Wikipedia talk:Notability

Adding one new thing to the current SNG text

The following discussion is closed. Please do not modify it. Subsequent comments should be made in a new section. A summary of the conclusions reached follows.

No consensus to add proposed text - jc37 02:44, 30 August 2021 (UTC)

---

If you have not been following ANI and events that have filtered down into WP:NSPORTS and WP:GEOLAND, basically, we have had problems with mass creation of stub articles based on information taken from simple databases (some with questionable reliability) and nothing else. Its clear due to how these have discussed at ANI that the community does not want mass creation of articles simply based on a database entry even if that meets an SNG, and as a result there's been discussion at both of these SNGs to try to see about how they can address this.

While whether these changes will actually be made to these SNGs, I do think the issue on mass creation on weak sources can be addressed here. I would suggest that after the current line Therefore, topics which pass an SNG are presumed to merit an article, though articles which pass an SNG or the GNG may still be deleted or merged into another article, especially if adequate sourcing or significant coverage cannot be found, or if the topic is not suitable for an encyclopedia. we could then add "Mass creation of short or stub articles based on simple lists or database sources that would pass an SNG is strongly discouraged." or something to that intent. --Masem (t) 13:13, 23 April 2021 (UTC)

I support it in principle but the wording is confusing. Perhaps shorten / clarify it to: "Mass creation of short or stub articles based on simple lists or database sources is strongly discouraged." North8000 (talk) 13:26, 23 April 2021 (UTC)

I agree with North8000 here - the issue isn't whether the stubs would pass an SNG; the point is that mass-creation of such articles is strongly discouraged. Newimpartial (talk) 13:32, 23 April 2021 (UTC)

My wording usually sucks to start so I am all for any improvements, but you get my point. --Masem (t) 13:36, 23 April 2021 (UTC)

Well, "my" version is just yours with some words taken out.  :-) North8000 (talk) 17:39, 23 April 2021 (UTC)

So, to make a specific proposal out of it that would be: Add the following to the end of the first paragraph: "Mass creation of short or stub articles based on simple lists or database sources is strongly discouraged." North8000 (talk) 20:02, 23 April 2021 (UTC)

Survey

• Support this, but I also think the slow creation of such stubs should be discouraged. Levivich ^harass/_hound 15:44, 24 April 2021 (UTC)

Well, maybe we do this "mass creation" to start and then move on to that. North8000 (talk) 13:49, 25 April 2021 (UTC)

No thanks, I'd rather encourage more stubs (if they are verifiable and reliable) if it means filling out the encyclopedia.--Ortizesp (talk) 15:23, 29 April 2021 (UTC)

Support North8000 wording amendment Davidstewartharvey (talk) 13:52, 25 April 2021 (UTC)

• Support amendment. My opinion is that almost any form of mass editing is disruptive, and should be discouraged... but we can start with this. Blueboar (talk) 14:06, 25 April 2021 (UTC)
• Support amendment. Articles that don't have content beyond minimal database-type facts are not in the best interest. Reywas92^Talk 19:15, 25 April 2021 (UTC)
• You can add WP:PROF to the SNGs for which this issue has come up. The ones I keep running into are the old sub-stubs from one particular user (under several user names) of the form "So-and-so of (employer) was named Fellow of the IEEE in (year)". They're automatically notable under WP:PROF, and usually easily expanded into at least a real stub, but they look so stubby and undersourced that people keep tagging them for notability or deletion (recent example). I'm not sure their creation was as "mass" as the geostubs, but there are quite a few of them and I think that form of article creation by expanding a line from a database without any individualized attention should be discouraged regardless. My only caution is that these databases can be a good source of missing article topics and I would not want to discourage individual hand-crafted article creations that actually expand on these database entries by bringing in other sources, even when the result is still a stub. —David Eppstein (talk) 19:31, 25 April 2021 (UTC)
  • The keyword in this addition is "mass", and points to both human-made and bot-made articles. Someone that makes the effort to create a non-stubby article based on one of these SNGs that normally lead to mass creation should be applauded, and that's the effort we don't want to stop; only strongly discouraging mass creation. --Masem (t) 20:04, 25 April 2021 (UTC)
    • Right. Anyway, support. —David Eppstein (talk) 20:42, 25 April 2021 (UTC)
• Support North8000's version, though not entirely sure if this is an RfC or general discussion. SportingFlyer T·C 19:59, 25 April 2021 (UTC)
• Support North8000's version. It's good WP:PAGEDECIDE guidance. – Finnusertop (talk ⋅ contribs) 20:03, 25 April 2021 (UTC)
• Support for reasons above. Xxanthippe (talk) 23:26, 25 April 2021 (UTC).
• Conditional support for North8000 version assuming there is clarification of what is meant by "mass creation". I'm not sure where to draw the line, but the problem has arisen in cases where people churn out articles at a rate as fast as one per minute. Cbl62 (talk) 00:04, 26 April 2021 (UTC)

Maybe something like "... mass creation of articles at a pace of less than five minutes per article ..." It is admittedly a bit arbitrary but I think it captures the type of mass cration that we are trying to discourage. Cbl62 (talk) 00:07, 26 April 2021 (UTC)

The fuzzy Wikipedian system leaves many things to interpretation. The current proposal is one step and then we could tweak and refine. North8000 (talk) 00:15, 26 April 2021 (UTC)

Yes, let the issue of what exactly constitutes "mass creation" be hashed out elsewhere, we just want to discourage editors from doing that. --Masem (t) 00:17, 26 April 2021 (UTC)

Oppose. Without some clarification as to what "mass creation" means, I think the proposal is vague, fundamentally flawed, and simply kicks the can down the road. Given the lack of willingness to clarify, I must change to oppose. Cbl62 (talk) 04:23, 27 April 2021 (UTC)

Respectfully, that can be determined in discussions when it comes up. This is how most policies and guidelines are worded and operate. North8000 (talk) 15:44, 27 April 2021 (UTC)

I simply believe that the proposal, without some definition of "mass creation", invites mischief. Short articles sourced to a database can be a legitimate and valuable starting point in the natural collaborative process of building the encyclopedia. Especially when created with some discretion. The problem that led to this proposal is the mass creation of one- or two-line cricket and Olympian micro-stubs based on an SNG that is not properly calibrated to GNG. I support the effort to constrain such mass article creation. I am concerned, however, that it will be misused if there is no definition or guidance on what constitutes "mass creation". Cbl62 (talk) 22:36, 27 April 2021 (UTC)

This is a guideline, the advice being added is simply a caution. Mass creation is a behavior problem and while the issue of using single source to meet an SNG to create articles en masse is perhaps the most likely way this happens, we'd still be evaluating behavior issues and not necessarily on this piece of advice. For example, going off GEOLAND, if some experienced editor saw that for some reason in a populous country that we failed to have documented all recognized towns over 25,000 people and went and mass created them as stubs based on a reliable sourced database, with this advice in place, there likely would be no major issue; this piece of advice is not meant to get in the way of those that know what they are doing. It is meant to warn editors that may not know the ropes to avoid going there, and hence the need to be explicit on what mass creation is defined as beyond its core principles. --Masem (t) 23:36, 27 April 2021 (UTC)

Cbl62 I respect and admire your carefulness on avoiding unintended consequences, we need more of that. But it's hard to imagine a situation where this would go awry. Plus one additional issue is that it would be so difficult and probably support-losing to create an explicit rule (vs. a general consideration amongst others to be taken into account) that such an effort could constitute a poison pill for this effort or create an overreach (via an explicit stand-alone rule) that could have more of the unintended consequences that you are being careful about. Sincerely, North8000 (talk) 15:50, 28 April 2021 (UTC)

• Support the proposed addition. More broadly, I also support discouraging the creation of single source, one sentence stubs in general, whether written rapidly or slowly. Cullen³²⁸ Let's discuss it 00:20, 26 April 2021 (UTC)
• Support, and agree with User:Cullen328 as well. JoelleJay (talk) 01:11, 26 April 2021 (UTC)

  @Cullen328 and JoelleJay: You mean like this one? Its creator has done that sort of thing before, and has a following of people who rather admire him. --GRuban (talk) 11:25, 22 July 2021 (UTC)

  GRuban, in that particular case, the one sentence stub did not languish untouched for years. Within a few hours, it was expanded to something much more informative. Cullen³²⁸ Let's discuss it 16:51, 22 July 2021 (UTC)

  I strongly suspect all creators of stubs intend for others to expand them. --GRuban (talk) 17:40, 22 July 2021 (UTC)

