villages" (that were not actually villages) based on the Iranian census caused for editors Persian Wikipedia (see here for them begging for Carlossuarez46 to stop) it simply cannot be said that the present standard helps the disadvantaged. Rather, it is a vehicle for mass-creation of poor-quality stubs that are of no use to anyone and which will not be expanded, but which also find their way into mirrors and become confusing artefacts on Google Maps thus trashing the information space. FOARP (talk) 16:22, 15 September 2023 (UTC)
• Option 1 as an "everything not mandatory is forbidden" problem. The basic problem with these guidelines -- that I'm currently in the progress of writing up something much longer and more centralized about -- is that too many people, on both 'sides' of the grander sweeping debate here, interpret "something is notable" as meaning "something must have a stand-alone article". This is explicitly contradicted by WP:NOPAGE, and editors have full freedom to 1. upmerge content better covered in a centralized place and 2. not mass-create articles, which is the real problem here (ACAS was hideously botched by the fact no one can agree what mass creation is, but I'd certainly support a flat ban on bot creation). "Notable vs not notable" is not "mandatory article vs option of a list", it's "option of article or list vs mandatory list".
  The practice we consistently see across "X thing where functionally any true case of it will be notable, so is nominally said to have default notability, stops being said to have default notability" is disparate impact. It's still extremely rare for, say, Anglosphere schools or train stations to be deleted, but much more common where sources are less accessible. This is complicated; as someone who's done a lot of scheduling for multiple areas of the main page, I've seen enough view patterns to know how reader interests intertwine with national/regional bias, and there's a reasonable case to be made that the idea we need to improve coverage through such article genres specifically (that is, these sorts of "practical individual elements", rather than much-harder-to-write articles about e.g. culture) is wrong. Both the arguments that mass creation somehow "helps other countries" and the one that deletion standards that specifically try to get at the mass creation problem don't have disparate impact are flatly wrong, observably wrong to anyone who watches the project with open eyes.
  Everything that all these recent arguments are trying to get at is the mass creation problem. Mass creation is the single most commons-burning thing on the project, because it takes completely valid, usable notability standards and sees them tightened in an attempt to prevent the "congratulations on your 2 billion new inaccurate stubs" failure mode. This is because people on both sides don't get that things can be neither mandatory nor forbidden. Notable things do not have to have articles -- but non-notable things have to not have articles. Tightening notability standards, specifically, in response to mass creation is the worst of all results, because the commons-burning articles remain while potential valid articles in the future are prohibited. (Low notability standards are in and of themselves encouraging; people are much more enthusiastic to work where they don't feel like they'll need to prove a borderline case. This means that a standard that "totally allows bad articles!" is one that allows many more good articles than one that's theoretically high enough to only allow the cream of the crop; people writing the cream of the crop overlap heavily with people who do not care to fight at AfD.) The response is to recognize that NOPAGE is a guideline and to be much stricter about merging permastubs. The real response is to recognize that notability standards are a poor proxy for quality standards and that's where all these cagematches really come from, but..."I don't think you're ready for that one yet, but your kids are gonna love it". Vaticidalprophet 16:54, 15 September 2023 (UTC)

  While I agree that WP:NOPAGE is under-considered, I don't think it is applicable here because there are no circumstances where we should have an article when information about them beyond statistics, region, and coordinates does not exist. BilledMammal (talk) 17:50, 15 September 2023 (UTC)

  Sure, we shouldn't have articles that don't demonstrate any reason for them to be articles. Note that this is a carefully separate statement to "where X doesn't exist" -- it's hard to determine in many cases whether one editor's absence of evidence for X is a true evidence of absence, and I've seen plenty of both merges and deletions where such sources could be found with very little trouble. I'm not enthused at all by that, but I also don't shed any tears for articles in the shape those are in when they get merged or prodded. Redlinks can encourage good articles more than bad articles do.

  But I don't see any clear reason why we shouldn't have lists, or other ways-of-covering-a-subject, for things of clear relevance to a reference work that nonetheless aren't best covered independently. We have many lists of settlements in X, cricketers in X, etc. These provide reference-work value and allow us to cover a greater range of subjects.

  I want to reiterate the "generous notability standards encourage good articles" point, because I think it's lost in a lot of discussions that fall for other myths like the idea inclusion-deletion has any "quality vs quantity" component. It is so much easier to write to a high standard when you don't have to worry someone could erase it on a technicality. The generosity of NBOOK bears fruit -- 1.7% of articles in WikiProject Books are GA/FA compared to 0.7% of the project as a whole. Medicine holds at about 1.25% non-stubs being quality-assessed, and the standards for a medical article to pass GA/FA are markedly higher than just about anywhere else (ask me how I know) -- but all diseases are notable. (And yet, we've had no trouble getting stuff deleted when it doesn't want or need a dedicated article, even if it's technically a described human disease.) Medicine also happens to have a significant majority non-stub articles despite such a low standard and an admitted history of mass creation, implying people are generally willing to expand articles past their minimum notability-demonstrating threshold in at least this topic area. Very few people enjoy arguing at AfD, and those who do enjoy it less for one's own articles. The unexpected consequences of raising notability standards -- as opposed to other standards -- need to be considered. Vaticidalprophet 18:26, 15 September 2023 (UTC)

  But I don't see any clear reason why we shouldn't have lists, or other ways-of-covering-a-subject, for things of clear relevance to a reference work that nonetheless aren't best covered independently. This change won't have any impact on whether or not we can create such lists; currently such lists are permitted under WP:LISTN with WP:GEOLAND being silent on the topic, and that status quo will continue regardless of whether there is a consensus for this change. BilledMammal (talk) 19:29, 15 September 2023 (UTC)

  Your specific contention is that NOPAGE (When creating new content about a notable topic, editors should consider how best to help readers understand it. Often, understanding is best achieved by presenting the topic on a dedicated standalone page, but it is not required that we do so; at times it is better to cover a notable topic as part of a larger page about a broader topic, with more context (and doing so in no way disparages the importance of the topic). Editorial judgment goes into each decision about whether or not to create a separate page, but the decision should always be based upon specific considerations about how to make the topic understandable, and not merely upon personal likes or dislikes.) is inapplicable here. How is it inapplicable? If there's too little to write about a place for a useful article, don't have one; if there's enough, write one if you want. ("Statistics" for a populated place are rather broadly construed, such that it's very possible to conceive of an all-'statistics' article providing unambiguous reference work value -- the example bot-stubs are not this, but that's exactly the problem with using inspecific and arguable terminology here.) Vaticidalprophet 19:39, 15 September 2023 (UTC)

  It's inapplicable here because it has no bearing on discussions about whether we should keep or delete a list; NOPAGE doesn't grant notability to the list (the "broader topic"), NLIST does. To put it another way: can you provide an example of a list that would be deleted if this change was implemented? BilledMammal (talk) 19:52, 15 September 2023 (UTC)

  ...Yeah, I think you're flatly misunderstanding me, because that's not what I'm talking about at all. Vaticidalprophet 20:11, 15 September 2023 (UTC)

Thebiguglyalien explained it better BilledMammal (talk) 20:29, 15 September 2023 (UTC)
The following discussion has been closed. Please do not modify it.
My understanding is that you are opposing this because you believe we should include this content in lists, and you believe that through NOPAGE the current wording permits this. I disagree with this, because NOPAGE is not relevant to deciding whether a topic is notable; instead, such lists are permitted through NLIST. If this is incorrect, please correct me. BilledMammal (talk) 20:18, 15 September 2023 (UTC) I think: Whether a populated place is covered solely through statistics (extremely broadly construed concept, for populated places) or not is something where being able to argue about it for ages, rather than shrug and move on to the next article, is not a net positive to the project or its encyclopedic coverage (there are multiple subsets of this one; whether people can easily find non-statistical sources depends on factors other than their existence, 'statistics-only' for a populated place is a concept such that it is not in all circumstances incompatible with a valuable article, etc) The only reason we care about this is because mass creation burns the commons/ruins everything, by producing a vast number of articles that don't have reference work value But "large number of stubs that won't realistically be expanded" is solvable through routes other than changing notability, because they can be upmerged, because notability is specifically "below this you can't have an article" rather than "above this you must have an article" Accordingly, the specific problem here is solvable through routes other than "raising the minimum threshold under which it's impossible to ever have an article" Vaticidalprophet 20:27, 15 September 2023 (UTC)

• Vaticidalprophet I think that the miscommunication here is this: the contention is that if it is permitted to create a stub about a village, then someone will create said stub, and this will include mass creation. And when someone suggests that these microstubs be merged into a single list, enough people will say "no, it's notable per WP:GEOLAND" and the microstubs will remain regardless of any list. As someone who advocates such listification, I've seen this happen. So either we'll debate every single microstub one at a time to decide if WP:NOPAGE should be applied (Option 1), or we need to say "secondary sources are required for individual articles" and then the deleted microstubs can be preserved in lists (Option 2). BilledMammal, is this a fair analysis in your view? Thebiguglyalien (talk) 20:24, 15 September 2023 (UTC)

  There's actually nothing wrong with mass creation here as long as whoever is mass creating has demonstrably verified that these are populated places. We've only gotten into trouble where a census is vague or unreliable, and we're really at the point where there shouldn't be that many places left to create articles for. SportingFlyer T·C 20:26, 15 September 2023 (UTC)

  @SportingFlyer: I hear your concern about uncritically relying on an external reference work, but I have to disagree that "we're really at the point where there shouldn't be that many places left to create articles for". Peek into any other Wikipedia language edition (that isn't written in Cebuano) and you'll find lots of opportunities for expansion. I still occasionally find the time to write articles about places in Vietnam; it's quite clear to me that we have a ways to go for places with a longer history and more population than analogous places in Europe or North America.

  Very often, when I need to learn more about a place I've stumbled upon in OpenStreetMap or OpenHistoricalMap, I end up needing to run a Chinese or German or Spanish Wikipedia article through Google Translate because there's no English coverage and the Wikidata item can only say so much.

  When someone creates a glut of pro forma stubs, the problem doesn't necessarily have anything to do with the subjects of those stubs themselves. It could have everything to do with the manner in which they were created. The problem with regulating this activity in a notability guideline is that it also casts a chilling effect on others who come along wanting to write about the same subjects with more care.

   – Minh Nguyễn ^💬 01:08, 16 September 2023 (UTC)

  Yes, it is; thank you. BilledMammal (talk) 20:27, 15 September 2023 (UTC)

  There are plenty of editors actively merging geo-microstubs into lists. I've seen this happen, probably in far greater number than the argument version (because the argument version revolves around individual articles while upmergers can get tens of articles at once easily). Most stuff people do on Wikipedia isn't discussed, they just make the edits. If you're watching discussions rather than edits, you're coming away with an inaccurate view of the project; this is true in pretty much every problematic area (RSN/RSPS is the classic example). Vaticidalprophet 20:30, 15 September 2023 (UTC)

  I suppose this is one of those instances where it's two different approaches to a similar result. The way I see this is that the benefit of cleanly setting the bar now outweighs any downside of setting a restriction against articles that (in my opinion) universally should be upmerged anyway. Thebiguglyalien (talk) 20:42, 15 September 2023 (UTC)

  @Vaticidalprophet - I upmerged some of the Turkish “neighbourhood” articles that were the subject of one of the discussions listed in the background section. Someone went through recently and restored every single one of them back to being “X is a neighbourhood in Y” microstubs they were before, making exactly the “they’re GEOLAND passes so they should have articles” argument @Thebiguglyalien refers to. FOARP (talk) 20:59, 15 September 2023 (UTC)

  Whoever that someone is did a dumb thing and should feel dumb for doing it. Folly Mox (talk) 20:12, 16 September 2023 (UTC)

  Sorry that was pointless and bitchy of me. This section is being edited too rapidly for me to strike or remove, but I'd like to. Apologies. I feel less grumpy having since eaten food. Folly Mox (talk) 20:44, 16 September 2023 (UTC)

  I'm also not convinced that low inclusion standards result in a high average quality - WP:WikiProject Geography, the WikiProject that is pertinent to this discussion, has just 0.15% as good or featured articles. BilledMammal (talk) 19:52, 15 September 2023 (UTC)

  @BilledMammal: As far as I know, most place articles aren't even listed under WikiProject Geography; there are lots of more local WikiProjects that cover geography as a matter of course. Regardless, I think this metric mostly just demonstrates the sheer ambition of a WikiProject Geography versus a WikiProject Elements (to unfairly pick on a random WikiProject). Minh Nguyễn ^💬 01:22, 16 September 2023 (UTC)

• Option 1. Nothing in the present guidelines prevents merging small articles into larger ones, nor deleting articles that fail WP:V. This means that there is not a problem that needs solving. I also completely disagree with the premise that articles which are primarily statistical are in some way a bad thing - they provide readers with verifiable encyclopaedic information about a topic they are looking for. Thryduulf (talk) 17:19, 15 September 2023 (UTC)

  If you had looked at the original discussion, most of us were against mass deletion of articles. The discussion started because "legal defined place" is impossible to define when you actually start to research around the world. The phrase Settlements and Administrative regions actually means most articles are now covered by GEOLAND.Davidstewartharvey (talk) 19:00, 15 September 2023 (UTC)

  @Davidstewartharvey how is that relevant to anything in my comment? Thryduulf (talk) 19:11, 15 September 2023 (UTC)

• Option 2, per FOARP. If information about them beyond statistics, region, and coordinates does not exist then we can not have an article on them as such an article provides no benefit to the reader; this change will ensure that we don't. Of course, content can still be included in an appropriate parent article for which such information does exist. BilledMammal (talk) 17:49, 15 September 2023 (UTC)

  Too add to this, Wikipedia is not a database; data should be put in context with explanations referenced to independent sources. If all we have is statistics, region, and coordinates - if all we have is basic data - then we lack the explanations to put the data in context and any such article is a WP:NOT violation. This proposal will align the guideline WP:NGEO with the policy WP:NOT. BilledMammal (talk) 18:56, 16 September 2023 (UTC)

  To expand further, if all we have are statistics, region, and coordinates then that almost always means that all we have are primary sources; WP:OR tells us Do not base an entire article on primary sources, and be cautious about basing large passages on them. For example, elsewhere in this discussion, an example of an article that those in opposition to this proposal believe should exist is Salisbury East, South Australia as they were able to expand it, but even the expanded version is sourced solely to the Australian census, a primary source.

  In addition, others have pointed out that such sources often have reliability issues resulting in our articles containing significant inaccuracies to the detriment of our readers; this change won't only align NGEO with NOT, it will also align it with two of our core policies, OR and WP:RS. BilledMammal (talk) 03:05, 17 September 2023 (UTC)

• Option 2 I started this whole shebang off when I asked the question about the existing wording. The question was what is a legal recognised place?. From there it became apparent that others saw the reasoning behind needing change. The current wording is not clear and we need it clarifying. I think the proposal is the best compromise.Davidstewartharvey (talk) 18:04, 15 September 2023 (UTC)
• Option 1 I certainly did not, and do not endorse this proposal. This is another symptom of the rampant deletionism/GNG witch-hunt taking place across the site (just see the cricket proposal at the top of the page). It also completely ignores WP:NEXIST. --Rschen7754 18:06, 15 September 2023 (UTC)
• Option 1 per Thryduulf. Dave (talk) 18:24, 15 September 2023 (UTC)
• Option 2 (Cautious support) There are several good changes in here on minor and housekeeping things. Also the one significant change is that it sort of sets a "1/3 of GNG" standard....merely that some non-trivial coverage is known to exist. I was involved in lengthy discussions related to this. I think that it was near-unanimous intent to not create any big upset of the apple cart regarding current articles but instead to avoid mass creation of immense amounts of microstubs which could be a huge problem in this area. I think that widespread sentiment would provide an extra layer of insurance against the former regarding current articles. I think that the wp:notability ecosystem has it's finger slightly on the scale towards geographic articles, and I think rightly so in view of these tending to be more enclyclopedic. On the flip side, the current standard sets us for being ripe for creation of hundreds of thousands of mictostubs or a random selection from amongst those. I think that the proposal treads the "best of both worlds" middle ground. North8000 (talk) 19:41, 15 September 2023 (UTC)

  I'm sort of middle of the road. I think that the net effect of our current system has been OK. My concern is that there are 2,000,000 - 4,000,000 yet-to-be written permastubs which could pass the current SNG.North8000 (talk) 03:50, 19 September 2023 (UTC)

• Strong Option 1 Articles about where people live are one of the key components of Wikipedia, which functions as a gazetteer. There is absolutely nothing wrong with having a page on a place where we can verify people live or have lived, which is a very easy test to apply. If the only thing we discuss at the moment is simple statistics, that does not mean the article cannot be potentially expanded in the future, even for very small, remote places - it is important to maintain a consistent international standard so we don't start redirecting places which aren't as well documented. I think a change that needs to be made based on recent AfDs are something along the lines of just being listed in a census is not de facto evidence of verification, due to different standards across international censuses. Any one bulk creating articles from censuses is disruptive. SportingFlyer T·C 19:50, 15 September 2023 (UTC)

  Problem: houses, streets, hotels, pest-control districts etc. are all “places where people lived” and for which statistical data can be provided. FOARP (talk) 20:35, 15 September 2023 (UTC)

  It's been flagrantly obvious for a long time what NGEO is meant to include, which clearly does not apply to houses, streets, hotels, pest-control districts... SportingFlyer T·C 23:24, 15 September 2023 (UTC)

  Yet it does not say anything like that, and in the Iranian case we ended up with many thousands of articles about exactly those kinds of things. FOARP (talk) 05:19, 16 September 2023 (UTC)

• Option 1. We've been a gazetteer from the start of the project. Unlike articles on BLPs, where conceivable harm exists from short articles about people of borderline notability that can be a target of bad edits/vandalism, there is no harm to having articles on populated settlements that exist. Espresso Addict (talk) 20:05, 15 September 2023 (UTC)
• Option 2. There is absolutely no reason to have an article about something if there are not reliable secondary sources covering it. The whole point of notability guidelines is to avoid articles about random topics of which there's nothing meaningful to write about. For this particular subject, GNG should be more than sufficient to determine whether the topic is notable, and I don't see the need for this SNG at all. Thebiguglyalien (talk) 20:11, 15 September 2023 (UTC)

  Then why have we allowed this exception for two decades? We decided a long time ago that we're a gazetteer. SportingFlyer T·C 20:33, 15 September 2023 (UTC)

  We never decided anything of the kind. There is no consensus, anywhere on Wikipedia, ever, that EN Wikipedia is a gazetteer (i.e., a geographical dictionary) per se. At most we have elements of gazetteer, just as we have elements of other non-encyclopaedic text books. Gazetteers are typically long lists of feature-names paired with statistical data. FOARP (talk) 20:40, 15 September 2023 (UTC)

  This is absolutely false. I've dug through it before, it occurred a long time ago, will try and find it again soon. SportingFlyer T·C 21:17, 15 September 2023 (UTC)

  This has been documented at WP:5P1 for decades now. --Joy (talk) 10:46, 1 October 2023 (UTC)

  I agree that Wikipedia should have gazetteer-like content. But I believe that this content belongs in list form, not as a separate encyclopedia article for every individual entry. Thebiguglyalien (talk) 20:44, 15 September 2023 (UTC)

• Option 2 per above commenters. Makes perfect sense to me. Edward-Woodrow :) _[talk] 20:17, 15 September 2023 (UTC)
• Option 1 While in principle I support option 2, I think the wording of the proposed replacement is exceptionally poor. The current version is tight and specific, while the proposed replacement is far too wordy. I think it it would be far better if the proposed changes were added to the current version rather than replacing it outright; it'd actually be quite simple and could be done in only one or two sentences. Curbon7 (talk) 20:41, 15 September 2023 (UTC)

  What is your preferred version? I’ve got to say we discussed the amendment for months with lots of waves of revision, but I’d also hate to lose a support !vote on wording alone. FOARP (talk) 21:02, 15 September 2023 (UTC)

• Option 1 Wikipedia is a gazetteer. From my understanding, the main issue people have with GEOLAND is it leads to many stub articles that will never really progress past a stub. However, as many have pointed out, these can be upmerged whenever needed. Plus as Espresso Addict added, there is no tangible harm that really arises from even these articles. ULPS ^{(talk • contribs)} 22:09, 15 September 2023 (UTC)

  Google Maps now contains random, non-existent “villages” based on them scrapping innaccurate data from Wikipedia. The Iranian case linked above is another example of harm that has come from mass-production of articles based solely on a database. FOARP (talk) 05:16, 16 September 2023 (UTC)

  If that is true, it is a reason to prohibit the citation of Google Maps as a source in Wikipedia articles, not other sources that don't do that. We are not responsible for what mirrors of Wikipedia choose to scrape. Even if we were responsible, that is an argument to restrain scrapping, not an argument to delete accurate articles just because they are presently short or their present content is statistical. James500 (talk) 05:27, 16 September 2023 (UTC)

  Ok, Google Maps sucks and shouldn't be used as a source. We should not be overcorrecting because of one bad incident that is somewhat related. ULPS ^{(talk • contribs)} 19:18, 16 September 2023 (UTC)

  What? It is our fault that Google Maps – a resource many people routinely use – is now innaccurate. Edward-Woodrow :) _[talk] 19:55, 16 September 2023 (UTC)

  I think James500's comment explains my opinion on this. ULPS ^{(talk • contribs)} 19:57, 16 September 2023 (UTC)

  Of course we're not responsible for which information other parties choose to source from us, or their lack of double checking before republication. But we are responsible for our own inaccuracies. "It's their fault for trusting us" is not a particularly strong argument. Folly Mox (talk) 20:24, 16 September 2023 (UTC)

  @FOARP: This is an oversimplification. Google Maps maintains their own proprietary database of places that they source from a variety of official databases, including GNIS, Geonames, and TIGER (all of them notorious for mixups and duplication). [1][2] The fact that you're seeing these places on Google Maps is a coincidence. We also know that they maintain their own database of place names, because sometimes they're well ahead of Wikipedia in acknowledging a place's new name. However, in underresourced languages, they pull place names directly from Wikidata. (Their Vietnamese place labels were full of disambiguators until I cleaned them up in Wikidata earlier this year.)

  Yes, Google Maps does look for a matching Wikipedia article if you ask it for more information about a place you see on the map. They aren't going to dump the user into the GNIS database, after all. You could argue that Wikipedia's coverage of the place lends it more legitimacy in the user's eyes than if it had remained a bare place name with no information, but not that we're leading Google astray.

