Policy | Technical | Proposals | Idea lab | WMF | Miscellaneous |
- Table of contents
- First discussion
- End of page
- New post
The proposals section of the village pump is used to offer specific changes for discussion. Before submitting:
- Check to see whether your proposal is already described at Perennial proposals. You may also wish to search the FAQ.
- This page is for concrete, actionable proposals. Consider developing earlier-stage proposals at Village pump (idea lab).
- Proposed policy changes belong at Village pump (policy).
- Proposed speedy deletion criteria belong at Wikipedia talk:Criteria for speedy deletion.
- Proposed WikiProjects or task forces may be submitted at Wikipedia:WikiProject Council/Proposals.
- Proposed new wikis belong at meta:Proposals for new projects.
- Proposed new articles belong at Wikipedia:Requested articles.
- Discussions or proposals which warrant the attention or involvement of the Wikimedia Foundation belong at Wikipedia:Village pump (WMF).
- Software changes which have consensus should be filed at Phabricator.
Discussions are automatically archived after remaining inactive for nine days.
Mass creation of pages on fish species
I was told through a talk page message to come here and submit this as a proposal even though it has been an ongoing process for a few months. I have been creating large numbers of pages on fish species and most of them are very short due to the relative scarcity of information on many of these species. I was informed that what I have been doing is considered "bot-like editing", so I am posting this here since apparently it needs to be discussed. I should note that I started this process back in December and this is the first time I have been notified that it is bot-like in nature, so this is the first time I am posting about it here or in any similar location.
I apologize if I am not carrying out this proposal process correctly as I am rather new here and frankly still a bit confused about this whole thing. This is not really a proposal, but I was told to verify that there is a consensus for creating such pages before continuing to do so. Lumpsucker (talk) 16:30, 30 August 2022 (UTC)
- WikiProject Tree of Life and WikiProject Fishes have been informed of this discussion. BilledMammal (talk) 02:55, 31 August 2022 (UTC)
- Thank you for bringing this here. To provide some context, I noticed that Lumpsucher had been engaged in the mass creation of articles on fish, and requested that they seek consensus for this, here or at another suitable location, as mass creation requires an affirmative consensus per WP:MASSCREATE and to avoid creating WP:FAIT issues.
- Of these articles, the majority are small, near-identical articles (eg, Hypostomus brevis, Hypostomus argus, Hypostomus johnii, Hypostomus agna) that I believe are better suited as entries in a list than as stand alone articles. BilledMammal (talk) 16:41, 30 August 2022 (UTC)
- I said in a recent discussion about dam articles that instead of stand-alone pages, it's better to have lists of dams. Same here. Pages like Hypostomus brevis, Hypostomus argus, Hypostomus johnii, Hypostomus agna would be better off if they were either listings in a table, or sections, of Hypostomus. One of the reasons is context: Hypostomus brevis doesn't tell me that this particular species of catfish is one of 100+ species of Hypostomus. It says brevis is 7.4 inches but I don't know if that's big or small for a Hypostomus species. It tells me that brevis is in the Parana River but it doesn't tell me if all these catfish in the Parana, or just rivers, or just South America, or what. Hypostomus gives me way, way more information -- even about the brevis specis -- than Hypostomus brevis. That's why it's better to combine these one- or two-sentence stubs about list entries into one list. Levivich 16:56, 30 August 2022 (UTC)
- Adding: for me it's not about WP:GNG--named fish species will likely meet GNG--it's about WP:PAGEDECIDE: even if the topics are notable, the brevity of the article content means "listification" will provide the reader with more information and context than stand-alone pages. Levivich 02:13, 31 August 2022 (UTC)
- Support creation of articles on fish species. There is consensus that they are notable and short articles on species have been accepted for a long time. BeanieFan11 (talk) 17:01, 30 August 2022 (UTC)
- This is why I have been doing it this way and not through lists. These sorts of articles have been the convention for quite some time as far as I know and I have yet to see a genus article where information on the species is presented in the form of a list. Lumpsucker (talk) 17:19, 30 August 2022 (UTC)
- Agree completely with Levivich. I think we're all aware of WP:SPECIESOUTCOMES here, which suggests species-level articles are typically kept. But to the question of what should we do: I think tiny stub articles serve our encyclopedic purpose poorly (per Levivich's points above). A more helpful way to include this information would be to host it at whatever taxonomic level we can support a half-decent well-contextualized encyclopedia article. In some cases that may be the genus; in others an even-higher order taxon. It just depends on how well-covered the organisms are in sources. Some flexibility may be required, as we're aiming to create a high-quality encyclopedia here, not a rigid rule set for where to host each tidbit of information. Ajpolino (talk) 17:12, 30 August 2022 (UTC)
- Oppose mass creation of stub articles, on fish species or otherwise. Agree with Levivich and Ajpolino (especially the "quality over quantity" point) and would add that the guiding principle should be this: database entries alone are insufficient to support an encyclopedic article, whether that database is of actors, fish species, geographic features or some other group. Remember that the notability guideline requires 'significant coverage," and databases do not provide this. I suggest that at a minimum the creator add as a reference the scientific paper in which the species was described (and maybe even some of the information from said paper). UnitedStatesian (talk) 17:29, 30 August 2022 (UTC)
- Unfortunately, a lot of these species description articles are paywalled. Quetzal1964 (talk) 12:21, 15 October 2022 (UTC)
- Support article creation We have precedent for creating species articles and guidelines confirming the appropriateness of this. That said, we could revisit the subject with a broader discussion. If anyone wants to start that discussion, they are going to have to do some background work. Things that I would expect to see are 1) list of previous discussions on this topic and 2) some summary of the previous discussions. I estimate this to be 1-2 hours of preparation for a good new discussion. Even with a new discussion though, I personally would not expect the community to reverse the precedent, because new technology is making information, data, and media more accessible at scale. I acknowledge that stub articles are problematic if they were to persist indefinitely without development, but my feeling is that we can reasonably expect good development of these articles soon. It has been some years since the project was active, but previously I contributed to the commons:User:Open Access Media Importer Bot which, among other things, pulled photos of species from open access scientific journals for upload to Wikimedia Commons. Some of those images are at commons:Commons:Open Access File of the Day. Technology for doing such activity has gotten much less expensive in the 10 years since that import project was running, so I expect something similar to happen eventually. Because of this, I do not see stubby Wikipedia articles as a problem, because I see them getting more content within years, and translated into other Wikipedia language versions not long after that. Bluerasberry (talk) 17:34, 30 August 2022 (UTC)
- Oppose The community has become wary of mass stub creation ( as these seem to be) even if in the past such stubs were acceptable. a far better solution would be to expand the...famiky? pages (the level just above species. then you can have searchable redirects to these. when a non stub species article can be created, just add the link then. User:Masem 18:25, 30 August 2022 (UTC)
- Genus, King Philip Could Only Find Girl Scouts Andre🚐 20:09, 30 August 2022 (UTC)
- Oppose. I have been guilty of this exact thing in the past (cfe. Synapturanus ajuricaba) but I believe that we should stop the mass creation of stubs in the future. If a genus has no associated species articles, then try covering all the species in sections on the genus article and split out the most substantial sections into their own article. casualdejekyll 18:38, 30 August 2022 (UTC)
- I see nothing wrong with this user's creations. I wouldn't even call it "mass creation"; the user is averaging two articles per day (since the first day the user created an article) on subjects that are all notable (and articles like these have had consensus for a long time). BeanieFan11 (talk) 18:41, 30 August 2022 (UTC)
- The problem isn't as much the quantity of the articles as the quality of the articles. They're stubs. Sourced stubs, sure, but if they weren't on such an obviously notable subject they don't have the sourcing to show the notability. casualdejekyll 18:50, 30 August 2022 (UTC)
- An article being a stub is not a problem. BeanieFan11 (talk) 18:52, 30 August 2022 (UTC)
- An article being a stub that does not pass GNG is. As UnitedStatesian said, database entries alone are insufficient to support an encyclopedic article. casualdejekyll 18:53, 30 August 2022 (UTC)
- Not all subjects have to pass GNG. NPOL, NPROF (I believe), and NSPECIES only require that subjects (that meet the criteria) need to be verifiable, which is the case with the articles this user is creating. BeanieFan11 (talk) 18:59, 30 August 2022 (UTC)
- only NPROF does not require any long term requirement to show GNG type presumed notability. NSPECIES is a general outcome and NOT an SNG Masem (t) 19:15, 30 August 2022 (UTC)
- That's kind of the issue at hand here I think. If they don't pass GNG they should be merged into a list. Andre🚐 20:01, 30 August 2022 (UTC)
- Not all subjects have to pass GNG. NPOL, NPROF (I believe), and NSPECIES only require that subjects (that meet the criteria) need to be verifiable, which is the case with the articles this user is creating. BeanieFan11 (talk) 18:59, 30 August 2022 (UTC)
- An article being a stub that does not pass GNG is. As UnitedStatesian said, database entries alone are insufficient to support an encyclopedic article. casualdejekyll 18:53, 30 August 2022 (UTC)
- An article being a stub is not a problem. BeanieFan11 (talk) 18:52, 30 August 2022 (UTC)
- The problem isn't as much the quantity of the articles as the quality of the articles. They're stubs. Sourced stubs, sure, but if they weren't on such an obviously notable subject they don't have the sourcing to show the notability. casualdejekyll 18:50, 30 August 2022 (UTC)
- Question: how does this fit into Wikispecies? Looking at a typical example Hypostomus multidens vs Hypostomus multidens. Wikispecies has about 800,000 pages which is more than 10% of Enwiki, in case the idea is to allow every page Wikispecies have a twin at Enwiki per WP:SPECIESOUTCOMES. -- GreenC 18:54, 30 August 2022 (UTC)
- Great point. Maybe they should go there. Andre🚐 19:59, 30 August 2022 (UTC)
- Wikispecies is, from what I've heard on WP:DISCORD, widely considered to be a ghost town and redundant to Wikidata. To use an analogy to another group of related things where there are clear winners and losers, Wikispecies is the Michigan's Adventure to Wikipedia's Cedar Point. casualdejekyll 23:05, 30 August 2022 (UTC)
- Oppose If there is only a line or two to say, they are better as a section or in a list at the genus level. Johnbod (talk) 18:56, 30 August 2022 (UTC)
- I agree that WP should cover these species… BUT… that does not mean we should have separate stand alone articles for each of them. If all we can say is one or two lines (effectively noting that they exist and not much else) then we should cover them in section or list within a larger, genus level, “parent” article. Blueboar (talk) 19:06, 30 August 2022 (UTC)
- I don't think an RFC has been started, but I am of the opinion that we do not need sub-stubs for subjects for which GNG has not been met. I think it is best to incorporate what information can be reliably sourced about those subjects into higher-level articles. - Donald Albury 19:52, 30 August 2022 (UTC)
- I concur with Donald Albury. I don't know a lot about ichthyology, but when it comes to plants, there are a ton of variations. So we don't need every article on a ficus. I count about 10 red links on the ficus page. Maybe all the ficuses (fici?) will eventually be written about, but possibly not, and that's totally OK. We should only write about species that have significant coverage in the published reliable work. Andre🚐 19:59, 30 August 2022 (UTC)
- Oppose, as said by multiple people already, these are better covered in a single article. Agree with Blueboar, the "topic" is still covered in the encyclopedia, just not with a separate article. The same goes for many other things in addition to minor species. MB 20:03, 30 August 2022 (UTC)
- WP:PAGEDECIDE leans against creating stubs when a broader article would serve readers better. Slywriter (talk) 20:36, 30 August 2022 (UTC)
- Comment: I think User:Casualdejekyll outlines exactly the right approach. Start by creating larger articles with sections, then divide out the sections as sufficient content develops. BD2412 T 21:35, 30 August 2022 (UTC)
- Support this creation of stub articles for species. This user is not using an automated bot, nor a creation rate to be considered as "mass creation". Using biological databases as references is valuable as an indicator of peer-reviewed secondary sources of the species notability - not like a sports almanac of trivial facts. Now, do I think a stub should be created for every species? Probably not, it would be impossible for a start. Many genus level articles could more readily summarize all the information for their species than a series of stubs. But for a topic such as fish, with many active editors, it is not inconceivable that someone else may come along and add an image or other information that will grow the article, although maybe not for 5-10 years. In the long-term approach (that is measured in decades, not months), species stubs are an asset and not something to be dissuaded from creating.
- The present creation of species stubs is well within established consensus and I would expect any serious proposal to attempt to reverse this be presented with both the proper background preparation (as detailed by Blueraspberry above) and a wide prior notification within the WP:TOL (Tree of Life) projects. Loopy30 (talk) 22:50, 30 August 2022 (UTC)
- Mass creation doesn't need to be at bot speed to be a problem. Masem (t) 23:11, 30 August 2022 (UTC)
- We would expect pre discussion of mass creation to be at a community wide location like WP:VPP. Wikiprojects are in general too insular for mass creation (though are a good first check). Masem (t) 23:25, 30 August 2022 (UTC)
The present creation of species stubs is well within established consensus
- could you link that consensus for me? As far as I am aware, no such consensus exists. BilledMammal (talk) 23:29, 30 August 2022 (UTC)- Listed in WP:OUTCOMES for a good decade now (see shortcut WP:NSPECIES); and I went through the entirety of the past 6.5 years of AfDs listed at Organisms deletion sorting archive and cannot find so much as a single AfD that involved an actual, valid, non-fictional species being closed as delete (individual plants/animals, plant cultivars, cannabis strains, breeds of domesticated animals, "dubious" and invalid taxa, fictional species/cryptozoology stuff, etc.? Sure. Actual valid species? Nope. Not as much as one of them in the past 6.5 years) or even no consensus.
- Of course it's possible that if a discussion happens, it turns out to show that consensus has, in fact, changed--but as things stand, there absolutely is plenty of evidence of consensus for the keeping-by-default of valid species articles. From that, it logically follows there is consensus for creation of valid species articles, too. AddWittyNameHere 01:31, 31 August 2022 (UTC)
- using OUTCOMES to support mass creation is absolutely the wrong approach (its circular reasoning) Masem (t) 01:40, 31 August 2022 (UTC)
- No, circular reasoning would be if I stated "this sums up that such articles are generally kept; therefore we shouldn't delete this specific example of it". What I'm saying is "this sums up that such articles are generally kept (and evidence shows that summary to remain correct even when it comes to the current state of things); from this follows that there is at utter minimum an implied consensus for the existence, and therefore creation, of such articles."