• Oppose The proposal supposes that simple lists or databases are inherently bad sources. This is incorrect because they may be quite authoritative and excellent. It's the quality of the information which matters, not its format. Like Blueboar says, it's mass-editing which is the real problem – grinding with tools like AWB to boost edit-count. Creating countless rules is another vice. WP:BURO, WP:CREEP, WP:IAR and WP:NOTLAW strongly discourage this too but the OP pays no attention. Andrew🐉(talk) 10:06, 26 April 2021 (UTC)
  • The addition does not say anything about the quality of databases. Simply that populating articles only from databases in mass is the problem, which is how the wording is focused. Someone who makes exactly one stub article from an RS database that otherwise meets a SNG is not going to have their hand slapped nor is that article likely to be rushed to be deleted. --Masem (t) 13:21, 26 April 2021 (UTC)
    • If an update is ok then several updates of the same sort will be ok too. The proposition seems to express hostility to mass updates which is not a notability issue. My !vote stands. Andrew🐉(talk) 09:58, 16 July 2021 (UTC)
• Support- I agree with North8000's proposed wording. Reyk _YO! 12:50, 26 April 2021 (UTC)
• Support- Especially with North8000's wording. -- Michael Scott Cuthbert (talk) 23:52, 26 April 2021 (UTC)
• Oppose - 1) Why whould "mass creation" be discouraged and not creation in general? What I see, the quality of mass created stubs is better compared to stubs created by unexperienced (stub) creators. 2) A large % of sports articles started as a stub and expanded later. Also stubs attracts a lot of new users, including myself. As there is a large importance of statistics and numbers on articles on sportspeople, good stubs are of high value. A sportpeople is known for her/his results; and those results are the perfect way to start a sports article. SportsOlympic (talk) 11:27, 29 April 2021 (UTC)
  • One-off creation of stubs isn't a problem - someone might be going through a book, see a name come up , affirm it meets a SNG but we have no article, and makes a stub with the minimum sourcing with the likely expectation they may come back to improve later. Mass creators, however, do not appear to consider returning to improve the articles they create and that's generally where the problem lies. Also, in terms of sportspeople, while you may claim their importance is in their results, WP is not a stats book and we have to focus outside what their numbers are and downplay the stats per WP:NOT#STATS. We can include them, but the article must be far more than just that. --Masem (t) 12:28, 29 April 2021 (UTC)
• Support, and would support it without "mass" as well. Fram (talk) 12:19, 29 April 2021 (UTC)
• Strong Oppose there's nothing wrong with stubs. There's nothing wrong with using databases as stubs. There is something wrong with misinterpreting data, which is what happened with WP:GEOLAND. I disagree with the aim of this proposal, and I'm not liking the direction the community is taking which is inherently limiting the scope of Wikipedia instead of expanding it.--Ortizesp (talk) 15:23, 29 April 2021 (UTC)
  • While there was one issue with GEOLAND and the use of a questionable database, the issues of mass creation were not related to just that. The community has clearly expressed concern that an editor that mass creates articles with no intention of going back to expand them beyond stubs, just because they can show the article passes an SNG, is potentially disrupting the work. --Masem (t) 15:45, 29 April 2021 (UTC)
    • I agree that it can be disruptive, but I don't think it is always necessarily disruptive. If there's a positive use case for mass creation in the future, I don't want to block out the possibility of its use if that makes sense. I'd also want to know what we are going to define as mass-creation.--Ortizesp (talk) 16:33, 29 April 2021 (UTC)
      • I think that extra consideration is given if it is a case of an individual taking the time to create an individual article. For example, there about 8.5 million species of plants and animals the don't have an article in Wikipedia. If one person spent time creating an article on one of them from database / tertiary sources, (even is just a stub) the community would almost certainly accept that. If someone created a bot to generate 8.5 million new article from a species database, the community would not. This documents that. North8000 (talk) 21:11, 2 May 2021 (UTC)
• Conditional support The premise is great. I wonder if it should be amended to change to "based on simple lists or database sources a single list or database source" to be more explicit that an article should contain multiple sources. I can envision a scenario where someone might create stubs following an election, or after an awards ceremony, where a single list is the basis for creation of the article, but there are other sources available.--Enos733 (talk) 16:13, 29 April 2021 (UTC)
• Support on the basis that I'm happier with it in, than without it. Nothing's perfect but this would be a small step in the right direction. Nigej (talk) 17:10, 29 April 2021 (UTC)
• Weak oppose whilst I have issues with mass creation of low-quality stubs, especially by experienced editors (heck, I've even seen AWB used to create one-sourcers by an experienced editor in the WP:FOOTY community), I don't think this is the best way to go about fixing it as I see no issue with "mass creation" of stubs since someone will come along and expand them. Also, this is already implied afaiac and basically how these SNGs work in practise. Microwave Anarchist (talk) 20:31, 29 April 2021 (UTC)
• Oppose. It's not a notability issue. --Michig (talk) 09:56, 2 May 2021 (UTC)
• Support - While I can see a role for database-generated content in the encyclopedia, that role goes through an interface to Wikidata, not the creation of stubs. Database-only sourced articles are bad for the human-edited part of the encyclopedia, and we should not allow the SNGs to be used to wave them through. These articles have wasted a lot of time at XfD, and will continue to do so if we do not make a rule against them. — Charles Stewart (talk) 10:29, 2 May 2021 (UTC)
• Support, this shouldn't even have to be clarified, but apparently it does. Seraphimblade ^{Talk to me} 23:39, 2 May 2021 (UTC)
• Support North8000's version as an improvement to the status quo, which should help with our problem with mass-created database stubs. While manually-created articles should be held to the same standards, Masem has a good point that manually-created articles don't appear to be as problematic, so it's fair to start with the mass-created ones. -M.Nelson (talk) 22:23, 7 May 2021 (UTC)
• Support North8000's version minus the word "mass", and support the overall spirit of Masem's suggestion. I think we can work out the wording and this is a big step in the right direction. Shooterwalker (talk) 01:04, 18 May 2021 (UTC)
• Comment Here's what would make me support: 1. We figure out what "mass" creation is, and 2. Change it to "based on simple lists or database sources a single list or database source" (got the idea from Enos733). Otherwise I will Oppose. BeanieFan11 (talk) 18:50, 24 May 2021 (UTC)

It's been open for a month....time to close? BTW while I understand that it's being called "per North8000" for identification purposes, I feel guilty unless I emphasize that Masem wrote it and I just shortened it.  :-) North8000 (talk) 20:59, 24 May 2021 (UTC)

• Support. --_{Piotr Konieczny aka Prokonsul Piotrus| reply here} 03:05, 28 June 2021 (UTC)
• Support. In particular, I prefer the original proposal to the modified proposal below, and I think the amended version suggested by Enos733 is perhaps the best of all. --JBL (talk) 00:05, 3 July 2021 (UTC)

Of course I'm involved, but is this a good summary?: Of course it's not a vote, but there were 21 supporting it and 8 opposing it. Included in the 21 is two conditional supports, one conditional on an even stronger version and one on some tweaks so maybe it's just 19 or 20. The most substantial support argument is in Masem's original proposal which is too substantial/thorough to recap here, and in posts assuaging concerns raised by those in opposition, and that it can be tweaked after it is in. The most common concern expressed by those in opposition is that lack of a specific meaning for "mass creation" could cause problems. The scope of the process was broader than that for many changes (40+ days on this prominent page with 29 participants) but narrower than a fully advertised RFC. IMHO the result is that there is a sufficient basis for putting it in, with the understanding that it can be tweaked afterwards. North8000 (talk) 18:18, 3 June 2021 (UTC)

See the subsection I just opened below. {{u|Sdkb}} ^talk 15:05, 4 June 2021 (UTC)

IMO we have a decision and should implement it and then it could be modified from there. North8000 (talk) 20:15, 8 June 2021 (UTC)

• Support Mass creations of stubs based on databases have wasted countless manhours of editor time cleaning them up.Jackattack1597 (talk) 00:05, 3 July 2021 (UTC)
• Comment WP:MASSCREATE does not 'strongly discourage' mass-creation, it just requires prior approval. I would be fine with adding a note to this guideline pointing to what existing policy (even if only rarely followed) is. If we're going to implement a new policy of "strong discouragement" then I don't know whether this discussion has been properly advertised, as Joe Roe mentions, through the relevant processes to get broad feedback on this change. ProcrastinatingReader (talk) 11:02, 6 July 2021 (UTC)

One note: this is a guideline, not a policy. North8000 (talk) 11:22, 6 July 2021 (UTC)

I don't know which part of my argument that affects, but in general I rarely see the distinction (c.f. WP:JUSTAGUIDELINE). WP:N is enforced as if it were a policy. If the text is in the page, it will be used in dispute resolution, irrelevant of whether there's a guideline banner at the top or a policy banner. At the same time, "strongly discouraged" is toothless, so it's likely this will have no actual effect. ProcrastinatingReader (talk) 11:25, 6 July 2021 (UTC)

• Oppose per my comment below. {{u|Sdkb}} ^talk 21:12, 15 July 2021 (UTC)

Notified: WT:Bot policy, WP:BOTN. {{u|Sdkb}} ^talk 05:59, 16 July 2021 (UTC)

• Support as entirely sensible. Stifle (talk) 09:34, 16 July 2021 (UTC)
• Oppose as redundant to the combination of WP:MASSCREATION and WP:MEATBOT, which sets a much stricter restriction than the given suggestion. I think the point here—evidenced by the fact that these policies have only been pointed out so late in the process—is that we don't need more rules, but more people to intervene faster on a case-by-case basis with people who are doing this. It's particularly undesirable when someone spends years doing something and only then does the community try to put a stop to it. Tell someone that what they're doing isn't a good use of effort a few hours in and they'll generally be reasonable. Tell them one thousand hours of hard work in and they will do or believe anything that lets them continue thinking that they haven't wasted an enormous amount of time. — Bilorv (talk) 11:56, 16 July 2021 (UTC)

  I feel like I'm kinda repeating myself but the issue is that only bot owners read the bot policy, whereas people more likely to be creating this kind of content read the notability guideline. It does no harm to mention the same thing here. If someone believes that we need to intervene sooner, then I don't see why they wouldn't support text being added here to mention the relevant parts of those policies. Whatever the case may be, the fact is that policy is never enforced, so clearly the status quo is not working. ProcrastinatingReader (talk) 12:12, 16 July 2021 (UTC)

  I'd support a sentence like "Users who are engaging in mass creation should take note of WP:MASSCREATION and WP:MEATBOT". I can see that none of the mass creators are reading the right policies but—and it seems you became in agreement of this half an hour after writing the above—a new, redundant rule is not the solution. Most of my comment is an argument that the main change needed is awareness of the rules, not redesign (so redesign is only good if its main function is increasing awareness of WP:MASSCREATION). — Bilorv (talk) 11:25, 20 July 2021 (UTC)

• Support North8000's version. Ealdgyth (talk) 12:28, 16 July 2021 (UTC)
• Oppose this specific proposal violates WP:POLCON because it suggests that such edits are simply 'discouraged', when in reality a policy already prohibits them without approval. The proposer of the most supported version said they think WP:MASSCREATE only applies to bots, which isn't true. It's likely many other early supporters were also unfamiliar with that policy. Consensus can certainly decide to make that policy obsolete, but it needs to consciously do so, and in this discussion it doesn't seem editors are intentionally trying to loosen existing policy because the whole point of this discussion was to add additional standards. Even some bot users seem to be unfamiliar with that policy, so clearly some text needs to be added to a more relevant PAG, but this text isn't that. ProcrastinatingReader (talk) 12:50, 16 July 2021 (UTC)

Couple of notes. IMHO the operative wording at WP:MASSCREATE is "Any large-scale automated or semi-automated content page creation task must be approved at Wikipedia:Bots/Requests for approval." On the second note, while there are cases where both would apply, IMHO for those the existence of a less restrictive constraint and a more restrictive constraint is not a conflict. Sort of like a national speed limit law that in essence says "never go over 80 MPH in this country " and one in a town that says "Don't go over 20 MPH on this street here". There is no conflict, they are merely two co-existing sets of constraints which must be complied with. Sincerely, North8000 (talk) 13:14, 16 July 2021 (UTC)

We already know that most content creators, and many bot operators too, do not read the bot policy, so this guideline is the only one they will read. Implying that such creation is "strongly discouraged" is then all they will know, and they'll think that's the bar. I think your analogy is better adjusted to something like: say the national speed limit (eg 30mph) is buried in the middle of a law nobody reads or knows about, and say on some street you place an advisory speed limit sign for 50mph. Well, drivers can't go between 30-50mph anyway, so the sign is just confusing and would make people think that they can go up to 50mph when they can't. It makes no sense to provide an advisory that is less restrictive than the requirement. ProcrastinatingReader (talk) 13:27, 16 July 2021 (UTC)

It's possible to have both an approval requirement and have the action be strongly discouraged, but in that case both should be mentioned. Institutional memory already sucks and there are countless examples of widespread confusion and misapplication when policy is split like that. ProcrastinatingReader (talk) 13:39, 16 July 2021 (UTC)

All good points, but IMO at the logical core of it, founded on a presumption which I think is generally not correct. Bot creators are the subset that is subject to both, and the statement / underlying assumption is that bot creators often don't read the bot policy. To me, someone creating bots (a small minority of editors creating big powerful wiki-dangerous stuff) without reading the bot policy is a pretty blatant problem specific to that specific area. Sincerely, North8000 (talk) 13:59, 16 July 2021 (UTC)

As I said below, the policy was created following this discussion: Wikipedia:Village_pump_(policy)/Archive_66#Automated_creation_of_stubs. That discussion was not about bot operators, it was about normal users mass-creating formulaic stubs. ProcrastinatingReader (talk) 14:03, 16 July 2021 (UTC)