   – Minh Nguyễn ^💬 23:45, 16 September 2023 (UTC)

• Option 1 per Thryduulf. Villages generally satisfy GNG. There is generally either extensive written historical records or extensive archaeology. There is a mountain of history books based on those. Even Roman and medieval villages that now exist only under the ground are the subject of an extensive archaeological literature, because they get excavated. Similarly, a parish used as a unit of local government etc is generally going to satisfy GNG from at least the middle ages onwards, due to the records and the history books based on those. Contemporary settlements are the subject of a gigantic mountain of newspapers. Since these topics generally satisfy GNG, they should be presumed notable. We don't want another spree of mass nominations based on the principle of "this village/civil parish satisfies GNG, but the sources are not in the article yet, so let's mass nominate it without a WP:BEFORE search". If, for the sake of argument, a region of local governement was particularly short lived due to boundary changes, it can be merged without affecting the general principle, because notability only creates a rebuttable presumption. The Iranian census incident is a red herring, because those places were not villages. James500 (talk) 22:49, 15 September 2023 (UTC)
• Option 1 per Vaticidalprophet. –Fredddie™ 23:58, 15 September 2023 (UTC)
• Option 1 Option 2 is blatantly anglocentric and comes across as a attempt to radically decrease Wikipedia's level of coverage on geographic areas outside of the anglosphere. There are multiple villages and/or significant settlements across much of India that may have only been listed in the census or other related database specific websites. Finding reliable third party coverage of these places are hard to come by especially based on just English search phrases (not to mention one place might have multiple transliterations which complicates searching for sourcing even further). Any such article sent to AFD will result in a near certain delete (due to the historically low participation of editors in this area). -- Sohom (talk) 00:33, 16 September 2023 (UTC)

  I doubt the intent of the proposal is "anglocentrism". I personally see it as WP:NOTDB issue. Edward-Woodrow :) _[talk] 12:21, 16 September 2023 (UTC)

  @Edward-Woodrow I've struck out the portions refering to the anglocentricness of the proposal, however, I think the meat of the oppose is valid, from my POV, it's less a NOTDB issue and more a issue with having the ability to delete articles that might have some coverage (say in one or two local newspapers) in non-english sources that might not be easily accessible on the internet/via a straight/obvious Google search (ping on reply) -- Sohom (talk) 20:06, 20 September 2023 (UTC)

  I mean, I think there will always be a bias towards the anglosphere on enwiki, not because we want one, but because, well, all editors speak english, so many live in the english-speaking world. There will always be a bias towards les pays francophones on frwiki, and so on. Edward-Woodrow • talk 20:16, 20 September 2023 (UTC)

  And english editors will have higher average difficulty in correctly reading, say, a Telugu source, just as Telugu Wikipedia editors have higher average difficulty in reading an English source. @Sohom Datta: forgot to ping. Edward-Woodrow • talk 20:18, 20 September 2023 (UTC)

  @Edward-Woodrow I do agree with what you are saying, but we should probably make a effort to counteract this systemic bias if we do get a chance to do so :) Having a bunch of small stubs that somebody can come in and improve is a much better status quo than a big red banner telling you that the article had been deleted recently in a AFD. (Which will discourage both new and semi-experienced editors from re-creating the article). Sohom (talk) 11:14, 23 September 2023 (UTC)

  @Edward-Woodrow: In a sense you're right, but if we have an opportunity to avoid entrenching and reinforcing this systemic bias, why would we pass up this opportunity? The example of Telugu is particularly interesting: in many South Asian languages, a significant portion of the reliable sources on a given topic will exist only as text inside an image due to inadequate or inconvenient support for the relevant character encoding or fonts. Google Search won't find this information. Google Image Search can OCR images but doesn't do so nearly as reliably in non-Western scripts (or even in non-Western Latin alphabets). @Sohom Datta makes a good point about transliteration being another confounding factor. Censuses and databases have the potential to help us overcome these issues because of their systematic nature. Minh Nguyễn ^💬 10:06, 23 September 2023 (UTC)

  Is it really helping systemic bias to have a guideline where having some "historical building" is enough to make a location notable? Because that sounds like a great way to get tens of thousands of stubs on minor neighborhoods and streets in the West where there's money, governmental stability, and interest in building preservation, rather than increasing coverage of notable non-Western communities. And anyway, why do we need separate standalones for any of these topics in order for us to address "systemic bias"? If they can be covered in the same amount of depth as part of a page on a larger municipality, with redirects to their section, what's the problem? An unwatched microstub is going to get waaay less attention and protection than a page on an upper-level administrative region. JoelleJay (talk) 18:55, 23 September 2023 (UTC)

  @JoelleJay: I think that historic building tangent below was a bit of a distraction from the core issues; English villages are relatively safe regardless of the outcome of this poll. The thread you've responded to is specifically about a concern that English Wikipedians are already at a disadvantage in finding appropriate sources due to language barriers. My point in response to @Edward-Woodrow is that we should not be defeatist about these challenges. Wikipedia can be an awesome tool for overcoming language barriers, but we need to send the message that editors with language skills are truly welcome and not being set up to fight more uphill battles than necessary. This is not about relaxing standards; it's about ensuring a less hostile work environment, as it were. Minh Nguyễn ^💬 06:19, 24 September 2023 (UTC)

  It's not as if Indian people and journalists are incapable of providing non-trivial coverage of their "significant" communities. Language barriers and language preferences are an inherent barrier to the development of certain spheres of topics in any given language specific Wikipedia due to the editing base which each may attract. Shockingly, we have might have less coverage of Burundian topics than by a hypothetical active Wikipedia.ku, but that doesn't mean universal standards still can't be applied, even if results are imperfect. A new standard would also apply to hundreds/thousands of places in the Anglosphere as well, but that doesn't mean it is Anglophobic. Yes, a person living in the Anglosphere (the average English Wikipedia editor) might have to work a little harder to find "notable"/includable communities under Option 2. But I think outright dismissing it as a total inequity assumes that people in other countries are incapable of writing their own newspaper articles on places that are politically, economically, or culturally significant to them. Ask yourself, how "significant" are these Indian villages if for example, after extensive search, it appears not even an Indian journalist has bothered writing about them? Should they still be included because it would make a white-majority internet community feel better about itself, since they think they have better judgement of what is important than the people of those nations? -Indy beetle (talk) 09:36, 17 September 2023 (UTC)

  @Indy beetle I am not arguing that people outside the Anglosphere are incapable of creating newspaper articles (not sure what gave you that impression). What I'm pointing to is the fact that finding that one article in a major newspaper about a populated area (that is notable for having a burger joint) in the United States of America is probably easier than finding the one articles in a Indian local-language newspaper (that pronounce the name of the place differently) that tells you that the place in question is a semi-major hub because it has the only hospital within the next few 100 or so miles. (if you are lucky). (Also, pls ping on reply) -- Sohom (talk) 20:06, 20 September 2023 (UTC)

  @Sohom Datta: My wording was inappropriate for my message, my apologies (though I have encountered others before on Wikipedia who seem to think they know a great more about what is important to a given people than the people themselves). I do not mean to suggest that you were exercising prejudice. What I am saying is that I think we'd be way better off writing articles about localities in any country because we have at least one secondary SIGCOV RS to help us do so (thus implying their relative significance) rather than assuming that just because it was listed as an entry in a census means it is notable. And I understand challenges exist with regards to trying to write articles about a topic/area with which most of our editors are unfamiliar, but as someone who has written about undercovered subjects, I think it's worth the extra work necessary to make the start to a proper article. I don't see how copying what the Indian census might say about a random small town but in different words is a great victory for CSB. Pretending that it is such a victory implies that we are either doubtful that the Indian media (or whoever) are incapable of distinguishing important places in their society or that we editors are incapable of digging up relevant secondary RS to evidence notability. I think that sets the bar way too low. Because inevitably we'll end up with a bunch of insignificant places while pretending that they're not, and that follows for both the Anglosphere and outside of it. Start crawling through obscure Indian place news articles and we might actually be able to CSB instead of just including everything in a census and pretending like that's good enough for an encyclopedia. -Indy beetle (talk) 04:03, 21 September 2023 (UTC)

  @Indy beetle I do agree that copying the census data verbatim is not a very desireable way of CSB, but, it is a better middle ground than deleting potentially notable articles due to unfamiliarity with the subject matter/lack of participation in AFDs. (If there was a hypothetical scenario where the BEFORE guidelines where much much stricter, this proposal would make some sense, but the current ones are not strict enough, requiring all but a cursory Google search which most of these Indian places will fail).

  Also wrt to the question of not being able to distinguish important/notable places, I think we also need to keep in mind that the sources available on the web for Indian places are from a very (relatively) small time period (1990-2023++) and a lot of the written records before this time period are probably not digitized and are hard to come by. I recently (last year) came across a fairly remote village that had a (currently decreipt) pakka bridge contructed across a river during the early 19th century during the British raj, if this was the United States of America (or somewhere in Europe), there would have been some kind of digitized written record and/or newspaper clippings of this (making the place notable), however, there is not one mention across the internet of this bridge being built (making the place non-notable by Option 2 standards). Sohom (talk) 11:08, 23 September 2023 (UTC)

• Option 2. "Gazetteer functions" still exist when individual settlements are covered within larger articles. That is in fact more like a gazetteer than what single-line GIS+census data stubs provide.
  JoelleJay (talk) 01:23, 16 September 2023 (UTC)
  • It is obvious that, if this proposal achieves consensus, there will be further proposals for mass draftification and then mass deletion without WP:BEFORE, as with the Olympians and the cricketers. Such proposals would massively disrupt the entire project, as articles on villages and units of local government are absolutely vital to Wikipedia's coverage of geography, history, government and just about everything else. Mass elimination of notable villages is not like removing a few BLPs or a few company articles for contemporary small businesses, but rather rises to the level of the most extreme disruption possible. If further proposals for mass draftification without WP:BEFORE are not prohibited (and simply making such a proposal without WP:BEFORE in respect of villages and other vital topics would need to be a potentially blockable offence), we are going to need more and stronger SNG presumptions of notability on this project, not less, in order to prevent such mass draftification proposals from seriously disrupting the project. At this stage, we have reached a point where we need to have a policy to restrain the making of proposals for mass draftification without WP:BEFORE, we need to have that policy immediately, and we need to have it before we can even consider any proposal to eliminate any existing SNG notability criteria. James500 (talk) 04:26, 16 September 2023 (UTC)

    User:James500, everything you've said in this comment seems predicated on the false idea that the only way to present information in Wikipedia is a standalone article. I don't disagree that it's important for the project to contain encyclopaedically relevant information on geographic subdivisions (and, to generalise, any topic type), but if we don't have enough for a third and fourth sentence on a topic, it's better all around – for the reader as well as editors – for the information to be presented in its larger context, rather than a tiny garbage substub. Folly Mox (talk) 15:16, 16 September 2023 (UTC)

    Please do not put words into my mouth that having nothing to do with what I actually said. The problem with refusing to perform WP:BEFORE has nothing to do with whether any page should have a standalone article. The problem is that someone who refuses to do WP:BEFORE is failing to do due diligence, is failing to check the factual accuracy of the claims he is making, is attempting to freeload and to make himself a burden on others, and is trying to get other people to do something that he ought to do himself. This is not an objection to articles having to satisfy GNG, but to an unfair fact finding exercise where it is impossible to prove that they satisfy GNG before the end of the RfC, due to the fact that there is not enough time (assuming that the RfC is not going last six months or more). James500 (talk) 22:10, 16 September 2023 (UTC)

    My apologies. I must have misunderstood your argument. Folly Mox (talk) 03:57, 17 September 2023 (UTC)

    I should also point out that draftification is not merger. I appreciate that the proposal refers to including "information . . . in a more general article", but there is no realistic prospect of that instruction actually being followed in a mass draftification proposal. The recent mass draftification proposals specifically proposed that WP:ATD-M would not be followed, that the information in the articles would not be merged into the "more general articles" that already existed, and that "more general articles" would not be created as merger targets where they did not already exist. It is obvious that, if there is a mass draftification proposal for villages, nothing will be merged, and that every article not expanded will be draftified and then deleted, because that is what was proposed in the previous two RfCs. James500 (talk) 22:30, 16 September 2023 (UTC)

    I should also point out that this proposal is not confined to two sentence "substubs". This proposal is worded in such a way that it will in practice generally delete any article that does not contain references that prove that it satisfies GNG, regardless of the length of that article and regardless of whether the topic does in fact satisfy GNG. That is effect of the reference to WP:BURDEN and the words "extent of coverage in secondary sources should be considered to ensure there is enough verifiable content for an encyclopedic article" and "cannot be developed using known sources" and "avoid creating very short articles", when read with criteria 8 of the deletion policy, and having regard to the fact WP:BEFORE is a dead letter at AfD, and the fact that instructions to merge non-notable pages are typically not followed at AfD, and those pages typically just get deleted. James500 (talk) 22:56, 16 September 2023 (UTC)

    Thank-you! Up-merging and table articles really can be more useful than a myriad of X is Y articles. Edward-Woodrow :) _[talk] 15:18, 16 September 2023 (UTC)

    There are a number of affordances of enwiki where this is not the case: notably the category system, effective hypertext cross-referencing and mouseover (or google) preview functionality. Only some of these limitations can be addressed by the use of categorized redirects and/or hard work. Wikipedia is not simply a "ball of text", even if some editors tend to see it that way. Newimpartial (talk) 16:43, 16 September 2023 (UTC)

    That's a good counterpoint. As a person who does not use categories (or find them useful) and is unable to see navigation popups or page previews due to being exclusively on mobile, I had not considered either of those. Folly Mox (talk) 20:32, 16 September 2023 (UTC)

    Is it? I mean, that seems more like a matter of taste than anything else. If the standard practice were to have bigger articles with boilerplatish prose and tables about subdivisions, wouldn't editors just complain that those articles read too much like databases? (If the question begins "Would Wikipedia editors complain...", the answer is probably "yes".) XOR'easter (talk) 16:34, 16 September 2023 (UTC)

    I really think 95% of this issue is aesthetic. I like the way stubs feel – there's something very pleasing and early-days-Wikipedian about a page that says "this is a thing we don't have a proper article on, maybe you want to write it?" Others, clearly, find them so distasteful they have to keep coming up with new compounds to express their contempt (substub, microstub, ickystub). That doesn't seem like something we can easily build consensus around, but it also doesn't seem like something worth restructuring vast swathes of the encyclopaedia over, given we're already so far down the road of many stubs instead of few lists. – Joe (talk) 16:58, 16 September 2023 (UTC)

    From a maintenance perspective: absolutely yes. When there is an IP-hopping vandal subtly changing numerical infobox parameters, or inserting nationalist alternate names in non-Latin scripts; or when a good-faith editor adds a bunch of entries sourced to non-English government sources that happen to be for gas stations, it is so much easier and quicker to identify the problem when it's occurring on a well-watched list article than on dozens of individual pages.
    From a category browsing perspective: categories with hundreds of entries and (sub)_ncategories are daunting, difficult to navigate, and devoid of context. If you're wanting to learn about major cities in some region, combing through hundreds of names to find the ones where actual info exists is a chore. A list sortable by population is so much better. JoelleJay (talk) 17:23, 16 September 2023 (UTC)

    And some kind of data structure of which one could make flexible queries would be even better.... XOR'easter (talk) 17:42, 16 September 2023 (UTC)

    Yeah well maybe that will happen when Wikipedia's search function doesn't practically make you convert any mildly complex wildcard query into monadic second-order logic... JoelleJay (talk) 00:57, 23 September 2023 (UTC)

    Heh, maybe if someone really wants prose about the place, it won't be long before they can consult the Abstract Wikipedia, powered by Wikidata and not subject to the English Wikipedia's notability guidelines. Minh Nguyễn ^💬 07:41, 18 September 2023 (UTC)

    Apart from the maintenance burden, it is probably mostly personal choice, although I'd characterise the personal choice based more on neurotype or something similar than aesthetic taste. I don't know if I can understand anything outside its context, so for me it feels like super teeny tiny articles are basically useless. Presumably there's another kind of brain that can only understand things when removed from context and made into a topic of discussion, for whom any list of things is deeply unhelpful. I find myself agreeing most here with Vaticidalprophet, but I don't think I'll be bolding a not-vote. Folly Mox (talk) 20:40, 16 September 2023 (UTC)

    I noted in Discussion that I was able to substantially expand two local articles based only on census results (Clayton, Victoria#Demographics, previously a one-paragraph section in a vaguely Start-class article, and Salisbury East, South Australia, previously an uncited two-sentence stub). While neither of these articles are passing GAN any time soon, they also provide clear (to me) reference work value and a substantial amount of context about those places. I definitely think there's a level of aesthetic-disagreement here, but I'm not a fan of stubs-for-their-own-sake myself and I still land quite clearly "this is a terrible idea". WAID has made the point a few times that "database" is not a natural category -- you can write a fleshed-out article for a location based on database sources. Vaticidalprophet 20:59, 16 September 2023 (UTC)

  • Sometimes I wonder, if we gave users the option to automatically display short articles as part of a list, would the inclusionist–deletionist debate just disappear? – Joe (talk) 11:59, 16 September 2023 (UTC)

    That's a really interesting idea. Some functionality where if an article is under a certain size, opening the link inserts its contents into the article you're already reading? Folly Mox (talk) 20:28, 16 September 2023 (UTC)

    @Joe Roe @Folly Mox: I made something like that once, as a memorial to the Encyclopædia Britannica, our timeless inspiration. [3] Minh Nguyễn ^💬 23:53, 16 September 2023 (UTC)

• Option 1. Generally agree with Vaticidalprophet. We do need improved processes for merging permastubs and other lower quality articles. Tightening notability standards to force articles slowlly through AfD (or future proposals) when merging is generally the ideal outcome is not the right way to accomplish this. —siroχo 05:58, 16 September 2023 (UTC)
• Option 2, because I don't see a point in having articles that are just copied from (possibly copyrighted) database entries. These articles will be peripheral and therefore unmaintained, posing accuracy and issues; there's no benefit for readers to come to us for these, when they can just check the original, properly-maintained, up-to-date, accurate database. Beyond that, we're also literally not a database, in the sense that people don't browse Wikipedia using SQL commands. This is not a WP:NOT-based argument about the encyclopedic value of content; I'm saying that turning database entries into microstubs is an inane way to present this information. Though I would remove paragraph 2 from option 2, because merging should be decided on a case-by-base basis, and there's no need to address it in the guideline. A small settlement won't always be relevant enough to mention in the article of a bigger area. DFlhb (talk) 09:16, 16 September 2023 (UTC)
• Option 1 (no change). WP:GEOLAND is a long-standing guideline with simple, easily-applied criteria. The proposed replacement is basically a restatement of the GNG except somehow both vaguer (how much "verifiable content" is enough?) and longer. It's going to produce more work for editors who write geographic articles, more work for editors who review them, more work for participants and closers at AfD, and... for what? To avoid another User:Carlossuarez46 incident? The problematic articles he created didn't meet GEOLAND as it's currently written; the problem there was an editor being sloppy with sources in a language he didn't understand, and that's not something we can address with new guidelines. Apart from these isolated incidents of editor misconduct, I've yet to hear a coherent rationale for these endless attempts to jettison our gazetteer function, other than that the proponents just don't like short articles about places. Which, guys, isn't enough... surely you realise that by now? – Joe (talk) 11:54, 16 September 2023 (UTC)
• Option 2 Upward merge or deletion is the natural course of action for tiny statistical entries that cannot be expanded and which would never feature in a serious encyclopedia anyway. The proposed change is of benefit to readers, editors, and just about everyone except stub enthusiasts, and puts the guideline in line with the current policies, that Wikipedia is not a database or a gazetteer. Avilich (talk) 14:22, 16 September 2023 (UTC)
  • How does the proposed change benefit readers? At the moment the reader entering the name of a village gets taken straight to the information (which may well be just one sentence, as most articles in most of the encyclopedias that Wikipedia put out of business were) that we have about it. If they get redirected to a list article they may have to read lots of information about other villages before getting to the one they were looking for. How is that a benefit? Of course the separate article now should have a link to the next largest unit if anyone wants to read about that. Phil Bridger (talk) 17:02, 16 September 2023 (UTC)
• Option 1 I find that Thryduulf pretty much said what I was going to say. I also agree with Joe Roe that the proposed replacement is like the GNG but worse. (It's hard to shake the feeling that the underlying rationale here is that we need to change the guidelines because we have too many articles that would be excluded by the guideline we would change to; it feels oddly circular.) I remain unconvinced that there is a real problem here; if there is, I think siroχo makes a good point that this is the wrong way to solve it. The word "database" seems to be thrown around like a boogeyman in these arguments. Sub-stubs about tiny villages read too much like database entries, we are told (but wouldn't a big page about all the villages in a province read like a database dump?). If particular databases turn out to be full of shoddy information about tiny villages, then we should mark them as unreliable for the purpose. The proposed replacement doesn't actually offer any useful guidance on this point, whereas the current text does: The Geographic Names Information System and the GEOnet Names Server do not satisfy the "legal recognition" requirement and are also unreliable for "populated place" designation. Why are we making the guideline worse?! XOR'easter (talk) 16:29, 16 September 2023 (UTC)

  To amplify: If there is a problem, then the proposal at hand is a bad solution because it is more ambiguous than the current guideline. XOR'easter (talk) 20:42, 16 September 2023 (UTC)

  More ambiguous that suddenly having to make yourself an expert on Iranian or Turkish law to work out whether a Mahalle or Abadi are "populated, legally recognized places", or working out whether inclusion in GNIS is the same as being legally recognised? And then wondering why practically-identical settlements in a neighbouring country that don't have any special legal-ish name aren't notable? FOARP (talk) 16:35, 17 September 2023 (UTC)

  Yes, because it throws out the specific point about GNIS. For anyone who wants to get started writing or editing articles about places, it is manifestly less useful than what we have now. XOR'easter (talk) 22:00, 17 September 2023 (UTC)

• Option 1 If you want to stop people from mass-creating bad articles, create rules around the mass-creation of GEOLAND articles. The current standard is perfectly fine. I support stricter standards around notability for people, especially BLPs, but the same concerns don't apply to places. Per Joe Roe, there's no need to create a bunch more work for no real reason. Galobtter (talk) 18:01, 16 September 2023 (UTC)
• Option 1 Wikipedia is an almanac. voorts (talk/contributions) 18:11, 16 September 2023 (UTC)

  I'm confused here. Since when has Wikipedia ever catered for the content typical of almanacs (i.e., " weather forecasts, farmers' planting dates, tide tables" etc.)? An even if Wikipedia were an almanac, why would that be relevant in this case? WP:5P is an essay and the inclusion of almanacs (literal indiscriminate collections of information) in it is why it remains so. FOARP (talk) 16:29, 17 September 2023 (UTC)

  Sure, we don't have weather forecasts, planting dates, or tide tables. But the lead of our article at Almanac also mentions dates of eclipses and religious festivals (which often have dates in their infoboxes, e.g. Easter and Ramadan). Later in the article it mentions many more topics that are covered in contemporary almanacs, many of which we cover too (e.g. demographics, winners of awards, and so on). Anomie⚔ 16:54, 17 September 2023 (UTC)

• Option 2 GEOLAND has always been too over-broad; anything that restricts it is welcome for me. Chris Troutman (talk) 19:03, 16 September 2023 (UTC)
• Option 1. Does this feel to anyone else like attempt No. 10,832 by a group of users to drastically reduce the amount of content we've got here? Per Wikipedia:Five pillars: Wikipedia is an encyclopedia: Wikipedia combines many features of general and specialized encyclopedias, almanacs, and gazetteers. This feels like its only going to result in the mass removal of historic places and locations in other countries, as we do not have nearly as good access to their sources. BeanieFan11 (talk) 20:14, 16 September 2023 (UTC)

  Does this feel to anyone else like BeanieFan11 getting quite close to the casting of aspersions? XAM2175 ^(T) 00:43, 18 September 2023 (UTC)

  @XAM2175, I do not feel that @BeanieFan11's comment is casting aspersions. It's describing the impression they get from the proposal, and while it might be an exaggeration and may or may not be accurate it is not an unreasonable interpretation of the proposal. Thryduulf (talk) 01:15, 18 September 2023 (UTC)

  Huh? I do not see at all how one could get that impression. BeanieFan11 (talk) 01:59, 18 September 2023 (UTC)

• Option 2 - This solves two problems: Ambiguity in the meaning of "legally recognized populated place" and the unreliability/unsuitability of sources that simply provide statistics. Time and time again we've seen the consequences of relying on databases and the like, even official government ones: They often either document things that aren't really settlements (such as census tracts that are drawn as a convenient way to count people) or are flat-out wrong as is the case with WP:GNIS. Aside from establishing notability, secondary sources are needed to confirm that the place is actually a settlement and not something else entirely.
  I'm baffled by the argument that these places meet GNG and yet requiring the inclusion of SIGCOV sources would prevent them from being created. If the sources exist, why not require editors to include them as we did with WP:SPORTBASIC #5? –dlthewave ☎ 20:36, 16 September 2023 (UTC)

  The problem is not the creation of new pages. The problem is that we have a lot of legacy articles on topics that do in fact satisfy GNG, but don't actually cite the sources to prove it, because the relevant SNG, WP:ARTN, WP:ASSERTN and WP:BEFORE told editors, and still do tell editors, that if a topic in fact satisfies GNG, it is not necessary for the article to prove that it satisfies GNG. In reality, that is no longer true. These days an article usually does have to actually prove that it satisfies GNG in order to avoid being deleted or nominated for deletion.

  It would take a certain amount of time to put the relevant sources into those legacy articles. However, without the SNG, we are not likely to get the time necessary to add those sources. We will probably get a large number of high speed WP:FAIT mass nominations or mass draftification proposals in quick succession. That is the way things are on this project at the moment. James500 (talk) 01:37, 17 September 2023 (UTC)

• Option 2 and just to pitch in here, a few years back I had to nominate >100 articles for deletion, created by a now-blocked admin from an old gazetteer that had created whole areas, towns and cities in the United Arab Emirates that had never existed in reality or had ceased to exist decades ago. Apps and delivery services scrape Wikipedia, leading to whole communities searchable online that had no basis whatsoever in reality other than a dumb one-line article in Wikipedia based on a single source that was itself based on a 70-year-old survey of a country that has been utterly transformed over the past five decades. One-line, one-source stubs based on Wikimapia, Google or any other (often user-editable with no oversight)) database sources should not pass WP:GEOLAND. Just to be clear here - we actually have a responsibility... Best Alexandermcnabb (talk) 15:38, 16 September 2023 (UTC)

  We can choose to have or not have mass creation whether this proposal passes or not. We can, and by your own evidence do, nominate articles for deletion for failing WP:V. What sources we regard as accurate or inaccurate is completely independent of this proposal. So even if those things are problems (which is not universally agreed) they can be solved without any changes to the proposed changes would not solve those problems. Additionally, notability is not temporary - just because something has ceased to exist does not mean we should not cover it (or are you proposing to delete subjects like Pangaea, Ur, Orleans, Oregon, Capel Celyn, etc, etc. We have a responsibility to our readers to provide accurate, neutral information about encyclopaedic topics. If other people are using our information for other purposes it is their responsibility to ensure it is accurate for that use, not us. Thryduulf (talk) 20:28, 16 September 2023 (UTC)

  I agree that we have a responsibility to provide accurate information. A big problem with maps and databases is that they nearly always lack sufficient context to establish what the place actually is/was, making it impossible to write an accurate article. Similar to the UAE situation, a few of us are currently dealing with "phantom communities" in northern Canada: Someone went through the national geo database making articles for every "flag stop" railroad station (these are spots in remote areas where the train will drop you off upon request) and since these "rail points" are also categorized as "unincorporated places", they went ahead and made an "unincorporated community" article for each one as well despite none of them having any sort of human habitation. This type of thing is extremely common in articles based on databases and is also very time consuming to clean up, especially since they end up being cross-referenced to other articles (Settlement A is located 9 miles from Settlement B). Requiring some minimum amount of secondary sourcing would prevent many of these situations. –dlthewave ☎ 15:08, 17 September 2023 (UTC)

• Option 2 (hesitantly). So many articles on communities are simply "x place exists in Y region in Z country." I respect our gazetteer functions, and I don't really want to see a bunch of random small place crap crammed into whatever is deemed the parent article (Do readers really need to know that the unofficial neighborhood of Foo in Bar Province was recorded as having 37 residents in the 1999 census on the Bar Province article? The redirecting and merging will inevitably occur if this is adopted...), but the definition of "community" and particularly, "Populated, legally recognized places" is so broad and undefined it is practical unworkable. Such a definition works relatively well if restricted to incorporated communities in the American sense (allowed to have a local government) but breaks down due to the lack of clear analogous versions elsewhere, and thus, it seems having been mentioned in some "official" document somewhere has been deemed sufficient. At least I was able to include some local history on the Moss Neck, North Carolina article when I wrote it due to a newspaper report on local history and a book, even if the material is only a paragraph long. Should that not be our minimum expectation? Otherwise we are simply demonstrating that places WP:EXIST without saying anything beyond that. -Indy beetle (talk) 09:11, 17 September 2023 (UTC)

  Agree that this is basically a standard designed primarily for North America (though it does not function so well there), and which quickly makes no sense at all when applied to other countries where even very small communities - potentially single buildings - can be "legally recognised" (e.g., the Iranian Abadi and Turkish Mahalle). The objection that the proposed changes are "anglocentric" is misguided.