- I don't disagree it could be a good idea to confirm whether that implied consensus is, indeed, still the wider en.wiki consensus. But it's not circular reasoning to recognize that, up until this discussion where the implied consensus was challenged, such consensus was effectively in place. AddWittyNameHere 01:54, 31 August 2022 (UTC)
- I believe it is very unlikely that a proposal to create a guideline granting presumed notability (and an exemption from WP:NOTDATABASE) to species would find a consensus; given this probable inability to get a formal consensus, I cannot see an implied consensus.
- I would also note that we're not discussing individual article creations; we are discussing mass creations. Actions taken at scale can be problematic and against consensus even when smaller number of actions would not be - for example, creating stubs on five WP:GEOLAND-compliant locations is not an issue, but creating stubs on 5000 is. BilledMammal (talk) 02:11, 31 August 2022 (UTC)
we are discussing mass creations
- Are we, though? I'd argue that 2-3 per day (as seems to be pretty much the average in this case) fits perfectly intoAlternatives to simply creating mass quantities of content pages include creating the pages in small batches
right from WP:MASSCREATION. AddWittyNameHere 02:22, 31 August 2022 (UTC)- Yes; the relevant line is
While no specific definition of "large-scale" was decided, a suggestion of "anything more than 25 or 50" was not opposed.
- Lumpsucker has created hundreds of near-identical articles on fish. BilledMammal (talk) 02:32, 31 August 2022 (UTC)- The problem with that line is that it is so imprecise as to be entirely useless: it gives absolutely no strictures other than a vague indicator of number. It doesn't say they have to be stubs. It doesn't say they have to be on similar subjects. It doesn't specify in what time-span.
- If we want to apply that line as-is, practically every content creator who has been around a while is a mass-creator by default, regardless of the quality, subject or frequency of their articles. We have over 10000 editors with at least 68 non-redirect mainspace page creations to their name. I highly doubt the intention was that they'd all be deemed mass creators who'd need prior approval before creating more articles. AddWittyNameHere 01:50, 1 September 2022 (UTC)
- For what it's worth, the just-opened RfC on mass-creation currently says
"Rapid and/or mass creation/deletion" refers to more than 25 per day
, with a note the definition may need workshopping during the RfC. By that definition, Lumpsucker very much is not engaging in mass-creation. AddWittyNameHere 01:56, 1 September 2022 (UTC)
- For what it's worth, the just-opened RfC on mass-creation currently says
- Yes; the relevant line is
- WP:OUTCOMES
is not one of Wikipedia's policies or guidelines, as it has not been thoroughly vetted by the community.
It's not an WP:SNG (which are guidelines). Levivich 02:03, 31 August 2022 (UTC)
- using OUTCOMES to support mass creation is absolutely the wrong approach (its circular reasoning) Masem (t) 01:40, 31 August 2022 (UTC)
- Support the "mass" creation of species stubs with the proviso that at least one non-database RS is used as a reference. As an example, I found a source for Hypostomus agna and added some tidbits of info, such that now the species article contains information beyond what is given in the genus article. The substub->stub conversion took less than 5 minutes. I've written hundreds of stubs on lichen species and intend to continue doing so until there's one for each of the 20,000 or so lichens. Esculenta (talk) 01:12, 31 August 2022 (UTC)
- Question Can anybody point me to some previous examples where editors have requested consensus/permission to mass create articles in a semi-automated way? I realize that is what WP:MASSCREATE calls for, but it's not something I'm aware of being followed in practice.Plantdrew (talk) 01:24, 31 August 2022 (UTC)
- @Plantdrew: The essay Wikipedia:Bot-created articles has links to relevant history here. Particularly for plants and animals, you'll typically find that Lsjbot has mass created stubs in other languages already for many species. In practice, I think the answer is that automated creation of articles is more or less not done on English Wikipedia anymore? Steven Walling • talk 01:49, 31 August 2022 (UTC)
- @Plantdrew: Wikipedia:Bots/Noticeboard#Dams articles Levivich 01:58, 31 August 2022 (UTC)
- That was the one and only time that was done (at least that I can recall). BeanieFan11 (talk) 02:04, 31 August 2022 (UTC)
- I found some examples in the BRFA archives (1, 2, 3, 4, 5, 6). –xenotalk 02:12, 31 August 2022 (UTC)
- Every one of those editors, like Lumpsucker, deserve our thanks for seeking community input on this. Levivich 02:44, 31 August 2022 (UTC)
- Xeno, thank you for digging up those examples. Those are all seeking permission for fully automated article creation. I assume Xeno would have discovered any requests for approval for semi-automated article creation, and by absence of examples, no such requests exist (aside from the recent case for Japanese dams). Plantdrew (talk) 15:32, 31 August 2022 (UTC)
- Because WP:MASSCREATE says
Any large-scale automated or semi-automated content page creation task must be approved at Wikipedia:Bots/Requests for approval
, I don't think the automated/semi-automated distinction is relevant. Levivich 17:58, 1 September 2022 (UTC)
- Because WP:MASSCREATE says
- Xeno, thank you for digging up those examples. Those are all seeking permission for fully automated article creation. I assume Xeno would have discovered any requests for approval for semi-automated article creation, and by absence of examples, no such requests exist (aside from the recent case for Japanese dams). Plantdrew (talk) 15:32, 31 August 2022 (UTC)
- Every one of those editors, like Lumpsucker, deserve our thanks for seeking community input on this. Levivich 02:44, 31 August 2022 (UTC)
- Support Animals and plants that have references to a species description are notable because there is plenty of highly reliable source material about them, even if the initial stubs are usually small. To get an accepted species description, a scientist typically has to go through peer review and have the description presented in an academic journal. An accepted species description is arguably even more reliable than say, census data used to populate stubs about tiny towns or census-designated places, and it's really easy to create one or two of these stubs manually every day once you know where to find the free online science databases and/or get a few books handy. In addition to the original species description, most of these are also covered in other reliable sources like monographs about a particular genus. Most of these scientific papers, even for rare and endangered species, are easily findable via Google Scholar or JSTOR if you search for the scientific names. TL;DR: Since there are almost always multiple highly reliable sources exclusively covering the subject these species pass the basic test of notability. Steven Walling • talk 02:08, 31 August 2022 (UTC)
- I think this argument is persuasive to a point. However, some old genus and species names have been around for a while and usually there isn't much information about them. Right? I mean could you break down a couple of really obscure fish for us? Or how about those red-linked ficus? Andre🚐 02:16, 31 August 2022 (UTC)
- Ficus is one of the largest genera of plants on the planet, and there are quite a large number of scientific papers and whole books written just about various species within it. We could most definitely create reliably sourced articles about almost all of them. Just like with all subjects though, if we can't find enough source material... then don't. Each article has to be considered on its own merits, but to propose that we categorically can't have stubs about individual plant and animal species is just plain wrong when you look at WP:N and more than a decade of discussion at AFD. Steven Walling • talk 03:20, 31 August 2022 (UTC)
- I didn't say we can't have stubs, but many of these are substubs. An article needs to say why it's important and why we should care, IMHO. Andre🚐 03:34, 31 August 2022 (UTC)
- Why? I'm not convinced an article needs to say why it's important. Ortizesp (talk) 16:09, 5 October 2022 (UTC)
- It does. Andre🚐 16:19, 5 October 2022 (UTC)
- @Andrevan, that only applies to people, individual animals (like racehorses), organizations that aren't schools, web content (e.g., specific websites such as YouTube), organized events (such as the Olympics), and musical recordings (such as The Beatles (album)). To quote the policy, "it does not apply to articles about...products, books, films, TV programs, software, or other creative works, nor to entire species of animals" (emphasis in the original). I'd add that it doesn't apply to geographical locations, either. WhatamIdoing (talk) 23:00, 5 October 2022 (UTC)
- Yes, that's true, and thank you for clarifying and pointing out that there is a specific carve-out for species and for the other examples mentioned, but not because those articles are exempt from the requirement of a credible claim of significance. It's because it's assumed that their notability requirements will cover that requirement. Because as we've discussed, to even be an independent species requires you to demonstrate significance by virtue of what a species is. Andre🚐 23:08, 5 October 2022 (UTC)
- The page you linked says "This is an explanatory essay about the several sections of Wikipedia:Criteria for speedy deletion." You seem to think it apples to every subject, and not just the ones that are relevant to "the several sections of Wikipedia:Criteria for speedy deletion".
- What makes you think that every page is required to have a claim of significance? You won't find a requirement for that in any policy. Subjects are notable (or not) no matter what state the article is in, including if there are zero sources and no claim of any significance whatsoever. As a pure matter of practicality, if you can explain why a subject matters and toss a few impressive-looking citations onto the page, then that will stop an editor who doesn't know anything about the subject from trying to delete the article, but the subject's notable (or not) no matter what's on the page. WhatamIdoing (talk) 23:38, 5 October 2022 (UTC)
- Yes, the page I linked is an essay explaining it, and not a policy. However an article which falls under a speedy deletion criterion of not having such claim of significance, for the categories to which such policy applies, will be deleted. And indeed the spirit of notability of being "worthy of note" is essentially synonymous in my view with a "significance," and it's kind of at the core of notability. Per WP:SUSTAINED,
a topic is "notable" in Wikipedia terms only if the outside world has already "taken notice of it".
for asignificant period of time
. The test of notability is not notability itself: notability is whether the world has taken note due to something worth noting. Why did you note it? That's what significance is. The intrinsic characteristic we're trying to test with tests of notability. Andre🚐 23:55, 5 October 2022 (UTC)- There is nothing in WP:N or any other guideline or policy that says an article about (e.g.) a book needs to say anything about its significance. Notability is independent of article contents.
- Notable book + article makes a credible claim of significance = notable
- Notable book + article makes no claim of significance = still notable
- The presence or absence of a claim of significance in the article is not a requirement except for the specific subjects named in CSD. WhatamIdoing (talk) 19:17, 6 October 2022 (UTC)
- Well, while that requirement doesn't exist explicitly, we do have a requirement of WP:SIGCOV, ie not a passing mention. Sources will end up having to explain the subject in some amount of depth. The rules are principles and the principle of notability, in my mind, is tied to the subject's significance, not only by the CSD policy. Notability is a property of the subject, not the article - but the subject still needs to have significance demonstrated by virtue of significant coverage in the material. Andre🚐 19:59, 6 October 2022 (UTC)
- I think you might be missing my main complaint. You wrote: An article needs to say why it's important. Notice which word I bolded.
- Now you are here telling me that the sources need to talk about the subject in some amount of depth. Sure, that's fine, but the sources (which need to amount to significant coverage) are not the article (which is what your original statement was about). The article does not need to say why the subject is important. WhatamIdoing (talk) 04:10, 7 October 2022 (UTC)
- Well, while that requirement doesn't exist explicitly, we do have a requirement of WP:SIGCOV, ie not a passing mention. Sources will end up having to explain the subject in some amount of depth. The rules are principles and the principle of notability, in my mind, is tied to the subject's significance, not only by the CSD policy. Notability is a property of the subject, not the article - but the subject still needs to have significance demonstrated by virtue of significant coverage in the material. Andre🚐 19:59, 6 October 2022 (UTC)
- There is nothing in WP:N or any other guideline or policy that says an article about (e.g.) a book needs to say anything about its significance. Notability is independent of article contents.
- Yes, the page I linked is an essay explaining it, and not a policy. However an article which falls under a speedy deletion criterion of not having such claim of significance, for the categories to which such policy applies, will be deleted. And indeed the spirit of notability of being "worthy of note" is essentially synonymous in my view with a "significance," and it's kind of at the core of notability. Per WP:SUSTAINED,
- Yes, that's true, and thank you for clarifying and pointing out that there is a specific carve-out for species and for the other examples mentioned, but not because those articles are exempt from the requirement of a credible claim of significance. It's because it's assumed that their notability requirements will cover that requirement. Because as we've discussed, to even be an independent species requires you to demonstrate significance by virtue of what a species is. Andre🚐 23:08, 5 October 2022 (UTC)
- @Andrevan, that only applies to people, individual animals (like racehorses), organizations that aren't schools, web content (e.g., specific websites such as YouTube), organized events (such as the Olympics), and musical recordings (such as The Beatles (album)). To quote the policy, "it does not apply to articles about...products, books, films, TV programs, software, or other creative works, nor to entire species of animals" (emphasis in the original). I'd add that it doesn't apply to geographical locations, either. WhatamIdoing (talk) 23:00, 5 October 2022 (UTC)
- It does. Andre🚐 16:19, 5 October 2022 (UTC)
- Why? I'm not convinced an article needs to say why it's important. Ortizesp (talk) 16:09, 5 October 2022 (UTC)
- I didn't say we can't have stubs, but many of these are substubs. An article needs to say why it's important and why we should care, IMHO. Andre🚐 03:34, 31 August 2022 (UTC)
- Ficus is one of the largest genera of plants on the planet, and there are quite a large number of scientific papers and whole books written just about various species within it. We could most definitely create reliably sourced articles about almost all of them. Just like with all subjects though, if we can't find enough source material... then don't. Each article has to be considered on its own merits, but to propose that we categorically can't have stubs about individual plant and animal species is just plain wrong when you look at WP:N and more than a decade of discussion at AFD. Steven Walling • talk 03:20, 31 August 2022 (UTC)
- I just did a search on JSTOR and Google Scholar for the first fish I mentioned, Hypostomus brevis. There were some passing mentions, but no WP:SIGCOV. In addition, WP:N isn't the only reason these articles are problematic; WP:NOTDATABASE also applies.