I think that in essence you are saying that general editors should know about "Any large-scale automated or semi-automated content page creation task must be approved at Wikipedia:Bots/Requests for approval." which is in the bot policy and know that it applies to manual edits based on intentions expressed in an archived discussion. And that they should go to "Wikipedia:Bots/Requests for approval" for approval of their non-bot creations. And that so the subject topic here is already covered and thus in conflict with the bot policy. I know that that is not the way you'd say it but I think it's time to agree to disagree and thank you for the exchange. Happy to discuss more if you wish but otherwise I'll leave it at that. Sincerely, North8000 (talk) 14:38, 16 July 2021 (UTC)

• Oppose. What is proposed is a behavioural guideline. Nothing to do with notability whatsoever. This guideline is long and complicated enough already without bringing in irrelevant issues that should be dealt with elsewhere. Phil Bridger (talk) 14:34, 16 July 2021 (UTC)
• Oppose any attempt now to automatically or semi-automatically create articles would need to go through an approval process and community consultation per WP:MASSCREATION, and the community would probably say no unless the case for the creations was very compelling. The bot policy is also a better place to put restrictions on automated editing than the notability guideline. Hut 8.5 08:04, 17 July 2021 (UTC)
  • This is not true, at least for semi-automated (more human than bot operation). See this recent ANI thread that continues the same concerns about one user's mass creation of articles of which they have been cautioned about. This is why the language suggested is not an absolute but a precautionary statement that we want to discourage this but editors that are aware of what they are doing take onus for any fallout from it. --Masem (t) 13:09, 17 July 2021 (UTC)
• Oppose -- If a topic meets an SNG as demonstrated by reliable sources, it may deserve a standalone article, and so, editors may create such articles. Anyone who disagrees may initiate a merge or deletion discussion. There is no need to preemptively stop editors from exercising this discretion. Now, if they are being mass-created with unreliable sources, that would be a behavioural issue. I should preempt my opposition to the "we may have needed it at the start, but no longer" line of thinking; one need only move on to topics from Africa and Asia to find just how much is missing. I am tempted to but will refrain from opposing this on the basis that strongly discouraged is probably the most unhelpful phrase one could encounter in PAGs. It always disadvantages good faith editors and advantages bad actors. It is likely to be disproportionately weaponised against content-focused editors who are averse to wikilawyering and don't make many friends in the drama department who would come to their rescue.
  As a latecomer who followed the CENT list, I agree with everything Joe Roe says below and thank them for getting this widely advertised and extended. I memorise shortcuts for PAG pages, I don't manually watchlist them. Regards! Usedtobecool ☎️ 19:24, 18 July 2021 (UTC)
• Oppose as it fails to indicate the manner of enforcement. Would such stubs be deleted? Then I oppose it on principle, as there's no real harm in letting a bunch of stubs on notable topics linger even if we wouldn't necessarily create them in the first place. Should the creator be sanctioned? Then it is a behavioral guideline and does not belong here. Or is it merely an unenforceable suggestion? Then there is no need to bloat up the page any further. -- King of ♥ ♦ ♣ ♠ 23:49, 18 July 2021 (UTC)
  • The point is a caution and not a rule or anything. What exactly would be done if someone mass created numerous articles against this caution really depends what the "damage" was and that's a behavior issue beyond the scope of content guidelines like WP:N. Guidelines are absolutely allowed to have advice like this. --Masem (t) 00:58, 19 July 2021 (UTC)
• Support either version. Mass creating articles by human or by bot should be discouraged, and I'm rather convinced that this is already covered by MEATBOT. From this discussion I think it's clear that the community generally opposes mass creations of permastubs even if they meet an SNG, but a lot of that light is getting lost in where to put that. I don't really care where it goes, but I agree that we should encourage editors to think about whether an article is worthwhile when all we know is that a place exists. I also appreciate the CENT listing as I didn't see this. — Wug·a·po·des​ 07:21, 19 July 2021 (UTC)
• Oppose any "discourage" language and new guidelines, but support directing users to existing guidance on bots and semi-automated editing. We want people to undertake this work, if done in an effective and sophisticated way. And we want people thinking about it to realize that this may be a high bar. For example, it would be great if the redlinks at Municipalities of Bolivia were each replaced by database-informed stubs, so long as the underlying dataset is reliable and citations are effective. But the community needs to have input on doing so effectively.--Carwil (talk) 17:55, 19 July 2021 (UTC)
• Oppose As noted by others above we already have other guidelines which state this. Moreover, in some occassions in which the source of said database is high and meet a SNG than it may warrant some creation (though at what point it becomes 'mass' is ambiguous).  Spy-cicle💥  ^Talk? 19:26, 19 July 2021 (UTC)
• Support. I am sympathetic to the concerns by some opposing users that this is just instruction WP:CREEP, but apparently we really do need to restate this due to such mass-creation by database without seeking consensus or any sort of notification first. I think it's a harmless and relevant reminder of the existing standards to be added to this page. I'm also worried that some of the opposes will be interpreted as endorsing such mass creation rather than affirming such activity is already discouraged, so even if this addition does not gain consensus, the fact that the opposes are split between "actually, mass creation off a single source is a great idea" and "we don't need this" should still reflect a general consensus to strongly discourage such articles. SnowFire (talk) 06:22, 22 July 2021 (UTC)
• Support English Wikipedia is not the place for this, but Wikidata is a great place for this activity. The d:Wikidata:Wikidata Bridge is looking ahead to supporting the automated creation of stubs in every language version of Wikipedia, as are projects like meta:Abstract Wikipedia. Without engaging in Wikidata this kind of activity is not helpful and also problematic; by starting with Wikidata, this activity becomes much more helpful. Blue Rasberry (talk) 14:56, 22 July 2021 (UTC)
• Support WP:AGF aside, mass-creating stubs on niche subjects with negligible probability of expansion and/or picking up by someone else (like those tiny villages with only 24 or so households) runs afoul of the WP:NOTDATABASE policy. I think the perpetuation of such practice would inflate our general stubs statistics, increase the backlog and is not helpful in reducing stubs. Brandmeister^talk 22:00, 23 July 2021 (UTC)
• Oppose until "mass creation" is more clearly defined. There's discussion about it at the bottom of this section but it's not clear what change to make to the guidelines yet. Dr. Universe (talk) 19:18, 25 July 2021 (UTC)

Modified proposal

I came here hoping to be able to close this discussion, so sorry North8000 that I'm instead about to extend it. But what I take from the above is that editors are concerned about the creation of low-quality stub pages, either because the database itself is questionable or because the extraction from it is done poorly. The original proposal, however, goes beyond what editors' actual concerns are, and instead uses language that would effectively stop the practice even in the (unfortunately rare) case that the database is high-quality and the extraction is done well, and even in circumstances in which a local consensus of editors might otherwise agree to move forward. Therefore, I'd propose modifying the addition to read: Consensus should be sought before the mass creation of short or stub articles based on simple lists or database sources. {{u|Sdkb}} ^talk 15:05, 4 June 2021 (UTC)

• Support^{See below} as proposer. Asking for affirmative consensus before allowing any mass-creations will resolve the concerns editors have expressed above, as editors will not support mass-creations if the source is at all questionable, the resulting stubs will be of low quality, or there are any other significant issues. {{u|Sdkb}} ^talk 15:05, 4 June 2021 (UTC)
• Question - Given this isn't actually dependent on notability, doesn't this make more sense to add to WP:MASSCREATE or WP:MEATBOT or the like first? — Rhododendrites ^talk \\ 15:30, 4 June 2021 (UTC)

  Oh wow. I really wish that those links had been brought up at the start of the long discussion above, as it likely would've averted the entire need for it. The concerns that are leading to all the support above are already covered by those pages, and they're already policy. I would guess that many !voters above were not familiar with what those sections say, and that they would've been much more inclined to view this as unneeded or as WP:CREEP if they were. I'm not precisely sure how we should proceed from here, but this discussion should certainly not be closed until we've had a chance to see how that new information affects the direction of consensus. {{u|Sdkb}} ^talk 17:35, 4 June 2021 (UTC)

• IMO we should close and act on the first one and then have an RFC to tweak it. North8000 (talk) 17:11, 4 June 2021 (UTC)

Trainwreck

I've just reverted the addition of the following text, citing this discussion:

Mass creation of short or stub articles based on simple lists or database sources is strongly discouraged.

This would be a major shift in policy, given that editors have been doing this to create hundreds of thousands of articles since the start of the project. I don't see a consensus for it here and moreover, I think it's time to declare this discussion a trainwreck. Although I'm sure it was started with the best of intentions, the proposal buries the lead, is under an uninformative heading, and as far as I can tell was not advertised anywhere outside this talk page. A change to a core policy as large as this needs a high level of consensus; at least a proper, well-attended RfC and a formal close. – Joe (talk) 12:20, 2 July 2021 (UTC)

First, this is a guideline, not policy; second, this is a highly visible page on WP and the above discussion shows strong attendance to that; and third, this reflects current practice (based on the events of AN threads that discouraged mass article creation from databases). It's also not an absolute piece of advice ("you must do...") but a cautionary warning that matches current practice against mass article creation. So no, this isn't anywhere close to the problems you believe it may have. --Masem (t) 13:34, 2 July 2021 (UTC)

I think we all know that, guideline or not, WP:N is vigorously enforced. If it says something is "strongly discouraged", people are going to use that to stop it happening. I don't think it reflects current practice—a lot of AN threads about someone doing something badly doesn't mean the community doesn't want it to happen at all—but if that is the case it should not be hard to show a clear consensus at a properly-formatted RfC. – Joe (talk) 14:50, 2 July 2021 (UTC)

As Rhododendrites noted above, we already caution against automated and semi-automated mass-creations in WP:MASSCREATE, so I don’t see the proposed language as being a significant change to policy/guidance… it’s more of a logical extension of existing policy/guidance, as applied to notability. Blueboar (talk) 14:17, 2 July 2021 (UTC)

WP:MASSCREATE is part of the bot policy. Extending it to non-automated, human edits is significant change to current policy. – Joe (talk) 14:50, 2 July 2021 (UTC)

It is already extended to human edits. See WP:MEATBOT… Whether they are done by a human or a bot, mass edits are discouraged. Not necessarily banned, but definitely discouraged. At minimum, any mass-creation needs to be discussed at the project level (or equivalent) and receive some form of community consensus. Blueboar (talk) 15:50, 2 July 2021 (UTC)

Are you able to point me to any support in the community over, say, the last ten years, for (non-automated) Mass creation of short or stub articles based on simple lists or database sources? Because I don't recall seeing any. Newimpartial (talk) 15:17, 2 July 2021 (UTC)

With the reminder that consensus can change (as this had been done well in WP's early days but does not have favored use now) --Masem (t) 15:32, 2 July 2021 (UTC)