  The present standard basically requires us to become experts in local law in order to identify which communities are and are not "legally recognised", and creates wide variance between otherwise identical communities in different countries depending on what the local law says. FOARP (talk) 12:32, 17 September 2023 (UTC)

  There's no need to clean up the "x place exists in Y region in Z country," though. Many of these are linked to articles in other local languages which may be able to give a user more information, or an editor a starting point on how to expand the article. Not only would we lose that with an up-merge, but up-merging to a list would also suggest to any one wanting to edit that article in the future that a stand-alone version would not be allowed, which is clearly incorrect. SportingFlyer T·C 12:37, 17 September 2023 (UTC)

  up-merging to a list would also suggest to any one wanting to edit that article in the future that a stand-alone version would not be allowed. I am frankly not seeing that, seeing as this proposal provides the guidance for when one can create a separate article (a mere single piece of secondary SIGCOV). Meanwhile, take a look at List of townships in North Carolina, which is a comprehensive list of all townships in North Carolina with each entry being Wikilinked (most being red but some being blue). From the best I can tell, every blue link article on a township is a simple regurgitation of census data or unsourced description of the township borders most certainly thanks to our current NGEO standards. Townships in North Carolina haven't had any meaningful legal/political significance of their own since the 1870s (please read the source material I just added on their history), and their continued de jure existence is an accident of history. A good handful of rural counties use the boundaries set by the townships for dividing up information or determining other boundaries (for their voting precincts typically, in my experience) and the census does sort some data along those lines, but I have yet to read a single news article which even mentions a township in a non-historical sense or meet a single North Carolinian who identifies themselves by or even knows in what township they reside or originate. Ansonville Township, Anson County, North Carolina, for example, is all primary sources and OR. How good is that? Under current NGEO standards, all ~1,000 NC townships could have standalone articles ("Populated, legally recognized places" they are, arguably), and yet they mean next-to-nothing to most people who live there, with similarly next-to-nothing secondary RS to show for them. This is about as good as having articles for every voting precinct, policing precinct, or firefighting district, in a given place. For those that are notable of course we can have our new standards which require the bare minimum, but why just assume that all of them are important when nobody else does? -Indy beetle (talk) 04:34, 21 September 2023 (UTC)

  Which brings up the question of how many microstubs have been created on obsolete, unknown-even-to-locals administrative configurations in other countries... JoelleJay (talk) 01:12, 23 September 2023 (UTC)

  @Indy beetle: For what it's worth, I think North Carolina townships aren't very representative of the kinds of civil townships we want a full complement of articles about. Some Midwestern states have much stronger forms of townships than North Carolina, to the point that every year some Ohio villages intentionally disband in order to assume township government. It's possible that North Carolinians saw articles on these "strong" townships and thought they could do the same for their state, unaware of the subtle distinction. Throughout these discussions, I've been encouraging people to consider notability of a class rather than notability of an individual subject, to avoid this kind of confusion, but I don't think the message has gotten through. As a counterpoint, the former Storrs Township, Hamilton County, Ohio, started out as not even a list item – just a redirect to an inaccurate mention buried in some other article. But now I'm quite confident that it establishes notability, ironically due to how everyone had forgotten about the place, over and over again! Minh Nguyễn ^💬 06:38, 24 September 2023 (UTC)

  • I'm not sure what you mean when you refer to the "notability of a class rather than notability of an individual subject". If anything, that seems to be the exact problem here. People assumed all US township articles are notable, and then created a handful of unsourced/primary stubs for NC and redlinked the other 900 or so. See right now at Wikipedia:Articles for deletion/Averasboro Township, Harnett County, North Carolina where a user is arguing that since a township was included in a local accommodations tax district and the census bureau records how many people there are, it is notable. -Indy beetle (talk) 07:48, 24 September 2023 (UTC)

    The same thing happened in the discussion that led to this RfC, where at one point there was a thought to whitelisting or blacklisting such overloaded terms as "township", unqualified. People need to understand that "the class" is not civil township, it's "civil township in North Carolina". A notable member of this class is an exception to the rule, whereas a notable township in Michigan is the rule. At best, proving notability for each individual Michigan township would be makework, leading to articles written for an audience of notability appraisers rather than the general public. Minh Nguyễn ^💬 23:23, 24 September 2023 (UTC)

• Option 2 Even if you think we're a gazetteer (we're not, he have have features of one), database-style content just giving existence of a place and its location can be provided in lists, rather than stand-alone articles that lack coverage beyond the database. Reywas92^Talk 14:12, 17 September 2023 (UTC)
• Option 1—as many have said there, this change would substitute some vague GNG-based guideline to solve a problem the wrong way. GNG is supposed to be a default rule for notability when a subject-specific guideline doesn't exist, or for cases that demonstrate notability another way that isn't captured by that SNG. Instead, there seems to be a drive to go around actual policy by subordinating all notability guidelines to GNG, to make them all subject-specific applications of it, which violates the spirit of the original rules. The current SNG at discussion here already works well enough on its own and does not need to be changed. If the issue is mass creation of articles, directly address that by limiting article creation. If the issue is stubs, and I will disagree that stubs are a bad thing, then address them individually through editing processes like proposed mergers and redirection. Imzadi 1979 → 14:33, 17 September 2023 (UTC)
• Option 2 I don't think that the current SNG works globally: in many cases, it's impossible to tell whether a place is legally recognized or not (per Davidstewartharvey). The proposal is not based on GNG. It's based on the long-standing notability criterion that is now at WP:GEONATURAL. That SNG has worked well for years --- I think it's applicable to places (after suitable modification achieved via discussion), and it will be better than what we have today. — hike395 (talk) 17:00, 17 September 2023 (UTC)
• Option 1. Seeing as the proper procedure for deletion will likely not be taken when mass-deleting articles that help keep Wikipedia a comprehensive geographic resource rather than a dulled directory of only the largest cities, it is better to maintain the status quo and have an actual discussion with people knowledgeable in the field. Frankly, the changes proposed reek of a Western worldview that runs counter to this project's goal to acknowledge and reduce systematic bias; many countries do not have such a rigid view of what is and is not a settled place. SounderBruce 20:13, 17 September 2023 (UTC)

  However if we keep the current rule, many parts of the early do not have a formal process of what is Populated, legally recognized places. Census data is not actually legal recognition in many parts of the world, so those villages, towns, administrative regions would then have to meet GNG under the current rules, which would then be a fail without WP:SIGCOV. Davidstewartharvey (talk) 21:18, 17 September 2023 (UTC)

• Option 1 some of the reliable census provide significant data to enable a start class article. Where there is hardly any information but what there is is verified in a reliable census or other reliable source the option of creating a list instead of seperate articles is a viable alternative to deletion providing no reliably sourced information is lost. Also WP:NOTSTATS is irrelevant to this debate as what it refers to is uncontextualised statistics and confusing lists whereas census information is obviously in context for the specific place and can be put into simple prose to avoid confusion, imv Atlantic306 (talk) 20:58, 17 September 2023 (UTC)

  Policy explicitly prohibits census data from contributing to notability because it is primary data. JoelleJay (talk) 00:30, 18 September 2023 (UTC)

  If this is true, I haven't seen any clear evidence for it. For example, WP:GNG states that "sources should be secondary sources", but the masthead at WP:N insists on reliable and independent sources without discounting primary ones. And WP:PRIMARY number 5 is not a comment about Notability; in fact, it is part of a list that is bounded by competing P&G considerations and does not represent a blanket prohibition whatsoever. Newimpartial (talk) 01:51, 18 September 2023 (UTC)

  Secondary or tertiary sources are needed to establish the topic's notability and avoid novel interpretations of primary sources.

  Further examples of primary sources include: archeological artifacts; census results...

  Do not base an entire article on primary sources, and be cautious about basing large passages on them

  Creating an article based entirely on census data is precisely what is being prohibited here. JoelleJay (talk) 03:05, 18 September 2023 (UTC)

  That third pullquote is the exact one I addressed (and contextualized) as WP:PRIMARY number 5. The first pullquote is in a paragraph that opens Wikipedia articles should be based and itself should, I believe, be interpreted with the same caveat for competing P&G considerations. I do not believe the community interptets this passage as independent primary sources being prohibited from contributing to Notability in the way you mean (although the GNG insists on a degree of secondary sourcing within its domain). Newimpartial (talk) 10:05, 18 September 2023 (UTC)

  You did not "contextualize" that quote with anything that changes its meaning. It doesn't matter whether it explicitly says anything about notability (especially since elsewhere the policy does state this) because notability directly follows from it. The prohibition is on basing articles on primary sources; if primary sources could be used to establish notability then they would necessarily have to be permitted when they are the only sources. And even if we took your tortured additive interpretation of SIGCOV, an article that is based on primary sources is still violating policy regardless of whether the topic is notable. JoelleJay (talk) 20:08, 18 September 2023 (UTC)

  I contextualized it by pointing to the phrase preceding it which says, essentially, that policy-based exceptions apply to all the bullets including the one quoted. This would be one of those policy-based exceptions.

  And the fact is that SIGCOV applies within the scope of GNG, and is therefore not directly applicable to this discussion. I am not sure why you are bringing it up. Newimpartial (talk) 21:23, 19 September 2023 (UTC)

• I would consider making the rules of presumed notability a bit more lax but otherwise I support. Crouch, Swale (talk) 21:50, 17 September 2023 (UTC)
• Option 1 per VaticidalProphet. The existing guidance is sufficiently clear as to what we'd like to include, and our sourcing guidance should be sufficient with respect to addressing specific instances of low-quality content. If it ain't broke... — Red-tailed hawk _(nest) 00:19, 18 September 2023 (UTC)

  I would say the fact we've recently had to delete thousands (tens of thousands? @FOARP?) of stubs on "villages" that were created over the last decade due to editors simply using the current guideline is evidence something is broke. And it's not just a problem of mass creation unless you consider everyone who uses government censuses as the bases of articles to be "mass creating". JoelleJay (talk) 00:37, 18 September 2023 (UTC)

  If someone is misreading a source, en masse, and mass-creating articles a-la-Iranian "village" stubs, then yes, that's a source-text agreement issue. That isn't an issue with the guidance here; that's an issue with one's understanding of a particular country's municipality system. — Red-tailed hawk _(nest) 00:47, 18 September 2023 (UTC)

  If the guidance says "populated, legally recognized places" and a government happens to classify gas stations as "populated, legally recognized places", then that's an indication "populated, legally recognized places" is too ambiguous. We had to add abadis to the guideline after the Iran fiasco; how many other such designations exist that would still qualify based on the current wording? JoelleJay (talk) 01:01, 18 September 2023 (UTC)

  Do you have any evidence that any government does classify gas stations as "populated, legally recognised places"? Unless you do then this is just unhelpful hyperbole. Even if it is true, then it is entirely possible to merge mention of said gas stations to a larger article without needing to change any policies or guidelines, let alone make such potentially disruptive changes as are proposed here. Thryduulf (talk) 01:19, 18 September 2023 (UTC)

  "Do you have any evidence that any government does classify gas stations as "populated, legally recognised places"? - Are petrol stations populated? Well, yes, since apparently the Iranian census counted people living at them. Is an Abadi a legally-recognised place? Well this rather hangs on what you think a "legally-recognised place" is since it is not actually a concept that exists outside Wikipedia, but the standard preferred by many is "mentioned in government documents/legislation", which abadis are.

  "it is entirely possible to merge mention of said gas stations" - are you seriously proposing to merge the fact that town X has a petrol station with a population of Y?

  PS - let's take a recently-created example, not something by Carlossuarez46 or the other historical mass creators - Babinac, Ivanska. Can you tell me how this is a "legally recognised populated place" if Agro-Industry Complex was not? And then consider that 56 "village" articles were created by the same creator and on the same day as Babinac, Ivanska, a day in May this year. FOARP (talk) 07:31, 18 September 2023 (UTC)

  This is tangential, but it's not unheard of for people to maintain permanent residences in the same structure as a business they operate. I saw this all the time when I lived in China; Iran may have a similar tradition. I'm not arguing that a business domicile like a petrol station or general store deserves any mention on Wikipedia, but it is possible for them to be "populated places". I've also "lived at work" (distinct to "working from home"), but never for more than a month, and usually when I also had an actual separate home with an address. Folly Mox (talk) 18:09, 20 September 2023 (UTC)

  I “lived at work” when I lived in a factory dormitory whilst living in China (Foxconn in Longhua, Shenzhen, but also the university campus when I studied Chinese whilst teaching). I think anyone can see that these are not “legally recognised, populated places” just because a census counted people living at them.

  Living in farms and shops is pretty common around the world, not just in Iran/China. FOARP (talk) 20:26, 20 September 2023 (UTC)

  I think anyone can see that these are not “legally recognised, populated places”. I don't think you can speak in such absolute terms. What is the line between single isolated farms and a dispersed settlement of farms? We should always follow the sources - if they describe somewhere as a populated place then we should treat it as a populated place. The proposal would make it much easier to delete encyclopaedic content about the place when (a) a stub article is not actually harming anyone or anything, and (b) nothing in the current guidelines prohibits merging when appropriate. Thryduulf (talk) 20:52, 20 September 2023 (UTC)

  @FOARP I'm glad you mentioned Babinac, Ivanska because I just visited the article and many alike with my bot. That is a third-level administrative division place in Croatia. People would indeed say: "I was born in Babinac, I went to school and church there, my grandparents' grave is there." I thought not much more could be said about the village, but I see the Hungarian wiki has quite some text. Also, out of 6700 naselje/settlements, some 3000+500 have been created so far. Are we gonna say to the rest they are not allowed to join any more? There should be a (reasonable) limit, for some countries a third-level division, for some 4th or 5th, but that should be decided per country. Some administrative divisions have a long history, and some, like these industrial complexes or whatever, do not. Ponor (talk) 09:30, 30 September 2023 (UTC)

  The problem with basing notability on what legal status a country gives a settlement is it can very widely from country to country and is very hard to interpret. Instead we should base our coverage on what there actually is that can actually be said about that settlement based on reliable, secondary sources. FOARP (talk) 09:42, 30 September 2023 (UTC)

  I don't disagree with you, @FOARP, ideally everything would be based on secondary sources, and our articles would have many words – if there were enough people to write them, and in most countries, I'm afraid, there are not. So now we're in a situation that over the last 20 years some 50% of third-level division articles (many, many stubs) have been created. Stubs are not much information, but are nevertheless more than zero information, which should be good. I don't see much difference in having one sentence, coordinates, population (trends), pushpin map, photo shown in a separate settlement article vs. having many of them listed like that in one (of their municipality, for example). That wouldn't even save us any storage space. If criteria are abused for some countries, then those countries should have additional criteria (like: would anyone say I was born in Agro-Industry Complex?) I had a short conversation with someone recently, and we (sort of) thought that (mass) creating something like Dubrava, Split-Dalmatia County would not be such a bad idea, given that some uniformity could be achieved, as opposed to "a mess" that has emerged over the years. I am "a few clicks" away from creating such articles, and would hate to miss the opportunity, but I'm quite sure many of those would never be expanded in any foreseeable future. That bot work would also save those two guys who were recently creating Croatian stubs quite some time, which they could use for some other work. Mixed feelings... Ponor (talk) 11:32, 30 September 2023 (UTC)

  I would object to the generation of one line stubs like that Ponor. Give them a few lines of sourced prose and I would support it though. xxx is a village, population xx can easily be presented in a list of villages by district. You're right though that the bot is more ideal for articles such as this and has a lot of potential. I tried to get FritzpollBot coded in 2008 which would mass create articles for everywhere but it never got the go ahead. ♦ Dr. Blofeld 16:50, 4 October 2023 (UTC)

• Option 1 per Joe, Thryduulf and VaticidalProphet. As appears above there are many subtexts to the proposal, none of which seem to be actually beneficial and some actively harmful. There is nothing intrinsically wrong with stubs, and as AfD shows, with a little effort GNG can be made to reject almost anything, in the relevant hands. Per Red-tailed hawk, as above, If it ain't broke..., except that I do think that the concept of legally recognised place is very unhelpful and should be revised / removed but that doesn't go near to justifying what is proposed here. Ingratis (talk) 01:01, 18 September 2023 (UTC)
• Option 2 No, we should not have an article on every 30-person village that is listed on the census. In-depth sourcing is essential to writing an actual encyclopedia article and without it a stand alone article should not exist. (t · c) buidhe 01:09, 18 September 2023 (UTC)
• Option 1 The single phrase provided information about them beyond statistics, region, and coordinates is known to exist. from the proposal is enough to make my hair stand up. If their are sources, just give them. Do not assume that there are sources. The Banner talk 01:35, 18 September 2023 (UTC)

  @The Banner, how is the proposal worse than the existing guidance Populated, legally recognized places are typically presumed to be notable, which permits an article even if it's known that no such sources exist? JoelleJay (talk) 03:11, 18 September 2023 (UTC)

  Return question: would you write an article about a place when you can not prove that it exists? Every editor in wiki-world would be permitted to ask for sources and they have to be given. The new proposal does lower that standard of sourcing. The Banner talk 10:34, 18 September 2023 (UTC)

  @The Banner What?? Under the current guideline, there is no requirement for sources containing anything beyond stats etc. An editor asking someone to provide such a source would be denied outright. The proposal, like the current guideline, expects that a stats/coordinates source that verifies a location's existence is already cited in the article; what the proposal asks is that for determining notability editors should additionally consider whether non-stats sources exist before creating the article. JoelleJay (talk) 20:22, 18 September 2023 (UTC)

• Option 1 is fine. The new proposed text is poorly written (what does it even mean for "information [to be] known to exist"?) and unclear. The current guideline provides some guidance about relevant criteria (population and legal recognition). The proposed text would replace that with a series of interpretational landmines which would only add to confusion and pointless endless debates (what is "known", "very short", "enough content"?). Moreover, by saying that "This criterion applies" etc., the new text would make all other places fall back to GNG, including places which the current guideline mentions: is that intentional? Nemo 04:49, 18 September 2023 (UTC)
• No changes - I don't like the idea of being told I only have two options as it puts the person suggesting it in a position of power over me. No changes need to be made. We're also having what seems to be monthly discussions around here where hundreds to thousands of articles risk deletion if new rules are made, and that needs to stop. KatoKungLee (talk) 12:18, 18 September 2023 (UTC)
• Option 1. When it comes to locations that exist (or have existed in the past), we should have an article on it to maintain our role as a gazetteer. We should presume notability on very few things, but places people live is one of them. Census data is enough, and if the census data is wrong, we can assess it on a case-by-case basis. Anarchyte (talk) 14:13, 18 September 2023 (UTC)

  But the current guidelines does actually cover that! Census data does not actually mean they are a legally recognised place. So therefore if we cannot prove they are legally recognised, we have to go to the second point of GEOLAND which means they have to meet GNG, where census data is clearly not useable to establish notability under wiki rules.

  The proposal means all Settlements and Administrative Regions can have an article if it is more than just census data, if not if can be put into the next higher area. They won't be losses, and with redirects, if more refs become available we can they reverse the situation. Davidstewartharvey (talk) 16:18, 18 September 2023 (UTC)

• Option 1. Wikipedia was, in a sense, built on geography articles citing database sources; one of the site's largest sources of content early on came from Ram-Man's articles on U.S. places based on census data. As Thryduulf said early on, these articles still provide useful info to readers, and I'm strongly opposed to a proposal to let users merge them without even trying to look for more sources (as the link to WP:BURDEN implies). TheCatalyst31 ^{Reaction•Creation} 17:39, 18 September 2023 (UTC)
• Option 1 per vaticidalprophet, Imzadi, SounderBruce, and TheCatalyst 31. --JackFromWisconsin (talk | contribs) 18:51, 18 September 2023 (UTC)
• Option 1 per Sohom, and SportingFlyer We already have a WP:SYSTEMIC bias problem that promotes the views of the scholarly West (particularly the Anglosphere) over everyone else. Option 2. and all these other attempts to get rid of our WP:SNGs will only exacerbate that problem. Typing an Indian town's name into Google and getting 0 results does not mean that secondary sources don't exist for it. The WP:NSPECIES proposal is problematic too. From 2007 to 2017, two editors mass-created articles on Lepidoptera. Most of them are still stubs, but a single editor has expanded most of the species endemic to New Zealand to the point where anything other than a !Keep outcome is impossible. Of course, most people from New Zealand speak English. An editor who speaks Portuguese probably could expand the stubs related to the Lepidoptera of Brazil. I am applying the same logic to settlements because I do not know of any settlement in the Anglosphere that genuinely isn't notable. Scorpions1325 (talk) 03:38, 19 September 2023 (UTC)
• Option 1 but I do see a problem with the "legally recognized populated place" criterion. Performing a Google search got few hits — it appears to be a term confined to Wikipedia. Perhaps an attempt for a clearer definition would help, however the proposed changes to the wording sought here do not. What I anticipate should option 2 gain consensus is deletions of many village articles in countries where English isn't the prime language and internet derived sources in English are not easily found. I base this on the lack of support for Indian and Nepalese schools at AfD. I'm not saying this is the intention of the proposed change, but it is a likely outcome and so restricts the encyclopedia's gazetteer function. Rupples (talk) 04:34, 19 September 2023 (UTC)
• Option 1 option 2 would explicitly recognise that administrative regions are presumed to be notable. The current wording doesn't explicitly say this and isn't usually interpreted that way. While the conventional high level subdivisions of a country are likely to be notable, low level subdivisions usually aren't unless they coincide with settlements. I'm sympathetic to the rest of the proposal but this part could actually open the door to more low-quality geographical articles being created. Hut 8.5 06:20, 19 September 2023 (UTC)
• Option 1 per Vaticidalprophet, Thryduulf, and Mxn below. Ed ^{[talk] [majestic titan]} 02:03, 20 September 2023 (UTC)
• Option 1 - largely following Vaticidalprophet but also concerns raised about WP:5P1: Wikipedia has an important function as a gazetteer. I also find the prose in the status quo more comprehensible. Suriname0 (talk) 15:28, 20 September 2023 (UTC)
• Option 3. A subject is notable if it has been extensively covered in reliable and independent sources (seems we've already got that somewhere). That is the point at which we should have a full article on it, otherwise there is nothing from which to write such an article. That does not, of course, mean that we cannot say anything at all about it absent that, but if all we know about some "populated place" are a few database factoids, include that in List of populated places in Example Administrative Region, not a separate article. That would still let us fulfill any "gazetteer" function, and indeed, that would look much more like a gazetteer than the current situation does. Seraphimblade ^{Talk to me} 22:22, 20 September 2023 (UTC)

  This is already possible with the status quo so there is the same as option 1. Thryduulf (talk) 22:26, 20 September 2023 (UTC)

• Option 1 this is a solution in search of a problem. If there is no consensus to merge settlement stubs, they don't need to be merged. Eluchil404 (talk) 03:05, 21 September 2023 (UTC)
• Option 1 sufficiently articulated by Vaticidalprophet above that I don't need to go repeat the same thing here. RecycledPixels (talk) 17:16, 21 September 2023 (UTC)
• Option 3. Suggest: Legally-recognized populated places are generally considered to be notable and can generally be written about, even if most of the text is based on map observational data, but the creation of short stubs where there is nothing but a database mentioning location online is strongly discouraged. The minimum requirement expected is a population figure or/and some basic facts (former settlements may not have population but other dat a for instanve) and basic location details as a precondition for a stub (like Aung Myay), where the stub can be fleshed out to resemble an encyclopedia article instead of an "xxx is a village" type database substub. In cases where there is no population or other data and it resembles a database, consider merging into a tabled list.