- We also need to consider the interests of the reader. Even for the ones that are notable, readers are better served by the inclusion of content at Hypostomus than they are by a standalone micro-stub. BilledMammal (talk) 02:23, 31 August 2022 (UTC)
- A Google Books search shows several mentions of that species in books (significant coverage doesn't have to be in sources you can read for free online after a cursory search) and there are probably even more sources out there that you and I can't find because we're not familiar with the journals that cover Hypostomus or other fish. Just because an article is small right now does not mean it's not notable. That's a logical fallacy and if we followed that argument we wouldn't have Wikipedia as it exists today. As for improving the genus article... we can have our cake and eat it too. There's nothing stopping expansion of the genus article and having a stub. There are thousands of examples of this in the encyclopedia right now. Since most of our traffic comes from Google, you'd have to redirect and merge millions of species stubs to the genus articles for your theory to work. Steven Walling • talk 03:12, 31 August 2022 (UTC)
- In addition, for older species which were previously in other genera, many of the older resources will resultingly have used the at-the-time valid binomial, not the current one. Searching for just the current one is typically a good way to miss the bulk of older sources. (Though in this case, the majority of older sources are likely not going to be available online, anyway: digitization for old non-English sources lags a fair bit behind that of old English sources, and this particular fish species occurs solely outside the Anglosphere) AddWittyNameHere 03:30, 31 August 2022 (UTC)
- A Google Books search shows several mentions of that species in books (significant coverage doesn't have to be in sources you can read for free online after a cursory search) and there are probably even more sources out there that you and I can't find because we're not familiar with the journals that cover Hypostomus or other fish. Just because an article is small right now does not mean it's not notable. That's a logical fallacy and if we followed that argument we wouldn't have Wikipedia as it exists today. As for improving the genus article... we can have our cake and eat it too. There's nothing stopping expansion of the genus article and having a stub. There are thousands of examples of this in the encyclopedia right now. Since most of our traffic comes from Google, you'd have to redirect and merge millions of species stubs to the genus articles for your theory to work. Steven Walling • talk 03:12, 31 August 2022 (UTC)
- I also did a search on Google Scholar for Hypostomus brevis. Looks like GNG is met pretty easily. There are additional sources available (Monographs about fish in Brazil, in Portuguese). Esculenta (talk) 05:40, 31 August 2022 (UTC)
- Two of those sources are clearly trivial coverage; the third, the Portuguese language source, might count towards GNG, but doesn't meet it on its own.
- However, that is getting off topic; the issue is that sources like these aren't being included when these articles are created. Instead, they are being mass created by duplicating database entries; we should instead be encouraging the editor to add this information to the genus article where it will be of more benefit to the reader than in a micro-stub - of course, if editors are able to find sources that permit them to create an actual article on the subject that would be of more benefit to readers than the genus article, then they are encouraged to do so. BilledMammal (talk) 05:56, 31 August 2022 (UTC)
- I also did a search on Google Scholar for Hypostomus brevis. Looks like GNG is met pretty easily. There are additional sources available (Monographs about fish in Brazil, in Portuguese). Esculenta (talk) 05:40, 31 August 2022 (UTC)
- Comment. What are we actually voting on here? The mass creation of articles on fish by Lumpsucker? If so, what is the scope; how many articles will be created, what genus of fish will be covered, etc? I suggest everyone hold off from voting until there is a formal proposal to vote on. BilledMammal (talk) 02:23, 31 August 2022 (UTC)
- We're not voting. This is a discussion. Andre🚐 02:28, 31 August 2022 (UTC)
- Sorry, meant !voting, in reference to the various bolded "support" and "oppose" !votes. BilledMammal (talk) 02:42, 31 August 2022 (UTC)
- On pages like this which have proposals on them, people start doing that sometimes. That's why in the old days people used to say voting is evil. And the reason why we put the ! (which means a logical not) in front a vote is to remind everyone that WP:NOTDEMOCRACY. In theory, though I guess it's increasingly less common in practice, the best arguments should prevail in a rough consensus (not a tally). All of that to say, nobody formally proposed anything and nothing's being voted on, that's the answer to your question. Andre🚐 02:45, 31 August 2022 (UTC)
- Sorry, meant !voting, in reference to the various bolded "support" and "oppose" !votes. BilledMammal (talk) 02:42, 31 August 2022 (UTC)
- I should probably clear something up. There is no vote being called here. This discussion here would not exist were it not for the fact that BilledMammal specifically recommended I start it here, as this topic is inherently not a formal proposal. Lumpsucker (talk) 02:59, 31 August 2022 (UTC)
- It might be a good idea if you made a formal proposal and we can open an RfC on the basis of it. In particular, what is the scope of your proposed mass creation; can you give us a rough idea of what articles you will be creating? (ie, about 100 articles on species within the genus x, about 50 articles on species within the genus y, etc) BilledMammal (talk) 03:03, 31 August 2022 (UTC)
- We're not voting. This is a discussion. Andre🚐 02:28, 31 August 2022 (UTC)
- I Support the creation of animals and plants that are able to be referenced to a species description. Less work for someone else to expand in the future. If a creator has a high error rate of content that is a separate issue regardless of number of articles created. Limiting stub creation makes it very complicated - what is the cap. How big does an article need to be before editor is allowed to make another etc. Massive headache and bureaucratic creep. Cas Liber (talk · contribs) 02:52, 31 August 2022 (UTC)
- Here's my suggestion: just stop creating new articles unless you can add 1-2 sentences explaining why they are notable and significant. Andre🚐 03:22, 31 August 2022 (UTC)
- creating articles of 1 or 2 sentences plus an infobox is exactly what is being argued against by BilledMammal, and running around wikipedia disrupting good editors. Gnangarra 04:06, 31 August 2022 (UTC)
- But I looked at the articles. Some, indeed, are ok. Hypostomus commersoni looks OK, it even has a pic, and it appears in the aquarium trade. It's not the most lengthy or interesting article but it seems to barely qualify. Hypostomus carvalhoi, Hypostomus chrysostiktos, Hypostomus crassicauda, Hypostomus simios Hypostomus corantijni, Hypostomus brevicauda. These don't assert notability or explain significance. It just says they're a fish, in a certain river basin, they grow to this size, and breathe air. The end. The links are to databases only. I don't see why those articles couldn't all be in the Hypostomus article. Andre🚐 04:15, 31 August 2022 (UTC)
- Stating its a species of fish, plant, insect, animal. fungi or is establishing notability and data bases are are the where you get the official taxonomy from along with common names, or synonyms and previous naming. Gnangarra 06:02, 31 August 2022 (UTC)
- That is not how notability works on WP. There is no special presumption of notability just because you can define its species. we need significant coverage, which 1or 2 sentences do not do. Masem (t) 07:23, 31 August 2022 (UTC)
- That is the very definition of notable for species, and has been for 20 years. If you want to overturn notability of species then this isnt the place Gnangarra 09:00, 31 August 2022 (UTC)
- "This" meaning this specific thread, probably not. But if the Village Pump isn't the right place, then I don't know what is. casualdejekyll 16:29, 31 August 2022 (UTC)
- That is the very definition of notable for species, and has been for 20 years. If you want to overturn notability of species then this isnt the place Gnangarra 09:00, 31 August 2022 (UTC)
- That is not how notability works on WP. There is no special presumption of notability just because you can define its species. we need significant coverage, which 1or 2 sentences do not do. Masem (t) 07:23, 31 August 2022 (UTC)
- Stating its a species of fish, plant, insect, animal. fungi or is establishing notability and data bases are are the where you get the official taxonomy from along with common names, or synonyms and previous naming. Gnangarra 06:02, 31 August 2022 (UTC)
- But I looked at the articles. Some, indeed, are ok. Hypostomus commersoni looks OK, it even has a pic, and it appears in the aquarium trade. It's not the most lengthy or interesting article but it seems to barely qualify. Hypostomus carvalhoi, Hypostomus chrysostiktos, Hypostomus crassicauda, Hypostomus simios Hypostomus corantijni, Hypostomus brevicauda. These don't assert notability or explain significance. It just says they're a fish, in a certain river basin, they grow to this size, and breathe air. The end. The links are to databases only. I don't see why those articles couldn't all be in the Hypostomus article. Andre🚐 04:15, 31 August 2022 (UTC)
- creating articles of 1 or 2 sentences plus an infobox is exactly what is being argued against by BilledMammal, and running around wikipedia disrupting good editors. Gnangarra 04:06, 31 August 2022 (UTC)
- the practice of creating stubs about any Notable topic has been standard since the very first edit occurred on Wikipedia 20 years ago. On any given day I photograph 10 species within 100 miles of here that dont have an article every single one of the are notable. Its dam sight easier to add a photo, and additional information when the article has already been started. As someone who has spent many years doing outreach at various Universities having these articles already created also makes it easier to teach others how to edit Wikipedia as they cant create articles in the first few days of creating an account, standard practice has always been to create stubs in areas where the students will be writing. You'll also note that these articles are one WMF KPI's for events if you look over at Outreach dashboards. Gnangarra 04:06, 31 August 2022 (UTC)
- I think the idea that instead of stubs, species should be listed in list articles makes perfect sense. That species are notable by default does not automatically imply that a standalone stub is the best way to present information on the species. If a given species' section expands too much, then we can split off a dedicated article for it. Jo-Jo Eumerus (talk) 10:09, 31 August 2022 (UTC)
- Support the continued creation of stub articles for the reasons given above by Loopy and others. I don't see how the 2-3 articles a day by Lumpsucker counts as automated or semi-automated, as each will be separately evualated when created. If the creation was automated using a bot I would want guidelines and limits, but this isn't the case. Stubs are useful as they make it easy for people to extend the articles, when article creation with appropriate taxoboxes, taxonbars and categories is more of a barrier. — Jts1882 | talk 12:03, 31 August 2022 (UTC)
If the creation was automated using a bot I would want guidelines and limits
Idk whether it could actually be automated (it sounds as if it might be possible to some extent) but assuming it could be, what guidelines and limits would you suggest? Selfstudier (talk) 12:42, 31 August 2022 (UTC)
- Comment. From the guidelines at the page header above: "
This page is for concrete, actionable proposals.
" - I would posit that the discussion has now strayed greatly from that requirement and has instead become merely discourse for any prejudices different editors wish to air (dislike of stubs, mis-characterization of non-automated user edits as "bot-like", perceived excessive article creation rates of 2/day, validity of species notability guidelines, or their preferred layout of genus/species/list articles). If there is actually any problem that can be identified here (ie. disruptive editing, creation of non-notable articles, incorrect facts included in the article text, etc) then that can be addressed as a separate issue. It is clear that WP:MASSCREATE does not apply in this situation, in fact the low-rate production of human-reviewed articles is specifically mentioned as a legitimate alternative to semi-automated or automated edits.
- The ongoing non-automated creation of new species articles by various editors has been the status quo here for twenty years and there is no need to pressure another editor for them to now ask permission to be able to create such articles. If you feel that this type of activity is detrimental to the encyclopedia, then the onus would be on you to gather the necessary information yourself (again, as recommended by Blueraspberry above) and propose a "concrete, actionable proposal" to overturn the status quo. Loopy30 (talk) 14:13, 31 August 2022 (UTC)
- The “Status quo” is that most articles on species are kept… but that does not mean that ALL articles on species should be kept. Most species will have enough sourcing to justify an article… but there will be exceptions. Those exceptions can be covered in larger articles. Blueboar (talk) 14:27, 31 August 2022 (UTC)
- It has not be proven that anywhere near the majority of recently created species articles are themselves lacking sources or not notable. At best its come to perhaps 1 or 2 individual articles which could be new species, synonyms of depreciated binomials, or poorly documented though even then these articles havent been proven as not meeting GNG. The sole complaint is that the volume of creating 2 or 3 articles on species a day is too much. Some editors are very good at replicating what is a whole of project agreed MOS for such articles and implementing that consistency. What we then have is the bonus that all species articles start with universally consistent format for other contributors to follow and integrates these with Wikimedia Commons, Wikidata, and Wikispecies while facilitating Outreach activities to bring new editors to Wikipedia. Gnangarra 14:51, 31 August 2022 (UTC)
- The “Status quo” is that most articles on species are kept… but that does not mean that ALL articles on species should be kept. Most species will have enough sourcing to justify an article… but there will be exceptions. Those exceptions can be covered in larger articles. Blueboar (talk) 14:27, 31 August 2022 (UTC)
- Support Fish species articles have an astronomically low potential for abuse; they aren't likely to be spammy or sources of promotion, they aren't likely to contain controversial information, they aren't likely to misrepresent things, they aren't likely to cause WP:BLP issues, etc. etc. Given the low potential of abuse and the fact that they will only be created based on published, reliable sources, I say go for it. --Jayron32 15:03, 31 August 2022 (UTC)
- That is a good point. There isn't a lot of fish POV pushing that I know of. Though IMHO, I prefer my fish & chips to be haddock over cod or sole. Andre🚐 15:31, 31 August 2022 (UTC)
- it is not that these are going to be spam targets or the like. WP has a problem with creation of underrepresented topics like women's studies and other minrities, which still tend to struggle due to poor non primary sourcing. we continue to thumb our noses when we allow stubby articles on things like fish species and sports players but these attempts to expand in underrepresented areas are expected to have more fully fleshed out articles. its why the community has soured of mass article creation that leaves stubby article behind even if these could be notable topics. it's lazy work. Masem (t) 19:59, 31 August 2022 (UTC)
- The problem I've noticed is that we keep these pages around of an expectation that "eventually someone will improve them" but it never really happens except in rare cases. See also Wikipedia:Don't_hope_the_house_will_build_itself, Wikipedia:An unfinished house is a real problem. casualdejekyll 20:10, 31 August 2022 (UTC)