My understanding was that WP:MASSCREATE applies to human editors. The problem in the RfC that led up to that section was a human one as well, Wikipedia:Village_pump_(policy)/Archive_66#Proposal:_Any_large-scale_semi-/automated_article_creation_task_require_BRFA. The bot policy does have other provisions relating to human editors too, such as WP:ASSISTED. However, few people (outside the bot circle) actually read the bot policy. It would probably be helpful to have in this guideline, as it is more read by the relevant groups of people. That said, I'm not sure that the Bot Approvals Group is best-placed to approve such tasks. It seems its role was envisioned as just rubber stamping the consensus of participants, but few non-technical people go to WP:BRFA, and AFAIK few people (in recent times) have actually sought BAG approval for making these changes. I think if we are going to formalise this provision, perhaps we should work out a better approval mechanism, as I'm not sure whether putting it under the BAG/BRFA system is going to encourage adhering to the process (at the same time, I don't know of another venue well-suited for this; a general discussion at WP:AN would be the next best option?). ProcrastinatingReader (talk) 16:03, 2 July 2021 (UTC)

(edit conflict) "Logical extension" is one way to put it; "redundant creepy fork" is another. I'm with Joe that I do not think it's possible to derive a consensus from this discussion. {{u|Sdkb}} ^talk 16:10, 2 July 2021 (UTC)

We had a 6 week discussion on that precise proposal. (April 23 - June 3) Me adding it was reverted on the basis that there was no close, which is fine. So now we need a close. And addition of a new idea or new thoughts do not negate that. Sincerely, North8000 (talk) 17:18, 2 July 2021 (UTC)

Beyond me how anyone can look at this discussion and not perceive consensus for the proposed addition. I strongly oppose declaring this a trainwreck (whatever that means) or tossing aside the work that the participants to this discussion put in to achieving consensus. I guess we can wait for a formal close if that's what's being demanded before editing the page. Levivich 18:06, 2 July 2021 (UTC)

To be valid, consensus needs to be determined with access to full context and background. We're talking about a change deeply connected to WP:MASSCREATE/WP:MEATBOT, but those linked weren't even brought up until almost all of the !votes had been cast (that fact is not unrelated to the fact that those noticeboards and other relevant forums were never notified of this discussion). I think it's very plausible that many of the support !voters would have opposed if they'd been aware of what guidance we already have in place. And there's no way of knowing how the discussion might have gone differently if the relevant parties had been properly notified. That casts enough doubt on the outcome to lower it below the high bar needed to make a change of this magnitude. {{u|Sdkb}} ^talk 19:22, 2 July 2021 (UTC)

Oh yeah, if only they knew X, all those support voters would have changed to oppose. Sure, that's a good reason to ignore what they wrote. Levivich 19:41, 2 July 2021 (UTC)

Sdkb, raising a point after the fact & hypothesizing that the point might have changed the discussion is not a valid way to cancel the results of the discussion.North8000 (talk) 21:54, 3 July 2021 (UTC)

The length of time and level of consensus is irrelevant if the discussion if the participant base is not broad enough. This is a major change in policy that needs community-wide consensus, not local consensus in a poorly-advertised, poorly-formatted discussion on a guideline talk page. – Joe (talk) 07:18, 6 July 2021 (UTC)

Further discussion is at "Close on this? subsection above. North8000 (talk) 19:02, 2 July 2021 (UTC)

Make that below. It got moved. North8000 (talk) 21:56, 3 July 2021 (UTC)

Close on this?

We had a 6 week discussion on that precise proposal. (April 23 - June 3) Me adding it was reverted on the basis that there was no close, which is fine. So now we need a close. And addition of a new idea or new thoughts do not negate that. North8000 (talk) 17:19, 2 July 2021 (UTC)

It would no be out of line to just put it back in if someone cares to. But I requested a close at WP:AN North8000 (talk) 18:56, 2 July 2021 (UTC)

I moved the request to Wikipedia:Closure_requests North8000 (talk) 00:03, 3 July 2021 (UTC)

I came her to respond to North8000's request and intended to close it. The discussion seems very clear-cut and the level of consensus is such that I would normally feel confident making a close that the discussion was clearly in favor of the proposed change but ...... I think that Joe Roe is right: Proposals for new guidelines and policies require discussion and a high level of consensus from the entire community.... This is not a new guideline and it does comport with other guidelines as others pointed out above. That said, WP:N itself is a guideline and I think that the change is significant enough that the the procedural policy applies. This discussion has a goodly amount of discussion but was "the entire community" aware of it? I suggest that it should remain open for a short-ish period of time while this is listed on WP:CENT. I doubt that will change the result but some-one else can close it after that is done and feel secure in their compliance. I was considering whether WP:NOTBURO argued for simply closing it and not delaying but I have to admit I'm not confident of that. There should be more opportunity for other editors to find and comment on this change. Eggishorn (talk) (contrib) 01:36, 15 July 2021 (UTC)

Cool, I'll list it at WP:CENT.North8000 (talk) 12:38, 15 July 2021 (UTC)

I added it at wp:cent. First time for me. Feel free to fix if I screwed up. North8000 (talk) 19:10, 15 July 2021 (UTC)

With that having been completed and sufficient additional time passed, I requested a closure at Wikipedia:Closure_requests North8000 (talk) 18:00, 19 August 2021 (UTC)

This is creep already covered by WP:MEATBOT and WP:MASSCREATE

I continue to be concerned that many earlier participants in this discussion may have missed the point raised above that we already have policies on the books that speak to the mass creation of stubs. WP:MEATBOT stipulates that bot-like editing requires bot-level approval, and WP:MASSCREATE states Any large-scale automated or semi-automated content page creation task must be approved at Wikipedia:Bots/Requests for approval. ... While no specific definition of "large-scale" was decided, a suggestion of "anything more than 25 or 50" was not opposed. It is also strongly encouraged (and may be required by BAG) that community input be solicited at WP:Village pump (proposals) and the talk pages of any relevant WikiProjects. Bot operators must ensure that all creations are strictly within the terms of their approval. Putting these together, the only circumstance in which a bunch of pages would be created from a database would be one where the community explicitly approved it. The community would not do so if the reliability of the database is at all questionable; therefore, the problems that prompted this proposal are already addressed in two policies, and per WP:CREEP we shouldn't be needlessly expanding a guideline. {{u|Sdkb}} ^talk 21:12, 15 July 2021 (UTC)

Duplication of existing policy/guidelines into other policy/guidelines with reference to the originating policy/guideline is completely acceptable as to remind editors that these concepts exist. Yes, both listed policy/guidelines establish those already, the point here is to remind editors that simply passing notability via a SNG in which easily can be met through a database would likely be in violation of both of those. It is not trying to establish a new take on this, but simply framing a question related to notability in context of those. --Masem (t) 21:25, 15 July 2021 (UTC)

would an in text “see also” link to MASSCREATE and MEATBOT be enough? Blueboar (talk) 21:36, 15 July 2021 (UTC)

The question posed is that mass-creation is discouraged, which seems novel to me. WP:MASSCREATE simply requires prior approval, and I don't see a reason that exact thing can't already be added here. Probably some would argue that the consensus is obsolete since it seems nobody actually follows it, and I'm not aware of any BRFAs for mass article creation, but it still seems to be in the policy books with a valid RfC with widespread support behind it, so it seems safe to reference the relevant portions of that policy in this guideline. ProcrastinatingReader (talk) 21:39, 15 July 2021 (UTC)

The subject mass creation can be done without bots. Those referred to items (WP:MEATBOT and WP:MASSCREATE) are about bots. Also, I know that you weighed in elsewhere as opposed on this. Do you think that the additional point that you made might be better reformatted as an opinion rather than formatting the argument as the title of a section? Sincerely, North8000 (talk) 21:43, 15 July 2021 (UTC)

They aren't about bots. The issue was users mass-creating articles, particularly permastubs. Read the RfC that led up to that policy: Wikipedia:Village_pump_(policy)/Archive_66#Proposal:_Any_large-scale_semi-/automated_article_creation_task_require_BRFA. Some users raised the concern that this requirement was being added under the bot process, but it was done so anyway.

The reason why it would be a good idea to add it here is because few outside of bot owners read the bot policy, which makes that an unnatural place for this rule to reside, and is probably why it's never followed. ProcrastinatingReader (talk) 21:46, 15 July 2021 (UTC)

I think the above RfC really shows editors support discouraging mass creation, which is pretty different from just prior approval. Plus, "prior approval" was really developed for bots/bot owners, which are already subject to supervision by an existing community; applying this to regular editors would be difficult to enforce (who would be determining whether a proposed mass creation is allowed (surely not BRFA or BAG?)? would they be required to monitor the edits? what would be the recourse if an approved editor wanders outside the scope of their proposal -- automatic indef soft-block like with bot operators?). The existing language in those policies also doesn't cover the "formulaic creation of stubs", which is really what the heart of the issue is. And as others said, an editor intending to create a bunch of microstubs from a database would definitely not think to look at the bot policy for guidance. JoelleJay (talk) 17:42, 16 July 2021 (UTC)

Comments and support in principle: I got lost trying to follow this. I don't know if they were added by a bot (likely) but apparently, thousands of stub articles were created from a single geo-location source (in one case by one editor) that have been shown to be unreliably producing articles of non-locations with neither historical nor current populations also bogging down AFD. Creating articles and expanding Wikipedia coverage is one thing, mass-creating dictionary entries with one source, can violate policies and guidelines, to include WP:notability and many errors, that cannot be fully supported or confirmed with one source. This does zero for advancing the reliability of Wikipedia.

I am all for limiting bureaucracy, and some editors here may have created many hundreds of potentially notable articles with just a few issues, but imho, how can adding some wording to SNG, that at the very least could help clarify any possible confusion concerning WP:MASSCREATE and/or Wikipedia:ASSISTED and WP:MEATBOT, be anything but helpful? The "trainwreck" is not this discussion, which clearly and according to all the basically thousands of non-notable junk articles that could take years (maintenance and AFD are far behind creation) to resolve, is not inappropriate.

WP:SNG has the included "WikiProject notability guidance pages should be treated as essays". Everyone here should know that unless a concern or issue involving a WikiProject evolves into a Wikipedia-wide issue by RFC or maybe at AFD, the projects enjoy a certain amount of autonomy. These WikiProject "guidelines" normally reflect current practices, backed by a sometimes large, yet still considered local consensus, as well as arguably a more broad Wikipedia consensus. This would de facto mean a Wikipedia Project (likely everyone here belongs to one or more) guideline would carry more weight than some possible "essay" with a very limited number of editors. However, the "notice" is there and from my past experience (at least with two) is also needed. Now, if pointing this out seems important (a tie to SNG and WikiProjects), to remind editors basically that a project cannot "establish new notability standards", then it does not seem unnecessary creep to somehow add something along mass-creation lines (content or links) to SNG as a tie-in concerning notability. If someone can show me where editors mass-creating articles, bot assisted not, should not be "reminded" that such creations are subject to notability guidelines.