  @Dr. Blofeld Your bolded !vote indicates you support option 2, but your comment indicates you support option 1. Thryduulf (talk) 12:32, 22 September 2023 (UTC)

  I don't believe in "inherent notability", but you'll find that you can write a fleshed out stub for most places. Aung Myay is an example which has barely anything online but clearly worth keeping. I think we should strive to have a minimum like that for all populated places, but if editors can't find anything to write about it then we should redirect to a tabled list. I see it as more an organizational rather than notability issues. I believe most populated places are notable and potentially will have content written about them eventually, but it's eradicating the empty xxx is a village stubs in cases where they can't be fleshed out. ♦ Dr. Blofeld 12:39, 22 September 2023 (UTC)

  A requirement for a population figure and basic location details might be a good idea as a precondition for a stub like that Aung Myay article... ♦ Dr. Blofeld 12:41, 22 September 2023 (UTC)

Thryduulf I've switched to my proposed Option 3 then, which I think is stronger than the others..♦ Dr. Blofeld 13:00, 22 September 2023 (UTC)

• Option 2 Proposal provides multiple examples of mass-created articles that rely on existing GEOLAND wording as justification. Even if you agree that Wikipedia is a gazetteer, that fails to explain why we need separate stubs for sparcely populated areas when their census and location data could be tabulated on the article for a higher level of administration. The new wording deemphasizes legal recognition to focus on significant coverage, in line with GNG standards BluePenguin18 🐧 ( 💬 ) 22:15, 23 September 2023 (UTC)

  As multiple people have noted above, nothing about GEOLAND requires separate stubs nor prevents merging to broader articles (or creating them as sections on a broader article in the first place). The proposal will however make it much easier to delete content instead of merging it. Nothing about the proposal will make mass creation easier or harder than it currently is, so it will not solve anything in that regard. Thryduulf (talk) 22:23, 23 September 2023 (UTC)

  But GEOLAND is ALWAYS cited as the reason for having these stubs, and when any effort to redirect stubs gets reverted. FOARP (talk) 09:31, 25 September 2023 (UTC)

  If any effort to redirect stubs gets reverted, then consider perhaps trying getting consensus to merge the content. Alternatively consider that if your actions with regards to these articles are always reverted that it might be your actions rather than the articles that are not in accordance with the wishes of the community. Thryduulf (talk) 09:41, 25 September 2023 (UTC)

• Option 1 - Changes may be needed, but those suggested in Option 2 ain't it. Beyond My Ken (talk) 01:41, 25 September 2023 (UTC)
• Option 1 per many above. I'm not seeing these articles as particularly problematic, and if we were to make a change I would rather start with upping the standards for new articles. ϢereSpielChequers 13:03, 25 September 2023 (UTC)
• Option 1. "... enough verifiable content for an encyclopedic article" in the second option seems to lead us immediately into whether or not we like the article. Also the problem in option 1 is not removed, namely that merging into lists can fail because sometimes a higher level entity can receive less coverage than lower level and doctrinaire deletion can result. Thincat (talk) 19:12, 25 September 2023 (UTC)
• Option 3 as proposed by Dr. Blofeld looks best to me. It's clearer and more concise than either the current or the proposed GEOLAND. Additionally, it addresses the issue of places for which only database information can be found. WP:CSC would certainly seem to allow the types of lists suggested under it. - Presidentman talk · contribs (Talkback) 21:43, 25 September 2023 (UTC)

  I do not think that proposal would be suitable for a former village. A medieval, roman or pre-roman village should be the subject of extensive archaeological and historical literature, but could in theory have no total population figure, not even an estimate. I do not see why we cannot base an article on the details of an archaeological excavation, or a book like The Lost Villages of England. I suspect that there might not be a total population figure for a depopulated medieval village, because there was no census at that time. I suspect that you are far more likely to find the number of taxpayers or tenants, the quantity and usage of farmland and buildings etc, the financial value of the land (eg the rent or tax value), or even the number of ploughs and farm animals. I suspect that the medieval authorities were more concerned about collecting rent and taxes than with adminstering the kind of social legislation and public services that require knowledge of population. And they had a very high rate of infant mortality. The poll taxes of 1377, 1379 and 1381 only applied to people over 16, and Beresford says the tax records of 1377 were the only thing close to a census before the chantry returns of 1545 (Lost Villages, 1983 and 1987 reprint, pp 286 to 288). James500 (talk) 03:44, 26 September 2023 (UTC)

  Or, to cite a more recent example: the tabled list in List of towns and villages depopulated during the 1947–1949 Palestine war comes with helpful context and historic population figures, but the horizontal scrolling and unwieldy references list demonstrate why a list is not always a superior reading experience to prose. (It's unclear to me whether the proposed "option 3" allows for historic population figures, but maybe that's the intent.) Minh Nguyễn ^💬 06:20, 26 September 2023 (UTC)

  I don't read option 3 as precluding "bas[ing] an article on the details of an archaeological excavation, or a book like The Lost Villages of England." For historic places, GNG would still apply. Presidentman talk · contribs (Talkback) 14:29, 26 September 2023 (UTC)

  I agree. There would be a different requirement for former settlements. Population isn't compulsory, my point was that the new articles should have some data and location info about it, bare minimum.♦ Dr. Blofeld 08:49, 4 October 2023 (UTC)

• Option 2. The current text is far too permissive and has led to the creation of huge numbers of articles that clearly fail WP:GNG. Mere statistical data does not make a useful encyclopaedia article, and we don't need directory entries for every single legally recognised place (per WP:NOTDIRECTORY). While I could quibble with parts of the proposed wording, it is a huge improvement over the status quo and I find the arguments for option 1 unconvincing. Modest Genius ^talk 11:04, 27 September 2023 (UTC)
• Option 2. The WP:GNG ought to be a hard minimum and no article, under any circumstances, should exist without the two sources it requires. Creating articles from raw data is essentially speculative in nature, opens us up to the common problems with such sources, inevitably involves speculation and WP:OR by editors, and should be completely and unambiguously forbidden in all circumstances; but it is particularly absurd to suggest that it could ever be acceptable to use raw data as the sole source for an article on an entire locale. If (as some people have speculated) a particular locale has an actual history and sources covering it, it ought to be easy to find them; and the requirement to do so will prevent mass-automated article creation, which is something we ought to be strongly discouraging anyway in this era of LLMs and other tools that will flood the encyclopedia with junk pulled from low-quality databases or worse if not stopped. People need to be willing to stop, slow down, and take the time to actually make articles by hand; I can understand the desire to use a Wikipedia article as a badge of importance or significance, or for people to feel like they are contributing large amounts, but the value in these things can only come from the actual human work and effort necessary to find and validate sources. People who pour databases directly into the wiki as raw otherwise-unsourced articles are doing serious long-term harm to our reputation. --Aquillion (talk) 09:50, 28 September 2023 (UTC)

  I think it's fair to expect an article to eventually involve enough secondary sources to satisfy GNG regardless of the subject matter. However, notability guidelines like GNG and GEOLAND are very often being applied much earlier, at the time someone first encounters an article, by inclusionists and deletionists alike. One doesn't need to use an LLM to generate geographical articles based on a geographical or historical database. Structured data makes it easier and cheaper to generate text with simple heuristics, as Rambot did so many years ago. But if the motivation behind this change is to prevent mass-creation, then why does the change apply equally to someone making a single article by hand, one step at a time? Minh Nguyễn ^💬 17:18, 28 September 2023 (UTC)

  I would just point out that neither the WP:N system as it exists in policy, nor the current set of AFD outcomes to date - nor even Option 2 as proposed here - requires GNG as a hard minimum for article creation of retention. While I recognise that Aquilion is describing what ought to be rather than what is, the fact is that enacting such a proposal as this !vote recommends would be a major change to the way the project currently works, and is decidedly over the skis of the current RfC.

  Also, the current proposal is not tightly targeted at automated or bot-like article creation, and represents a concern quite tangential IMO to those surrounding bot-like editing (speaking as someone opposed to bot-like editing in toto). The idea that two independent, reliable sources are required to document that a place exists and is officially recognised and populated - the long-standing criterion for GEOLAND - does not seem plausible prima facie and would represent a substantial departure from long-running consensus while serving no clear purpose for the encyclopaedia. Newimpartial (talk) 19:10, 28 September 2023 (UTC)

• Option 2, there absolutely must be something more than being mentioned in a database or listing or appearing on a map for creating a stand-alone article. Contrary to what many above have mistakenly claimed, Wikipedia is NOT a gazetteer -- although it DOES include some features of a gazetteer. Mere existence of a place is never adequate for a stand-alone article. Of course, such places can be mentioned in other articles with applicable redirection, such as to an article for a larger area or higher administrative area. And there may be cases (now relatively rare) where some limited presumption of notability might apply for certain categories of places. But these should have clearly defined criteria, not a vague Jedi hand-wave of 'legally recognized' AND this 'presumption of notability' means that notability can be established if effort is expended. As a corollary, I would support allowing challenges to such 'presumptions of notability' for specific articles that have seen no improvements. older ≠ wiser 10:32, 28 September 2023 (UTC)

  All this can be achieved without any changes to current policies or guidelines. Thryduulf (talk) 11:12, 28 September 2023 (UTC)

  Perhaps it could be achieved. But as it stands currently, there are perennial problems that continue to recur. I think something needs to change. older ≠ wiser 11:24, 28 September 2023 (UTC)

• Option 2 it'd be redundant to restate all the reasons already articulated above, but Option 2 better aligns with our goals (we have elements of gazetteers, we are not specifically one; we try and focus on secondary sourcing and multiple references to meet V and GNG; geostubs, especially mass-created ones, have long been demonstrated to be flatly wrong and error-prone in a way more carefully considered and created articles are not, et al.) Der Wohltemperierte Fuchs ^talk 12:08, 28 September 2023 (UTC)

  There are so many newspapers, so many geography, history and architecture books and periodicals, so many historical records, and so many archaeological publications dealing with villages that it appears on the face of it to be very, very, very obvious that villages do generally satisfy GNG. So far, no actual evidence has actually been presented of any actual village that demonstrably does not satisfy GNG. Not one. No one has said, for example, "I did a WP:BEFORE search of all the relevant newspapers, and could find nothing that could be considered significant coverage of this village".

  The Iranian census articles, and others like them, are a red herring. None of those articles was actually about a village. Not one. If we have a problem, it is with editors using unreliable sources, or with editors mistranslating sources. This proposal will not do anything to fix that, and may actually make it worse. James500 (talk) 16:40, 28 September 2023 (UTC)

  I think this is missing the point. The WP:BURDEN lies with those adding the information. And it is actually quite decidedly NOT very, very, very obvious that villages do generally satisfy GNG -- else we would not be having this discussion (and very many other recurrent similar discussions). older ≠ wiser 17:31, 28 September 2023 (UTC)

  WP:BURDEN says nothing about GNG or presumptions of notability in SNG. WP:ARTN says in express words that GNG has nothing to do with the "referencing within a Wikipedia article" and WP:BEFORE says that a search for sources must be conducted before claiming that topic does not satisfy GNG. That said, an enormous amount of positive evidence has been physically produced onwiki that villages satisfy GNG, both in many articles, and in many community discussions and talk pages. Conversely, no one has ever produced a single counter example of a village that does not satisfy GNG, let alone a single counter example of a country or other region that has no such abundance of books, newspapers, periodicals, publications or records. There must come a point where the volume and strength of evidence that villages satisfy GNG, and the total lack of contrary evidence, means that it has become objectively obvious that villages generally satisfy GNG. And whether that is objectively obvious has nothing to do with the subjective perceptions of any individual, and therefore nothing to do with the reason we are having this discussion. James500 (talk) 19:40, 28 September 2023 (UTC)

  Ah yes, the old canard of extreme inclusionists that shifts responsibility for improving an article away from the ones who create the crap in the first place. older ≠ wiser 20:36, 28 September 2023 (UTC)

  Except that it isn't. I am no more happy than you about the creation of articles on subjects such as Iranian census "abadi" or GNIS railway sidings where nobody ever lived (which is the sort of thing that seems to have precipitated this discussion). But the point is that none of those "crap" articles that you refer to were actually about villages. Because those "abadi" and railway sidings etc are not villages. GEOLAND never authorised the creation of any of those articles, even in its original form, because they are not "populated places". James500 (talk) 21:11, 28 September 2023 (UTC)

  No, I was not thinking of either abadi or railway sidings. The part I find most problematic in GEOLAND is the infuriatingly vague notion of "legally recognized". It means that just about any place that appears in any sort of government sanctioned document is thus legally recognized, which is a load of bollocks. 01:26, 29 September 2023 (UTC) older ≠ wiser 01:26, 29 September 2023 (UTC)

  There is overwhelming long standing consensus that the purpose and effect of GEOLAND is to create a presumption that villages are notable. The fact that a place appears in a "government sanctioned document" absolutely does not make it a "populated place" that is a "settlement or administrative region" within the meaning of the present wording of GEOLAND. I do not believe that anyone in this RfC is trying to claim that "populated place" includes any place that appears in a "government sanctioned document" or anything like that. I do not believe that such a claim would have any chance of gaining any traction whatsoever in any community discussion at AfD or anywhere else. James500 (talk) 12:29, 29 September 2023 (UTC)

  And yet, there continues to be recurring extended discussions regarding various divergent interpretations and abuses of the supposedly clear language in GEOLAND. older ≠ wiser 14:29, 29 September 2023 (UTC)

  A strawman interpretation is not a divergent interpretation. James500 (talk) 21:27, 30 September 2023 (UTC)

  And what is the strawman (I haven't bothered to respond to your insertion of irrelevancies -- if that is what you meant, you have created your own strawman). Or are you suggesting that all of the voluminous discussions on GEOLAND that present interpretations you do not agree with are strawmen? older ≠ wiser 02:08, 1 October 2023 (UTC)

  Agree the idea that the issues with GEOLAND are just "strawmen" really doesn't fly. We have regular, extended debate about whether the following constitute "legal recognition":

  • Mentions of locations in official documents - over the years I've seen people cite brief mentions of locations in geological reports, weather reports, accident reports, minutes of state-senate proceedings, etc. at AFD as evidence of legal recognition.
  • Listing of locations in census statistics (including locations with zero population). Censuses simply aren't intended to a be a list of legally-recognised communities, they are just used as a way of counting the population.
  • Mention of locations in official Gazetteers (including those with systematic errors - which in reality is all of them though GNIS and GEONet Names Server are the only ones where it's recognised at present).
  • Postal addresses. I've regularly seen it asserted at AFD that having a postal address constitutes "legal recognition".
  • Post offices, including temporary post offices.
  • Railway stops.
  • Listed/protected buildings within the community.

  At heart legal recognition is problematic because it outsources notability assessment to national legislation that varies wildly from country to country and does not coincide with actual coverage. FOARP (talk) 09:43, 1 October 2023 (UTC)

  Can you provide diffs or links to support these examples? I need to know, in each case, what kind of place the article under discussion at AfD actually was. If the place was a village, the precise form of legal recognition is completely irrelevant, because villages generally satisfy GNG. If, on the other hand, the place was something like, for example, a solid waste disposal district, no amount of legal recognition would make that district satisfy GEOLAND, because such a district is simply not a populated place. James500 (talk) 19:24, 1 October 2023 (UTC)

  Sure:

  • In this AFD part of the keep argument was based on the location being included in a postal address: Wikipedia:Articles for deletion/Al-Nuwaiseeb. The result was no-consensus so the article was kept.
  • In this AFD part of the keep argument was based on a weather report, and a website that referenced the place having a supposed gay district: Wikipedia:Articles_for_deletion/Kittlers, Arkansas. The article was deleted.
  • In this AFD evidence of a gas-condensate field and of legislation renaming locations was presented as evidence that the article was a GEOLAND pass: Wikipedia:Articles_for_deletion/Kirpichli,_Turkmenistan_(2nd_nomination). There remains no actual evidence of a population at the location, no buidlings can be found at any of the locations given for it. The discussion was closed as no-consensus and the article kept.
  • In this AFD the keep argument was based on the minutes of a senate meeting, a post office, and various bare-mentions of the place-name in news stories: Wikipedia:Articles for deletion/Emmet, North Dakota. The close was no-consensus.

  Let me know if there's a specific example you're interested in. FOARP (talk) 17:54, 3 October 2023 (UTC)

  Well, if one considers any discussion in RS of a building or person or event in a given location to be "direct and significant" coverage of the location, regardless of whether the location itself is merely named in passing, then perhaps GNG is "very obviously" satisfied... Especially if you ignore the requirements that a source must be secondary, independent, and non-routine. JoelleJay (talk) 17:40, 28 September 2023 (UTC)

• Option 1 I fully get the reasoning behind this proposal. And I generally don't object to a merge (an actual merge, not a redirect) when all we have is very basic stats and no real coverage. But at the least we need the redirects and at the most having stubs gives us a much better chance of useful things being added when sources appear or are otherwise found. Hobit (talk) 13:21, 28 September 2023 (UTC)
• Option 1. This is an area where the use of sourcing-based notability standards really does create a significant Wikipedia:Systemic bias that our current standards avoid. We should not throw that away chasing something (the mass-stub-creation problem) that is not really about notability. —David Eppstein (talk) 06:46, 30 September 2023 (UTC)

  @David Eppstein - if you're going to assert systematic bias, that really does have to be evidenced. It is trite to say "people who write in English will find secondary sources harder to find when not written in English", but it is just as valid to say that the present standard massively biases our coverage towards those countries that have easily-accessible electronic databases. From a little study (I clicked on "random" 30 times and studied the one-sentence Geostubs found) by far the biggest beneficiaries of the present standard are the more developed countries of Central and Eastern Europe (particularly Poland, Slovenia etc.), because they tend to have very granular statistical units in their censuses. The imagined benefit to the poorest countries of Africa and Asia simply doesn't exist - many of these countries don't even take censuses. FOARP (talk) 21:01, 30 September 2023 (UTC)

  As far as I can tell, every country in the world has had at least one census from 1932 onwards. All but seven have had at least one census from 1990 onwards, and of those, one is about to have a census: [4] [5]. James500 (talk) 01:06, 1 October 2023 (UTC)

  I don't think CE and EE countries are more granular (most follow the Country>County>Town>Settlement division). As a counterexample, take Poquott, New York or Belle Terre, New York, neither of which have their school, church, grocery store, library... it's 2½ streets and lots of trees. That'd be (Country)>State>County>Town>Village level for the US. The two articles are not stubs, but are poorly sourced, if at all — even their census references are like www.census.gov/**nothing**. Would we keep them or merge them into, IDK, Port Jefferson, New York (where they go for coffee) or Brookhaven, New York? Ponor (talk) 09:16, 1 October 2023 (UTC)

  People on here often cite excessive US coverage as a "gotcha". This is predicated on the idea that the other editors in the discussion are Americans (I am not) and that they approve of the examples of excessive coverage in the US (I do not). Actually the US is the hardest hit by mass-creation from GNIS (the official US gazetteer), which is now recognised as an unreliable source - but that was only recognised in late 2021 after 20 years of articles being created based on it. GNIS-based articles are the subject of a massive clean-up that has been going on for years and will likely be going on for years more. See WP:GNIS for further details on this.

  Frankly I'd prefer to get the articles right first. FOARP (talk) 09:52, 1 October 2023 (UTC)

• Option 2 or Option 3 larely per David Fuchs and Seraphimblade. If we can't find a non-database source about a geographic division, we shouldn't have an article about it. --Guerillero ^{Parlez Moi} 18:32, 30 September 2023 (UTC)

  Option 2 says nothing about "databases" (unless we are using the word "database" as a misnomer for something that is not actually a database). This proposal would not exclude the use of a "database" as the only source for an article, since a "database" could satisfy the proposed criteria. This proposal is not confined to the use of "databases" as sources, and would affect a much wider range of sources. James500 (talk) 19:28, 30 September 2023 (UTC)

• The proposers need to provide a rationale that is more coherent than this, which amounts to "let's swap in this new text for the old one and if you want to know why, follow this rabbit hole of half a dozen discussions". If you want to change a status quo that is this entrenched in the ideas and practices of the project, the explanation needs to be way more accessible than this. The most obvious issue is why WP:INDISCRIMINATE doesn't suffice to fix the apparent issues. (Oppose) --Joy (talk) 10:51, 1 October 2023 (UTC)

  @Joy - thanks for your feedback. I did try to express the case in the best way I could in my !vote. RFC questions are required to be formatted neutrally without advocating for either side of the argument, so it is not possible to express in the RFC question itself why a change is needed.

  However, the problem is how to apply WP:IINFO to the geographical space. I and other people who tend to agree that the GEOLAND standard essentially ignores WP:IINFO would point to GEOLAND allowing articles about places that are essentially individual farms, houses, streets, etc. so long as they are in some way “legally recognised”.

  The argument on the other side appears to come in two flavours.

  The first is that, based on WP:5P, as well as being an encyclopaedia, Wikipedia is a gazetteer (i.e., a geographical dictionary) and so the geographical space is essentially exempt from WP:IINFO. The directory-style level of coverage of a gazetteer, where ALL named items in an area are included, regardless of whether there is anything beyond statistics to write about them, is therefore permitted.

  The second is that these articles can in any case all be expanded and so are not WP:IINFO. Typically the lack of sourcing is hand-waved away in this case - that some can be expanded is sufficient for them all to be kept.

  My answer to the first argument is that WP:5P does not say that Wikipedia is a gazetteer, just that we have features of one. My answer to the second is that the sourcing should be in the article for them to be kept.

  I hope that explains this case without requiring, as you say, people to go down the rabbit hole. FOARP (talk) 12:08, 1 October 2023 (UTC)

  I suppose the main question would be whether we think that e.g. a statistics bureau is inherently biased in presenting its results through censa, and can't be considered an impartial observer of human settlement, but a government agency that effectively promotes such settlement; ditto for a geodesic bureau for mapping etc. Mentions of these places in secondary sources will habitually be cursory, in passing, often also seen through the lens of statistics (for example an article about archeological excavations in an area will mention a number of villages there, but it doesn't really constitute significant coverage of the villages themselves). I think I've previously had fairly similar discussions about biographies of TV presenters - when a major national television station publishes dozens of articles about their own shows that mention these, does that constitute insignificant and/or biased coverage or is it just a matter of fact that they put people on a map (heh). The passage of time could also play a part - if e.g. a village has been attested to in a dozen censa, and there's a local road there, classified since many years ago, does this dilute the argument against documenting it - the encyclopedia might as well describe it because it does seem to be a real-world phenomenon of a modicum of long-term significance. --Joy (talk) 12:23, 1 October 2023 (UTC)

  These things are issues, but even more of a problem with both censuses and gazetteers, is that neither was ever meant to be used as a list of "legally recognised populated places". For gazetteers the emphasis is typically on providing a directory of names for locations, with any other information provided being secondary (hence GNIS, the official US gazetteer, listing a lot of places as being "populated", which actually aren't populated and never were - this part of the directory didn't matter to the compilers as it wasn't its core purpose). For censuses, they're supposed to be a survey of the whole population rather than just those living in a particular place, and the people who compile them don't typically care if the geographical points used in the census correspond to actual communities (hence the Iranian census having lots of "villages" that were actually shops, pumps, farms, bridges etc. that people were counted near).

  I think the TV presenter analogy you've raised is quite relevant - it's as though people were going through IMDB and saying "they're on IMDB so they must be notable, sources will be added later". FOARP (talk) 19:33, 1 October 2023 (UTC)

  That seems to be region-specific, because for example in Croatia the legal recognition of populated places is largely unambiguous, and the national authorities that conduct a census work off of organized geo data on settlements that is largely rooted in centuries-old village and town boundaries. In addition there's a parallel system of cadastral municipalities that usually further attests to places and locations. The census also includes a fair bit of nuance about who lives where, and whether they're there on a permanent or a temporary basis. Despite all this, we still haven't made articles for nearly all of those relatively well-documented places, but you're telling me about pumps and bridges :D So this looks to me more like a matter of deciding to which level do we employ a specific source to determine this threshold of legal recognition for standalone articles. Surely there's some sort of a standards organization that analyzes statistics bodies that could serve as a guide as to which of them are more reliable than others? --Joy (talk) 06:56, 2 October 2023 (UTC)

  The regional variation is entirely a product of the GEOLAND requirements that places be "legally recognised" - the law differs from country to country by a very great degree. Unfortunately I'm not aware of any standards body that oversees national censuses/gazetteers worldwide. I'm not even sure how you could standardise something that varies so wildly.