- exactly. if it takes a bit more work to write a full two paragragh stub that includes demonstration of meeting the GNG, that would be different, but these here are all two sentence stubs with no clear notability, nor any expectation they will be improved to meet that. Masem (t) 20:17, 31 August 2022 (UTC)
- The problem I've noticed is that we keep these pages around of an expectation that "eventually someone will improve them" but it never really happens except in rare cases. See also Wikipedia:Don't_hope_the_house_will_build_itself, Wikipedia:An unfinished house is a real problem. casualdejekyll 20:10, 31 August 2022 (UTC)
- Support creation of stubs for species. Of course it would be better if more substantial articles were being created. But it is absurd to suggest that an editor can create no more than 25-50 stubs in the course of their Wikipedia career (or needs to seek permission to do so). Plantdrew (talk) 16:54, 31 August 2022 (UTC)
- oppose creation of taxonomy-only stubs. If at least some description were put in for each one I would be supportive. Mangoe (talk) 19:50, 31 August 2022 (UTC)
- None of these are taxonomy-only. Lumpsucker (talk) 20:21, 31 August 2022 (UTC)
- Oppose. I while ago I did a lot of work Assessing hundreds (or thousands?) of fish articles. To do this I looked at their pageviews, and there are huge numbers of fish articles that receive about one pageview per month, and there were some that had one pageview in five years. These stubs are not helpful, and arguably non-encyclopedic. Abductive (reasoning) 20:57, 31 August 2022 (UTC)
- Support editor's preference to create articles. Species have long had implied notability, and there can be no WP:N objection to these stubs. The project is entirely manual and at the stated rate of creation displays neither bot nor bot-like activity; the fact that some articles are very similar speaks more to the similarity of the species involved than evidence of bot-like editing. Beyond that, nobody has invoked any policy or guideline prohibiting the work, which means it should be left to the editor's own discretion. Every editor is free to edit in whatever way they think fit, provided they do so in accordance with policy and guidelines. We shouldn't be attempting to create new rules on the hoof in this thread to force an obviously committed editor to comply with some other editors' "lists are best" preferences. Suggest, by all means; compel, no. MichaelMaggs (talk) 22:17, 31 August 2022 (UTC)
- This discussion is a waste of time. First off, creating 500 short articles in the space of 9 months isn't "mass creation" in any meaningful sense of the term. Leaving the creator what looks like a templated warning about bot [?!] editing and urging them to get permission at the village pump [??] was poor form. Still, when undertaking a big project that's going to affect a large number of articles (even if the project is slow-going), it's a good idea to first post about it in a relevant place, like WT:FISH. This is not so much asking for permission as soliciting suggestions (this can, for example, help identify the best taxonomy sources: you wouldn't want to spend months working on something only to later find out your work was based on outdated information). So, what are we trying to do here? Some sort of rehearsal for Wikipedia:Arbitration Committee/Requests for comment/Article creation at scale, or just an exercise in venting our frustrations at the low-quality mass stub creations of those high-profile recently banned editors in other topic areas? Rethinking how we present information about obscure species will certainly be worth it, and I second some of Levivich's ideas above; however, treating the species only in a list at the genus articles is not the done thing, and we can't expect people to suddenly have to start asking our opinion before doing things the same way they have been done in the past 20 years. Any talk of species articles needing to "demonstrate notability" is nonsense (as clear from reading Steven Walling's comment). Personally, I don't think it's a good idea to create stubs that very few people will ever read (obscure species are among the least read articles in the encyclopedia), and I wouldn't be spending my time on that, but it's up to individual editors to decide how they're going to spend their time, and they shouldn't be pushed into asking for approval for medium-scale manual creations. Uanfala (talk) 22:30, 31 August 2022 (UTC)
- After watching this discussion unfold I have to agree with you. I was under the impression that there was a logical precedent for sending editors here and describing their edits as "bot-like", but it seems as though that may not be as true as I thought it was. After reading this message I will probably continue editing in the way that I have been editing. If any editor here believes that there is a logical reason to modify an individual page of mine, they are free to do so. I encourage expanding and editing small articles of all sorts. Lumpsucker (talk) 23:44, 31 August 2022 (UTC)
- Probably best to wait for Wikipedia:Arbitration Committee/Requests for comment/Article creation at scale. Levivich 23:48, 31 August 2022 (UTC)
- I don't exactly know how all of this works, so I might take a month or so off from editing to see what happens there. Lumpsucker (talk) 00:03, 1 September 2022 (UTC)
- @Lumpsucker: Please don't be discouraged from creating articles. Anyone can start a discussion on the Village Pump, it doesn't mean you are required to do anything, unless there's been a consensus declared. One of Wikipedia's core principles is to be bold, so the last thing in the world is that you should think you have to wait for someone's permission to edit. Especially since there is no agreement via an articles for deletion discussion that any of these articles are not notable and should be removed, you're doing helpful work. My only suggestion would be to try find a second source up front for each stub, not just a taxonomy database. If you aren't comfortable proceeding alone, WikiProject Fishes or other editors like myself who are supportive of writing species articles would be happy to help review and expand stubs. Steven Walling • talk 19:52, 1 September 2022 (UTC)
- Steven is correct, you may continue creating articles, but you might find that if you create an article sourced only to fishbase and another generic database site, it will end up at AFD. Andre🚐 20:09, 1 September 2022 (UTC)
- And to just to be clear Lumpsucker, if you're not familiar with AFD, that's also just like this discussion. Any random editor can propose any article for deletion at any time, and then we discuss it as a community. It would be extremely unusual for a legitimate named species to actually be deleted. What will more likely happen is that someone will just help expand the sources and then it will be kept. So don't let Andre's threatening talk stop you. Steven Walling • talk 21:04, 1 September 2022 (UTC)
- Before creating any other articles, I am planning on expanding a few of my smallest articles as I already have been doing. Some of those were only created with a single source and are less than 1 KB in length, and I am looking to improve most of the articles of that size that I made. Lumpsucker (talk) 00:12, 2 September 2022 (UTC)
- Steven is correct, you may continue creating articles, but you might find that if you create an article sourced only to fishbase and another generic database site, it will end up at AFD. Andre🚐 20:09, 1 September 2022 (UTC)
- @Lumpsucker: Please don't be discouraged from creating articles. Anyone can start a discussion on the Village Pump, it doesn't mean you are required to do anything, unless there's been a consensus declared. One of Wikipedia's core principles is to be bold, so the last thing in the world is that you should think you have to wait for someone's permission to edit. Especially since there is no agreement via an articles for deletion discussion that any of these articles are not notable and should be removed, you're doing helpful work. My only suggestion would be to try find a second source up front for each stub, not just a taxonomy database. If you aren't comfortable proceeding alone, WikiProject Fishes or other editors like myself who are supportive of writing species articles would be happy to help review and expand stubs. Steven Walling • talk 19:52, 1 September 2022 (UTC)
- I don't exactly know how all of this works, so I might take a month or so off from editing to see what happens there. Lumpsucker (talk) 00:03, 1 September 2022 (UTC)
- Probably best to wait for Wikipedia:Arbitration Committee/Requests for comment/Article creation at scale. Levivich 23:48, 31 August 2022 (UTC)
- +1 to @Uanfala. WhatamIdoing (talk) 23:45, 31 August 2022 (UTC)
- After watching this discussion unfold I have to agree with you. I was under the impression that there was a logical precedent for sending editors here and describing their edits as "bot-like", but it seems as though that may not be as true as I thought it was. After reading this message I will probably continue editing in the way that I have been editing. If any editor here believes that there is a logical reason to modify an individual page of mine, they are free to do so. I encourage expanding and editing small articles of all sorts. Lumpsucker (talk) 23:44, 31 August 2022 (UTC)
Also, I wonder how many people have actually looked at the articles in question? I looked at this one, which I transclude here so that you won't even have to go to the trouble of clicking through to an article:
(I'd tell you that I picked the name randomly, but it'd be more accurate to say that I picked it arbitrarily, on the basis of its bird-like name.)
I see above people like Levivich talking about "one- or two-sentence stubs" and Blueboar saying "If all we can say is one or two lines", but this is four sentences plus an infobox. That's not what I'm seeing in these articles.
Hypostomus chrysostiktos has four sentences, an infobox, and three sources (one of which is a scholarly paper). Hypostomus ericae has four sentences, an infobox, and two sources. These are typical. Then there are the not-so-typical ones, like Pseudacanthicus pirarara, which has 500 words, four sources, and an ==In popular culture== section. Pseudancistrus megacephalus has 250 words and 11 sources. Chaetostoma platyrhynchus has 250 words and 7 sources. Pseudoqolus, with six sentences and four sources, is about a genus.
So I wonder: If you actually look at the articles, why would anyone want to discourage this editor? WhatamIdoing (talk) 23:46, 31 August 2022 (UTC)
- I removed the transcription, as I found it very confusing reading through the page only to encounter an article; I hope you don't mind.
- However, to be clear we don't want to discourage articles like Pseudacanthicus pirarara and Chaetostoma platyrhynchus; they aren't mass created, they aren't WP:NOTDATABASE violations, they demonstrate notability, and they are more useful to the reader than the higher level article would be.
- What we do want to discourage is articles like the ones I referenced above, and the ones like Hypostomus ericae, which are mass created on the basis of a template, and where the information would be more useful for the reader in a higher level article. BilledMammal (talk) 23:54, 31 August 2022 (UTC)
- Hypostomus ericae has five sentences, three sources, and an infobox. One of the sources is a scholarly journal article. Why would you want to discourage the creation of an article like that? WhatamIdoing (talk) 03:09, 1 September 2022 (UTC)
- This was the version at the time of my comment. Database-replications like that are not useful for readers; they are better off being sent to a parent article where that information can be included until an editor has time to write a more comprehensive article. BilledMammal (talk) 03:17, 1 September 2022 (UTC)
- Yes, it's longer now though. So it shouldn't so clearly be AFD'd or merged. So if every new stub started out this long, I bet there wouldn't be a dispute. Just don't create it if you can't make it this long. Andre🚐 03:20, 1 September 2022 (UTC)
I bet there wouldn't be a dispute
; you would be right. If that was what was happening, I wouldn't have asked them to come here and get approval for mass creation, because they wouldn't be engaged in mass creation. BilledMammal (talk) 03:48, 1 September 2022 (UTC)- At the time of your comment, H. ericae contained four sentences, two sources, and an infobox. That is not normally a size that we are concerned about. The (historical) page defining substubs said "Substubs are usually no longer than a dictionary definition, and usually contain information that anyone would know." I don't think that describes any of these articles. They are longer than a dictionary definition, and almost none of the information in them is something that anyone would know.
- Your assertion that articles "like that are not useful for readers" is your personal opinion. I personally benefited from a series of equally short (or worse) species article just a couple of weeks ago (for a plant, rather than a fish). As a result, my personal opinion is that your personal opinion is wrong. I could believe that these articles a not interesting to very many readers, but that's not the same as them never being useful to anyone at all.
- I think that the general principle you should be considering is related to Wikipedia:Deletion is not cleanup. It's actually okay to create a species stub with a few sentences, two sources, and an infobox. Not all articles have to be long ones, and they definitely don't have to be long articles on the day that they're created. WhatamIdoing (talk) 05:47, 1 September 2022 (UTC)
- Yes, it's longer now though. So it shouldn't so clearly be AFD'd or merged. So if every new stub started out this long, I bet there wouldn't be a dispute. Just don't create it if you can't make it this long. Andre🚐 03:20, 1 September 2022 (UTC)
- This was the version at the time of my comment. Database-replications like that are not useful for readers; they are better off being sent to a parent article where that information can be included until an editor has time to write a more comprehensive article. BilledMammal (talk) 03:17, 1 September 2022 (UTC)
- Hypostomus ericae has five sentences, three sources, and an infobox. One of the sources is a scholarly journal article. Why would you want to discourage the creation of an article like that? WhatamIdoing (talk) 03:09, 1 September 2022 (UTC)
- Oppose mass creation of any group of articles, period. Mass creations can sometimes lead to mass Prods or mass AfDs. The latter of which can get messy. GoodDay (talk) 00:02, 1 September 2022 (UTC)
- Agree with Uanfala and WhatamIdoing. If someone wants to go around and create a bunch of species articles like Hypostomus robinii, then great. That really shouldn't be problematic. If it would only be a single line derived from basic taxonomy and/or if we were talking about some tens or hundreds of thousands, then sure we should have a discussion, but this should be a non-issue. It sure seems like some people are just against the idea of creating stubs at all, and the specter of "mass creation" (500 manually created articles, with acceptable sourcing, created over 9 months is not what I'd call "mass creation" of the sort that's controversial) provides a convenient mechanism for action. — Rhododendrites talk \\ 03:24, 1 September 2022 (UTC)
- I think there probably wouldn't be an issue if substubs such as this "article," weren't created: "Hypostomus chrysostiktos is a species of catfish in the family Loricariidae. It is native to South America, where it occurs in the Paraguaçu River basin in Brazil. It is typically seen in blackwater portions of rivers with rocky substrates at elevations of 50 to 662 m (164 to 2172 ft) above sea level. The species reaches 26 cm (10.2 inches) SL and is believed to be a facultative air-breather." Andre🚐 03:29, 1 September 2022 (UTC)
- You keep using this word "substubs" like an article with just 3-4 referenced sentences is too short to exist, which is not something there is consensus for. 18 years ago there was in fact a discussion about the essay Wikipedia:Substub which not only ended up with a consensus that very short articles are fine, but we don't even mark such articles separately from other stubs, even though according to WP:STUB there is no consensus on how big a stub is. There's no minimum required length for a Wikipedia article, only that it passes the bar for verifiability and notability. Steven Walling • talk 04:42, 1 September 2022 (UTC)
There's no minimum required length for a Wikipedia article, only that it passes the bar for verifiability and notability.