Maybe, just some "see also" at the top of the section or adding a subsection for this (linking to the above two or three locations) and also maybe including the "Wikiproject notice". Mass-creations, no matter how they are done, when not according to policies and guidelines, as well as current practices, sometimes arguably creating articles with zero notability, may help show Wikipedia is growing, but "sometimes" there does need to be some restraints to include the quality along with quantity. Since I rarely edit policies or guidelines (once that I know of) I would like to see something done. Maybe even listing this in the "See also" infobox. Even before the amendment proposal (trying to find common ground) there was large support (yes I know just a count) and collaboration is how we solve things. One editor stated, "There are databases of clearly notable subjects...". When there is some discussion somewhere on a subject involving a database, especially when there is "presumed notability", this changes things. Those editors may know their bot-assisted creations will likely survive AFD. Not all editors are familiar with "the rules" though. To this end, I feel this is worthy of discussion that could use a satisfactory conclusion. -- Otr500 (talk) 12:39, 23 July 2021 (UTC)

What is "mass creation"?

I think it might bear mentioning somewhere in any putative wording what constitutes "mass creation". The noticeboard threads I can remember seeing about this all concern people who have made thousands of stub articles. It seems obvious to me that this is what the phrase is meant to refer to, but those without the benefit of having read hundreds of AN/I threads will not have any idea what it indicates. It's easy to imagine someone interpreting this as encompassing them making three articles in one day (and that's it, they don't make any more after that). It would be prudent to avoid creating potential chilling effects on stuff that we don't think is a problem. The language at WP:MASSCREATE, for example, says that "25 to 50" was proposed and didn't see opposition, and clarifies that doing them in smaller batches could be acceptable, so that might be a good starting point. jp×g 22:03, 15 July 2021 (UTC)

• I would like to point out as an alternative that "mass creation" of drafts should not pose a problem, so long as the drafts are then brought up to article quality before being moved to mainspace. I have used this methodology myself, creating over 1,500 draft-space pages on missing U.S. state supreme court justices (who are inherently notable), and then moving these to article space individually as sources are found to support inclusion. I think that this would be an excellent alternative to mass-creating stubs in article space. The objection I can see being raised is that these drafts might not be developed, and might then be deleted as abandoned, but I don't think that's necessarily a worse outcome than the permastub lingering in article space. BD2412 T 22:11, 15 July 2021 (UTC)
  • Certainly a process like that, limited to your userspace until read for mainspace, is good, as long as one shows continued efforts to bring the remaining drafts to mainspace minimum quality. We usually aren't going to use any policy (short of hard copyright failures and clear BLP violations) to judge user-space drafts. Now, I would argue that using the Draft: space approach for something like this would be an issue since that's more of a formal process and while Drafts in this draft space shouldn't be removed due to notability, I would not want to see that flooded with mass creations based on slim notability factors. --Masem (t) 22:43, 15 July 2021 (UTC)
    • Project-space is also an option. I prefer not to start projects of this scale in userspace, as I feel that this discourages other editors from working on them. BD2412 T 23:56, 15 July 2021 (UTC)

Most of Wikipedia's policies and guidelines include terminology that is open to interpretation rather than being explicit. Far from being a flaw, this is essential to some of the things that make Wikipedia work. First, with somewhere north of 350 official policies and guidelines with est over a million words, that overlap with each other, such is a necessity. More importantly, it allows other factors to be taken into consideration when making such decisions. North8000 (talk) 23:32, 15 July 2021 (UTC)

• For something like the justices articles mentioned above, the option I would go with is “BATCH creation”… create batch of 5 or 6 stubs, get them up to snuff… and then move on to creating another batch. It may take a bit longer to reach the end goal (articles on all the justices), but it’s less disruptive. Blueboar (talk) 00:25, 16 July 2021 (UTC)
  • I don't think mass-creating drafts in draftspace is disruptive at all. Who does it affect? Another option that I have used sometimes is to start by creating a list article with the items as entries on the list, and then break them out into separate articles as the list entry becomes well-enough developed. BD2412 T 19:09, 16 July 2021 (UTC)
    • Whereas we don't usually prune out stuff in userspace, Draft space is meant to be used short term (6 months per footnote #2 at WP:DRAFTS) development of an article. So flooding draft space with articles that will all likelihood not be in a state for mainspace will waste a lot of people's time in cleaning up those drafts. Doing something lke that in userspace which doesn't have the time factor. That said, if one does the "create a few from a batch and expand them" and uses draft space for those 5-6 articles at a time, that's fine, that's a reasonable number to be worked on and cleared out in time. We don't want though someone dropping hundreds of stubs in a short period in draft space. --Masem (t) 13:14, 17 July 2021 (UTC)

Different proposal: Just reiterate WP:MASSCREATE and WP:MEATBOT in this guideline

As I said above, while policies already exist relating to this behaviour (WP:MASSCREATE and WP:MEATBOT) they are never followed in practice, which means the status quo is not working. I suspect it's likely because most content creators do not read the bot policy, and so are unaware of the mass-creation rules which were intended to apply to all users. Rather than the above wording about "discouraging" mass-creation, I'm proposing reiterating that policy in this guideline, without either encouraging or discouraging it but simply mentioning that it requires approval and consensus.

Specifically: Any large-scale article creation that appears automated or semi-automated must be approved in advance at Wikipedia:Bots/Requests for approval. A common example is the formulaic creation of stubs (eg: X is a Y that Z), that appears to effectively import an external database into Wikipedia. ProcrastinatingReader (talk) 12:25, 16 July 2021 (UTC)

Copy pasting from an already approved policy? Sounds promising. This might be the easiest path forward. –Novem Linguae (talk) 04:43, 19 July 2021 (UTC)

• Support; this addresses the core issue much better than the initial proposal. {{u|Sdkb}} ^talk 14:52, 19 July 2021 (UTC)
• The only change I would have that makes this relevant to WP:N is to alter this part A common example is the formulaic creation of stubs (eg: X is a Y that Z) that may meet an SNG's criteria, but that appears to effectively import an external database into Wikipedia.. or wording along those lines. --Masem (t) 15:03, 19 July 2021 (UTC)
• The issue isn't that people don't follow MASSCREATE and MEATBOT; it's that those sections don't actually say what many people want them to say. If someone mass creates pages without the use of a tool, without making errors, and without doing anything clearly against consensus, those sections are satisfied. Or, at worst, discussion is shifted to what does/doesn't have consensus. All of this enthusiasm for setting rules about mass creation needs to just migrate over there, or to a more central location, because notability is tangential. — Rhododendrites ^talk \\ 15:43, 19 July 2021 (UTC)

  Notability is not even tangential. It is a totally separate issue. What is proposed is a behavioural guideline, but this is a content guideline. Phil Bridger (talk) 16:03, 19 July 2021 (UTC)

• The notability issue is this: is a subject/topic notable simply for being in a database with other subjects/topics that might be notable? When one mass creates based on a database (acting in a bot like way) you make an OTHERSTUFF assumption… one that is not necessarily valid. Blueboar (talk) 17:07, 19 July 2021 (UTC)
  • And it is important that there are criteria that exist in SNGs that can be read to support crawling through a DB to create stubby articles that would pass the SNG (notably many in NSPORT, like WP:NGRIDIRON#1) We're not saying those SNGs are wrong (though there's been separate discussion to fine tune them better to also prevent mass creation), but that just because the SNG gives an easy potential route to mass creation, it should not be done. But this also is based on exactly how the mass creation is being done, and the closer to a bot-like manner, the more problematic it is. To Phil's point above, a user that systematically mass creates articles but is taking the time and effort to do additional validation or improvement is not the problem. It is the bot-like nature we're running up against. --Masem (t) 17:15, 19 July 2021 (UTC)
  • The notability issue is this: is a subject/topic notable simply for being in a database with other subjects/topics that might be notable - Then mass creation isn't relevant. If we shouldn't create articles based just on a database, we should say that. Doing so a hundred times or one time doesn't really matter for notability purposes. If there's consensus that this shouldn't be done, that would save the time of changing MASSCREATE/MEATBOT. — Rhododendrites ^talk \\ 18:22, 19 July 2021 (UTC)

• IMHO this new proposal starts with and is based on an incorrect statement. By it's own wording this applies only to automated and semi-automated creation. Also agree with Blueboar that the proposed location in the main proposal is suitable. Sincerely, North8000 (talk) 17:20, 19 July 2021 (UTC)
• Oppse as worded. There are databases of clearly notable subjects, such as those detailing our missing state supreme court justices. There must be some level of permissibility to manually engage in quick series of edits that could be perceived as "bot-like" without risking getting hauled into some nightmare of bureaucracy. BD2412 T 18:47, 19 July 2021 (UTC)
  • Key words are "automated or semi-automated", with the understanding that we do not have hard definitions of when that threshold is crossed, by design, though usually we look at rates of editing to determine this (with article creation on the order of a few minutes or less per article as a problem). The method you described in how you are expanding judges would not fall under that because it fails what we'd normally call semi-automated. That's where the pointed to the existing policy would be used to describe this more. --Masem (t) 18:57, 19 July 2021 (UTC)
    • Does using AutoWikiBrowser to populate articles count as "semi-automated"? Does it matter if the text is drafted elsewhere in the project and AWB is only used to quickly move that text to articles? BD2412 T 19:37, 19 July 2021 (UTC)
      • The use of AWB alone wouldn't be that. We'd judge if articles had the appearance of mass creation by auto or semi-auto tools by the sustained rate of creation, how little content they include, and if they look like the result of simple mail-merge type information. A process that uses AWB to put articles to mainspace, but where it is clear that time and effort was made on the article before populating it in mainspace wouldn't be an issue, given that likely this wouldn't even hit the creation rate threshold. But the exact bounds on all those points are vague and impossible to define further, just that all those factors and likely others are considered as part of it. --Masem (t) 20:22, 19 July 2021 (UTC)

      The use of AWB to create articles like:

      1. "Joe Dobbs is a Supreme Court Justice from Ottowa. (cite to http://scjustices.com/joedobbs)"
      2. "Joe Bloggs is a Supreme Court Justice from Seattle. (cite to http://scjustices.com/joebloggs)"
      3. "NotJoe Dobbs is a Supreme Court Justice from Dallas. (cite to http://scjustices.com/notjoe)"

      (etc...) is almost certainly a violation. If they don't read like they're literally written by a script, probably not. Do you have specific examples of article revisions you're concerned about? ProcrastinatingReader (talk) 20:30, 19 July 2021 (UTC)

      • @ProcrastinatingReader: At one point we had a bot called PolBot which literally pulled the entire Federal Judicial Center database on federal judges and made about 3,000 articles overnight. They generally looked like this to begin with, which I think is fine. I personally used AWB to create about 1,800 draft-space stubs for state supreme court justices, which can be found at Wikipedia:WikiProject Missing encyclopedic articles/United States judges and justices. They typically started like this, and many still are in that shape, but most are improved from that, and over 600 have been sourced and promoted to article space. Some are extensive (e.g., Charles Erasmus Fenner) and some remain along the lines of James P. Alexander, which is certainly closer to your example above. I see no problem with articles like Alexander; it clearly states the basis for encyclopedic notability, with a reliable source, and therefore does not belong in draft space even though it was created there. BD2412 T 19:20, 22 July 2021 (UTC)