  You correctly identify a big issue with the way GEOLAND shapes our coverage of villages: as there is no requirement to seek secondary coverage, the emphasis is on going through databases and reproducing stub articles on every item in them in mechanistic fashion (see here for a non-untypical day of article-creation for a single editor in this space, where they created 330 article in one sitting, ignoring duplicates and items with names like "school"). The assumption is the secondary sources will be added later, but these will not even exist in many cases, and analysis of stubs by user BilledMammals shows that relatively few are ever improved, and if they are improved it is the original author who does it. FOARP (talk) 08:23, 2 October 2023 (UTC)

  No evidence has been presented that the "legal recognition" requirement has any effect whatsoever. As far as I can tell, the 330 articles listed are generally villages or depopulated former villages, that have articles in the Russian Wikipedia. As far as I can tell, they generally have or had the population that would be expected of a village. No evidence has been presented that they were selected on the basis of their legal status. The batch creation looks more like a transwiki of a batch of articles that already existed on the Russian Wikipedia. The claim that coverage does not exist for Russian villages is not plausible for a country that, amongst other sources, has 45,000 registered local newspapers and periodicals as of 2015: [6]. James500 (talk) 23:35, 2 October 2023 (UTC)

  "No evidence has been presented that they were selected on the basis of their legal status" - Every single one was selected because it is listed on the Russian census as a selo. They were written in the same alphabetical order that they are on the Russia census.

  "the 330 articles listed are generally villages or depopulated former villages" - including the duplicate ones? Including the one that literally says it's just a factory in its name and appears to be exactly that on the map?

  "The batch creation looks more like a transwiki of a batch of articles that already existed on the Russian Wikipedia." - yet they overwhelmingly lack ALL of the context that can be found on RU wiki, so no, that's not what was being done.

  "The claim that coverage does not exist for Russian villages is not plausible for a country that, amongst other sources, has 45,000 registered local newspapers and periodicals as of 2015" - Yet Russia has more than 150,000 selo, of which 20,000 are uninhabited. Simple maths says that many selo will be located in places without any actual coverage. Of course historically we run up against the simple problem of high levels of illiteracy. FOARP (talk) 07:50, 3 October 2023 (UTC)

  I'm not quite sure what to do with people who apparently copy&paste from ruwiki with archived sources, but it doesn't sound like a change to a guideline would fix that, though. Even if we AFD'd all these, WP:POTENTIAL can easily be argued, as well as the benefit of building the web of properly disambiguated location links. --Joy (talk) 08:05, 3 October 2023 (UTC)

  WP:Potential can be argued for articles that have it. There's a long series of e.g., Russian "village" articles that are really railway stations and have never been anything else that clearly don't have it. See, e.g., 2826 km, 2647 km, 2797 km, 2779 km - all Russian "village" articles for railway stations that don't even have a proper name, just the kilometre-number along a railway. We have hundreds of articles for such places, all justified by GEOLAND. FOARP (talk) 17:20, 3 October 2023 (UTC)

  I don't think those four articles are justifiable by the current guideline, that stuff is just plain unhelpful. --Joy (talk) 11:53, 4 October 2023 (UTC)

  However, since the guideline gives a presumption of notability to any "legally recognised populated place", and these are Russian "rural settlements" for which data is included in the census, these would likely be kept under the current guideline. There are more than a hundred Russian "village" articles named after kilometre marks that are clearly just railway stations. There are many others that are railway stations but which aren't named after kilometre-marks and which don't mention the fact that they are railway stations but instead claim to be villages (e.g., Shomyrtly - clearly just a railway station). Cleaning up all the places that are not really villages in Russia, but are at presently listed as such, is going to be years of work since - in the absence of secondary sourcing which virtually none of these locations actually have - it requires an editor to check manually what the location is on the map.

  Anyway, I hope this explains why the present standard is, in my view, unsatisfactory and needs to change. Obviously there are arguments on the other side, which I have tried my best to reproduce fairly, though the advocates of the present system may disagree with that. I understand if your vote is still to oppose, but at least you haven't had to go down the rabbit-hole on all the previous discussions to get a full picture of things. FOARP (talk) 12:29, 4 October 2023 (UTC)

  As I have explained in some detail elsewhere, having examined the evidence presented so far, I am not convinced the "x number of kms" articles actually are railway stations (as opposed to being settlements named after railway stations, which is not the same thing). Further to what I said there, the etymology of a place name does not necessarily prove that the place is not a settlement. For example, the logical conclusion of this kind of etymological argument would be that the article Oxford must be about a ford used to drive oxen across the River Thames, and that the article Cambridge must be about a bridge across the River Cam, and that neither could possibly be a university town, because the place name does not mention a town or a university. James500 (talk) 19:08, 4 October 2023 (UTC)

  Sorry, but this is simply wrong. Your explanation consists of waving your hands and saying that these are “dispersed villages” (i.e,. not actual villages) and ignoring the fact that our articles (and the Russian articles) declare them to be railway stations that are part of other settlements. Talking about camps/dormitories/military barracks only emphasises that these are not villages. FOARP (talk) 20:25, 4 October 2023 (UTC)

  (1) You are attributing to me a number of statements, assertions and opinions that I did not actually make or express. (2) I refer you to the comments made by Ymblanter and Ezhiki at WT:NGEO. (3) I am not prepared to further discuss Russian rural localities at the village pump for the time being, in view of the present progress of the discussion that is taking place at WT:NGEO. James500 (talk) 21:44, 5 October 2023 (UTC)

• Option 1. The risk of eliminating coverage of notable places, especially historical settlements that have since been depopulated, due to lack of ready access to sources and relative lack of editors with relevant language competence, outweighs the risk of having articles on non-notable places. It would damage the encyclopedia by making unnecessary holes in the web of coverage that is one of the major advantages of our being an online encyclopedia. Consider the case of someone born in a place where the only current online sources are government databases—it's better to be able to link to the specific place rather than to a larger administrative unit (which may not even have existed at the time of their birth). Or consider a reader researching regional architecture, collectivization of farming, or former shtetls. For countries whose census categories are problematic for us because they don't correspond to English-language classifications of populated places, specific discussions and actions should take place, as happened with Iranian abadi. Fix the specific problems, don't pre-emptively enshrine our natural bias toward readily available and recent information. Yngvadottir (talk) 02:49, 3 October 2023 (UTC)
• option 2, but definitely NOT option 1 Whatever the deficiencies of option 2's wording, the reality that many people seem determined to deny is that the current wording has quite consistently produced articles which were untrue. The fear of missing formerly populated places has not been borne out in deletion discussions; the presenting problem is that every time someone looks at some sort of official listing of places, we get a stream of poorly-verified articles of which some greater or lesser proportion state that a town or whatever exists when it does not. It's no good to say that a place might have been populated in the past when it cannot be shown that this is so; we have found many places that are no longer there, but we know so because there is documentation of the disappearance of the place besides a juxtaposition of line in a database and a blank spot on the map. Personally I think the "contains elements of a gazetteer" claim is vacuous in that any work containing data on places by name does so, no matter how incomplete its coverage is. But at any rate, the real problem is that the current language encourages people to skip the verification which GNG would require that would show that the articles they are creating are inaccurate. It therefore puts too much onus on those who are trying to keep WP honest. Mangoe (talk) 21:13, 3 October 2023 (UTC)

  @Mangoe: What percentage of stubs on populated places are untrue, would you estimate? – Joe (talk) 08:54, 4 October 2023 (UTC)

  I understand something like ~20% of the ~80k Geostub articles written by Carlossuarez46 have now been deleted. 13,147 were deleted in one batch deletion, ~1,000 articles had been deleted before his resignation, and probably another ~2,000 have been deleted through PRODs and bundled AFDs (e.g., these ones: 1 2 3 4 5) since the batch deletion was completed. That would be a reasonable, very rough estimate for inaccurate Geostub articles in general in my view.

  I do not think that any of the present efforts at mass-creation of Geostubs by others have been any better that C46's. Indeed the mass-creation based on GEONet Names Server (GNS) data is probably lower-quality, since it was not checked against census data, and GNS is based on old US military maps that have often been compiled in a way that compounds systematic error (see the errors in South Korea referenced in our article on GNS). A lot of articles were written based on Geonames, which is an even worse source as it's user-created. We've discussed the efforts of one editor on Russian Geostubs elsewhere, but there are others doing essentially the same thing in other areas also.

  The essential problem is exactly as Mangoe says - without ever checking a gazetteer/census-source against secondary sources, you have no idea what the actual nature of the location actually is/was. FOARP (talk) 13:05, 4 October 2023 (UTC)

• Option 1. I'm also open to Blofeld's Option 3 proposal. I agree with TheCatalyst31's comments above and think this is a solution in search of a problem and will only exacerbate systemic bias. Calliopejen1 (talk) 04:01, 4 October 2023 (UTC)

Discussion (Changes to GEOLAND)

Extended discussion and responses here please... FOARP (talk) 16:25, 15 September 2023 (UTC)

So it looks like FOARP brings this up every year because they clearly don't like the idea that Wikipedia exists in part to detail places where people live. A couple years ago there was a unilateral change which was then brought to the larger community and thoroughly rejected. This would have the same effect - by making NGEO slightly more difficult to apply at AfD, we would tighten our notability guidelines and lose one of our purposes as an encyclopedia.The The idea wikipedia is a gazetteer was boldly added by Unitedstatesian in 2008 and reflected how NGEO worked at AfD (this was later made a guideline.) Even though it was boldly added, its place was further confirmed in discussion in 2009. In reality, right now, the NGEO test is very easy: if we can prove people live or lived in a place, it's eligible for a stand-alone article, similar to how gazetteers work. This would require us to be able to say something about the place, which would instantly put a number of places at risk of deletion, especially in non-English countries or in places where you can't just web search for a village.The reason is clear: even though we may not be able to say anything about a verified populated place now does not mean we'll never be able to: the fact people live in a locale means there's something notable about that locale, even if there's just an oral history and it's not on the internet yet. A lot of the deletion discussions I've been a part of have been about verifiability, for instance the American GNIS stubs, and I can see a change leading to a lot of negative outcomes where we would delete verifiable places without easily accessible online information. SportingFlyer T·C 21:41, 15 September 2023 (UTC)

SportingFlyer - You said you were going to find a source to support your assertion that Wikipedia is a gazetteer. None of these links do so. Even WP:5P (an essay) only states that it includes elements of gazetteers. FOARP (talk) 14:49, 16 September 2023 (UTC)

It's in there. It was boldly added to reflect geography policy at early AfDs and then was further accepted by the community through discussion. SportingFlyer T·C 21:31, 16 September 2023 (UTC)

It's in where? A simple quote will do. FOARP (talk) 12:47, 17 September 2023 (UTC)

Again it goes all the way back to the first edit of NGEO, and the concept was endorsed as much a little under two years ago: [7] SportingFlyer T·C 09:45, 18 September 2023 (UTC)

... which only ever said that Wikipedia includes "combines many features of ... gazetteers", not that it is a gazetteer. FOARP (talk) 10:27, 18 September 2023 (UTC)

However given how many people in this and related discussions base their argument on Wikipedia being/including a gazetteer I'm not convinced that the distinction you are making is one that is supported by community consensus. When it comes to the inclusion of information about populated and formerly populated places, I don't think the distinction is either meaningful or useful. Thryduulf (talk) 14:20, 18 September 2023 (UTC)

I think people should recognise this “Wikipedia is a gazetteer” argument for what it would logically require: that Wikipedia give the same level of coverage to geographical features that gazetteers do. This means extending our coverage to every single named feature, regardless of notability or whether there is anything verifiable to write about it beyond statistics. FOARP (talk) 03:25, 22 September 2023 (UTC)

We should have coverage of every populated place we can verify. That doesn't necessarily mean a standalone article, but nothing in current policy or guidelines requires that so it's not a problem. Whether we have a similar coverage of every named feature, if that's what the community wants then I see no issue with that. Thryduulf (talk) 08:25, 22 September 2023 (UTC)

"We should have coverage of every populated place we can verify" - both my house and the street it is located on are verifiable and populated. They even have "legal recognition" in the sense that they are listed in the official census/gazetteer. FOARP (talk) 09:34, 25 September 2023 (UTC)

Okay, replace the word "place" with the word "settlement". Although I would not be opposed to verifiable "list of streets in X" articles, with links to notable streets and/or notable buildings on that street.Thryduulf (talk) 10:39, 25 September 2023 (UTC)

We should have coverage of every populated place we can verify. We already do; what we don't have is coverage of every populalated place on the most granular level, but why should we? What benefit would this level of granular and indiscriminate coverage bring? BilledMammal (talk) 10:23, 25 September 2023 (UTC)

I didn't say anything about indiscriminate coverage or excessively granular detail, so you are attacking a straw man. The current policies allow for the appropriate level of coverage in the appropriate place, which is sometimes a standalone article and sometimes inclusion on a broader article. The proposal would preferentially delete rather than merge content, especially where English-language sources are not trivially available on Google increasing systematic bias. Thryduulf (talk) 10:37, 25 September 2023 (UTC)

You said We should have coverage of every populated place we can verify. The implication was that you won't be satisfied with coverage of populated places at anything but a very granular level - or indiscriminate - level. Am I incorrect? BilledMammal (talk) 10:56, 25 September 2023 (UTC)

Yes you are incorrect, "coverage" is not a synonym of "indiscriminate or excessively granular coverage", nor is it a synonym of "every populated place must have an individual article" or any of the other straw men being argued against. "Coverage" means "we should have information about the subject on an appropriate article", with the most appropriate article depending on the situation. Thryduulf (talk) 11:07, 25 September 2023 (UTC)

Yeah, I think the primary result of this would be "more uncertainty" rather than "better project". "Statistics-only" for a populated place is extremely broad; it's possible to write a significant amount about a place based solely on its census stats. I could pull out any neighbourhood in my city and assemble its census stats into a legitimate-reference-work-addition. In fact, just to test my point, I've done that and massively expanded Clayton, Victoria#Demographics. There's a lot more to the article than this section -- but if you replaced the whole article with that section and a lead saying Clayton is a suburb in Melbourne, Victoria, Australia, 19 km south-east of Melbourne's Central Business District. It is the home of Monash University, you would have a substantially more explanatory article than the one that currently exists. I could write more there, but I very consciously restricted myself only to census results and to one article based off census results (because its secondary analysis drew a comparison the results themselves don't). The problem we have right now is that there are mass-created articles for which it's unrealistic to assume mass expansion, which is a different problem to what you can write based on something. Vaticidalprophet 22:36, 15 September 2023 (UTC)

Purer example: I remembered Salisbury East, South Australia being a stub. Excitingly, it was even an uncited stub! It's not now. I think this tells you a fair bit about the place. Vaticidalprophet 23:03, 15 September 2023 (UTC)

The process was not started by FOARP, but my me. If you had looked at the actual discussion instead of just reading the blarb at the top, the discussion was started on the basis of the actual wording of the current GEOLAND. Legal recognised place is virtually impossible to actually evidence in large parts of the world. This is why the phrase has been changed to Settlements and Administrative Regions. From that discussion the issue of permastubs came up. Now I don't have a problem with stubs if they are accurate, but census data being the only evidence to create an article is not really good enough for an encyclopedia. Glasgow University History of the British Census says "It is essential to bear in mind when examining census data that the census it is not a completely accurate record." And that is in a western country where census data is seemed to be accurate, as pointed out in the original discussions how many census in parts of the world give a name to a nomad village (such as Abadi) which has disappeared by the next census and a new one appears? As pointed out by Vaticidalprophet, he took an article which was just statistics and expanded hugely, so it is possible in a large number of cases. However sometimes you can't. Hare Green, a hamlet in Essex is a perfect example. It was just evidence by a map and census data. I have filled out with all that I can find to expand it - some historical houses and some info about the roads. None of it would meet GNG. To be honest I live in Essex and had not heard of it, and if it was redirected to Great Bromley and added there we would not lose the information.Davidstewartharvey (talk) 06:33, 16 September 2023 (UTC)

Hare Green doesn't seem a great example to me. There are two 17th-century listed buildings, suggesting a long history which will be covered in (offline) local history books. The population is over 700, which is a small village, not really a hamlet. The settlement will appear in local newspapers, probably for hundreds of years, and certainly in multiple historical directories, which will probably turn out to include material on local public houses and businesses. 'I've never heard of it' is never a reason for deletion.

When academic sources state that the historical UK census is not completely accurate, what's generally meant is that old handwritten records can be difficult to decipher and people gave their names in different forms, so tracing individuals can be tricky, not that a whole village was incorrectly entered! Espresso Addict (talk) 07:31, 16 September 2023 (UTC)

I am not saying something shouldn't exist cos i have no heard of it! I have done more than internet searches - as I said I am local and found very little beyond architecture. Because something has existed doesn't mean anything interesting exists and needs an article about it based on that it just exists. If the information was added to the Great Bentley page and the page redirected there, we are not losing anything as there is not much to lose.

The Glasgow University comment is on the British Census, but it counts for census' around the world. How many people put down Jedi as their religion? And although you maybe bound by law to complete the census in parts of the world, they are actually not considered legal documents to confirm settlements actually legally exist in many parts of the world either.Davidstewartharvey (talk) 09:32, 16 September 2023 (UTC)

@ Davidstewartharvey: Hare Green satisfies GNG and, on account of the listed buildings, GEOFEAT. Your expansion is far from exhaustive. You have said virtually nothing about the two listed buildings, despite the fact that there is more coverage. You have not even mentioned the ancient earthworks: [8]. No apparent use of the coverage in GBooks, GScholar and the Internet Archive. You have made little use of the newspaper articles in GNews, which contain more information than you have added. You have not added any of the newspaper articles in the British Newspaper Archive [9] [10] [11], or in Newspapers.com [12] [13]. I don't see any use (beyond what is included in GNews) of the websites of the essex newspapers like this one: [14]. (GNews only contains articles from the last few years, whereas newspaper websites often go back to the 1990s). More importantly, Great Bromley itself certainly does satisfy GNG by a wide margin, but would be in real danger of an AfD or draftification under this proposal, because the sources are not in the article yet. I can put up with a place like Hare Green being merged, but I can't put up with a place like Great Bromley being mistakenly deleted or draftified, which would be the probable outcome of this proposal.

As for the census: There is no reason to assume that any reliable source is completely accurate. It would not surprising if every reliable source in the world contains errors. If we rejected every source that contained at least one error, it would not be surprising if we end up with no sources at all. I happen to be aware that sources say that even the New York Times contains some errors (such as in its coverage of the Russian Revolution) and is therefore not completely accurate. James500 (talk) 10:11, 16 September 2023 (UTC)

How does the current article meet GNG? Firstly maps can't be used to define notability, the census data is not from ONS but a database site based in Germany - very questionable. The road references - one a local newspaper and the other's primary source as it is the official legislation. The references on BNA - sales of properties and livestock, a barn fire with a death, are these really enough to establish GNG? And as to GEOFEAT, I remember that there was a discussion a few years back that agreed a concensus that Grade II properties are were not accepted as notable based just on a listing. This was based on the fact there is 343,004 Grade II listed along in England. In addition the Henge is not in Hare's Green but in Little Bentley.Davidstewartharvey (talk) 15:47, 16 September 2023 (UTC)

I read that discussion about GEOFEAT. IIRC, it was not an RfC, established no consensus, either to change the wording of the guideline or otherwise, and it was based on a pseudohistory argument that because the population of the UK is smaller than that of the USA today, the UK should have fewer listed buildings than the USA. In fact, buildings are not normally listed in the UK unless they were built before 1840, and are not automatically listed unless they were built before 1700. The UK had a much larger population than the USA before 1840, and a vastly larger population before 1700. The UK has a much larger number of pre-1840 buildings, and a vastly larger number pre-1700 buildings, than the USA. Therefore the UK should have a much larger number of listed buildings than the USA, because the UK has more historic buildings than the USA, and the UK's historic buildings are generally more important than those of the USA. The reality is that the NRHP largely consists of late 19th and 20th century buildings that would never get listed in the UK in a million years, because their historical importance is very low. The number of grade II listed buildings in the UK is reasonable for a country that has a massive concentration of medieval and early modern buildings of immense historical importance, something that does not exist in the USA, which has no medieval buildings and almost no early modern buildings. James500 (talk) 00:11, 17 September 2023 (UTC)

BAR British Series 175 says that the penannular ditch is in "Hare Green, Great Bromley". Unless you have another source, it is in Hare Green. James500 (talk) 00:51, 17 September 2023 (UTC)

there are two henges one in Great Bromley map ref 51.860498N Longitude: 1.054156E ^[1] and one in Little Bentley map ref Latitude: 51.884112 Longitude: 1.062716 ^[2] Other henges in the Tendring area is at Great Wigborough Latitude: 51.7942 / 51°47'38"N

Longitude: 0.8369 / 0°50'12"E, Tye Henge in Lawford Latitude: 51.936516N Longitude: 1.036039E, though the closest to Hare Green is Little Bromley Latitude: 51.905055N Longitude: 1.035369E. The book is the only reference to a Henge in Hare Green. Davidstewartharvey (talk) 16:56, 17 September 2023 (UTC)

And further more upon checking the website https://www.citypopulation.de/en/uk/eastofengland/ which is the source of the population data, it quotes states Source: UK Office for National Statistics (web).

Explanation: All population figures and depicted boundaries are based on output areas officially assigned to the 2022 built-up areas. Output areas often include some unbuilt parts. However, tabulated area figures refer to (typically smaller) actual built-up areas in order to present a more realistic population density. Some of the older population figures are approximate values. Well when you check the data that is currently on the site (supposedly 2021 data) it states Hare Green has a population of 678. However Great Bromley is not in there list, and when you check the data on the ONS ^[3] it says the parish of Great Bromley has 1100 people, with no further breakdown available. So the reference is actually duff!Davidstewartharvey (talk) 17:45, 17 September 2023 (UTC) Davidstewartharvey (talk) 17:45, 17 September 2023 (UTC)

Trivial mentions do not count towards GNG. This is very explicitly stated. Hare Green being mentioned once as a location in local news stories (failing NOTNEWS) is not detailed, direct coverage of Hare Green. It is indirect and incidental. List entries for individual historic structures that happen to be in Hare Green are also not direct significant coverage of Hare Green. And literally no one would be trying to get Great Bromley deleted with this proposal, what a ridiculous leap. JoelleJay (talk) 21:32, 16 September 2023 (UTC)

Your argument is that a village cannot be notable for significant coverage of its history or its buildings. In other words, you are arguing that a village can never be notable, since there is no other coverage that a village could receive (unless you want coverage of the quality of its subsoil). I do not think that your argument is within the letter, let alone the spirit, of GNG or NGEO. And all that your argument could achieve, at most, is a page move to something like History and buildings of Hare Green, which will put you in violation of WP:COMMONNAME. James500 (talk) 02:01, 17 September 2023 (UTC)

If a building that happens to be in Hare Green receives SIGCOV that doesn't actually discuss Hare Green itself it is not coverage of Hare Green. This should be exceptionally clear from our P&Gs. And the buildings in question don't even have SIGCOV themselves, they just have a context-less, nearly-prose-less description of the building and items inside.

TM 02 SE GREAT BROMLEY HARWICH ROAD HARE GREEN (north side)
House. C17 with later rear extension. Timber framed and rough rendered. Red plain tiled roof. Central red brick chimney stack. 2 storeys. 2 window range of small paned vertically sliding sashes. Central 2 panel 2 light door, simple surround. Interior features include vertically boarded and panelled doors, some with HL hinges. Inglenook fireplace with cast iron fireback reading "1640", "Fairfax" and depicting a man on horseback. Iron meathooks to rear kitchen ceiling beams.

If you seriously think that is direct significant coverage of Hare Green then I seriously question your understanding of our policies. JoelleJay (talk) 03:18, 17 September 2023 (UTC)

(1) You have not answered what I said to you at all. You have instead twisted my words and put words into my mouth.

(2) WP:SIGCOV does not contain a definition of the word "directly" (other than "so that no original research is needed to extract the content") and it is not acceptable for you to claim that your personal opinion about how that word should be interpretated is community consensus unless you can produce evidence of that alleged consenus. Could you, for example, provide me with a link to an RfC on the meaning of the word "directly" in WP:SIGCOV which confirms that your interpretation of that word actually is supported by community consensus?

(3) At this point, I really suspect the fact that you are still making comments like "nearly-prose-less" tells me everything I need to know about the level of community support for your opinions. I happen to remember that your theory (and it is still only a theory) that significant coverage must consist of what you refer to as "prose" was discussed during WP:ACAS. The community pointed out to you that GNG contains no such requirement, and your personal interpretation that GNG requires what you refer to as "prose" was totally rejected by community consensus. Your proposal in Q14 was simply rejected out of hand by the community.