- it also needs to meet the requirements of WP:PAGEDECIDE and not violate WP:NOT. "Microstubs" or "Substubs", whatever you prefer to call them, often struggle to do so. BilledMammal (talk) 04:45, 1 September 2022 (UTC)- I never said there were length requirements for articles, but articles do need to be encyclopedic and informative. The reason why substubs don't exist anymore per se, as I remember it, and like you said it was a very long time ago, is the advent of bright line notability requirements and referencing requirements. Back then, you could just make pages that were 1 or 2 sentences long and put
{{substub}}
, no references, and it would still be debated at VFD (as it was called then), and plenty of pages existed for a while before getting deleted that were all kinds of fictional characters and stuff that we today, will merge into a list article to avoid too much fancruft (to use another old wiki slang term). As I remember it, the reason why the template and the categories for substubs were done away with had a lot to do with there not being many substubs that could justify their existence. There subsequently was the requirement that articles had to assert notability. And now of course there is NPP, drafts, etc. Andre🚐 04:49, 1 September 2022 (UTC)
- "substubs" aren't a thing (well, other than a rhetorical tool). I don't see a problem with starting Hypostomus chrysostiktos the way Lumpsucker did. Expand it if you can, propose a merge if you want. All regular editorial decisions remain in place. It doesn't "create work" for other people because nobody is obliged to improve anything and as it stands it's completely policy compliant. — Rhododendrites talk \\ 19:47, 1 September 2022 (UTC)
- Substubs were a thing but let's not get hung up on terminology. I think tagging substubs stopped being necessary in part because WP:CSD began to cover A7,A9,and A11. Per this essay, Wikipedia:Credible claim of significance. You're right, no policy is being violated, it's not "making new work," but that is separate from whether all of these articles created should be kept at an AFD and not merged, which I think is a worthwhile question. Not saying Lumpsucker was the bad guy, but he should take note of the community's feedback good and bad for the article work. Andre🚐 21:02, 1 September 2022 (UTC)
- You keep using this word "substubs" like an article with just 3-4 referenced sentences is too short to exist, which is not something there is consensus for. 18 years ago there was in fact a discussion about the essay Wikipedia:Substub which not only ended up with a consensus that very short articles are fine, but we don't even mark such articles separately from other stubs, even though according to WP:STUB there is no consensus on how big a stub is. There's no minimum required length for a Wikipedia article, only that it passes the bar for verifiability and notability. Steven Walling • talk 04:42, 1 September 2022 (UTC)
- If someone wants to create a bunch of species articles like the current status of Hypostomus robinii, then great; we wouldn't be here if that was what was happening. However, that isn't what was happening; this is what was created, and that is an issue. BilledMammal (talk) 03:39, 1 September 2022 (UTC)
- Exactly, so, if the fish team could simply agree to make all the articles be up to that par or close to it, we'd have no problems. Andre🚐 04:51, 1 September 2022 (UTC)
- Hypostomus robinii as originally created was a perfectly good stub. It managed in its four sentences to include eight or nine separate facts about the species, in addition to comprehensive taxonomic information. It's now been further expanded to Hypostomus robinii. The encyclopedia is working as it should. MichaelMaggs (talk) 08:36, 1 September 2022 (UTC)
- I'm astounded that the initial version of Hypostomus robinii is being presented as an example of a bad stub. This thread started over concerns about semi-automated editing. Some of Lumpsucker's fish articles could have beeen generated by just copy-pasting a template and changing the species, river system where it occurs and standard length. The initial version of H. robinii includes a common name, diet, and more detail on habitat and distribution than just naming a single river system. I'll acknowledge that the only sources are databases, but it wasn't created in a semi-automated way by copy-pasting and changing a few variables (and FishBase has it's range as Central America:Trinidad, while the article places Trinidad in the Caribbean which I think is more accurate and shows Lumpsucker didn't just blindly insert variables from FishBase). Plantdrew (talk) 16:56, 1 September 2022 (UTC)
I don't see anybody presenting Hypostomus robinii as an example of a bad stub. It's not even relevant to this discussion, as has been pointed out above. Levivich 17:56, 1 September 2022 (UTC)
- I'm astounded that the initial version of Hypostomus robinii is being presented as an example of a bad stub. This thread started over concerns about semi-automated editing. Some of Lumpsucker's fish articles could have beeen generated by just copy-pasting a template and changing the species, river system where it occurs and standard length. The initial version of H. robinii includes a common name, diet, and more detail on habitat and distribution than just naming a single river system. I'll acknowledge that the only sources are databases, but it wasn't created in a semi-automated way by copy-pasting and changing a few variables (and FishBase has it's range as Central America:Trinidad, while the article places Trinidad in the Caribbean which I think is more accurate and shows Lumpsucker didn't just blindly insert variables from FishBase). Plantdrew (talk) 16:56, 1 September 2022 (UTC)
- Hypostomus robinii as originally created was a perfectly good stub. It managed in its four sentences to include eight or nine separate facts about the species, in addition to comprehensive taxonomic information. It's now been further expanded to Hypostomus robinii. The encyclopedia is working as it should. MichaelMaggs (talk) 08:36, 1 September 2022 (UTC)
If someone wants to create a bunch of species articles like the current status of Hypostomus robinii, then great; we wouldn't be here if that was what was happening
. True. What happened is someone manually started an article on a notable subject with two sources. Someone else came along and improved it. This is a wiki, after all. I would like it if people would improve an article they create rather than leave it a stub, but unless you find consensus for that, this sort of article isn't allowed. I didn't support Lugnuts' creation of single-sentence stubs based on databases because the databases didn't provide any real information and didn't actually ensure notability. An identified species will have some actual paperwork behind it, and indeed the starting version of this article has more than a single sentence with statistics. It may not be ideal, but it's completely in line with wikipolicy. — Rhododendrites talk \\ 19:47, 1 September 2022 (UTC)- These several recent ANIs with mass creation and the upcoming RFC mandated by sending about mass creation and deletion are signs the community dies not want someone to mass create stubs with questionable notability. A two paragraph article may be a stub but if it shown 403 or 4 good sources for significant coverage, we are far less likely to conplain. I think it is more dose don't want editors creating mass articles with low level effort (like just pulling from databases or finding one source and calling it good). mass creation with more effort will draw less attention. they may be stubs still but they are in a better place for expansion Masem (t) 20:03, 1 September 2022 (UTC)
- Exactly, so, if the fish team could simply agree to make all the articles be up to that par or close to it, we'd have no problems. Andre🚐 04:51, 1 September 2022 (UTC)
- I think there probably wouldn't be an issue if substubs such as this "article," weren't created: "Hypostomus chrysostiktos is a species of catfish in the family Loricariidae. It is native to South America, where it occurs in the Paraguaçu River basin in Brazil. It is typically seen in blackwater portions of rivers with rocky substrates at elevations of 50 to 662 m (164 to 2172 ft) above sea level. The species reaches 26 cm (10.2 inches) SL and is believed to be a facultative air-breather." Andre🚐 03:29, 1 September 2022 (UTC)
- It is not clear that a problem exists. The topics are notable in that a reliable source definitively exists for every recognised described species. Two articles per day on clearly notable topics is not mass creation, and most of the articles discussed here in any detail appear to be better than sub-stubs by a significant margin. While there are debatable advantages to the utility of conbining such articles, not necessrily as lists per se, but as sections in an article on a higher taxon, (actually my personal preference), the articles in general appear to be fully legitimate, with possible exceptions which have not been specified. The editor was requested to seek comment here, which they have done. I thank them for that as it displays the attitude of collegial discussion and open-mindedness we officially strive for. Unfortunately some of the comment lacks grounding in policy, guidance or tradition, but that is pretty normal, as everyone thinks they are an expert. Expressing a personal preference, opinion or bias as a !vote gives it no extra weight or validity. Cheers, · · · Peter Southwood (talk): 05:29, 2 September 2022 (UTC)
- [ec] It might be helpful if some of our editors did a little research on the process of taxon description and publishing, and the inclusion criteria of taxonomic databases before expressing their opinions on notabilty or lack thereof of a published taxon. · · · Peter Southwood (talk): 05:40, 2 September 2022 (UTC)
- Thank-you for the suggestion that we shoul learn more about Taxanomic databases before commenting. it is quite a fascinating topic. To summarise paid and volunteer work to create a taxanomic machine readable database that links physical material, media, and the written word.
- Problems are poor countries/organisations, uncontrolled/non-updated/non-validated subsets, secondary databases inability to create 'semantic and syntactic information that would improve the fitness of these data", and difficulties with occourence datasets ( Birdwatching etc) [1] [2][3][4]
- Best practise seems to be to link dynamicially Amphibia Web Wakelamp d[@-@]b (talk) 13:10, 20 September 2022 (UTC)
- Oppose creation of species articles cited only to database listings. These species can be covered in list articles. Support the creation of species articles that cite and summarize significant prose coverage of the species in reliable sources. Cullen328 (talk) 05:56, 2 September 2022 (UTC)
- Support. The sentiment that "mass creation from databases is bad" is close to right, but not quite there, in my opinion. Turning an accurate entry in an external public database into a stub is, in my opinion, neutral or very close to it. It doesn't really improve things—the reader could probably just as easily pull it from the original database, probably in a more useful format than Wikipedia prose—but it's not really harmful, either, the offense to some editors' taste notwithstanding. The damage occurs when the external database is not accurate (or if the mass creator is not accurately transcribing that information). In the case of biology stubs, this usually takes the form of a species being listed under a name that is no longer current or a name that was never really published. In the case of the sports stubs, my impression is that there have been widespread errors (depending on the creator) in birth and death dates, which events they competed in, standing etc. In the case of the geography stubs, places that are not, and never have been, concentrations of population have been misrepresented as being so (a single house or a landmark being misrepresented as a village). Propagating and disseminating external error, rather than correcting it through the power of many eyeballs, is directly damaging to the encyclopedia. All that said, I see no problem with the initial version of Hypostomus robinii that BilledMammal deprecates. In particular, I would note that it's sourced to two databases that appear to be substantially independent of each other. In my experience (I would note that I sink a fair bit of my leisure time into helping correct non-Wikipedia taxonomic databases), that's likely to avoid most of the damage caused by the really problematic mass creations described above, and also to limit the rate of creation to something relatively modest. Choess (talk) 15:53, 2 September 2022 (UTC)
- Oppose Bulk creation of database stubs. More substantive content and sourcing is expected for each article. A List of Hypostomus species that compiles content for comparison and context would be welcome. Reywas92Talk 18:21, 2 September 2022 (UTC)
- Strongly Oppose From their point of view we are creating a Wikipedia:Wikipedia_clones designed to reduce their traffic, and break thier community. We are doing the same thing on reptiles and other small databases. We also don't have as much information, our cite does not mention that we use thier cite, and we will have no one actively updating the information. Wakelamp d[@-@]b (talk) 04:49, 3 September 2022 (UTC)
- what an amazing waste of effort this discussion is producing, the simple answer is put the data base in Wikidata like we are asked to do then generate list articles from there. No where in the discussion has it been shown these stubs arent notable subjects, nor that the creation of them is causing any disruption to the project its a simple case of "I dont like it" therefore noone can do it. Gnangarra 14:18, 3 September 2022 (UTC)
- Totally agree - except I would actually get wikidata to negotitate with the site that is maintaining it, If a user searchs for it, they get shown a wikidata created page, Wakelamp d[@-@]b (talk) 14:59, 4 September 2022 (UTC)
- Support The OP's recent creations such as Hypostomus dardanelos and Parotocinclus hardmani look perfectly reasonable and respectable. A separate page for each such species is sensible as the page will then readily support an infobox and appropriate imagery such as a picture of the creature and map of its distribution. Lumpsucker's editing of such subjects is commendable as it seems quite fluent and encyclopedic. It does not seem especially bot-like or high volume and so no special process or permission is required. As for species, there are certainly a lot of them but so it goes and it's not a problem. Our policy is WP:NOTPAPER which states clearly that "there is no practical limit" and so we should not invent one just to be difficult. Andrew🐉(talk) 20:23, 3 September 2022 (UTC)
- Oppose - send the bot-like mass creations to WikiSpecies – we need articles at en.WP not taxosentences that are basically definitions or more like WikiData that Google can use.
Let the WMF capitalize on these BOTs and then they can use that money to get en.WP the tools we need to build this encyclopedia as the encyclopedia it was intended to be.As for all the support votes - if this Proposal passes, all the editors who voted support will be automatically added as NPP reviewers on assignment and your dedicated job will be to review all BOT-like creations.[FBDB] Atsme 💬 📧 21:15, 4 September 2022 (UTC) Correction: bot-like not bots. Apologies for the error and thank you for pointing it out to me Lumpsucker and Andrevan. I really would like to see WikiSpecies used for this purpose. Atsme 💬 📧 03:15, 5 September 2022 (UTC)- I don't know if I am understanding this message correctly, but this discussion is not about bot-created articles. Lumpsucker (talk) 22:26, 4 September 2022 (UTC)
- Moral support for Lumpsucker and since nothing was being proposed, nothing is being opposed. Lumpsucker has already stated they will improve these articles and will abide by policy. Andre🚐 22:32, 4 September 2022 (UTC)
- Oppose until a WP:BRFA has been passed, then support. Generally oppose bot-like edits with user accounts. AKAF (talk) 09:55, 6 September 2022 (UTC)
- @AKAF: The question is how would you define bot-like edits? That seems to be the crux of the disagreement, because to me the original version of this article that was used as an example for this discussion doesn't look bot-like at all–it has two sources and original content not written using a template. The current version has since been expanded to be decent stub that's even better. The original creator was doing 1-2 of these a day, which is plenty slow enough that the other editors in WikiProject Fishes or related projects can review them and hardly the scale that only a bot or script could achieve. Steven Walling • talk 21:01, 7 September 2022 (UTC)
- @Steven Walling: The original version was actually written using a template:
NAME is a species of catfish in the family Loricariidae. It is native to South America, where it occurs AREA OF OCCURRENCE. The species reaches LENGTH cm (LENGTH inches) SL and is believed to be a facultative air-breather.
- Compare it to the other examples I provided, like Hypostomus argus and Hypostomus johnii. BilledMammal (talk) 02:08, 8 September 2022 (UTC)
- Ok but I still don't agree that's a bot-like set of articles created. There's a human being reading the source material and applying that knowledge to original article content, which is subsequently being edited and expanded by others. How is this harmful rather than helpful? We now have more verifiable knowledge in the encyclopedia that has been reviewed by humans, not bots which are dumb and don't understand the meaning of sources. That's the problem with bot creation (that they're often incorrect, because the bot doesn't actually understand the sources) and why we don't allow bot-created articles generally. This is totally different in outcome. Steven Walling • talk 17:37, 8 September 2022 (UTC)
- If the source is reliable, a bot is actually better than a human; the human makes transcription mistakes, the bot does not.