        @BD2412: PolBot had a bot approval. The MASSCREATE policy only says that this stuff needs an approval. It's right that stuff like this get consensus before starting, especially since it has several issues and is very obviously bot-like. At minimum users should articulate a clear plan detailed for the creation, its scope, compliance with copyright policy, etc. I agree that James P. Alexander is not problematic, but the systematic mass-creation of articles like that should be subject to prior discussion and consensus, as it can easily get disruptive. There's no harm in starting a discussion to first discuss if pages from the University of Texas can be imported. At minimum it just saves a larger headache down the road, if it turns out there was a problem with the mass-creation that flew under the radar for years and misled the readers. ProcrastinatingReader (talk) 11:32, 24 July 2021 (UTC)

        Perhaps such a policy should take into account the experience and ability of the editor undertaking the task. When I created draft stubs for the missing justices, I had already been using AWB for approved tasks for many years, and had already created hundreds of articles in the area. I am also unclear on what constitutes "mass" creation. I did these by state, so some states had 100+ missing justices, and some had fewer than two dozen. There are instances where editors have bot-created tens of thousands of stubs. Where is the dividing line? BD2412 T 18:02, 24 July 2021 (UTC)

        I think pages in draftspace are unlikely to be much of a problem. As for the dividing line, unsure. My position is that this prohibition is already policy. It's a strange prohibition due to a) the unclear dividing line; and b) the lack of enforcement mechanisms. But seeking consensus in advance rarely hurts; for creations of hundreds to tens of thousands of BLPs I think it's a reasonable ask. The last unapproved mass-creation at AN was a then-sysop, so IMO I don't think experience/ability of the editor undertaking the task is a good exemption to add to the rule. ProcrastinatingReader (talk) 20:16, 24 July 2021 (UTC)

• Oppose; this is a very severe warning with a very vague definition. If we're going to say "must", we shouldn't say "appears" twice. jp×g 09:08, 27 July 2021 (UTC)

  That wording is intentional; "appears" is a consequence of WP:MEATBOT. It's technically impossible to know whether an edit is performed by a human or a bot. When you're talking about edits that import a database into Wikipedia it's impossible to know whether they're semi-automated or not, and it's not really relevant either. ProcrastinatingReader (talk) 11:52, 27 July 2021 (UTC)

The discussion above is closed. Please do not modify it. Subsequent comments should be made on the appropriate discussion page. No further edits should be made to this discussion.

Do any of the SNGs overrule WP:GNG?

I just need a straight yes or no answer, and where it says so. Howard the Duck (talk) 14:20, 27 July 2021 (UTC)

I don't think the word "overrule" really applies. There are two ways for an article to be presumed justified in having an article, either through the GNG or the applicable SNG. In certain cases - notably NPROF and ORG - an SNG does explicitly set aside the GNG in some respects, but in other cases the two usually operate in parallel. As to "where it says so", the situation is generally set out in WP:SNG, which came out of an RfC within the last year or so. WP:NPROF and WP:NORG set out specific cases where the GNG is set aside. Newimpartial (talk) 14:44, 27 July 2021 (UTC)

"[T]he situation is generally set out in WP:SNG". I read that, and I saw "articles which pass an SNG or the GNG may still be deleted or merged into another article, especially if adequate sourcing or significant coverage cannot be found, or if the topic is not suitable for an encyclopedia". So what does that mean? You can shout WP:GNG, then at the very most, delete the Wikipedia article but merge the information into another article? I'm not really deleting information, just the actual article, then merge that information somewhere else. Howard the Duck (talk) 15:11, 27 July 2021 (UTC)

The thing is, there is no guarantee for a topic to have its own article, based on either SNG or GNG. WP:NOT is also a consideration, and lots of reliably sourced topics run afoul of NOT. Topics within the domain of an SNG that don't meet it generally should not have an article, and topics outside the SNG domains (or in weakly presumptive SNG areas, like NSPORT, that demand a GNG pass as part of th SNG) that don't meet the GNG should generally also not have an article. But topics that *do* are not guaranteed an article, and topics that are borderline may be given an extended opportunity for sourcing to be improved, depending on the situation.

And yes, the best way to deal with sourced information that doesn't seem appropriate as an encyclopedic topic is often a merge discussion, rather than AfD. If we could all learn one thing from the Notability Talk page, I would hope for that to be the thing we learn. Newimpartial (talk) 15:33, 27 July 2021 (UTC)

No, they still have to pass GNG, it is just that SNG's give them a bit more time to be worked on.Slatersteven (talk) 15:12, 27 July 2021 (UTC)

This does not reflect how the SNGs and GNG have ever worked, either in policy or practice, much as certain editors would wish it to be so. Newimpartial (talk) 15:35, 27 July 2021 (UTC)

A way to see it is that of NPROF and NORG, they still uphold the principle of the GNG - we're looking for significant coverage of the topic from multiple independent, secondary sources. NORG modifies this by demanding more specificity to the sourcing to avoid cases of weak RSes or where potential COI exists and eliminating trivial mentions that might be okay in other areas. NPROF modifies this that not only significant coverage of the the academic could be considered but also coverage of the academics' research itself, and that that can be shown through accomplishments and references/citation counts. Neither change the ultimate target of what notability is trying to achieve, in-depth coverage of a topic based on independent and secondary sources at the end of the day, just tuned slightly for their specific functions. So I wouldn't call it an overriding feature --Masem (t) 15:17, 27 July 2021 (UTC)

I actually "read" WP:NPROF, and yes it does uphold GNG; if anything it clarifies what GNG means for people such as academics. I haven't read NORG, but by the way you described it, it does look like what NPROF does. For these two cases, the SNG builds upon GNG.... now, we all know which SNGs say they overrule GNG and I'd be interested if their SNG does indeed overrule GNG, and if that's allowed. Howard the Duck (talk) 15:26, 27 July 2021 (UTC)

Given NORG/NCORP as the one SNG that probably implies or outright states it overrides the GNG, it only does this in the definition of what is acceptable sourcing because it imposes a tighter subset. But the goals it is still trying to achieve remain consistent with the intent of notability (in-depth coverage via ind. sec. sources), hence I don't consider it a true override. An overriding SNG hypothetically would be one that says "All X can have a standalone article without question", which no SNG actually states. Eg even WP:GEOLAND states that we presume notability for all recognized populated places but does not actually assume no-questions-asked notability. --Masem (t) 15:30, 27 July 2021 (UTC)

People who cite WP:GEOLAND push the idea though that their SNG overrules WP:GNG. User:NemesisAT is even proud of making a Wikidata entry Wikipedia article passing that SNG, while not passing GNG, recently. So what gives? Howard the Duck (talk) 15:32, 27 July 2021 (UTC)

Howard the Duck, and one of the problems with this approach is that the articles are occasionally built on sourcing so flimsy and erroneous that, when you actually go and check whether the tiny village in question actually exists (on Google Maps, say), you can't find any evidence that it does. Reyk _YO! 06:57, 28 July 2021 (UTC)

Masem, the reason I don't like the way of talking about the principle of the GNG, as you did here, is that it lends itself to the mischaracterization Howard just made. Call the principle of reliable, independent sourcing "Notability" and the discussion becomes much more clear. The WP:GNG adds to that the notion of multiple sources and WP:SIGCOV, but those do not apply where the GNG does not apply. So WP:NORG replaces WP:SIGCOV with WP:CORPDEPTH and "multiple" sources with WP:SIRS, both much higher standards.

So, Howard, have you read the RfC that produced the current SNG text? That was am RfC to create language to reflect the status quo of WP:N as a whole, and it was pretty clearly found in that discussion and its close that the relationships between the SNGs and the GNG differ a great deal; it was also recognized that many of them provide a presumption of notability quite separate from the GNG criteria. Newimpartial (talk) 15:48, 27 July 2021 (UTC)

I mentioned "significant coverage" as part of what notability is looking for, and that while NORG may tighten or limit bounds of that, its still the same basic concept as it is all around being able to eventually write an in-depth article meeting the core content policies (NOT/V/NPOV/NOR) from independent and secondary sources. NORG just makes very specific exclusions on sourcing aspects that would otherwise be accepted elsewhere. --Masem (t) 15:55, 27 July 2021 (UTC)

But there is all the difference in the world between "significant coverage" as a concept and WP:SIGCOV as a guideline. Coverage requirements for academics in WP:NPROF and for numbers in WP:NUMBER define "significant coverage" in those domains, but their logic is even more different from WP:SIGCOV than WP:SIRS is from "multiple, reliable sources". And I also disagree with you about the purpose of NORG - it is not the additional challenges of writing verifiable ORG articles that makes it necessary to demand more and better sources up front; it is the risk of promotion and inappropriately biased sourcing that makes it necessary to create barriers against the creation of articles that are NOT encyclopaedic. Newimpartial (talk) 16:05, 27 July 2021 (UTC)

Its far easier to explain the intents of the SNGs as being field-specific implementation of the GNG rather than overrides (as originally asked). I agree that NORG has many deviations from the GNG to seem vastly different, but it has the same goal at the end of the day as I said, with the added goal of what you said, to prevent the proliferation of COI-filled articles. It does that by narrowing what it defines as significant coverage and what are appropriate sources (including AUD), but all that can still be read as seeking "significant coverage from independent, secondary sources", for all purposes. This approach is a KISS-style approach to explaining the differences and keeping the idea from the previous wording issues that the SNGs and GNG tend to work hand-in-hand. --Masem (t) 16:10, 27 July 2021 (UTC)

I might have believed that at one time, but I certainly don't now. The problem is this: if we use "GNG" to mean both the general principle of Notability and the specific criteria embodied in SIGCOV and "multiple, reliable sources", that only confuses people. Yes, the SNGs are based in some sense on Notability as a general criterion of sourced information on a topic, but most of them do so while superseding certain aspects of the GNG criteria. As long as people try to figure out the SNGs as if they were fiddling around with the minutiae of GNG tests, rather than (usually) establishing their own, domain-specific criteria, we are going to have more confused discussions like this one. The question of what categories of communities in the Philippines can be reliably shown to be officially recognized and populated ought to be settled in its own merits, not through a "hail Mary" attempt to separately apply the GNG in addition to GEOLAND. Newimpartial (talk) 16:24, 27 July 2021 (UTC)