(4) If you don't stop making unevidenced claims about alleged community consensus about the interpretation of particular words in guidelines, I could put a stop to that by starting an RfC myself on the meaning of "directly", to find out what the community really thinks. James500 (talk) 05:35, 17 September 2023 (UTC)

1. I'm the one twisting and putting words in your mouth!? You literally said Your argument is that a village cannot be notable for significant coverage of its history or its buildings. In other words, you are arguing that a village can never be notable, since there is no other coverage that a village could receive (unless you want coverage of the quality of its subsoil). That is either an intentionally outrageous mischaracterization of what I said or your reading comprehension is unsuitable for this project. I didn't make a single statement that claimed to restate your arguments whatsoever; the closest would be saying If you seriously think... which is very clearly not saying "You seriously think".
2. What do you believe "directly" in "Significant coverage" addresses the topic directly and in detail means if it doesn't mean the topic must be discussed directly! Do you believe that a source's SIGCOV is distributed to all topics that are named within it?? A source on a movie that trivially names who its director was before going into detail that has nothing to do with the director's role is not SIGCOV of the director, obviously.
3. I don't know how you are reading that the community "totally rejected" the interpretation that secondary coverage should be in prose from a proposal in the very specific context of limitations on acceptable sources for mass creation that the closers said had no discernible consensus. Most of the opposes were against "disallowing mass creation of articles that are based on any database" or were somehow misinterpreting the proposal as asserting that a database isn't RS, rather than making any statement on the prose-non-prose database distinction.
4. You are more than welcome to ask the community whether "directly and in detail" includes all coverage adjacent to a trivial mention of a topic regardless of whether any coverage is actually specifically about the topic. JoelleJay (talk) 02:50, 18 September 2023 (UTC)

It does not matter whether you expressly claimed to restate my words or not (and the words "If you seriously think" and "This should be exceptionally clear" were close enough), if you answered my comment in a way that clearly implied that I had said that the coverage was "direct significant coverage of Hare Green" or clearly implied that I had said that coverage of "a building that happens to be in Hare Green" discusses "Hare Green itself" and "is . . . coverage of Hare Green". If I ask what kind of coverage a village can receive other than coverage of its history and buildings, I should not get a response that implies that I have argued that such coverage satisfies GNG, when I did not comment on whether that coverage satisfies GNG at all in that particular post. I am prepared to accept that you did not realise what you were implying, because I am also starting to have doubts about your literacy and, in particular, whether you understand that you can imply something by means of juxtaposition. (If you put your comment beneath my comment, the juxtaposition can, in of itself, imply that your comment is talking about my comment). That said, I really think it would be better if we both stop questioning each others' literacy and just walk away from this conversation, because it is clear that neither of us wants to talk to the other. Goodbye. James500 (talk) 03:37, 18 September 2023 (UTC)

You asserted Your argument is that a village cannot be notable for significant coverage of its history or its buildings. In other words, you are arguing that a village can never be notable, since there is no other coverage that a village could receive. 1. I did not say anything close to that. I said passing mentions in primary news do not count and existence of listed buildings with zero coverage of the location do not count. 2. there is no other coverage that a village could receive is a claim you are making about what kind of coverage exists for villages. Since my statement only applied to the latter of your two options of "coverage of its history" and "coverage of its buildings" (I never said coverage of a town's history doesn't count, unless you are suggesting that primary trivial newspaper mentions qualify?!), your comment was read as claiming the coverage of the buildings in this particular case is sufficient for notability. So I responded with a rebuttal of this claim. JoelleJay (talk) 20:47, 18 September 2023 (UTC)

When I looked at the talk pages of the notability guidelines, I saw the discussion on WT:N that you are obviously referring to. I should point out that a discussion between three people (yourself, Masem and Whatamidoing) does not create, and is not evidence of, any kind of community consensus. Especially when at least three people (myself, Expresso Addict and Whatamidoing) now object to your interpretation. In short, there is no consensus for your interpretation of GNG. James500 (talk) 23:51, 17 September 2023 (UTC)

How is the mere existence of old buildings enough to assume a location has had SIGCOV? If these sorts of things are essentially conferring notability then that is a serious source of systemic bias since obviously that's only going to work for the places where society has been stable enough that random buildings have survived and there have been modern efforts to document and preserve them and there is a presumption of a written historical record providing significant primary coverage that has itself received secondary analysis (because we certainly cannot be using 17th century sources as the bases for any articles ever). JoelleJay (talk) 03:35, 17 September 2023 (UTC)

[E]ven though we may not be able to say anything about a verified populated place now does not mean we'll never be able to, and even though we will eventually be able to say more about these populated places when interested editors decide to add content, that does not mean we can't upmerge everything shorter than (say) four sentences into a list article now, leaving behind redirects for future expansion. Folly Mox (talk) 15:39, 16 September 2023 (UTC)

@SportingFlyer: comment on content, not contributers. Edward-Woodrow :) _[talk] 18:48, 16 September 2023 (UTC)

I haven't personally attacked anyone. Thank you. SportingFlyer T·C 21:33, 16 September 2023 (UTC)

I didn't say that. I reminded you not to comment on contributors. Saying that it looks like FOARP brings this up every year because they clearly don't like the idea that Wikipedia exists in part to detail places where people live. A couple years ago there was a unilateral change which was then brought to the larger community and thoroughly rejected. This is not a personal attack, but it is unnecessary commentary on a fellow editor. Edward-Woodrow :) _[talk] 22:38, 16 September 2023 (UTC)

They have consistently tried to make this change over the past few years, without gaining any consensus for it. It's getting tiring. Not sure why that's not allowed to be said. SportingFlyer T·C 12:38, 17 September 2023 (UTC)

@SportingFlyer - You're basically stating that this is my proposal: it was not, very little of the proposed amendment was from me. I linked the discussion where the proposed amendment was developed, and named the editors who contributed to that discussion. "it's getting tiring" - two vaguely-related discussions on the subject in three years tires you? "without gaining any consensus for it." - the proposed edit is the result of a consensus.

We can see from other discussions (NSPORTS, SCHOOLOUTCOMES, NASTRO) that previously over-broad notability standards can be reined in. Proposing the same be done in other areas is just part of the healthy give-and-take of policy on Wikipedia. FOARP (talk) 07:22, 18 September 2023 (UTC)

I'm intrigued by the proposed ten-per-day rate limit on AfD. If I understand correctly, some here fear mass creation of geographical microstubs. So why not put a rate limit on that instead? That could dampen the motivation for someone to run up their new article count by harnessing a poorly understood database. You don't want the waitstaff to take orders faster than the kitchen can fill them, but you also don't want the waitstaff to clear the tables faster than the diners can finish their meals! :^D – Minh Nguyễn ^💬 01:56, 16 September 2023 (UTC)

There was a discussion on the subject that got derailed by disagreement over the borderlands of what constitutes mass creation. IMO the obvious thing to ban is automated mass creation -- there are some arguments about certain other forms of high-volume article creation, but there's no number that a human can reach that people agree is problematic (and really shouldn't be). Discussion didn't establish scope, though, so it got bogged down in those, and until then we retain all these everything-not-mandatory-is-forbidden arguments. Vaticidalprophet 02:15, 16 September 2023 (UTC)

” there's no number that a human can reach that people agree is problematic (and really shouldn't be).”

One of the still-operating mass-creators in the GEOLAND field pumps out articles at a rate of up to 330 articles a day based in Russian census data. I can’t conceive of any situation in which that wouldn’t be problematic. FOARP (talk) 06:00, 28 September 2023 (UTC)

• Comment please split these mass deletion/merge RFCs out of VPR in the future. See User Talk:BilledMammal#Page size Mach61 (talk) 04:35, 16 September 2023 (UTC)

I think some kind of forced limitation (maybe using an edit filter) of five articles per day, and no more than 100 articles per month, would slow down any mass-creations like this. Edward-Woodrow :) _[talk] 18:52, 16 September 2023 (UTC)

It's certainly possible to legitimately create articles past that rate, though the day one moreso than the month. The problematic sort of mass creation is well beyond what can be done by an actual person; there are people who take an unusually wide view of what mass creation is problematic, but the actual motte of "mass creation produces data integrity and sometimes sensitive-topic-maintenance problems" is implausible for anything human-doable. Vaticidalprophet 20:53, 16 September 2023 (UTC)

I concur with Vaticidalprophet: "two sources = must have an article"-type thinking is a major problem. Mass creation can only be addressed directly, not through changes to notability guidelines. It merits a trip to ANI for WP:NOTHERE. I've thought a good bit about mass creation, and I don't think there's a realistic way to address it except requiring that notability be demonstrated inline rather than "presumed". I don't think automated creation is the real problem. And it'll mean that we don't have microstubs that are useless verbatim reprints of external sources, whether database entries or non-SIGCOV passing mentions. DFlhb (talk) 10:07, 16 September 2023 (UTC)

Please explain it to me like I'm an idiot: why is people writing lots of encyclopaedia articles a problem that we need to address at all? I thought that was kind of the point of this place? – Joe (talk) 14:00, 16 September 2023 (UTC)

Writing lots of articles isn't a problem; writing lots of low quality articles is. I have been considering proposing expanding WP:SPORTSCRIT #5 to all topics; it would effectively prevent mass creation, as the work required to identify at least one source makes it non-viable, and creates a (very low, but extant) minimum quality requirement. BilledMammal (talk) 14:12, 16 September 2023 (UTC)

So the kind of stub discouraged in the proposed guideline (Village X is in County Y, Province Z of Country A, at coordinates C, with a population of P.) is low quality? Why? – Joe (talk) 14:19, 16 September 2023 (UTC)

Because it replicates database-style content, like taxon sub-stubs. Edward-Woodrow :) _[talk] 14:36, 16 September 2023 (UTC)

And because it is likely erroneous. We saw that with the US "village" articles based solely on GNIS. We saw that in the Iranian "village" case. We saw that with Dr Blofeld's Indian "villages". We saw that with Lugnuts Turkish "villages". These database sources just aren't made for the purpose that some editors on Wikipedia think they are made for. FOARP (talk) 14:46, 16 September 2023 (UTC)

That's a reason to designate those sources as unreliable or at least unsuitable for certain purposes. I don't see how it's a reason to rule out that whole type of article. XOR'easter (talk) 16:42, 16 September 2023 (UTC)

We’ve already ruled out two of the world’s largest and most heavily-used Gazetteers (GNIS and GEONet Names Server) for certain information. We also ruled out a major class of entity in a census (abadi in the Iranian one) and are currently discussing doing the same with the Russian one.

I’m not sure how many times this needs to be done before we start to look at whether the rule by which every item on these directories warranted an article was actually a helpful one. FOARP (talk) 20:48, 2 October 2023 (UTC)

Except the rule doesn't allow every item on those directories to have an article. The current requirement is that two criteria are met:

1. The subject is a populated place
2. The subject has legal recognition

The abadis, railway junctions, etc. fail point 1. The current rule also doesn't require that every place that does meet both criteria has it's own article. Thryduulf (talk) 20:53, 2 October 2023 (UTC)

I'm just going to point out that if a rule is continuously causing issues then there is a problem with the rule that needs to be addressed, even if the issues are caused the rule being misunderstood or misapplied. BilledMammal (talk) 21:58, 2 October 2023 (UTC)

Except that if the instances where the rule is continuously causing issues is when it is being misapplied in violation of another community norm (in this case, the one against bot-like article creation), then it is unclear to me that the problem is best understood as belonging to the rule itself. Newimpartial (talk) 22:05, 2 October 2023 (UTC)

@Edward-Woodrow: WP:NOTDATABASE is a shortcut to our policy Wikipedia is not an indiscriminate collection of information, which in brief states that data should be put in context with explanations referenced to independent sources. I can't see how stubs like this contradict that policy? If anything, a stub is rather the opposite of an "indiscriminate collection of information". – Joe (talk) 16:01, 16 September 2023 (UTC)

Government censuses are not independent secondary sources. JoelleJay (talk) 21:35, 16 September 2023 (UTC)

Independent secondary sources are only a requirement of GNG. Verification is the requirement of NGEO, and government censuses do verify the existence of a populated place (yes, with spot exceptions based on the country). SportingFlyer T·C 09:12, 18 September 2023 (UTC)

Secondary sources are a requirement for NOR. JoelleJay (talk) 17:27, 25 September 2023 (UTC)

This is not correct. Browse Wikipedia talk:No original research for numerous discussions of this. Jc3s5h (talk) 18:41, 25 September 2023 (UTC)

It has recently become clear to me that JoelleJay and BilledMammal read WP:NOR as enshrining a requirement that secondary sources be included in all articles, whether or not those articles offer synthetic claims (which of course, if they do, secondary sources are required) and regardless of whether or not they derive their claim to Notability from GNG (which encourages secondary sourcing). I have tried to point out that this interpretation of NOR as documenting a high-level consensus that supercedes WP:N is not a generally-held community view, but to date I don't feel that either editor has really heard those comments as constructive. Newimpartial (talk) 21:41, 25 September 2023 (UTC)

@Jc3s5h, NOR explicitly says articles cannot be based on primary sources and that non-primary sources are needed to establish notability (and not just through GNG). How could an article based only on primary sources comply with those policies? I'm not saying that primary sources can never be used. JoelleJay (talk) 00:10, 26 September 2023 (UTC)

Incorrect, read what WP:NOR actually says and you'll see that primary sources are perfectly acceptable for the verification of facts. All analyses and interpretive or synthetic claims about primary sources must be referenced to a secondary or tertiary source and must not be an original analysis of the primary-source material by Wikipedia editors. but when there is no verification or interpretation happening a primary sources is fine. Thryduulf (talk) 18:42, 25 September 2023 (UTC)

What WP:OR says in that primary sources can be used in an article, but for an article to exist secondary or tertiary sources must also be used; Do not base an entire article on primary sources and Secondary or tertiary sources are needed to establish the topic's notability and avoid novel interpretations of primary sources.. BilledMammal (talk) 19:20, 25 September 2023 (UTC)

Secondary or tertiary sources are needed to establish the topic's notability You will note that this is talking about notability, not verifiability. Thryduulf (talk) 19:34, 25 September 2023 (UTC)

The full line is Secondary or tertiary sources are needed to establish the topic's notability and avoid novel interpretations of primary sources. And you'll note that this discussion is about when to have an article, not just verifiability - and I'm genuinely confused why you thought it was limited in that way. BilledMammal (talk) 19:56, 25 September 2023 (UTC)

This was a response to @JoelleJay asserting that Secondary sources are a requirement for NOR, which they are not. As for the full quote, I addressed that in my previous comment when there is no verification or interpretation happening a primary source[] is fine. Thryduulf (talk) 20:31, 25 September 2023 (UTC)

And to comply with our policies on origional research, articles must have secondary sources per the quotes I provided. JoelleJay's statement appears to be correct? BilledMammal (talk) 20:38, 25 September 2023 (UTC)

NOR could be read to require some secondary sources but it does not disallow an article that has a mixture of primary and secondary sources. JoelleJay's statement can be read as a claim that NOR entirely disallows primary sources. And the phrase "no verification" in Thryduulf's post makes no sense to me; it does not violate NOR to provide two citations that support the same claim, one of which is secondary and the other primary. Jc3s5h (talk) 20:53, 25 September 2023 (UTC)

Reading my comment again, I think I meant to write "synthesis" rather than "verification". An article consisting of entirely primary sources with no synthesis or interpretation of those sources is entirely compatible with NOR. In context, I read JoelleJay's statement as saying that NOR requires all articles to have secondary sources, which it does not. GNG requires secondary sources to exist, but doesn't require them to be present in the article (or even easy to find). Thryduulf (talk) 21:19, 25 September 2023 (UTC)

What, should I have said "non-primary" instead of "secondary"? If an article cannot be based on primary sources, and shouldn't even have large passages based on them, then of course non-primary sources are required by NOR? And if non-primary sources are needed to establish notability, then how can an article that has zero non-primary sources be established as notable? JoelleJay (talk) 23:26, 25 September 2023 (UTC)

@FOARP: "Likely"? Really? Do you have any data to back that up? AFAIK the errors made by the editors you cite were a small fraction of even their own output, never mind the hundreds of thousands of similar articles created by editors who have never made any waves. – Joe (talk) 16:04, 16 September 2023 (UTC)

The errors in GNIS and GEONet Names Server are par for the course - and discussed at length in the links in the background section. FOARP (talk) 21:18, 16 September 2023 (UTC)

Yes, and those two databases are specifically excluded by WP:GEOLAND (but not your proposed new version). – Joe (talk) 13:01, 20 September 2023 (UTC)

Those sources are already addressed at WP:RSP; while it's probably a good idea to keep the restriction in NGEO as editors creating articles based on it sometimes forget that other policies exist and apply, removing it won't permit editors to create articles based on them. BilledMammal (talk) 13:06, 20 September 2023 (UTC)

Because it is absent of context. What other villages exist in County Y, Province Z? How are they distributed spatially? Is P around the average population of this set, or an outlier? If I have to navigate to a different article to understand basic questions like these, I'd feel comfortable characterising such a stub as low quality. How is it high quality? Folly Mox (talk) 15:32, 16 September 2023 (UTC)

We're discussing notability, not quality, and one of the important reasons why you might have a stand-alone article for town X in county Y in province Z even though there's nothing else written in the article is because you may be able to link it to other articles in other Wikipedias where you may be able to find more information. I just looked up a place in a foreign language Wikipedia that doesn't have an English language article - there should be nothing wrong with a little stub in English on that particular place. SportingFlyer T·C 08:39, 17 September 2023 (UTC)

Our notability guidelines shouldn't be encouraging the creation of low quality articles; they particularly shouldn't be encouraging the creation of articles that are so low quality that they violate policies like WP:NOT and WP:OR.

Our purpose isn't to be a directory to foreign language articles. BilledMammal (talk) 09:03, 17 September 2023 (UTC)

What in that policy actually rules out being a directory of foreign-language articles? It says that Wikipedia is not a directory of everything in the universe that exists or has existed; the set of things that have Wikipedia articles in other languages is obviously more restricted than that. None of the specific no-no's (genealogy for the sake of genealogy, product pricing, etc.) either apply directly or are analogous enough to be applicable. I know that Wikipedias in different languages have different inclusion standards, and just because the Ruritanian WP includes an article doesn't mean we have to have an exact counterpart, but geographic locations are a case where parallel pages do make a considerable amount of sense. At the very least, WP:NOTDIRECTORY isn't a knock-down argument against it. XOR'easter (talk) 22:20, 17 September 2023 (UTC)

Wikipedia articles are not simple listings without contextual information showing encyclopedic merit. If the only justification for an article like this to exist is to link to external resources, then its existence violates WP:NOTDIRECTORY as it is nothing more than a simple listing without contextual information showing encyclopedic merit. BilledMammal (talk) 22:35, 17 September 2023 (UTC)

With NGEO though we've decided through long standing editing practices that the entire set of populated places is notable. The NOTDIRECTORY part you quote links to LISTCRIT which says selection criteria should be unambiguous and supported by reliable sources. While this isn't a list, Giddawa qualifies for both of those things. Unlike a cricketer of questionable notability, it's easily verifiable, and we're not improving the encyclopaedia by deleting it. SportingFlyer T·C 09:22, 18 September 2023 (UTC)

I think you misunderstand the point I'm making; I'm merely saying that links to foreign language articles isn't a justification to keep articles like the one I linked, per WP:NOTDIRECTORY, not that there aren't other reasons to keep it. BilledMammal (talk) 09:26, 18 September 2023 (UTC)

@SportingFlyer - “we've decided through long standing editing practices that the entire set of populated places is notable.” - This simply isn’t true. There have been large-scale deletions and redirection of verifiable “populated places”, because those “populated places” turned out to be statistical artefacts (eg the Iranian Abadi articles about pumps, bridges, factories, farms etc.). The GEOLAND standard requires “legal recognition” and doesn’t apply to every populated place - the problem is this is still over-broad and requires different coverage from one country to another depending on differing legal systems. FOARP (talk) 06:13, 28 September 2023 (UTC)

Joe, I've seen you around, you're no idiot. Anyway: my motivation is that obscure articles lack scrutiny, so the risk of inaccuracies is too high. Contributions aren't vetted, a-la Britannica. Given our natural lack of formal review, and our dearth of contributors, I think maintainability is a prime concern. We're the 7th most visited website, and false info entails real-world harm, including citogenesis, which makes inaccuracies very hard to fix.

According to WP:STUB, stubs should be "capable of expansion". I'm not trying to make an appeal-to-authority ("appeal-to-guideline"?), but it raises interesting questions: when people mass-create articles, do they take into account expansion potential? If not, are non-expandable stubs desirable? And given that non-notable articles need to be deleted eventually, can't we avoid their creation in the first place, saving the time of both the creator and the AfD-nominator?

Months ago BilledMammal showed that stubs are rarely improved by anyone other than their creator. So our long-held assumption, that stubs are a way to incubate articles until someone improves them, and that "stubiness" is temporary, is no longer true for many stubs. Wikipedia's changed, and the low-hanging fruits are gone; why shouldn't our practices adapt? DFlhb (talk) 16:04, 16 September 2023 (UTC)

Thanks for explaining. I asked you to treat me like an idiot because reading statements like yours (you're far from the only one), I genuinely feel like I am missing some fundamental piece of the puzzle. It simply does not make sense to me that creating stubs can be a problem in and of itself, and yet many people seem to state this as if it is self-evident. Rather than assuming that you're stupid (or I am), I assume there's a more basic misalignment of views underlying this disagreement.

Reading your response, it definitely seems like a big factor here is whether or not you feel like Wikipedia is in a consolidation phase, struggles with maintainability, and/or cannot tolerate more 'potential articles'. I've been around here a while, created a few dozen articles, and have a list of ones I'd like to create long enough that I'll never realistically finish it, so I'm a bit more optimistic on that score. Unfortunately it does sound more like a difference of perspective than something we'll come to an agreement on. – Joe (talk) 16:38, 16 September 2023 (UTC)

Right; it's a difference in views/approach, not something that can be easily worked out. And maybe I'm wrong on all this; still trying to read around, and open to changing my mind. DFlhb (talk) 17:53, 16 September 2023 (UTC)

There's way more low-hanging fruit than you think. (My current drafts and protodrafts include a Pulitzer winner, multiple recent popular works of media, and a major active political figure in an enwiki market -- and that's just what's redlinked, not what's in bad shape.) The trouble we have is not that we're in a maintenance phase -- quite the opposite -- but that we don't have the manpower for the expansion we need. The myth of the maintenance phase makes this worse, because people who've been exposed to it in the first place are people who are already part of the project so should be expanding articles. This is one reason I keep banging the drum of no seriously Wikipedia is not paper, in that how something is covered can change on a regular basis and that's a sign of the project functioning well. Increasing notability thresholds specifically is the worst of all worlds, because it decreases rather than increases the opportunity for editorial judgement on how to cover things that may or may not have article potential. Vaticidalprophet 21:24, 16 September 2023 (UTC)

@Joe Roe: I also find the idea of rate-limiting article creation to be unfortunate: it runs counter to what the readership imagines to be the ethos of Wikipedia. But I can also see how mass creation of stubs, if not well-planned, can have the same effects as mass vandalism, namely, a loss of trust on the part of the reader and a loss of morale among editors. On some other wikis, Lsjbot is the logical conclusion to this form of vandalism.

What I struggle with when evaluating this proposal is that we're treating the subject's notability as a proxy for the article's quality as a proxy for the editor's behavior. How am I supposed to explain to a prospective Wikipedian that they should not write a stub about a real place until they've called around about offline sources to prove that an article "can be developed using known sources", and that this hurdle exists because some people once misused databases to run rampant over the site? This is why we can't have nice things?

To me, the turmoil over databases like GNIS (the U.S. federal government's official gazetteer) actually shows that an individual basic error can persist for a long time in a prominent, somewhat mission-critical reference work without causing the sky to fall. We want Wikipedia to be better than that, but not at the cost of cutting off the motivation for people to transform themselves from readers to editors.

My other big hobby is contributing to OpenStreetMap, which has somehow managed to institute an automated edits code of conduct that requires getting buy-in from the community. This policy is the lodestar of mappers who identify as "craft mappers" and view automated edits with varying levels of suspicion, while the people who fill in the map at scale often chafe at the red tape. The policy is as robust as Swiss cheese, but it does send a message to people in a hurry that community health comes first.

Nowhere in this calculus is the idea that a contributor cannot micromap some kind of map feature in detail just because someone once botched a bulk import of it (which does happen, to be sure). Contributor behavior doesn't fundamentally change the nature of the real world that we document. A wiki map is very different than a wiki encyclopedia, but regardless a wiki is truly defined by what it can be, more than what it cannot be.

 – Minh Nguyễn ^💬 00:55, 17 September 2023 (UTC)

• Comment please split these mass deletion/merge RFCs out of VPR in the future. See User Talk:BilledMammal#Page size Mach61 (talk) 04:35, 16 September 2023 (UTC)

  You've already said that; see also the reply on my talk page. BilledMammal (talk) 19:39, 17 September 2023 (UTC)

• Proposal There are really two problems being discussed here:
  – The argument "Populated, legally recognized places" is vague
  – People will mass-create entries from a database.
  We're ignoring the fact these will introduce further problems:
  – "Settlements and administrative regions" is not actually less vague than what we currently have: how is a settlement actually defined differently than a populated, legally recognized place? This won't actually change anything at AfD but could lead to notable places being deleted. For instance, Hillat Abdul Saleh might fall afoul of the new rule and might get listified, but this would sever the link between the article and the Arabic-language article, which is better developed.
  – Nowhere on the project do we delete notable under-sourced entries, which is what this proposal would actually do, which I could see leading escalating deletion drama.
  I think we can all agree that there are a finite number of notable places that fall under WP:NGEO, and that there are still a lot of places eligible for articles which haven't been written yet. Back in the WP:NSPORT days used to have a master list of fully professional leagues where the rules for each country were continuously vetted. Instead of enforcing a new rule that's not based on our long-standing notability guidelines, why not attempt to define what would make a place notable per NGEO for each country? Most countries should have some sort of legal definition of a settlement, we just need to be able follow it consistently (like @Alexandermcnabb:'s odd UAE census problem.) SportingFlyer T·C 20:18, 17 September 2023 (UTC)

  Back in the WP:NSPORT days used to have a master list of fully professional leagues where the rules for each country were continuously vetted. Instead of enforcing a new rule that's not based on our long-standing notability guidelines, why not attempt to define what would make a place notable per NGEO for each country? That didn't work for NSPORT; why would it work here? BilledMammal (talk) 20:34, 17 September 2023 (UTC)

  This is an entirely different situation. Not only is there a limit to the number of maximum number of articles under NGEO, but NFOOTY wasn't the NSPORT problem. SportingFlyer T·C 23:16, 17 September 2023 (UTC)

  Not only is there a limit to the number of maximum number of articles under NGEO Can you clarify what this maximum is?