- As for why we don't want micro-stubs like these, Levivich and Casualdejekyll has said it well. Readers are better served by a list or general article than they are by these microstubs. I would also mention that we aren't just here to build a large encyclopedia, we are here to build a quality encyclopaedia, and the proliferation of micro-stubs like these reduce the quality. BilledMammal (talk) 23:52, 8 September 2022 (UTC)
- "micro-stubs like these reduce the quality"
- This is false and not in alignment with core editing policy. Having a small, well-referenced article does not detract from the quality of other articles. More importantly, it's inherent to the nature of the wiki that we have articles of varying size and quality. Wikipedia is a work in progress and perfection everywhere is not required. Steven Walling • talk 16:26, 9 September 2022 (UTC)
- I have to agree with this - how do micro-stubs reduce the quality of good articles? Regardless of whether you think such an article should be merged or kept, I don't see the additive production of small articles reducing the quality of the overall work. Andre🚐 16:28, 9 September 2022 (UTC)
- I agree that @Lumpsucker 's 500 articles are not an issue ( and I thank them for their work), but this discussion is about the general case. The reduction in the quality of good articles is in perception. If a study was done on Wikipedia accuracy, 86% (6.5/7.5 Millions would be based on stubs/starts/unassessed)
- We shouldn't create something that we will not update Nearly all updates on stubs are by bots, but with no content as @Andrevan' s example indicated. We have a higher Google page rank, so readers will view our less accurate version, at the expense of the community that we online database strip mined. Wakelamp d[@-@]b (talk) 02:22, 11 September 2022 (UTC)
- I think you made a good point Wakelamp about the smaller communities. Since Lumpsucker is trying to create Wikipedia-quality entries, hopefully, there will still be a place on the Internet for Fishbase to create entries for things that may not make a good Wiki entry at present, or maybe ever. Some species may not be attested much. Who knows why. There's no rule in natural history that says everything will be documented verifiably. Andre🚐 02:25, 11 September 2022 (UTC)
- I have to agree with this - how do micro-stubs reduce the quality of good articles? Regardless of whether you think such an article should be merged or kept, I don't see the additive production of small articles reducing the quality of the overall work. Andre🚐 16:28, 9 September 2022 (UTC)
- Ok but I still don't agree that's a bot-like set of articles created. There's a human being reading the source material and applying that knowledge to original article content, which is subsequently being edited and expanded by others. How is this harmful rather than helpful? We now have more verifiable knowledge in the encyclopedia that has been reviewed by humans, not bots which are dumb and don't understand the meaning of sources. That's the problem with bot creation (that they're often incorrect, because the bot doesn't actually understand the sources) and why we don't allow bot-created articles generally. This is totally different in outcome. Steven Walling • talk 17:37, 8 September 2022 (UTC)
- @Steven Walling: The original version was actually written using a template:
- @AKAF: The question is how would you define bot-like edits? That seems to be the crux of the disagreement, because to me the original version of this article that was used as an example for this discussion doesn't look bot-like at all–it has two sources and original content not written using a template. The current version has since been expanded to be decent stub that's even better. The original creator was doing 1-2 of these a day, which is plenty slow enough that the other editors in WikiProject Fishes or related projects can review them and hardly the scale that only a bot or script could achieve. Steven Walling • talk 21:01, 7 September 2022 (UTC)
- Oppose mass creation but this one is borderline on being mass creation Maybe slow down and do a thorough source search for material for each article and put the material in. North8000 (talk) 21:12, 7 September 2022 (UTC)
- Support. 2 articles a day is nowhere near mass creation. — Qwerfjkltalk 20:35, 8 September 2022 (UTC)
- oppose Levivich and BD2412 have the right idea here. -- LCU ActivelyDisinterested ∆transmissions∆ °co-ords° 22:16, 8 September 2022 (UTC)
- Uh-Oh we want to be careful here because "Support" is being used for two opposite opinions ("SUPPORT OP's suggestion of reigning hn these articles" but also "SUPPORT the continued creation of these articles"; and vice versa for "Oppose". Just pointing this out. Herostratus (talk) 01:57, 9 September 2022 (UTC)
- Well, I think these articles are fine. I think the argument is between "let's have lists" and "let's have individual articles". I think that people who don't think we should publish this material at all are in a minority compared to people who do but can't agree on the format.
- So, it's basically a matter of opinion. It's the Wikipedia not ExxonMobile, so it one editor wants to create a long list and another a bunch of small articles, I'm just glad to have the info. I'm not big on telling "you can't do that, you have to do it my way" to people who have just spent hundreds of hours creating content, and might stop if annoyed. Generally and withing reason, stare decisis is the rule for formatting, giving deference to the person who did, I don't know, the actual work of creating the encyclopedia, and also to prevent sterile warring about one editors personal preferance over another. We don't need a rule for everything, and we don't need to be busybodies.
- On the merits, I happen to like the articles better. That's just me. A proponent opines "One of the reasons [for having lists instead] is context: Hypostomus brevis doesn't tell me that this particular species of catfish is one of 100+ species of Hypostomus". Well it could, so go add it. But it's a reasonable point otherwise. But if there are 100+ entries on "List of Hypostomus species" when we get to them, that's a pretty long article. If you're including the infobox (much shorter, granted) and a picture, you're getting into a lot of these, and 100+ pictures for short entries makes it hard to have good formatting -- and easy reading. And the References section will be hundreds of entries long. Finding the references for only say Hypostomus brevis in particular will harder to do. Many of the refs will be just page numbers, and then you have to and find the first ref to the book I suppose. Navigation -- either your Table of Contents will be 100+ entries long which is rididulous, or you'll have to figure out some other scheme.
- I just don't think it's easier to navigate a large article that is 100's entries long than 100+ individual articles. This sort of thing is studied somewhere I'm sure, and I think it's very likely that a human factors engineer or user interface designer would barf at a TOC that large. Sure you can divide the species into groups -- by Subgenus, or geographical location, or whatever -- but then accessing an individual species is hard.
- But then, you can make the list shorter if you omit some information. Why omit inforation that we already have -- I don't think our readers can't handle these amounts of detail. So, sure you can omit say "...and is believed to be a facultative air-breather" from maybe the articles for a whole subgenus and put it in a section header ("All of the following are believed to be facultative air-breathers"), but then you are spreading out information about the entity into two places -- not a service to the reader. Or you can just have subgenus sections and just describe things common to all the species in that subgenus, or something. If that's how you roll. I don't.
- Another thing about lists is that they can be attractive nuisences forjust destroying some information. You know, like "Geez this list is long. Do we really need say 'The fish was first formally described by John Treadwell Nichols in 1919'? That doesn't tell anything about the species itself, and we can make the list a lot more managable if we leave out the discovery info". I'm not super on board with that either. Herostratus (talk) 01:57, 9 September 2022 (UTC)
- This is a very good point. I am a reader. I am not "served" by enormous lists, which take a while to load on my machine and are virtually impossible to edit for more than a few characters without the page hanging. Gnomingstuff (talk) 12:53, 16 September 2022 (UTC)
- Support -- The crusade against valid encyclopedic content just because the articles happen to be short was absurd from the beginning and remains absurd (not least because of the mountain of undetected vandalism, PR-speak, fan-speak, un-copyedited, and generally embarrassingly written content that is actually embarrassing to the project, as opposed to neutrally-written, academic, inobtrusive entries on, in this case, fish). The vilification of the people who make them is worse. There should not have to be a topic where one can support or oppose improving the encyclopedia! Gnomingstuff (talk) 12:45, 16 September 2022 (UTC)
- Oppose mass creation of "sub-stub articles" (per Donald Albury), on fish species or otherwise, per Levivich and Ajpolino and UnitedStatesian. Rationale: 1)- Data bases have errors. There is no actual secondary sources with significant coverage to confirm the information. I became weary of all the GNIS populated places articles that, with no proper outside sourcing, showed things like empty fields. What we end up with is just copied wording from a database to an article page, 2)- by creating these glorified dictionary entries there is too much important information missing and what is included in pieced from the database information. I know it is not the actually reasoning but since Wikipedia is gungho on stats and it does look good for article creation, 3)- it makes sense to put these in proper lists where silly things like Wikipedia:Reliable sources:
Wikipedia articles should be based on reliable, published sources, making sure that all majority and significant minority views that have appeared in those sources are covered
(NPOV), covers content and not notability. 4)- Article that are basically prose of database information is not encyclopedic. We can create short uninformative stubs of the "92 genera and just over 680 species" of the Loricariidae but how does that help Wikipedia except possible showing it is still growing. Why do we want to sacrifice quality for quantity? 5)- It would be better to "try covering all the species in sections on the genus article and split out the most substantial sections into their own article.", per casualdejekyll. Agree with "two sentence stubs with no clear notability, nor any expectation they will be improved to meet that." per Masem -- Otr500 (talk) 05:49, 18 September 2022 (UTC)- @Otr500, you say that "Data bases have errors". Are you aware of any errors in these databases? Telling editors that they shouldn't rely on these sources because similarly formatted sources in a completely unrelated field have problems would be an example of a Faulty generalization.
- As for whether it contains SIGCOV, as an example, I went through one reputable fish-species secondary database to see how much of an article I could write. You can read the results here. The short answer is that if you define SIGCOV as meaning 'contains enough information to write more than a doomed permastub', then this database definitely contains SIGCOV. WhatamIdoing (talk) 17:36, 19 September 2022 (UTC)
- I have seen errors in FishBase. Lumpsucker (talk) 18:12, 19 September 2022 (UTC)
- Reply: (actually to WhatamIdoing) I did not yet look at the example (yet) for I was looking elsewhere. Hypostomus johnii contains two sources, both databases (ITIS and FishBase) and "IF" this (I assume scholarly) abstract is considered reliable then the 2007 update (redescription) seems to indicate that Steindachner (1877) may have confused the Hypostomus johnii with congeners in the São Francisco River. "If" this is true then the article, using just database information, is flawed. A preliminary survey (Burgess, W.E., 1989) apparently centered only on Siluriformes (catfish) and not on individual species. This is the very first example I randomly picked. I suppose if you found one that could be considered SIGCOV, I would have to say (the only one I checked so far) does not pass.
- In no way do I want to hinder an editor from creating or improving articles. I am just attempting to show there should be more to an article that the use of only a database, which in essence is just a database copied into prose form possible to satisfy WP:SPECIESOUTCOMES:
at least a brief description must have been published in a reliable academic publication to be recognized as correct or valid.
, but I question if a database satisfies "academic publication" when there is no description and an editor has to create prose from the content. - When Wikipedia was striving to be a world encyclopedia it was common to create an article with one source, no source, or the source located in the "External links". We have nothing to prove now except improving the quality of even stub articles and the accuracy of any created article. I don't assume an article (even stub) is "doomed" because it only has a database listing.I think moving forward an editor should do at least some research and not just create I think that 900 articles (I am pretty sure I read that) a year , almost 2 1/2 articles a day, to some, can be considered mass creating if just copying material into prose from a database. Although there is no hurry for improvements I can't see any way a vast majority of these can ever be be checked/expanded, especially since the author has commented continuing the status quo, knowing consensus has previously considered species notable for inclusion.
- "If" consensus deems an editor can create as many articles as desired, especially using only database content converted to prose, I would hope that editor reads the negative rational here. casualdejekyll stated: "They're stubs. Sourced stubs, sure, if they weren't on such an obviously notable subject they don't have the sourcing to show the notability", and that is a sad realization. Since sourcing on any considered notable subjects (we can leave out "presumed" and add fish species to automatically notable), would be exempt from some standards, we can just create 1000's and not worry about any sourcing at all or accuracy. If there is still a presumption of notability then reliable sourcing is still required.
- In defense of the singled out editor one can look at the project Wikipedia:WikiProject Fishes/Assessment#Quality scale (Stub: more detailed criteria): "The article is either a very short article or a rough collection of information that will need much work to become a meaningful article.", and "may be little more than a dictionary definition" to see justification for stubby-stubs. However, it also includes "Although Stub-class articles are the lowest class of the normal classes, they are adequate enough to be an accepted article, though they do have risks of being dropped from being an article altogether." This appears to support that there is presumed and not inherit notability (arguemnts to avoid).
- WP:SIGCOV states:
"Significant coverage" addresses the topic directly and in detail, so that no original research is needed to extract the content.
and I can't see justification in a database listing and 1000's of articles created from them. The editors comments below might have missed the one I checked as I see some issues. Donald Albury (below) seems to have hit the nail on the head, having "no objection to using a database to find subjects for creating articles, as long as each article is also supported by at least one reliable source other than the database." A list could be created for the articles that did not follow that criteria or could be placed in a higher-level article. I do agree with above statements that articles on species would very likely not be deleted at AFD, although some other course of action could be the result. -- Otr500 (talk) - Special important Note: If we are concerned only with article creations concerning the above editor, the issue is solved per positive comments from the editor that stated:
Before creating any other articles, I am planning on expanding a few of my smallest articles as I already have been doing. Some of those were only created with a single source and are less than 1 KB in length, and I am looking to improve most of the articles of that size that I made.
I hope these improvements will include more than database content. That is all I need here since I do not foresee starting any AFD on species. "If" anyone does start an RFC on "mass article creation", please ping me, on species or any other subject. -- Otr500 (talk) 20:52, 19 September 2022 (UTC)- @Otr500:, the link to that abstract doesn't work for me, but I assume it's the same as publication as this one althought that was published in 2017 rather than 2007 (but if H. johniihad be redescribed in 2007, another redescription in 2017 would probably be unnecessary). The authors examined the same specimens (allegedly) from the São Francisco River that Steindachner did, and did not find them to be anything other than H. johnii. It may be the case that H. johnii did occur in the São Francisco River in 1877, but has since been extirpated; it may be the case that the specimens weren't actually collected in the São Francisco River but were mislabelled as such (which should not be a problem for Wikipedia; Wikipedia:Verifiability, not truth). The cover of Burgess's book promises a "complete checklist of the genera and species".
- The reliable academic publication that SPECIESOUTCOMES refers to is, in this case, Steindachner's 1877 original description of the species, available on BHL. It's in German and runs not quite 2 pages. Few species articles currently cite the original description directly (but there's almost always an indirect citation with the authority in the taxobox). Original descriptions may not be available online; BHL has a lot of taxonomic literature that is out of copyright, but uploading a publication that describes a single species isn't going to be a high priority for them. Other original descriptions may not be out of copyright. Many are in a foreign language (particularly Latin in the 19th century). The original description of a species described a long time ago may not be sufficient to distinguish it from the larger number of related species that are known now (this is why H. johnii was redescribed in 2017).
- SPECIESOUTCOMES begins with
Species that have a correct name (botany) or valid name (zoology) are generally kept.
. So far in this thread, nobody has mentioned why species articles are sourced to taxonomic databases. The database is the source that shows that a name is correct/valid. Taxonomy is subjective to a degree; taxonomists can disagree about whether two populations represent different subspecies, or aren't worth recognizing as taxonomically distinct at all; going the other way they can disagree about whether there are two species or one species with two subspecies. If we are going to present a list of species regarded as valid/correct in a genus article, that list needs to come from a single source, and in 2022 that source is usually going to be a taxonomic database (we can't just piece together a list of species in a genus from the original descriptions; there are several times as many species that have been described as are currently recognized by a consensus of taxonomists). Original descriptions are primary sources; the act of describing a species means that the person doing so thinks it is valid. Taxonomic databases are secondary or tertiary sources that show that somebody other than the original describer agrees a species is valid. - @WhatamIdoing:'s example doesn't mention the most important fact; it is a valid species, according to FishBase. Probably because that's taken as a given; taxonomic databases list invalid species as well as valid ones, and absolutely nobody is churning out sub-stubs for species that aren't considered valid. There are two high quality taxonomic databases for fish; the other Catalog of Fishes. For the vast majority of fish species, both databases are in agreement as to their validity. WikiProject Fishes has chosen to follow FishBase (in the early years of Wikipedia, it wasn't possible to link directly to species records in Catalog of Fishes). Every article for fish species should cite FishBase; that's the baseline to establish that the species is valid. Fish articles ought to cite the original description (if it was described recently); or a detailed rediscription (if it was described long ago). If there is active disagreement among taxonomists about the validity of a species, that ought to be mentioned as well (e.g. in article for species X "Catalog of Fishes recognizes species X and Y as distinct, but FishBase considers Y a synonym of X). Plantdrew (talk) 21:22, 20 September 2022 (UTC)
- I think this is that problem of most editors not knowing the subject area. Otr500 writes I question if a database satisfies "academic publication". The database isn't the academic publication, but these two databases don't include species unless there really was an academic publication. The real-world process for the would-be fish discoverer is:
- See a fish.