• If you read the SNGs, most say that notability is “presumed” if certain conditions are met. Why do they say this? Because there is a very strong likelihood that IF those conditions are met, passing GNG will be possible. SNGs should not be seen as an alternative to GNG, but a strong indication that GNG can be met (so keep looking for sources). Blueboar (talk) 15:39, 27 July 2021 (UTC)
  • Perhaps unfortunately, thus has never been true as a general rule. A close reading of the text of the SNGs and the GNG over time shows that both a GNG pass and a pass for most SNGs offer the presumption of meriting an article; it has only ever been a few SNGs (like NSPORT) where an SNG pass offers the presumption of a GNG pass (I have referred to those elsewhere as "weakly presumptive" SNGs). Newimpartial (talk) 15:48, 27 July 2021 (UTC)
    • And what makes it worse is that people look at the words "rebuttable presumption" that the sources necessary to meet WP:V and WP:N are going to turn up somewhere, and interpreting them as an exemption from those sourcing requirements. Reyk _YO! 15:53, 27 July 2021 (UTC)
      • I'm not sure what "rebuttable presumption" you are talking about, here. Obviously articles have to have adequate sources, but what count as adequate sources differ by domain. WP:NNUMBER, for example, specifies a sourcing requirement for numbers that would be absurdly narrow (and demanding) in any other domain. WP:GEOLAND, by contrast, does not demand such a high standard. Newimpartial (talk) 15:57, 27 July 2021 (UTC)
    • Another subject area where articles are almost always kept at AfD is railway stations, despite not always meeting WP:GNG. NemesisAT (talk) 15:58, 27 July 2021 (UTC)
      • This comes back to what the "presumed notable" and "rebuttable presumption" means. We allow SNGs to establish cases that clearly don't meet the GNG (significant coverage in independent + secondary sources) on the basis that meeting those conditions, that significant coverage can be found given time and effort. EG: GEOLAND's first point that any recognized populated place is presumed notable is that there's likely written history about that place that can be obtained from local sources at that place, and thus require the effort to physically go there and seek those out (not everything is online). Thus, when people AFD these types of articles that are presumed notable, this is where WP:BEFORE kicks in - they have to show an appropriate challenge to that presumption of notability by demonstrating, as best they can, no further sourcing likely exists. And most AFDs of SNG-meeting articles are kept due to an insufficient BEFORE approach - eg just searching online and finding nothing doesn't cut it. However, if the person did a reasonable BEFORE (had local access to a library and reporting a negative search) and found no more sources forthcoming, now the onus is on those wishing to keep to produce those sources, overcoming the presumption of notability. --Masem (t) 16:05, 27 July 2021 (UTC)
        • I'm going to say again: what you describe isn't how anything actually works. We allow SNGs to establish cases that clearly don't meet the GNG (significant coverage in independent + secondary sources) on the basis that meeting those conditions, that significant coverage can be found given time and effort. No, we don't. NPROF isn't premised on the idea that if we try hard enough we will find that profs who published frequently-cited articles will also be covered in RS human-interest pieces that would meet GNG. Rather, it has created a non-GNG Notability/sourcing standard for its domain. Newimpartial (talk) 16:15, 27 July 2021 (UTC)
          • As I said above, NPROF can be normalized in my scheme by recognizing that for an academic profession, it is not just their person but also their research that is considered part of the same topic, and it is usually that research that is going to get the secondary significant coverage rather than the person themselves. So it remains consistent with the GNG in that regards. --Masem (t) 16:41, 27 July 2021 (UTC)
            • Consistent with the principle of the GNG - in other words, source-based Notability - absolutely, yes. Consistent with the GNG as a set of criteria - not at all. I no longer see any point in trying to kludge the latter; it just confuses people IMO. Newimpartial (talk) 16:44, 27 July 2021 (UTC)
              • On the facet that SNGs are providing unique criteria towards the GNG that the GNG otherwise doesn't have, yes, I agree. but as Johnbod states below, I don't consider this "overriding" the GNG. Again, that would be completely ignoring sourced-based demonstration of significant coverage completely, which no SNG (not even NPROF) does. A criteria may seem to spell that out, but that's always tied to a presumption of notability that a sourced-based demonstration has to be met at some point. --Masem (t) 16:53, 27 July 2021 (UTC)

I don't disagree with any of that, and I positively agree with King of Hearts' paraphrase of Johnobod below. But I just don't think using "GNG" to mean both the WP:N principle and the criteria specified in the GNG section is at all helpful. Newimpartial (talk) 17:06, 27 July 2021 (UTC)

• "straight yes or no answer": in theory no, in practice yes. It doesn't say so anywhere. Johnbod (talk) 16:05, 27 July 2021 (UTC)
  • I think WP:SNG is reasonably clear in saying "sometimes yes". Newimpartial (talk)
    • On the contrary, it is very careful to avoid any clear talk of anything like "overruling". Johnbod (talk) 16:24, 27 July 2021 (UTC)
      • That was just good policy scribing on our part. :P As I stated near the top of this discussion, I wouldn't use "overruling" in any case. But the language is closer to "sometimes yes" than "in theory no", IMO. I don't think the theory of "no" is articulated anywhere in WP:N; it seems to be an oral tradition. Newimpartial (talk) 16:29, 27 July 2021 (UTC)
  • +1 exactly as I would put it. In theory, GNG just means that articles need to be based on reliable sources, and we need enough content to write an article about something, which is just common sense. Under this view, the SNGs are just tailoring the sourcing requirements to each subject. In practice, GNG refers to a very specific type of sourcing (primarily news articles and certain types of books), and SNGs can be said to "override" GNG in that for certain subjects, either sources that are disallowed under GNG are allowed or vice versa. -- King of ♥ ♦ ♣ ♠ 16:51, 27 July 2021 (UTC)

No. Or rather, as with all our myriad policies and guidelines we aim to reconcile through consensus in each specific case. Alanscottwalker (talk) 16:20, 27 July 2021 (UTC)

• Why do I think, given the OP's strident language both in posts and edit summaries, that the "need" for a "straight yes or no answer" is for ammunition to beat people over the head with at AfD? Ravenswing 16:26, 27 July 2021 (UTC)
  • Well, they wouldn't be able to beat anyone with a "yes" (which is more true than a "no", within that false dichotomy, IMO - I don't see how anyone could read NPROF or NORG and see how they are routinely used, and then conclude that the GNG is not set aside when it is supposed to be). Newimpartial (talk) 16:31, 27 July 2021 (UTC)

    I'm not sure it's a case of "when it's supposed to be", but I otherwise agree that "yes" is more true than "no", and that neither of them is entirely accurate.

    If you think about AFD, especially for borderline cases (imagine: an academic best known for marrying a celebrity; a business best known as the location of a media event), we could easily have one editor correctly argue that the GNG ought to be applied, and have another correctly argue that an SNG ought to be applied. Part of forming a consensus at AFD is forming a consensus about which set of guidelines is most appropriate. WhatamIdoing (talk) 23:50, 27 July 2021 (UTC)

    PROF is simple because it is orthogonal to GNG; it never restricts the applicability of GNG to its subject, e.g. Wikipedia:Articles for deletion/Angela Zhang (scientist). So whether someone should be considered an academic or not doesn't matter. They can be treated as an academic for the purposes of PROF, and as a non-academic for the purposes of GNG. If at least one passes, then the subject is notable.

    On the other hand, for restrictive SNGs like NORG, I can envision a dispute over whether it is subject to the SNG or not. Fortunately, unlike people, usually something either is an organization or it isn't, so for NORG this issue does not really come up in practice. -- King of ♥ ♦ ♣ ♠ 00:29, 28 July 2021 (UTC)

It could be from frustration that a simple categorical answer does not exist. First, by "GNG" do they mean the wp:Notability page including the meta-policy header, or just the sourcing-GNG criteria in it? Next, note that wp:Notability itself says that meeting the SNG bypasses the sourcing-GNG. But then the "principle" type wording of SNG's gives deference to the sourcing-GNG. IMHO Wikipedia:How Wikipedia notability works describes the overall reality. North8000 (talk) 17:02, 27 July 2021 (UTC)

• NPROF is the only SNG I'm aware that explicitly has no hierarchical relationship to GNG: a subject is considered notable through meeting the NPROF criteria or through meeting GNG, with neither guideline affecting the other (unlike NORG) or even overlapping in criteria. Other SNGs are explicitly subordinate to GNG, e.g. NSPORT (The topic-specific notability guidelines described on this page do not replace the general notability guideline. and Q2: If a sports figure meets the criteria specified in a sports-specific notability guideline, does this mean they do not have to meet the general notability guideline? A2: No, the article must still eventually provide sources indicating that the subject meets the general notability guideline. Some SNGs are hand-wavy about the relationship, but ultimately have criteria that are substantially similar to GNG anyway (like NEVENT). JoelleJay (talk) 05:06, 28 July 2021 (UTC)
  • NPROF is at one end of a continuum of relationships, NSPORT is at the other end. But the rest are not clustered at the NSPORT end of the continuum: NNUMBER and NORG set such high sourcing requirements that it seems absurd to me to regard them subordinate to the GNG, while NBIO has two kinds of criteria without a hierarchical relationship between: WP:BASIC is equivalent to GNG in its domain, but the specific criteria like WP:NAUTHOR operate separately from a GNG/BASIC-style logic. Newimpartial (talk) 12:21, 28 July 2021 (UTC)
    • Well, it rather seems a potential seriatim inquiry (sub(ordinate) in place).
    • Is the Conservative Party "generally notable"? Yes. Stop. No, can't decide? Is it notable as an organization?, etc. (if something is found in several places, like in the "Big Authoritative Book of British Politics" norg, does not matter)
    • Is Al Gore generally notable? Yes, stop. No, can't decide? Is he notable as a professor? or, Is he notable as author? or, Is he notable as a politician? or, etc. etc.
    • Is Princess Anne or Tiger Woods generally notable? Yes, stop. No, can't decide? Is she or he notable as a sports person?, etc.
    • There is no need to go further than the (blunt, general, usually obvious) GNG, unless there is something specific in the subject and in the narrow nature of the found sourcing. -- Alanscottwalker (talk) 15:31, 28 July 2021 (UTC)
      • That proposed logic is completely incompatible with WP:N. For ORGs and NUMBERs, using the GNG as a blunt, general, usually obvious tool gives a whole lot of "false positives" for notability, while for PROFs and some other speciality BIOs it gives plenty of false negatives. So I'm afraid that doesn't work, except in cases (like SPORTS) where a well-informed person already knows it is going to work. Sigh. Newimpartial (talk) 16:19, 28 July 2021 (UTC)
        • You've, taken "blunt, general, usually obvious" out of context. So, no, not false positives, unless by that you mean, you don't agree with the outcome of any particular Afd. The Authoritative Books published on the number Pi, still make Pi presumably worthy of an article, generally so. -- Alanscottwalker (talk) 13:11, 29 July 2021 (UTC)
          • If some fool well-intentioned admin were to close an AfD for a number as "keep" based on newspaper coverage, where peer-reviewed publications by mathematicians on that number did not exist, that would be a "false positive" for Notability and an incorrect result. That is a policy question, not a matter of personal opinion, just the same as if somebody closed an ORG AfD without considering WP:AUD requirements for sources. Newimpartial (talk) 13:31, 29 July 2021 (UTC)
            • If some well intentioned editor or admin don't know that notability is not a policy question, then they are likely not to understand PAGs. It is in no way a false positive to find notability based on numbers written as the subject of authoritative books, as GNG demonstrates. -- Alanscottwalker (talk) 20:35, 29 July 2021 (UTC)
              • • No; if the books don't meet the requirements of WP:NNUMBER, they don't contribute to the Notability of numbers, even if they are otherwise RS. Newimpartial (talk) 21:44, 29 July 2021 (UTC)
                  • The use of RS is always a matter of suitability and context including in the GNG. "Otherwise RS" have no meaning, unless you mean irrelevant RS, but than those would not be authoritative on the topic. -- Alanscottwalker (talk) 21:58, 29 July 2021 (UTC)