  NFOOTY wasn't the NSPORT problem NFOOTY was the largest problem with NSPORT, with the community widely considering it far too inclusive. BilledMammal (talk) 23:28, 17 September 2023 (UTC)

  Sure. Every week there are new young football players playing games who fall under GNG/would have fallen under NFOOTY. NGEO doesn't have this problem. Even though new towns and cities can and are being built, the set of topics that pass NGEO are more or less fixed - for instance, unless something exceptional happens, we will never need a new NGEO article for the country of Liechtenstein.

  Two asides. First, the football project worked hard to maintain standards and to cull players which didn't meet GNG. It wasn't entirely perfect, but it mostly worked well - but there are a lot of footballers in this world. Second, the article of Mäls - the Liechtensteiner hamlet I pulled up - would be eligible for deletion under this new guideline, but what good would that do? It's also easy to find the town's complete history online, and deleting it on content grounds would mean we no longer had a full set of Liechtensteiner populated place articles, which is just one of the reasons why I don't understand this proposal. SportingFlyer T·C 23:42, 17 September 2023 (UTC)

  I don't think comparing football players to countries is the most equivalent comparison; no one is saying we shouldn't have an article on every country - obviously, we should - but people are saying we shouldn't have an article on every football player, or every listed building - and obviously we shouldn't.

  A better comparison to football players would be listed buildings; in the UK, one in every 50 buildings is a Grade II listed building. Every week, new buildings are being listed, and most of them don't warrant an article.

  It wasn't entirely perfect, but it mostly worked well - but there are a lot of footballers in this world Let me remind you of John Charlton; it worked so poorly that it took three AfD's to delete an article on a footballer who was clearly not notable. BilledMammal (talk) 00:02, 18 September 2023 (UTC)

  No, there's no real comparison to footballers or buildings or otherwise - we've decided as a community that this set of articles is notable if they are verifiable, which is different than almost everything else on the website. There's still absolutely no problem with that approach. The problem here is that some external databases are too loose for our article set, and mass creators were not bothering to actually verify the places they were creating articles for, not that the notability guideline is incorrect. And changing the guideline doesn't solve either of those problems. SportingFlyer T·C 09:25, 18 September 2023 (UTC)

  Nowhere on the project do we delete notable under-sourced entries this is not for lack of trying, especially with regards Olympians, cricketers, etc. where draftifcation without review for notability is being persued (because straight deletion was rejected by the community). I have no doubts that if this proposal passes the same will be true for populated places. Thryduulf (talk) 20:49, 17 September 2023 (UTC)

  because we have already looked at that as part of the initial discussion. Take the UK. Towns incorporated under charter in medieval times can be found to be legally recognised, cities are granted charter by the monarch so are legally recognised. Administrative Regions are also legally recognised by government legislation setting them up, but then every town and village not covered under these would, under the current GEOLAND sng, have to meet GNG. Multiply that by how many countries in the world, many of which don't actually have legal recognition. Or they have complex system for Administrative Regions (ie. Development zones) which would probably full under the same rules irrigation districts. Trying to simplify the wording was what I started the whole process off to do. Davidstewartharvey (talk) 21:11, 17 September 2023 (UTC)

  So, if the question of when to write a stand-alone article is intrinsically complicated, why is simplifying the wording the right move? That just further separates the guideline from the material it's supposed to be giving guidance about. XOR'easter (talk) 22:05, 17 September 2023 (UTC)

  I don't think anyone believes an irrigation district is covered by NGEO? Plus, how many populated named places in the UK still lack a stand-alone article, and how many places of any size in say England wouldn't have legal recognition somewhere? I also looked up a couple very small English places, which are clearly named places. A couple of them are sourced and manicured stubs with photos, but the sources don't necessarily pass GNG. Why would we possibly want to delete those? SportingFlyer T·C 23:30, 17 September 2023 (UTC)

  If you can find evidence that there is legally recogised status for villages in the UK, then please share as several editors have tried and failed. Based upon the current rule, we could delete these articles, something which I think is wrong as they do exist, but we have a consensus on wikipedia that is not enough. So I see the new proposal a lot softer than what could happen if we actually followed our current policy.Davidstewartharvey (talk) 16:29, 18 September 2023 (UTC)

(I moved a couple comments here up to the survey section, as they appeared to be intended as votes rather than comments. If I am mistaken, please feel free to move them back.) – Minh Nguyễn ^💬 17:35, 20 September 2023 (UTC)

You'll find that you can write a fleshier stub on most legally recognized populated places. Even if it's map observation. The problem is the sub stubs where all there is is "xxx is a village" and you do a google search and no population data for it, nothing but databases. In those cases, I absolutely support the proposal to encourage merger into tabled lists. What about an Option 3. Legally-recognized populated places are generally considered to be notable, but the creation of short stubs where there is nothing but a database mentioning location online is strongly discouraged. The minimum requirement expected is a population figure and basic location details as a precondition for a stub (like Aung Myay), where the stub can be fleshed out to resemble an encyclopedia article instead of an xxx is a village type database substub. In cases where there is no population data and it resembles a database, consider merging into a tabled list.♦ Dr. Blofeld 12:02, 22 September 2023 (UTC)

A tale of two countries

Reading the votes so far, I observe that there's a commonly held fear that the proposed changes would insert extra scrutiny of cited sources into the process of assessing notability. Whereas currently, there is of course an expectation of reliable sources, but that's orthogonal to the subject's inherent notability to the extent that we can ascertain it. I'm also surprised that relatively few concrete examples have been explicitly named.

On the bright side, the United States will probably be OK, though many civil townships, census-designated places, and unincorporated places will need to be expanded, probably using information from (non-census) databases, to rescue them from deletion. Since the guideline will no longer explicitly discourage GNIS, it will be quite handy for this purpose. I bet most who like to conjure up the GNIS boogeyman don't know that the database very often includes historical details cited to reliable sources, if you can resolve the source codes. For example, our entry on Rudd, Iowa, omits the most interesting thing about it, the story of how it got its name. [15] Meanwhile, our encyclopedia is completely oblivious to Howard, Idaho [16], and Green Castle, Johnson County, Iowa. [17] But under the existing guideline, these places would have a decent shot at GNG anyways; the sources listed in GNIS are just the tip of the iceberg.

Not every country is so lucky. Turning to the other country that I write about, Vietnam, I certainly wonder if readily available online sources will be adequate for a place like Tân Hà commune (equivalent in stature to a township in the U.S. or a parish in England). The article already has a good start; it just needs a Vietnamese Wikipedian to translate to English. One need not worry that GeoNames has hallucinated this place. Otherwise, how would the infobox include a photo of the Tân Hà Commune People's Committee headquarters (equivalent to a town hall)? But it contains very little text "beyond statistics, region, and coordinates" and relies almost entirely on a government gazetteer and official website. The one addition is its administrative history, citing the relevant government edict (a primary source). Digging around online for a hot second, all I could find are content mills punctuated by the occasional news article that, if included, would surely provoke howls of WP:RECENT and WP:TOPIC and WP:COATRACK regarding this 40-year-old commune. [18]

Since this is a region that has been settled for millennia, there's likely a wealth of information about this commune's administrative predecessors and the local human geography in general. Much of it will be offline, perhaps in rare books collections in Vietnam, France, and the Vatican. Some of it will be written in chữ Nôm and thus difficult to search for. All of this information can be added in due time. Unfortunately, it will be even more difficult to find sources that technically pass muster as independent secondary sources. This is a country that, by law, lacks an independent press or publishing industry, so the kinds of sources we often take for granted may not exist, through no fault of the community in question. Honestly, under a strict reading of our guidelines, the best chance this place has is that a tourist from the West decided to write about it (most likely misspelling its name).

Other than the photo, this is quite typical of a Vietnamese commune article, at least to start out with. The Vietnamese Wikipedia originally banned articles on communes and everything below them, in order to encourage editors to write about districts and provinces first. But in 2007, the community decided that the ban was counterproductive and voted to allow communes under a guideline very similar to the existing one here. I'd imagine that Vietnam is not nearly the only country that would experience difficulty meeting the standard proposed here, even as the particulars may differ. I'm not entirely sure how I'll vote. Either way, I think the guideline will still be an unfinished one, subject to the usual inconsistent interpretations.

 – Minh Nguyễn ^💬 00:35, 20 September 2023 (UTC)

If y'all could decide soon, I getting ready to start on 1.6 million new permastubs on villages in China and India which qualify under the current N:Geo that don't have an article yet. After that I'll move on to the electoral discticts. :-) Sincerely, North8000 (talk) 17:51, 20 September 2023 (UTC)

Sounds like we need to agree on a rate limit on permastub creation by @North8000. ;^) Minh Nguyễn ^💬 17:59, 20 September 2023 (UTC)

And spoil my fun? :-) At two per minute going 24/7 it will still take 1 1/2 years to make them but they would live forever because it would millions of person-hours at AFD to get rid of them. :-) North8000 (talk) 21:14, 20 September 2023 (UTC)

Ah, but we could set up a bot to delete any articles created by you. Sure, there might be some collateral damage (like this B-class article with ~300 pageviews this month) but it be worth it. Probably. Edward-Woodrow • talk 21:18, 20 September 2023 (UTC)

Not so fast - I picked 12 at random and was able to expand all of them to Start Class by adding postal code, area code, distance to the nearest airport and a brief history of the province in which they're located! –dlthewave ☎ 22:03, 20 September 2023 (UTC)

If there's anyone notable from the village, one can just copy I mean, "summarize" their article, probably doubling the length! You're right, there is potential for growth. We could also transliterate the village name into various languages. Edward-Woodrow • talk 22:07, 20 September 2023 (UTC)

My dog chewed that patch up after I took the picture for Wiki, so Wiki got me to create my only memory. I was also riding high when SS Edmund Fitzgerald was the article of the day but a making zillion permastubs is is appealing enough to make me give up collecting string. :-) North8000 (talk) 11:56, 21 September 2023 (UTC)

Hey, at least WP:N8KSTUBS will give China and India something to agree on. Levivich (talk) 22:39, 20 September 2023 (UTC)

• As it looks like Option 1 is the most likely outcome of this discussions, could we then do an experiment to see if their is a wider concensus regarding GEOLAND? Take one of the current permastubs and take it to AFD and see if those in that circus agree that they meet GEOLAND or not? Especially the point about legally recognised, and if not they they meet GNG?Davidstewartharvey (talk) 17:23, 23 September 2023 (UTC)

  What would that prove? Others have already pointed out above that current guideline provides for editorial discretion about merging articles. For places that are not "legally recognized", such as unincorporated settlements, the current standard rests on WP:GNG, which is not dissimilar to what the proposed wording would apply to all places. Minh Nguyễn ^💬 17:42, 23 September 2023 (UTC)

  Because many editors take a census as meaning that it's a legally recognised place, and so therefore allowed - i.e. a whole raft of permastubs. Davidstewartharvey (talk) 18:44, 23 September 2023 (UTC)

  Firstly, why is a short article such a bad thing? Secondly, there is nothing in any policy or guideline that says legally recognised places must have stand-alone articles - they simply mean that the content should not be deleted. The community has rejected, time and again, proposals to turn Articles for Deletion into Articles for Discussion and/or otherwise fold in discussions of mergers. If you nominate something at AfD that policy says should not be deleted you're obviously going to get people turning up to say that it should not be deleted (doubly so if people who look beyond the first page on google turn up sources to expand the article with). If however you nominate that same content for merging to a suitable location you'll get support from many of the same people who would !vote keep at AfD. Thryduulf (talk) 19:29, 23 September 2023 (UTC)

  I actually think we should merge AfM into RM, but that's a different debate. BilledMammal (talk) 19:31, 23 September 2023 (UTC)

  Mergers (and splits) should certainly have a process that generates an outcome in a finite period of time. I'm not sure if RM is (or isn't) the right answer, but as you say that is a different discussion. 19:46, 23 September 2023 (UTC) Thryduulf (talk) 19:46, 23 September 2023 (UTC)

The present standard means we are literally publishing articles about non-notable houses

Not just that, but articles explicitly pointing out that the people who live in them live alone, and a long way from any neighbouring settlement. See Novino for one example chosen from many of single-occupant "villages" in Russia that were mass-created based on Russian census data. But it's a "legally recognised populated place" since it's a counting-unit in the Russian census, right? Does this strike anyone else as blatantly wrong, ikky, and potentially dangerous? FOARP (talk) 10:30, 27 September 2023 (UTC)

Does this strike anyone else as blatantly wrong, ikky, and potentially dangerous? Not really, especially if we approach this without the appeal to emotion. The content would probably be better merged somewhere (but I don't immediately know where) but that's not really your point. More importantly, your issue seems to be (again) with the mass creation of articles you don't like. Thryduulf (talk) 10:44, 27 September 2023 (UTC)

It's not just an appeal to emotion - the danger in advertising the location of random single-occupant houses and saying how far they are from potential help is obvious. FOARP (talk) 11:17, 27 September 2023 (UTC)

the danger in advertising the location of random single-occupant houses and saying how far they are from potential help is nothing other than an appeal to emotion. Per Joe below it is also factually incorrect. Thryduulf (talk) 11:35, 27 September 2023 (UTC)

What part of the description is wrong except that it apparently now is unpopulated? We have many other such articles showing a single-occupant miles from anywhere. FOARP (talk) 12:15, 27 September 2023 (UTC)

FOARP, while I've occasionally been frustrated by the way you've gone about trying to change policy in this area, I generally think you've got a really sharp analysis of notability and WP:NOT, which has changed my mind more than once. So please don't take it the wrong way when I say you've gone absolutely off the fucking deep end trying to rationalise this. Are you seriously suggesting that there are nefarious people trawling the English Wikipedia looking for isolated houses in Central Russia? Or that this something that someone who's chosen to live alone in the middle of the forest is going to be worried about? You don't like stubs about obscure places. That's okay – others disagree, but it's okay. You really don't need to tie yourself in knots trying to come up with justifications for it. – Joe (talk) 12:09, 27 September 2023 (UTC)

"Are you seriously suggesting that there are nefarious people trawling the English Wikipedia looking for isolated houses in Central Russia?" - would you be OK with Wikipedia having such an article about your house? Or would precisely these concerns come to your mind? FOARP (talk) 12:17, 27 September 2023 (UTC)

It wouldn't make a blind bit of difference, because people can already find my house by looking at a map. Or walking down the street. Or asking the postman. I'm utterly baffled at why you would think that the location of human settlements is a secret. – Joe (talk) 12:48, 27 September 2023 (UTC)

Being a secret or not does not mean we should publish articles publicising conveniently for anyone who might be interested (which, being the internet, means everyone, including criminals) the locations where people live alone in isolated locations. Why are we publishing articles that are about single houses, simply because they are mentioned on a census - something never intended to be used as a list of "villages" - as a line-item? FOARP (talk) 12:57, 27 September 2023 (UTC)

These houses are isolated because they are farms which need to be surrounded by agricultural land rather than other houses. Most of the articles don't even have co-ordinates. There is a very low risk of anyone abusing Wikipedia to research the best place to steal a turnip. Certes (talk) 13:04, 27 September 2023 (UTC)

Which of course leads us to many of these places actually being businesses/organisations... FOARP (talk) 13:07, 27 September 2023 (UTC)

The addresses of businesses are even more readily discoverable than the addresses of individuals, so your problem that isn't a problem is actually even less of a problem. Thryduulf (talk) 13:33, 27 September 2023 (UTC)

That's a problem for a separate reason; NCORP, not NGEO, applies. BilledMammal (talk) 13:44, 27 September 2023 (UTC)

To expand on what BM just said: the issue here is that business/organisations can also be the reference point used by the census. So, if a farm, factory, school or whatever is used as a reference point, hey presto! We have an article about a totally non-notable organisation without the need for any secondary sourcing at all. And yes, this does happen with Russian census data - not just farms, but factories (like "Metallist" in the same district as Novino) and schools (like "Shkolny" - literally "school" - again in the same district as Novino and apparently just the school for the local village). FOARP (talk) 14:26, 27 September 2023 (UTC)

According to ru:Новино (Кольчугинский район), it's actually no longer populated. However, in the past it was larger, peaking at a population of 103 in 1905, and there are historical records of it going back to at least 1637. In other words, it has a verifiable history longer than most US cities. With access to offline Russian sources like local newspapers, I'm sure you could write a quite lengthy article about Novino. We can at least say that it's definitely a (former) village, not a house. – Joe (talk) 11:02, 27 September 2023 (UTC)

See, this is the problem with machine-translating a term like "деревня" and saying "ah, that's a village". In reality, it can also be translated as "hamlet" and only ever referred to the smallest variety of rural settlements. The historical records referred to are to a single person's property - a single line-item. For "local newspapers" I would be very surprised actually if there is any coverage at all of Novino per se - but if there were that should be produced in the article for it to be kept. Single-line-item listings in census surveys do not create anything that can actually be written about. The reference to US cities is random - perhaps you could explain why we should discuss US cities here? FOARP (talk) 11:28, 27 September 2023 (UTC)

I can read Russian well enough without machine translations, thanks. Whether you call it a village or a hamlet, it's a human settlement which, as our current policies wisely recognise, tend to have deep and well-documented histories even if they're (currently) small or abandoned, and even if those sources are hard for English Wikipedians to immediately lay their hands on. I assume (or hope!) it's uncontroversial that we should have an article on major US cities, hence my observation that even tiny villages elsewhere in the world can have a more enduring notability. – Joe (talk) 11:56, 27 September 2023 (UTC)

I can understand your concern about systematic bias, but the fact is that we still have to actually have at least some of that "deep and well-documented history" to write an article. We cannot simply make stuff up based on nothing more than a number in raw census data and use that to support an article whose existence implicitly suggests the existence of an entirely speculative history - that is textbook WP:OR. It is equally possible that eg. the numbers you are pointing to are an error in data input (not uncommon when dealing with raw data, and a major reason why we should not be using it as the sole source for entire articles.) --Aquillion (talk) 09:43, 28 September 2023 (UTC)

One follow-up statement here: Russia is every bit as much a settler-colonist state as the United States. New York (founded 1624) is centuries older than some major Russian cities, including Vladivostok (1860), Khabarovsk (1858), Magnitogorsk (1724) etc. Even St. Petersburg was founded in 1703.

The concept that individual farms in Russia should have articles about them because of some (undemonstrated) village-history, and are all or even mostly ancient settlements, just because Russia is located in Eurasia and not the Americas, doesn’t appear in evidence. FOARP (talk) 23:56, 29 September 2023 (UTC)

And if the history you speculate about based on census data actually does turn out to be true, and someone actually the sources you speculate might exist, we could, one day, have an actual article using them, based on actual WP:RSes sufficient to pass the WP:GNG. But without that we don't actually have enough to write an article; the fact that you had to resort to flat speculation and (bluntly) WP:OR based on census numbers shows why we shouldn't use that sort of raw data as the sole source for an article in the first place. It is possible that your speculation is correct and that this is an actual village with an actual history; it is also possible that the census figures represent something else entirely or are simply inaccurate, as such figures often are. Without better sources we cannot know for sure and therefore we should not write an article that plainly (as you implicitly acknowledge) implies to the reader things that the sources we have available do not actually support. --Aquillion (talk) 09:43, 28 September 2023 (UTC)

I'm an inclusionist but we must have limits. My house doesn't have an article, because it's not notable, and nor are those houses. I came across this problem when listing disambiguation pages that might need to be created. One example (typical of many dozens) is Davydkovo, a name shared by 1 2 3 4 5 other places, none of which have a population above ten. It is difficult to see how these articles benefit Wikipedia. They are non-notable topics, which get a free pass to article status because they are places and we have an axiom that every place is notable. It's time to challenge that axiom. Certes (talk) 12:07, 27 September 2023 (UTC)

@Certes - Agree. On the specific point of mass-creation of articles about identically-named places, often the issue is that the census has divided an existing settlement into more than one listing. For example in the same district as Novino there is Litvinovo and Litvinovo (settlement), Vladimir Oblast - which is the real settlement here? Probably neither as a glance at the map appears to show them as neighbourhoods of Kolchugino.

But they appear in the census so... presumed notable? But that's how our present standard works. FOARP (talk) 12:38, 27 September 2023 (UTC)

Yes, the creator seems to be following our current guidelines, which is why I've been tidying up the same-name confusion rather than proposing for deletion. In case it's of interest, I'm running User:Certes/Reports/Qualified titles lacking disambiguation one initial at a time. It's currently on L, which is atypical due to "List of..." false positives, but the intro links to A–K results. After A–C, I realised that most of the work was on Russian places too tiny to be the most important pages I could improve on Wikipedia, so I started skipping them. (Now Turkey dominates, but those places have populations in three figures rather than one, and there are often surnames etc. to include in the dabs.) Certes (talk) 12:55, 27 September 2023 (UTC)

Ah yeah, the Turkish Mahalle/Neighbourhood articles. We even had an AFD that decided to up-merge the ones Lugnuts created, which was carried out, and of course the up-merging has since been reverted with GEOLAND used as the reason. FOARP (talk) 13:05, 27 September 2023 (UTC)

FWIW, I don't like the idea of going after all the articles created by Nikolai Kurbatov, who states "I am a disabled with bipolar disorder. The last years i have constant depression, which destroys the desire to live, and daily productive work is the only joy in my life." BeanieFan11 (talk) 12:13, 27 September 2023 (UTC)

Wikipedia is not therapy, however much sympathy one may have for that editor (who I have refrained from naming for exactly this reason). FOARP (talk) 12:19, 27 September 2023 (UTC)

I also noticed that comment but intended to refrain from naming names. I am concerned about the effect deletion might have on authors. It's tempting to sacrifice Wikipedia's usability to provide therapy where it's needed but that's a slippery slope, especially if two editors' needs require Wikipedia to bend in different directions. Certes (talk) 12:31, 27 September 2023 (UTC)

This reminds me of an administrative area that I once wrote about. For years, Mill Creek Township, Hamilton County, Ohio, had a population of just a mother and her son, surrounded by many dead people. If this article had been killed off for what was in its original revision, I don't think I would've ever known about the township in order to breathe new life into it.

Incidentally, as of 2020, there were nine incorporated cities, villages, etc. in the United States that had a population of zero. There may be a story behind each of them that's more notable than some incorporated places infinitely larger. For example, in reality, South Park View, Kentucky, is just a nondescript office park, but it's legally recognized by not only the Census Bureau but also the governor's office as a city. This is an objectively unusual phenomenon worth documenting.