- Figure out that it's never been described in a peer-reviewed scientific journal.
- Publish a description that contains a description sufficient for other scientists to differentiate your new fish from other fish plus your choice of name (which must follow certain rules) in a peer-reviewed scientific journal. (Fun fact: until 10 years ago, only journals printed on paper were acceptable. It is still true that only "reputable" peer-reviewed journals are acceptable.)
- FishBase reads your journal article and makes a record for the fish.
- From the Wikipedia editor's perspective, if it's in one of these databases and marked as having a valid name (a "valid name" is a special thing in the real world, and it is not the same as having, say, a grammatically correct name), then Wikipedia editors [should] know, with 100% certainty, that steps 1, 2, and 3 have happened in the real world. Even if you don't have the peer-reviewed scientific journal article, you know that it was published, because publishing that peer-reviewed article is the only way to get a valid name.
- But if you feel a need to check on whether this mandatory process really was followed for any particular entry that claims to have a valid name, then using Entomocorus benjamini as an example, FishBase says, in the very first line/title of the database record that the original description was written by Carl H. Eigenmann in 1917, like this:
- Entomocorus benjamini Eigenmann, 1917
- All four of the words are links. If you click on the last one (the year), it gives you the full bibliographic citation for the original peer-reviewed scientific journal article. Skeptical editors don't have to wonder whether there was an academic publication. You just have to click on the link in the database entry and find out. WhatamIdoing (talk) 01:09, 21 September 2022 (UTC)
- I think this is that problem of most editors not knowing the subject area. Otr500 writes I question if a database satisfies "academic publication". The database isn't the academic publication, but these two databases don't include species unless there really was an academic publication. The real-world process for the would-be fish discoverer is:
- I have seen errors in FishBase. Lumpsucker (talk) 18:12, 19 September 2022 (UTC)
- Support. I've never understood people who devote themselves to deleting useful properly sourced information. I looked at all the fish articles mentioned in this thread and don't see a problem with any of them. Some are short; so what, lots of articles start off short. Some even stay short because there aren't many sources, but that has never been an excuse for deletion if what is there is notable and properly sourced. Moreover, the claim that a species of life can be non-notable is frankly obscene. If Wikipedia had an article on every known species on Earth, that would be the most extraordinary achievement in the life of the encyclopedia. Lift your horizons. Zerotalk 08:48, 18 September 2022 (UTC)
- You're way too long in the tooth to be referring to merger as "deletion" or ignoring altogether WP:PAGEDECIDE (the section of the notability guideline dedicated entirely to the idea that just because a topic is notable doesn't necessarily mean it's best to have a stand-alone page). Levivich (talk) 14:32, 20 September 2022 (UTC)
- Comment I have not expressed an opinion on this set of creations, and I do not have the time available to review the articles, but, since my name has been invoked, I want to clarify that while I would like to see a consensus of the community to prevent articles being created based solely on a database, I have no objection to using a database to find subjects for creating articles, as long as each article is also supported by at least one reliable source other than the database. I also feel that list articles are fine for subjects for which sufficient material from reliable source has not yet been found to support a stand-alone article. YMMV. - Donald Albury 20:49, 18 September 2022 (UTC)
- WIkidata is looking at negotiating to be the central hub in DE Wakelamp d[@-@]b (talk) 08:18, 24 September 2022 (UTC)
- Strongest possible support, the antagonism towards microstubs and stubs, and deletionist policies in general go against the spirit of Wikipedia. If the pages are too short for your liking, then expand them. If you can't expand them, then move on with your life and expand the next article. Nothing is gained from removing information.--Ortizesp (talk) 16:12, 5 October 2022 (UTC)
- But, there are many stubs that will never be expanded, and are basically useless, or even harmful. Having an article about a place for which nothing can be found, except a name on a map, is not helping the encyclopedia, yet I was only able to remove one after spending many hours searching for anything about it. I might add that the stub in question (Hasan, Florida) had existed for years and had been edited 72 times, mainly adding and deleting unsourced, often nonsense, material, before it was deleted, a waste of a lot of time and effort. What is so terrible about asking someone who is creating an article to provide a citation or two to establish notability. Donald Albury 18:47, 5 October 2022 (UTC)
- There are many longer articles that, despite having been expanded, are useless and even harmful, and yet we don't call for deleting longer articles. Maybe size isn't the key point?
- You can "provide a citation or two" – or twelve, for that matter – without actually expanding the article. Also, no matter how many citations you provide, someone with a different POV is still free to declare that it doesn't establish notability. I remember seeing a major hospital at AFD. It was the size of Royal Derby Hospital, attached to a university and a medical school, and it had a bunch of independent sources, but that didn't stop an editor from trying to claim that it wasn't really notable. (He lost; the article was kept.) I think the real question here is not about whether it's a stub, but whether can you write an article to convince me that the subject is notable, when I already don't want to believe you because I think your favorite subject is unimportant and useless. WhatamIdoing (talk) 23:08, 5 October 2022 (UTC)
- @Gnangarra and Gnan: Maritime park for Editathons : Based on your comment "You'll also note that these articles are one WMF KPI's for events if you look over at Outreach dashboards." Should fish species be segregated from bot creation , so it is available for editathons now and in the future? (The Meta bot created drafts of thousands of women bios earlier this year.)
- How do these articles get updated based on new information? And would people be in favour of automated updates of existing articles - say to add new journal references?
- But, there are many stubs that will never be expanded, and are basically useless, or even harmful. Having an article about a place for which nothing can be found, except a name on a map, is not helping the encyclopedia, yet I was only able to remove one after spending many hours searching for anything about it. I might add that the stub in question (Hasan, Florida) had existed for years and had been edited 72 times, mainly adding and deleting unsourced, often nonsense, material, before it was deleted, a waste of a lot of time and effort. What is so terrible about asking someone who is creating an article to provide a citation or two to establish notability. Donald Albury 18:47, 5 October 2022 (UTC)
- @AddWittyNameHere: " digitization for old non-English sources" made me think that I would be rapt if WMF raised a 100 M a year if they helped fund scanning of old journals/supported small databases/Bought ancient Author rights, or paid for ethno-groups to scan their languages BUT ONLY IF they wished)
Wikidata lists
Should mainspace lists where the contents are pulled from and maintained at Wikidata be allowed or disallowed? Fram (talk) 07:25, 19 September 2022 (UTC) (edited same day at 13.35, see start of discussion section)
Summary (Fram)
There is a number of lists where the contents are pulled from Wikidata through Template:wdtable row, with specific subtemplates for different types of content. The result can be seen e.g. here. The Wikipedia-only version of the same list can be seen here. Previous RfCs (see Wikipedia:Wikidata#Appropriate usage in articles have already agreed that "Wikidata should not be linked to within the body of the article except in the manner of hidden comment" and that it is "not appropriate to use Wikidata in article text on English Wikipedia ", but allowing Wikidata in infoboxes and de facto also many in external links templates. Previous lists where not only the contents, but even the entries were Wikidata driven have been disallowed in the mainspace as well.
The new type of lists has a number of disadvantages compared to enwiki-based lists, i.e.
- it isn't maintainable here but requires going to another website with another interface, making it harder for most people (for the data)
- requires editing templates or creating new ones for the layout
- Has issues with e.g. sorting, see for example here where (with the current Wikidata data) sorting on the "opened" column gives a random "Apr 1998" data inbetween the blank dates, and sorts "17 Apr 1968" before 1917 and so on. This example shows also a typical issue with getting data from Wikidata like this, the formatting. Wikidata has "April 1998", so I suppose the "Apr 1998" entry is formatted in a template. This makes it again harder for regular editors to maintain or layout such articles.
- Similarly, at Talk:List of dams in Saga Prefecture an editor asked to remove the image column from the article, as they were unable to do this under the Wikidata format. This required the creation of a new template, instead of simply editing the article. Fram (talk) 07:41, 19 September 2022 (UTC)
- There are other more minor issues, like the name appearing in the list being the Wikidata label, not the enwiki article name, or the sourcing being inadequate (Wikidata items referenced to some Wikipedia version, often outdated (e.g. some of the entries on List of islands of the Isles of Scilly use the 2001 census instead of the 2011 census our articles use, indicating the glacial speed of update Wikidata often has) Fram (talk) 09:03, 19 September 2022 (UTC)
- See for example List of learned societies in Australia, with issues from the start (e.g. one entry with the incorrect Wikidata title instead of the correct enwiki title, and entries which don't even belong there like the Austronesian Formal Linguistics Association), and then made worse by an editor who probably couldn't figure out how to correctly add an entry[5]. This type of list is not editor-friendly at all. Fram (talk) 12:38, 19 September 2022 (UTC)
Summary (MSGJ)
Thank you for starting this discussion and inviting me to present an alternative viewpoint. First, some background for those who may not be familiar with Wikidata. This Wikimedia sister project, launched in 2012, is designed to hold data for use by Wikipedias and other sister projects. Its use on the English Wikipedia is not at all new - it has been used extensively in infobox templates for many years now - so its use in data tables and lists should not be surprising to Wikipedia editors. I really thought and hoped that the "us and them" attitude towards Wikipedia and Wikidata had diminished over the years.
Being designed for this purpose, Wkidata offers many advantages over conventional wikitext for storing reliable data, including:
- Numerous constraints which can catch incorrect data, such as incorrect units, a date of death before a date of birth, etc.
- A very user-friendly interface (almost certainly easier than editing wikitext for new editors). For example, compare how you would update the height of a dam on compared with on Wikidata.
- Ability to use powerful queries to find information.
- And probably most importantly, improvements to the data by one project will be of benefit to all projects.
All data on Wikidata can (and should) be referenced, just as it is on Wikipedia. All changes can be monitored via RecentChanges (if you have the appropriate option selected).
The use of a template to produce the rows of the table has several advantages, including:
- The template allows any column to be overridden by locally defined content. For example on List of Welsh mathematicians the "Notes" column is entirely local content.
- The wikicode to produce rows and columns, which is complex for many editors, is conveniently separated from the content of the table.
- A column can be added or removed from the table by making a single change to the template, rather than dozens of changes to the wikitext.
- The pencil icon links straight to where the data is stored.
Finally, a word on the previous discussions about the use of Wikidata on Wikipedia. A table is not classed as "article text" so the prohibition on using Wikidata for article text is not relevant here. And, regarding the linking to Wikidata, the only link is the pencil icon mentioned earlier which is widely used and accepted in infobox templates. — Martin (MSGJ · talk) 17:04, 19 September 2022 (UTC)
Discussion (Wikidata lists)
- @Fram: Two things: this is asking whether they're allowed, not whether they should be allowed. Is this about getting clarity of where past decisions have landed us, or deciding whether they should be allowed? If it's really just asking whether they're allowed, the list of reasons why they're bad seems out of place. If it's asking whether they should be allowed, you may want to edit the initial statement. The other suggestion is sort of dependent on the first, but you may want to separate the summary of where consensus stands from specific arguments about what our policy should be. The latter isn't so much a summary as arguments supporting one outcome. — Rhododendrites talk \\ 13:00, 19 September 2022 (UTC)
- Well, it's kind of a double question; are they allowed viz-a-viz the previous RfCs, and should they be allowed or not? I guess the second is more important than the first, as it's not intended as a "you did something that wasn't allowed" but more of a "this is how we'll proceed from now on". I'll change the RfC accordingly. Fram (talk) 13:35, 19 September 2022 (UTC)
- Initial thoughts (should I be creating my own section per above?): I have mixed feelings about Wikidata lists. On the bad side: technical limitations. Some of the things Fram lists seem like they could be fixable, but for others it's a matter of better integration of Wikidata in Wikipedia (in the sense of editing). The current templates, which send would-be editors to a Wikidata page with no explanation, are a bit clumsy (but, granted, an early step in the process). On the good side: I can see at least two good uses for Wikidata in Wikipedia lists. The first is as a starting point. If you want to make a lists of dams in a given place, that's something Wikidata has data for, and pulling from Wikidata could save a lot of time vs. hunting it down and formatting it yourself. Then you could convert it to wikitext and move on. I don't expect that's very controversial, though. The second case is when Wikidata pulls from databases that are more easily kept up-to-date than a Wikipedia list. We have an awful lot of out-of-date lists, and if it's an appropriate topic, why not let Wikidata gnomes keep it up to date? We just need more sophisticated templates to allow for flexibility in display and for fixing errors without sending someone on a journey to Wikidata. So I guess part of my answer (although per above I'm not sure which question I'm answering) is: yes, at some point these are useful, and I'd encourage people to shift the discussion from a binary yes/no to figuring out (a) in what contexts they're useful, and (b) if the current setup is inadequate, what changes to the interface and/or templates would be needed to ensure we can take advantage of this data in those cases when it's useful? — Rhododendrites talk \\ 13:23, 19 September 2022 (UTC)
- My understanding from past RFCs is that we continue to be extremely wary of Wikidata, and have limited its use… but that, within those limits, it can be used.