On WP, whether something counts as RS for Notability does indeed depend on context. Within the WP:GNG framework, WP:SIGCOV offers the relevant specification, but this is overruled by WP:SIRS in the case of WP:NORG and by some rather specific requirements internal to WP:NNUMBER. Any admin ignoring these more specific requirements in favor of SIGCOV, in an AfD decision where NORG or NNUMBER applies, is likely to make an error, even if the sources in question are relevant and comply with SIGCOV. :) Newimpartial (talk) 22:06, 29 July 2021 (UTC)

• No. GNG does not contain only SIGCOV, and no guideline, whether GNG or SNG, can possibly overrule the appropriateness and context requirements. -- Alanscottwalker (talk) 22:27, 29 July 2021 (UTC)
  • • • I don't believe your Conservative Party example is correct. The GNG says nothing about WP:AUD, so if we used GNG as a first pass a lot of local businesses would be getting through without needing to pass AUD. WP:NORG is not an alternative to GNG, it is the only acceptable version of GNG for organizations. -- King of ♥ ♦ ♣ ♠ 18:24, 28 July 2021 (UTC)
        • What? The GNG uses an organization, IBM, as an inarguable example of passing the GNG. That's not a false positive, it's plain as day. -- Alanscottwalker (talk) 19:18, 28 July 2021 (UTC)
          • If you think the GNG, rather than NORG, applies to IBM, then I don't think you understand WP:N very well. Newimpartial (talk) 13:37, 29 July 2021 (UTC)
            • Parts of guidelines are not hermetically sealed from the rest of the guideline. Understanding N is understanding its introduction, as several have pointed out. -- Alanscottwalker (talk) 20:46, 29 July 2021 (UTC)
              • • Yes, and the introduction tells us that Notability can be shown based on either the GNG or the applicable SNG. SNGs such as NNUMBER and NORG tells us that in those domains, a GNG pass on it's own is not sufficient to establish Notability. This isn't science rockets. Newimpartial (talk)\
                  • • As others have indicated, guidelines do not operate with such imagined rigidity. -- Alanscottwalker (talk) 22:06, 29 July 2021 (UTC)

I look forward, then, to your next (gelatinous?) AfD close. Newimpartial (talk) 22:08, 29 July 2021 (UTC)

One thing is certain, it won't and never has taken rocket science to do. -- Alanscottwalker (talk) 22:36, 29 July 2021 (UTC)

• Just as an overall statement: There is a reason that notability (as well as the GNG and the SNGs) is a guideline - there is no rigor or like in its application, nor should there be. It is meant to lay out concepts to be concerned when notability comes up - primarily at AFD - so that consensus-based discussion can occur. It is important that we have a general agreement on the relationship of the GNG and the SNGs, but this all still comes up as part of guidance and not absolute "rules" like WP:NOT or WP:V. To that end, that's why its probably not as important to be precise about this relationship (particularly given how diverse it is between the various SNGs) but instead keep an overarcing concept to the relationship. --Masem (t) 16:30, 28 July 2021 (UTC)
  • I think the reason why we codify guidelines is to help us assign weight to !votes in discussions. Ultimately, there are three factors that go into a closer's decision: 1) the raw !vote count; 2) whether the arguments are compliant with policies and guidelines; 3) a subjective assessment of whether the arguments are reasonable from general principles like WP:5P. If everyone sees the same sources and the dispute is over whether the coverage is long enough to be considered "significant", then it's pretty much going to go down to a !vote count; it is not the closer's job to tell people they were wrong in their subjective evaluation of sources. If one side has put together a strong argument that the other side has failed to refute, then the strength of the argument will win out. Finally, in certain situations (very rarely in AfD), a very well-argued IAR could be accepted despite what policies and guidelines say. -- King of ♥ ♦ ♣ ♠ 19:51, 28 July 2021 (UTC)
• For many years now this has been argued relentlessly. AFDs usually end with articles kept if the any of the subject specific guidelines are met, even without the general notability guidelines being met. A scientists is notable for their accomplishments, even if there never much written about them. Astrological articles, species, and various other obviously encyclopedic articles get kept as well. We are a legitimate encyclopedia not just a collection of popular culture articles that get media attention for a time. It clearly reads: A topic is presumed to merit an article if: It meets either the general notability guideline (GNG) below, or the criteria outlined in a subject-specific notability guideline (SNG) listed in the box on the right. Some who wish to change it may argue otherwise, but this is just how its always been. Your accomplishments make you notable, not the coverage you get for your accomplishments. Dream Focus 16:37, 28 July 2021 (UTC)
• Every time this comes up, it seems like the entire page Wikipedia:Notability is confused with "the GNG". The latter is a section of the former. "The GNG" doesn't define what "notability" means on Wikipedia; the introduction of Wikipedia:Notability does (to the extent that such a squishy concept can be defined). That introduction also spells out that an SNG can serve as a replacement for "the GNG". XOR'easter (talk) 21:31, 28 July 2021 (UTC)
• Yes. It says so at the beginning of N: A topic is presumed to merit an article if: 1. It meets either the general notability guideline (GNG) below, or the criteria outlined in a subject-specific notability guideline (SNG) listed in the box on the right; and 2. It is not excluded under the What Wikipedia is not policy. That "or" in that sentence is what editors will point to when arguing that meeting either GNG or an SNG means a topic is presumed (not guaranteed) to merit an article. So it's not that either SNG or GNG "overrules" the other (or predicts the other) but rather that they are two alternative and separate pathways to the presumption of notability (which, again, is not a guarantee). I don't agree with this, but this is how it is, AFAIK. Levivich 11:54, 29 July 2021 (UTC)
  • 'Yes, it does not overrule', seems similar, to 'no, it does not overrule', with different emphasis. It does seem certain there is more than one path, there is the general path and the specific paths; on the other hand, the construction of these paths use several similar concepts to inform, so it seems rather they are intertwined paths which, in a given circumstance, come together and diverge and come together and diverge, and come together each looking for the place to arrive at the presumption. -- Alanscottwalker (talk) 12:34, 29 July 2021 (UTC)

  On the other hand, some SNGs explicitly do defer to the GNG and state they only predict (GNG) notability. JoelleJay (talk) 17:29, 29 July 2021 (UTC)

So now what will be your position regarding barangays? Are all 42 K+ barangays – the local divisions of 1,634 municipal settlements of the Philippines (cities and municipalities or Philippine towns) – automatically notable because of WP:GEOLAND? Because for years various barangay articles have been AfD-ed or redirected due to various reasons, like lack of sources or unencyclopedic content (like directory, educational listings, or tourism-oriented listings which all violate WP:What Wikipedia is not). See Wikipedia:Tambayan Philippines/Frequent discussions/Articles on barangays for the related cases. JWilz12345 (Talk|Contrib's.) 13:09, 29 July 2021 (UTC)

That wasn't the question asked here, nor is this the place to ask it. If you wanted to discuss that question on a policy page, I would suggest the talk page for WP:GEOLAND. Newimpartial (talk) 13:26, 29 July 2021 (UTC)

I think that the answer to "Do any of the SNGs overrule WP:GNG" is that in the context of wiki-notability, it is an ambiguous question and thus no conscientious answer can be given here. It's ambiguous because any simple answer could be interpreted / applied in many many different ways. And if that weren't an issue, trying to accede to the demand for a single word answer from a choice of two words makes it doubly impossible to conscientiously answer. North8000 (talk) 17:48, 29 July 2021 (UTC)

No guideline overrules any other guideline. Both the GNG and SNGs should be seen as providing 'rule of thumb' guidance on when a topic is likely to be considered notable (i.e. worth having an article on). --Michig (talk) 18:07, 30 July 2021 (UTC)

Sure, but their scope of application is different, and they each sometimes explicitly defer to the other in certain domains (e.g., NSPORT is only weakly presumptive and requires an eventual GNG pass, deferring to the latter; NORG takes clear precedence over the GNG within its domain, and so does NNUMBER). Newimpartial (talk) 18:13, 30 July 2021 (UTC)

Television appearances

Having an interesting discussion at this AFD (and will ping Darth Mike, with whom I am having the discussion). Curious as to whether people believe that a significant number of television appearances (over time) would be considered significant coverage for the purposes of meeting our notability requirements. We seem to agree that simply being on television isn't enough and that thousands of people are on television every day for mundane reasons. In this instance, the subject has been invited as a special guest to appear on a number of variety shows (breakfast television and whatnot) to showcase her skills and talents. She is arguably the focus of that portion of the broadcast in each instance. There is ancillary print coverage in unreliable sources. The question is, do broadcasts of that nature constitute significant coverage? St★lwart¹¹¹ 05:54, 24 August 2021 (UTC)

• When we say coverage, we mean that a source (written, video, or otherwise) is about the subject of the article. A television appearance is not coverage of the guest any more than a written publication is coverage of its author. – Joe (talk) 07:01, 24 August 2021 (UTC)

  To be clear, she's not the host (the "author"), she's the subject of the coverage. St★lwart¹¹¹ 07:07, 24 August 2021 (UTC)

  It doesn't matter what the role is. Unless it's some sort of documentary from which we could extract actual biographical information, it's not coverage. And according to the article on Lilia Stepanova she "performed" on these television shows; they weren't programmes about her. How could you possibly use a contortionist act on Jimmy Kimmel as a source for a biography? I think this is one of those common cases where people find it difficult to accept that a person who is familiar, noted, or famous, can still not be notable by our definition. – Joe (talk) 07:17, 24 August 2021 (UTC)

• Nope, this would not be sufficient for notability, though would hint that there aught to be sources that explain why this person was a frequent guest on that show, which those sources would be useful as demonstrating notability. There are, for example, lots and lots of D-list actors in television productions that have copious appearances but would not meet notability guidelines just for that. --Masem (t) 13:25, 24 August 2021 (UTC)

Sports proposal

Just FYI, there is a proposal at Wikipedia_talk:Notability_(sports)#Formal_proposal:_Olympic_athletes to restrict presumptive notability to Olympians who have won a medal, instead of the current guideline which grants notability to anyone who has ever competed in one. Ravenswing 13:19, 29 August 2021 (UTC)

"If a tree falls in a forest and no one is around to hear it, does it make a sound?"

I see all too often people confusing the significant of X with its notability which is based on the reception of X by human societies. I wonder if the tree-example would help editors catch the essense of notability guideline easier and remember it for a longer time. We would be saying notability depends on whether ppl have heard the sound, not on the falling tree(s). Cinadon36 06:55, 18 September 2021 (UTC)

In essence that is my take, notability is "does anyone give a damn about it?".Slatersteven (talk) 13:07, 18 September 2021 (UTC)

thats the shorter version! 😀Cinadon36 13:17, 18 September 2021 (UTC)

RfC notice for establishing Wikipedia:Notability (television) as policy

This is a notice that an RfC has been started requesting comment on if the draft of Wikipedia:Notability (television) should be implemented as policy and a WP:SNG. Comments are welcome at the discussion, here. - Favre1fan93 (talk) 16:35, 27 September 2021 (UTC)