 – Minh Nguyễn ^💬 09:13, 28 September 2023 (UTC)

It's possible that a few of the Russian hamlets may once have been thriving cities. Anyone who finds a reliable source saying so is very welcome to apply for a REFUND or rewrite the article. However, in the absence of any indication to the contrary, it seems much more likely that each article represents what is, was and has always been a farm supporting one family and zero notability. Certes (talk) 09:55, 28 September 2023 (UTC)

Precisely this. We have many, many geographical stubs for which there is a prima-facie presumption that they have never been genuine settlements (e.g., ones named after the thing that they are - e.g., "XXXX's farm", "YYYY station", "School", "Paper Works") but are maintained under the presumption that, since they are or were line-items in a census, then they must all be Nineveh & Tyre, each a Troy waiting to be discovered by archaeologists. FOARP (talk) 13:00, 28 September 2023 (UTC)

How many, roughly speaking? Under the proposed changes, how many days would it take to delete them via AfD? Minh Nguyễn ^💬 17:28, 28 September 2023 (UTC)

We have 3,668 stubs on places with a population of 0–9. Some of those should be retained, and many articles not in that set could be deleted, but it's a ballpark figure. Certes (talk) 18:55, 28 September 2023 (UTC)

So it'll take at least a year roughly. Is the proposed change to the notability guideline, coupled with the proposed rate limit on AfD, the most effective vehicle for addressing the BLP-ish privacy concern urgently raised at the top of this section? Minh Nguyễn ^💬 19:05, 28 September 2023 (UTC)

No; notablity and privacy are unrelated. If there is a concern (I'm unconvinced) then it should be addressed urgently and separately whether the offending text is in Obscurokov or London. Certes (talk) 19:16, 28 September 2023 (UTC)

In the abadi clean up we deleted 4000+ zero-population “village articles for Iran alone. FOARP (talk) 00:02, 30 September 2023 (UTC)

I was mindful in particular of the Russian selo articles in my !vote above. We should not presume it was always a sub-hamlet, a single house, or whatever it is now; that's classic unconscious bias. The "places" that were administratively created out of railroad stations are a problem that should be dealt with after a specific discussion about such places. As in the case of the Iranian abadi, which I referred to. Yngvadottir (talk) 02:49, 3 October 2023 (UTC)

This is why GNG will always be a better standard when we're talking about a class of articles thousands strong where the best we can say is that sourcing might exist but have zero evidence that it does. For Wilton, North Carolina, an unincorporated community, the only historical info I was able to find was that it had a political meeting in the 1860s. Compare that to Moss Neck, North Carolina, a place of almost no significance now but had a nice history to uncover. As someone who has tried to dig up some information on small places like those at issue, I'm fully convinced that there isn't enough consistency in coverage for these exceptionally small places (and various artificial products of census keeping practices) to warrant making broad assumptions of notability. Joe's sweeping assertion that small "human settlement[s] ... tend to have deep and well-documented histories even if they're (currently) small or abandoned" is patently false. It's simply a matter of some do and many don't. Certes hits the nail on the head: "It's possible that a few of the Russian hamlets may once have been thriving cities. Anyone who finds a reliable source saying so is very welcome to apply for a REFUND or rewrite the article. However, in the absence of any indication to the contrary, it seems much more likely that each article represents what is, was and has always been a farm supporting one family and zero notability." -Indy beetle (talk) 04:39, 3 October 2023 (UTC)

Exactly this. Assuming that a place that is unpopulated or which does not have a substantial population was always like that unless evidence is produced to the contrary is not an example of “unconscious bias” - it is simply how logic works: the burden is on the person making the claim to prove it. Rather, assuming that every statistical item in a list was once a thriving village with its own newspaper is simply illogical when so many of them are very obviously railway stations, factories, mines, sidings, lumber-camps, barracks etc. etc. FOARP (talk) 08:04, 3 October 2023 (UTC)

The other thing is that failure to find adequate information isn't necessarily a permanent problem. If that information turns up, an article can always be written or recreated. It's more important for articles to be accurate than it is for them to exist. Mangoe (talk) 21:18, 3 October 2023 (UTC)

Notifications

• Notified to the WP:NGEO talk page here. FOARP (talk) 16:29, 15 September 2023 (UTC)
• Notified to WP:CENT in this edit. FOARP (talk) 16:38, 15 September 2023 (UTC)
• @Davidstewartharvey, Bookku, Certes, North8000, Horse Eye's Back, Selfstudier, Crouch, Swale, Visviva, BilledMammal, Hut 8.5, Bkonrad, Newimpartial, Hike395, Donald Albury, Viridiscalculus, Mxn, Avilich, Dlthewave, Choess, Joe Roe, Ezhiki, Rschen7754, Atlantic306, XOR'easter, and Imzadi1979: Ping here, since I don't believe the above ping went through (I didn't receive it). BilledMammal (talk) 04:31, 16 September 2023 (UTC)
• Rosguill
  Notified: WT:NPP. ~~~~

The following discussion is an archived record of a request for comment. Please do not modify it. No further edits should be made to this discussion. A summary of the conclusions reached follows.

---

Should the process described in this discussion be adopted in some form; to wit, should unreferenced articles be subject to a PROD-like process where, after a two weeks (or other suitable time), the article is deleted? 19:11, 6 October 2023 (UTC)

• Support as proposer: I see this as an extension of WP:BURDEN, which is part of Wikipedia's policy on verifiability. That policy states: The burden to demonstrate verifiability lies with the editor who adds or restores material, and it is satisfied by providing an inline citation to a reliable source that directly supports [...] the contribution. [...] Any material lacking an inline citation to a reliable source that directly supports [...] the material may be removed and should not be restored without an inline citation to a reliable source. Unsourced articles, as suggested in this essay, is not valuable to a serious encyclopedia.

In earlier times, Wikipedia needed to expand, to cover a variety of topics, so we could beat out competitors such as Britannica and Citizendium. That time has passed. We should now put more emphasis on quality, ensuring that our articles are trustworthy, well-written, neutral, and verifiable.

Obviously, the nominator of an unsourced article under this process should take time to conduct a quick WP:BEFORE; but again, the burden to demonstrate verifiability lies with the article creator. Article reviewers and other editors shouldn't have to break their backs scratching up sources for a now-destroyed theatre in Moscow.

This process would allow editors, using an ((unreferenced prod)) tag or similar, to nominate such unreferenced content (viz., articles) for deletion. These proposals would have the regular notifications to WikiProjects and user talk pages, giving involved editors plenty of time to add at least one reference that supports article content.

This would be a separate template from the existing ((unreferenced)) tag, so articles tagged with that template would not all suddenly be nominated for deletion. However, this would allow us to chip away at the backlog of 119 000 or so unreferenced articles more quickly, as the backlog would no longer grow – new articles would be tagged with the new PROD.

Given the two weeks to add at least one reference to the article, I find it unlikely that many articles would actually be deleted under this process.

WP:TLDR? A gloss of my 340-word comment: I think this proposal is an extension of our policy of WP:BURDEN. Article creators and interested editors are given two weeks – fourteen days! – to provide at least one reference that supports the article content. This would also help reduce the large backlog of unreferenced articles. An important point to note is that a separate template would be created; articles already tagged with ((unreferenced)) would not be all suddenly up for deletion. Edward-Woodrow • talk 19:11, 6 October 2023 (UTC)

• Oppose. Many people who are not well versed in our policies create articles without references. Why should those of us who know better not look for sources before nominating for deletion, as is the situation now? This proposal seems to imply that articles are owned by their creators. They are not. This seems to be another step away from becoming an encyclopedia towards becoming a Google mirror. Phil Bridger (talk) 19:59, 6 October 2023 (UTC)

  @Phil Bridger: Why should those of us who know better not look for sources before nominating for deletion, as is the situation now?. I have said no such thing. In fact, to clarify any confusion, I have said the opposite: Obviously, the nominator of an unsourced article under this process should take time to conduct a quick WP:BEFORE... Edward-Woodrow • talk 20:21, 6 October 2023 (UTC)

  So what change are you proposing from the current situation? As it is a nominator can look for sources and if none are found can nominate for deletion via AfD or PROD. Phil Bridger (talk) 20:45, 6 October 2023 (UTC)

  Oppose as per your take Immanuelle ❤️💚💙 (talk to the cutest Wikipedian) 02:48, 7 October 2023 (UTC)

• Question: what's the difference between the process proposed here and the existing process of simply making a proposed deletion under WP:DEL-REASON 7 (Articles for which thorough attempts to find reliable sources to verify them have failed)? Is this proposal just meant to reaffirm that the above is a valid reason for proposing deletion? Shells-shells (talk) 21:04, 6 October 2023 (UTC)

  No, it will make it easier to delete articles for which the nominator cannot be bothered to make any attempt to find reliable sources to verify them. That would be the outcome of applying WP:BURDEN to article deletion (to which that policy has never previously applied). James500 (talk) 22:19, 6 October 2023 (UTC)

  That is the point of the proposal, yes. To expedite and facilitate removal of unsourced content, which is worthless and of no value to the project. Edward-Woodrow • talk 22:48, 6 October 2023 (UTC)

  No, unverifiable content is worthless. Unsourced content that is verifiable is not necessarily worthless, and might be very valuable. In my experience, most of the unsourced articles I have examined were very easy to find sources for, and those sources generally verified their content. James500 (talk) 01:02, 7 October 2023 (UTC)

  That's been my experience too, particularly since the Wikipedia Library. Espresso Addict (talk) 03:59, 7 October 2023 (UTC)

• Oppose, I think. If the process is what I think it is (the precise RfC proposal is very poorly explained here), it won't do anything about the backlog, just delete a few random untagged unreferenced articles whose creators have disappeared. Some people are going to want to extend the process to the entire unreferenced backlog, which would be a major disaster. —Kusma (talk) 21:22, 6 October 2023 (UTC)

@Shells-shells, Immanuelle, Phil Bridger, Kusma, James500, Espresso Addict, Pppery, Thryduulf, Red-tailed hawk, BilledMammal, Oknazevad, and David Eppstein: If you're interested here's an example:

• I recently came across Chaimite, Mozambique, an unreferenced article. Initial searches turned up databases and Wikipedia mirrors/forks/scrapers, or passing mentions that only proved its existence. The was created in 2010, by the way.
• If it was recently created, I would draftify it... but it's 13 years old.
• Under this proposal, I would tag it with ((unreferenced PROD)), notify a WikiProject or two, and move on.
• Given the phrasing of DEL-REASON 7, though, I will continue my attempts to verify the information, probably eventually taking it to AfD or regular PROD.

Edward-Woodrow • talk 17:02, 7 October 2023 (UTC)

Or I will spend effort the page creator should have spent getting the article sourced. Edward-Woodrow • talk 17:05, 7 October 2023 (UTC)

No, it is not effort that only the page creator should have spent. As I have said before the page creator does not own the article, and this is a collaborative project, so anyone with an interest in the subject, including you, should spend the effort. Phil Bridger (talk) 18:07, 7 October 2023 (UTC)

I think this is a fine example of our current system working properly. Someone created the article as a translation from a pt article that had a source, but failed to include it. Another editor came across it later and sourced it. Encyclopedia built. Espresso Addict (talk) 22:14, 7 October 2023 (UTC)

This seems helpful. Worth at the very least a redirect to the parent Chibuto District. Deleting articles like this would make our systemic bias worse. —Kusma (talk) 17:10, 7 October 2023 (UTC)

None of that explains why standard prod, AfD, merging or redirecting are insufficient. Especially for places in parts of the world less well covered by the English-speaking internet, "initial searches" are wholly insufficient to declare an article unsalvageable. Thryduulf (talk) 17:29, 7 October 2023 (UTC)

Exactly. This would make systemic bias a lot worse on wikipedia Immanuelle ❤️💚💙 (talk to the cutest Wikipedian) 17:58, 7 October 2023 (UTC)

@Edward-Woodrow: This proposal was explicitly stated not to apply retroactively. Under this proposal, you could not tag a 13-year-old unreferenced article as you propose to tag it. This is a bad example. The proposal has not even gained consensus and already we see severe WP:CREEP in it. —David Eppstein (talk) 18:15, 7 October 2023 (UTC)

No, it would not apply to articles already tagged with ((unreferenced)), something that article was not. Edward-Woodrow • talk 18:30, 7 October 2023 (UTC)

That's not a helpful distinction to make. We can apply this to old articles, but only if nobody has bothered to tag them for cleanup? Why? Another reason to oppose this proposal. —David Eppstein (talk) 17:17, 9 October 2023 (UTC)

• Explanation for @Kusma, Phil Bridger, and Shells-shells: I apologize that I did not sufficiently explain the proposal in my RfC statement or my !vote. The proposal is different from PROD in the following ways:
• The tag may not be removed if the problem is not fixed (like BLP PROD)
• The tag allows a longer period for adding sources.
• PROD is for uncontroversial deletion; this is for unsourced articles. Not necessarily different, but not necessarily the same.
• Re DEL-REASON 7: The attempts need not be thorough, just a quick WP:BEFORE to make sure that the problem can't be fixed super-easily.

I swear there was another difference, but my mind is blanked out at the moment. Edward-Woodrow • talk 21:53, 6 October 2023 (UTC)

• Oppose. This proposal is not compatible with WP:ATD, criteria 6 and 7 of WP:DEL-REASON, WP:NPOSSIBLE or WP:BEFORE. Unreferenced tags are very often erroneously placed on articles that are referenced. Many erroneous tags still remain on those articles. There are editors (including an entire wikiproject) trying to clear the backlog of unreferenced articles by adding sources, and an artificial two week deadline would seriously disrupt their efforts. If an editor is trying to systematically work through the unreferenced tags on, to pick a random example, articles about the chronology of Ireland (which has many obviously notable articles to which tags were inappropriately added instead of sources), it is not helpful to waste his time by forcing him to look at yet another prodlist every day. If a nominator actually cannot find sources, there is no good reason for using a "sticky" prod instead of an ordinary prod. James500 (talk) 22:16, 6 October 2023 (UTC)

  @James500: I think there is a misunderstanding. The proposal would not apply to articles already tagged with ((unreferenced)). Unreferenced tags are very often erroneously placed on articles that are referenced. Many erroneous tags still remain on those articles. There are editors (including an entire wikiproject) trying to clear the backlog of unreferenced articles by adding sources, and an artificial two week deadline would seriously disrupt their efforts is not a problem. Edward-Woodrow • talk 22:46, 6 October 2023 (UTC)

• Support 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...} 22:51, 6 October 2023 (UTC)

  As written, the proposal has a grandfather clause for articles already tagged as unreferenced. —Kusma (talk) 13:33, 7 October 2023 (UTC)

  That's not what Edward-Woodrow said on the VPI thread. My understanding is that someone has to redundantly add the ((unreferenced PROD)) tag, but otherwise they are subject to the same process. * Pppery * _{it has begun...} 17:04, 7 October 2023 (UTC)

• Oppose Tagging or AfD (with deletion sorting) allows a reasonable chance for other editors to find sources to build a collaborative encyclopedia. Long experience with AfD suggests that many nominators either don't do references searches or do them very ineptly, and prods would receive even less scrutiny -- in my experience hardly anyone is organised enough to diligently comb through the messy list of prods. Espresso Addict (talk) 00:14, 7 October 2023 (UTC)
• Strong oppose per most people above. It is already too easy to nominate articles for deletion without even attempting to see if the problems are (a) real, (b) fixable, and/or (c) better resolvable by something other than deletion (e.g. merging or redirecting), we should be strengthening the protections against this not weakening them. Thryduulf (talk) 00:27, 7 October 2023 (UTC)
• Comment. Many NPPs will draftify unsourced articles if they are new enough (falling within the 90 day draftify rule). So there is kind of a process to handle these problematic articles already. –Novem Linguae (talk) 01:30, 7 October 2023 (UTC)
• Support. Unreferenced articles are incompatible with our core policies, including WP:V and WP:OR. This helps address that issue. BilledMammal (talk) 01:37, 7 October 2023 (UTC)

  No, unverifiable articles are incompatible with V and NOR, not unreferenced ones. This proposal does not help to add sources to verifiable but unreferenced articles. James500 (talk) 01:50, 7 October 2023 (UTC)

• Oppose. If the article is wholly unreferenced, and you think that no reliable citations for it exist, then you can already PROD/AfD the article. If the article is wholly unreferenced, and you can find citations, then stubbify the article and use them. The point of Wikipedia is to build an encyclopedia; rather than deleting wholesale, it may be better to have a one-to-two paragraph-long entry on a thing that's well-sourced (like the old paper encyclopedias used to have). And, if we can't find anything to substantiate an article, then WP:DEL-REASON#7 is more than sufficient. There's no need for a special carveout here. — Red-tailed hawk _(nest) 04:17, 7 October 2023 (UTC)
• Oppose. Typically new unsourced articles get draftified, and the proposal explicitly covers only new articles, so this process is unnecessary WP:NOTBURO and WP:CREEP. —David Eppstein (talk) 07:14, 7 October 2023 (UTC)
• Oppose. Fails to keep in line with the spirit of WP:BEFORE, and confuses the difference between article condition and notability, which is a property of the subject entirely separate from the state of the article. oknazevad (talk) 15:40, 7 October 2023 (UTC)

  and confuses the difference between article condition and notability Can you elaborate? I haven't mentioned notability anywhere in my proposal. Edward-Woodrow • talk 16:40, 7 October 2023 (UTC)

  The only true reason to delete an article is if the subject is non-notable or unverifiable. The present state of an article is not a valid deletion rationale. This basically would allow for PRODing articles based solely on their current state of referencing. That is not a proper reason to delete. oknazevad (talk) 21:56, 7 October 2023 (UTC)

  WP:5P5. Edward-Woodrow • talk 22:30, 7 October 2023 (UTC)

  I don't see how what I said disagrees with the five pillars in any fashion. Can you explain what you mean? oknazevad (talk) 03:09, 9 October 2023 (UTC)

• Support as per the discussion for new articles only. This is a situation where realistically only established editors can create articles like this, new or IP editors continuing to try and publish unreferenced articles aren't given such leniency. This theredore only brings established editors inline with what is community practice elsewhere. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 17:28, 7 October 2023 (UTC)

By "established editors", you mean those with autopatrolled? If an autopatrolled editor is creating articles entirely without sources, then the autopatrolled right should be removed. Espresso Addict (talk) 22:04, 7 October 2023 (UTC)

• Oppose, per several others above. BeanieFan11 (talk) 18:17, 7 October 2023 (UTC)
• Oppose. Creating unreferenced articles is undesirable and far from ideal IMO, and merely adding a citation is not at all difficult even for a new editor, who should read Help:References. However, I am unconvinced with this proposal. If I am understanding correctly, the grandfather clause is only for articles with an existing unreferenced tags and this applys to all other articles. Firstly, for new pages, the existing new page patrol system is sufficient already in sorting out problematic pages that are non-notable, unverifiable, or meet other deletion reasons. Instead, this proposal will increase the number of hasty proposed deletions mostly by non-NPRs who are more inexperienced with deletion criteria and especially NEXIST and BEFORE to little benefit. Another concern is that the rise of hasty/inappropriate PRODs will be accompanied by mass deprods that theoretically address the concern by adding a source, but then the source would be a poor one (i.e., obvious SPS and the like). There is some more benefit for older pages created years ago and are not subject to NPP, but I think that AfD is sufficient, and that inappropriate and hasty "unreferenced prods" and mass deprods that only adds a clearly unsuitable reference for older pages is still a concern. Overall, this proposal will escalate the contentiousness for little benefit. As such, due to these concerns I therefore oppose this proposal. VickKiang (talk) 20:55, 7 October 2023 (UTC) Updated on 20:55, 7 October 2023 (UTC)
• Oppose unless the process limited to certain subjects that can either cause real-world harm (e.g. medical treatments), or are severe "spam magnets" (e.g. currently available commercial products). We should not delete a page like weak operator topology because no one qualified to read and understand the potential sources has gotten around to fixing the page yet. That's tearing down a house only because it isn't finished. This site is, and will always be, a work in progress. Suffusion of Yellow (talk) 23:25, 7 October 2023 (UTC)
• Oppose This seems aimed at a nebulous idea of a problem rather than a well-understood actual problem. XOR'easter (talk) 00:20, 8 October 2023 (UTC)
• I don't really see the point TBH, there aren't that many deletion nominations due to being totally unsourced that I see the need for a lighter process. Alpha3031 (t • c) 03:09, 8 October 2023 (UTC)
• Oppose. Like some other opposers above, and mindful of WP:CREEP, I don't think there is sufficient need for introducing this new process, subtly different from existing ones yet also similar and overlapping in purpose. I hesitate to add to our assortment of deletion or quasi-deletion procedural mechanisms, which is already confusing to not-too-experienced editors. Adumbrativus (talk) 09:19, 8 October 2023 (UTC)
• Oppose - Proposal does not respect WP:DEMOLISH or WP:NODEADLINES. We don't delete unreferecend articles at AfD so doing so via prod is inappropriate. ~Kvng (talk) 19:17, 8 October 2023 (UTC)

  @Kvng: WP:DEMOLISH and WP:NODEADLINES are essays, not policies or even guidelines. I could have equally cherrypicked essays supporting by point of view: Support as nominator: respects and supports WP:REALPROBLEM, WP:DEADLINENOW, and WP:CHEWINGGUM Edward-Woodrow • talk 20:11, 8 October 2023 (UTC)

  And regarding your second point... I direct you to WP:5P5. Edward-Woodrow • talk 20:11, 8 October 2023 (UTC)

  User:Edward-Woodrow, I think we can all agree that a lack of sources is not a good thing, but why are current procedures insufficient? Some of us do some actual work by looking for sources when we don't have to spend time on these endless discussions mostly started by people who want someone else than themselves to do the work. Phil Bridger (talk) 20:34, 8 October 2023 (UTC)

  Some of us do some actual work by looking for sources when we don't have to spend time on these endless discussions mostly started by people who want someone else than themselves to do the work. Comment on content, not contributors. Edward-Woodrow • talk 21:18, 8 October 2023 (UTC)

• Oppose This proposal has clearly had a lot of thought and effort go into it, but I am not sure it is the right solution. As mentioned above, new unreferenced articles are almost always either draftified or tagged with WP:BLPPROD or ((Expand language)) (latter two if applicable), so I don't think new unreferenced articles are really a big issue. Older unreferenced articles should not be dealt with this way, either; in most cases, it would merely be a shortcut to deletion by supplanting the need to go through the proper processes (and thus WP:BEFORE checks). Curbon7 (talk) 21:27, 8 October 2023 (UTC)

  There are also old articles that haven't been tagged yet. Edward-Woodrow • talk 21:28, 8 October 2023 (UTC)

  My understanding of the proposal was that this would only be used for articles created as of the date this is implemented and not to existing unreferenced articles, no matter if those were tagged or untagged. Is that an incorrect reading? Curbon7 (talk) 22:23, 8 October 2023 (UTC)

  That is in fact an incorrect reading; and given that many people have read it this way, clarification is probably needed.

  By "does not apply retroactively", I mean that ((unreferenced)) would not be replaced with the PROD template; i.e., the contents of the backlog (the articles tagged with the ((unreferenced)) template) would not suddenly all be up for PROD. Edward-Woodrow • talk 22:28, 8 October 2023 (UTC)

  Thank you. In accordance, I have further clarified my position. Curbon7 (talk) 00:01, 9 October 2023 (UTC)

  Thanks. Edward-Woodrow • talk 00:17, 9 October 2023 (UTC)

• Motion to close this is approaching a point of WP:SNOW opposition; and in fact there is no clear proposal made here as the VPI discussion did not come close to finding consensus for any specific change. This discussion should be speedily closed. Walt Yoder (talk) 22:36, 8 October 2023 (UTC)
• Support in principle. A page w/o a single cited source is not an encyclopedic article. It is a 21st century specie of graffiti and should be dealt with accordingly. Completely unreferenced articles are incompatible with WP:V and WP:CITE. Their existence undermines the credibility of the project and should not be allowed. With the stipulation that deletion should only occur after a reasonable time in which sources can be added, something along this line is long overdue. -Ad Orientem (talk) 03:18, 9 October 2023 (UTC)

  Unreferenced articles are fully compatible with WP:V, it is unverifiable articles that are not. The two are not the same thing. Thryduulf (talk) 08:43, 9 October 2023 (UTC)

• Strong support (I see someone has written "strong oppose" above so maybe my "strong support" cancels that out otherwise I would just have written "support") I don't know whether Google or other search engines have manually favored all Wikipedia articles as well as particular ones but if they have then the existence of so many unsourced articles would be an argument for them removing that. To get rid of the backlog of unsourced articles under existing rules is likely to take at least a decade I suspect. If people want unsourced info they can just ask ChatGPT or similar nowadays, so there is no reason for us to keep our unsourced articles. Anyway the "prod" process allows an admin to get a "soft deleted" article back into draft if necessary I understand. Chidgk1 (talk) 06:30, 9 October 2023 (UTC)
• Procedural comment Can someone please add this to WP:CENT? I would do it myself, but cannot figure out a good wording for it. Curbon7 (talk) 07:53, 9 October 2023 (UTC)

  @Curbon7:  Done, I did my best with the wording but it is hard to phrase, I agree. Feel free to tweak it, of course. Edward-Woodrow • talk 13:25, 9 October 2023 (UTC)

  Not sure if it's needed to put an RFC that is snowing on T:CENT. But I guess it couldn't hurt. –Novem Linguae (talk) 13:38, 9 October 2023 (UTC)

• Oppose, due to significant overlap with normal PROD and AfD (despite the clarifications above), which are already frequently used to accomplish the same things as this proposal's stated goal. Excess overlap hints at a poorly-designed proposal. DFlhb (talk) 11:47, 9 October 2023 (UTC)
• Oppose. Unreferenced articles that went through WP:NPP would be draftified or PRODed 15-30 minutes after the article is created. As far as I know, NP patrollers are fairly zealous to act on unreferenced articles. Second, PROD in most cases are silent deletion process. New editors would just panic and didn't know what to do, while experienced editors who can help didn't know that somewhere in Wikipedia an article that they can rescue is being deleted. ✠ SunDawn ✠ (contact) 13:48, 9 October 2023 (UTC)
• Oppose seeks to overrule far-better thought out processes in WP:CREEP. Completely unnecessary. ~~ AirshipJungleman29 (talk) 14:23, 9 October 2023 (UTC)
• Oppose. I believe that our current processes are already sufficient for addressing this issue. The comments above by Red-tailed hawk and VickKiang lucidly describe my thinking on the matter. ModernDayTrilobite (talk • contribs) 15:42, 9 October 2023 (UTC)