- That said, I don’t think we have been very clear as to what exactly those limits ARE. We need to spell them out clearly. Do we have a guideline or policy section covering this? Blueboar (talk) 13:55, 19 September 2022 (UTC)
- I don't think its specific use in lists/tables has been put to RfC before. The closer of 2013 discussion wrote "There is a valid point raised that while running text is clearly not suitable for Wikidata use, it might be worth discussing use in tables specifically – but no consensus regarding this has been reached in this discussion." — Martin (MSGJ · talk) 18:09, 19 September 2022 (UTC)
- I would think that this is disallowed per the previous RfC, which stated that it is
not appropriate to use Wikidata in article text on English Wikipedia
. A list article is still an article. I've actually come across some of these lists before and wondered why they were using {{Wdtable row}} instead of {{Wikidata list}}. The presumable reason is that {{Wikidata list}} produces an error when used in mainspace due to the results of that RfC and others. The use of {{Wdtable row}} in mainspace articles strikes me as a clumsy workaround to skirt consensus. Spicy (talk) 16:31, 19 September 2022 (UTC)- Yes, a list is an article but the values in a table are not "article text", which I take to mean prose. Some tables combine values and prose, e.g. List of Welsh mathematicians, and the template will only produce the content for the values and not for the prose. — Martin (MSGJ · talk) 18:12, 19 September 2022 (UTC)
- I'm of a mind that we shouldn't encourage them, and don't personally use them, but if someone wants to put together say a rather exhausting list of plant species or something, Wikidata may simplify the process. If someone finds a clever way to use them, why stop them? I agree with Rhododendrites that we should focus on where they're best used and how to improve their usage. CaptainEek Edits Ho Cap'n!⚓ 16:54, 19 September 2022 (UTC)
- I agree with Rhododendrites - tables pulled from Wikidata are suitable for some uses and not from others. Where they are suitable I see no justification for either prohibiting or requiring their use (treat it like ENGVAR or citation styles: either version is acceptable, don't change without both a good reason and consensus). Where they aren't suitable obviously they shouldn't be used, but I hope nobody is advocating for that. We should work on making the integration better so that the problems identified are fixed rather than saying the first version is not perfect so go away and never come back again. I also suggest that developing a set of guidelines about where Wikidata tables are and are not appropriate for use to be a much better use of editors' time than arguing about whether they should or should not be used at all. Thryduulf (talk) 22:30, 19 September 2022 (UTC)
- In theory, I'm fairly open to tables containing entries from Wikidata. In practice, I'd like to see more working and non-working examples. I think there are a few other language Wikipedias with deeper Wikidata integration, but perhaps I am mistaken and that is all just infoboxes. There are also a lot of things where information is rather fuzzy, making Wikidata difficult to use. It is easy to annotate uncertain dates and debates around them in wikitext; learning how to do that on Wikidata is rather hard and certainly unintuitive for people used to Wikipedia. —Kusma (talk) 08:18, 20 September 2022 (UTC)
- Thanks for your comment. I am of the opinion that only a clear-cut value can be appropriately used from Wikidata. Anything which requires clarification or explanation should be locally defined. There are a couple of approaches I have used in these cases.
- For example on List of castles in Ireland#County Clare, the imprecise build date of Ballyhannon Castle is locally specified via
|c5=c. [[1490 in Ireland|1490]]<ref>{{harvp|Westropp|1899|p=351}}</ref>
- And on List of lighthouses in Scotland, I clarified the build date of Southerness Lighthouse by adding a footnote to the value from Wikidata via
|c5+={{efn|Built in 1748 but not lit till 1800. Rebuilt in 1844.}}
- For example on List of castles in Ireland#County Clare, the imprecise build date of Ballyhannon Castle is locally specified via
- — Martin (MSGJ · talk) 16:01, 21 September 2022 (UTC)
- Thanks for your comment. I am of the opinion that only a clear-cut value can be appropriately used from Wikidata. Anything which requires clarification or explanation should be locally defined. There are a couple of approaches I have used in these cases.
- I'll perhaps expand on my rationale later, but fundamentally, I support Wikidata-derived tables being allowed. It all comes down to implementation — if done well, the appearance to readers will be exactly the same as a manually generated article, and the Wikidata-derived one will be far more future-proof. {{u|Sdkb}} talk 06:03, 21 September 2022 (UTC)
- The only argument presented by people in favor of using Wikidata continuously (as opposed to using it once to generate a list) seems to be making things easier to keep up to date. But most uses of Module:Wikidata table are timeless: lists of dams, lighthouses, and castles, etc. don't need much updating (other than adding new entries if they are built, which still needs to be done manually in the Wikidata version), so in those cases that argument is not convincing. On the contrary, there's been no refutation to several of Fram's points above which amount to the fact that it will almost always be possible to further optimize such a list with local tweaks because humans are better at this then co. So what's the harm in detaching from Wikidata and letting that be done? I'm not seeing it. For the few that aren't timeless (List of Brazilian mathematicians, List of Welsh mathematicians and List of Polish mathematicians seem to be the only ones), there is a slightly better case for using Wikidata, and I haven't come to a strong opinion either way. * Pppery * it has begun... 17:43, 21 September 2022 (UTC)
- For me it is not only about the creation of the lists, but the future updates and improvements too. I do not expect these lists to stay the same for ever more - I really hope they will be expanded with more information and better references. I believe this is both easier and better in the Wikidata version. Easier, because of the structured environment which lends itself to data import and verification. Better, because the knowledge will be shared among all language Wikipedias. — Martin (MSGJ · talk) 19:19, 29 September 2022 (UTC)
- I would have thought the prohibition applied to lists and tables as well, but if that's not the consensus I would support extending the prohibition to lists and tables. No problem with using Wikidata to initially populate a table or list, of course; the onus is on the editor to make sure it's reliable data as with any edit. The problem of "sneaky vandalism", as I think Fram named it years ago, is real -- changes to Wikidata are not easily visible which means you may not know when your article has been vandalized, and even if it's detected a user who can edit here may be unable or unwilling to learn the quite different interface there. Re Rhododendrites' comment that it would be OK to have a table sourced to live Wikidata, knowing that the gnomes over there would keep it up to date -- I don't think anyone would be OK without outsourcing our table data to, say, Fishbase, for certain tables, so why would we be OK with it with an intermediary? Mike Christie (talk - contribs - library) 22:40, 21 September 2022 (UTC)
- Sneaky vandalism is a problem that can occur with any type of list. With watchlist integration, while not perfect, I can effectively patrol all changes to data which affect these articles. So I do not really accept that changes on Wikidata are invisible or hard to catcher than on Wikipedia. — Martin (MSGJ · talk) 19:23, 29 September 2022 (UTC)
- When I last tried "Show Wikidata edits in your watchlist" it added an incredible amount of noise that made the watchlist unusable. Has that changed? Johnuniq (talk) 01:25, 30 September 2022 (UTC)
- Yes, a couple of years ago. WMDE did the work, so it's probably documented at Meta-Wiki. The volume and relevance of notified changes seems to depend on the subjects that you're watching, so I suggest trying it out and seeing what you think for yourself/your own subjects. I feel like volunteer-me gets a surprisingly large number of notifications for Apology (act) and Apologia, but less than I expect for other subjects, like Lymphoma. Most of the changes I see are changes to the linked articles (e.g., someone creating an article at another language's Wikipedia about lymphoma). Whatamidoing (WMF) (talk) 22:32, 30 September 2022 (UTC)
- It's probably fair to say that there are still too many entries that make it through the filter. I don't need to know when someone changes the Bangladeshi label or adds a sitelink to the Hebrew Wikipedia. Really the only ones needed are those which affect the display of the relevant article, although an option to display all changes might be useful for some editors. — Martin (MSGJ · talk) 14:34, 1 October 2022 (UTC)
- Yes, a couple of years ago. WMDE did the work, so it's probably documented at Meta-Wiki. The volume and relevance of notified changes seems to depend on the subjects that you're watching, so I suggest trying it out and seeing what you think for yourself/your own subjects. I feel like volunteer-me gets a surprisingly large number of notifications for Apology (act) and Apologia, but less than I expect for other subjects, like Lymphoma. Most of the changes I see are changes to the linked articles (e.g., someone creating an article at another language's Wikipedia about lymphoma). Whatamidoing (WMF) (talk) 22:32, 30 September 2022 (UTC)
- When I last tried "Show Wikidata edits in your watchlist" it added an incredible amount of noise that made the watchlist unusable. Has that changed? Johnuniq (talk) 01:25, 30 September 2022 (UTC)
- Sneaky vandalism is a problem that can occur with any type of list. With watchlist integration, while not perfect, I can effectively patrol all changes to data which affect these articles. So I do not really accept that changes on Wikidata are invisible or hard to catcher than on Wikipedia. — Martin (MSGJ · talk) 19:23, 29 September 2022 (UTC)
- Question, if I wish to change or amend something in a table that uses Wikidata to generate info, would I have to go to Wikidata to make the edit, or can it be done using the edit mode here in WP? Say something simple like a style correction. Blueboar (talk) 01:46, 30 September 2022 (UTC)
- Taking the eventualist view, with future interface improvements, yes, you'll be able to stay on Wikipedia. Also, it's worth noting that tables generated through Wikidata are less likely to have errors to begin with because there are fewer elements being created manually. {{u|Sdkb}} talk 03:47, 30 September 2022 (UTC)
- So your "yes, you'll be able to stay on Wikipedia" is actually a "no". Deciding on the current status of such lists based on what might happen one day, perhaps (Wikidata is 9 years old, so it's not as if such changes happen rapidly) is not a good idea. So, @Blueboar: you indeed have to go to Wikidata to edit the info, Wikipedia edit mode won't help you. And Sdkb, your "less likely to have errors" doesn't seem to make much sense either, the elements are created manually at Wikidata or manually at enwiki, no reason why one would be less likely to have errors (on the one hand, Wikipedia has more editors and thus more vandals: on the other hand, vandalism on Wikidata is much more likely to stay undetected, see e.g. here where it took a full month until someone noticed that the page "Punjabi" had been moved (retitled) to "josh saunders", or here where it took more than a month for someone to notice that "Aaron Ramsey" was moved to "Penalty in the UEL final". Fram (talk) 07:37, 30 September 2022 (UTC)
- The reason Wikidata-derived tables have fewer errors is that, when you're adding a row through Wikipedia, you need to add both the data and the formatting, and each of those is a potential spot to mess up. I'm sure we've all seen tables that have an extra column used only in one row because someone put an extra "|-". When you're adding information on Wikidata, however, all you have to worry about is the data. And even there, constraint violations can help identify errors that Wikipedia would not have been able to flag, and data imports can help a single experienced editor add large amounts of high-quality information (rather than relying on piecemeal contributions by many different editors, any one of whom might mess up). {{u|Sdkb}} talk 15:26, 3 October 2022 (UTC)
- That's a rather anti-wiki position you take there. "relying on piecemeal contributions by many different editors" is what makes Wikipedia, and excluding these editors from pages is a good argument against Wikidata lists, not for them. (Never mind the countless times experienced editors made completely incorrect or botched mass updates of info on Wikidata of course). Fram (talk) 16:07, 3 October 2022 (UTC)
- I'm absolutely a fan of crowdsourced editing or I wouldn't be a Wikipedian. What I'm not for is forcing information that can be more efficiently handled in bulk to be handled piecemeal. That's why I oppose the wholesale deletion of template namespace, and also why I support the use of Wikidata. Neither of those things make me anti-wiki. Regarding the potential for errors in Wikidata imports, that potential exists in normal editing, too. I think the anti-wiki position would be to say that we should prohibit a type of editing entirely just because it's not done properly 100% of the time. {{u|Sdkb}} talk 18:18, 3 October 2022 (UTC)
the wholesale deletion of template namespace
Strewth!! Was that actually proposed? When? — GhostInTheMachine talk to me 19:15, 3 October 2022 (UTC)- I'm using it as a hyperbolic analogy here, but there are certainly examples of resistance to template usage for things like census data that I'd say fall at a milder point along the same spectrum. {{u|Sdkb}} talk 20:54, 3 October 2022 (UTC)
- I'm absolutely a fan of crowdsourced editing or I wouldn't be a Wikipedian. What I'm not for is forcing information that can be more efficiently handled in bulk to be handled piecemeal. That's why I oppose the wholesale deletion of template namespace, and also why I support the use of Wikidata. Neither of those things make me anti-wiki. Regarding the potential for errors in Wikidata imports, that potential exists in normal editing, too. I think the anti-wiki position would be to say that we should prohibit a type of editing entirely just because it's not done properly 100% of the time. {{u|Sdkb}} talk 18:18, 3 October 2022 (UTC)
- That's a rather anti-wiki position you take there. "relying on piecemeal contributions by many different editors" is what makes Wikipedia, and excluding these editors from pages is a good argument against Wikidata lists, not for them. (Never mind the countless times experienced editors made completely incorrect or botched mass updates of info on Wikidata of course). Fram (talk) 16:07, 3 October 2022 (UTC)
- The reason Wikidata-derived tables have fewer errors is that, when you're adding a row through Wikipedia, you need to add both the data and the formatting, and each of those is a potential spot to mess up. I'm sure we've all seen tables that have an extra column used only in one row because someone put an extra "|-". When you're adding information on Wikidata, however, all you have to worry about is the data. And even there, constraint violations can help identify errors that Wikipedia would not have been able to flag, and data imports can help a single experienced editor add large amounts of high-quality information (rather than relying on piecemeal contributions by many different editors, any one of whom might mess up). {{u|Sdkb}} talk 15:26, 3 October 2022 (UTC)
- And in at least one proposal to implement a list from Wikidata that I examined, the one list of items here would have been replaced with code that got items from dozens of different pages at Wikidata. In other words, the attack surface for vandalism would have been multiplied dozens of times, and manual checking of Wikidata items would have been dozens of times more difficult than looking at wikitext here. Johnuniq (talk) 09:12, 30 September 2022 (UTC)
- So your "yes, you'll be able to stay on Wikipedia" is actually a "no". Deciding on the current status of such lists based on what might happen one day, perhaps (Wikidata is 9 years old, so it's not as if such changes happen rapidly) is not a good idea. So, @Blueboar: you indeed have to go to Wikidata to edit the info, Wikipedia edit mode won't help you. And Sdkb, your "less likely to have errors" doesn't seem to make much sense either, the elements are created manually at Wikidata or manually at enwiki, no reason why one would be less likely to have errors (on the one hand, Wikipedia has more editors and thus more vandals: on the other hand, vandalism on Wikidata is much more likely to stay undetected, see e.g. here where it took a full month until someone noticed that the page "Punjabi" had been moved (retitled) to "josh saunders", or here where it took more than a month for someone to notice that "Aaron Ramsey" was moved to "Penalty in the UEL final". Fram (talk) 07:37, 30 September 2022 (UTC)
- Taking the eventualist view, with future interface improvements, yes, you'll be able to stay on Wikipedia. Also, it's worth noting that tables generated through Wikidata are less likely to have errors to begin with because there are fewer elements being created manually. {{u|Sdkb}} talk 03:47, 30 September 2022 (UTC)
- I remain extremely sceptical about the value of using Wikidata directly in articles for anything, although obviously compiling a list/table using Wikidata, and then disassociating it, is completely acceptable. As is using it in project space, and potentially talk pages. It makes editing the content nearly impossible for those not used to Wikidata, and it makes watching the page for changes completely impossible. ETA: For instance, List of Welsh mathematicians, linked above as an example, contains entries with three inline sources for things like date of birth, presumably unnecessary (and if the sources disagree, this should be footnoted), and has abbreviated months, which is not Wikipedia style but can't be edited within the Wikipedia page. The repeated edit links are also obtrusive. Espresso Addict (talk) 05:44, 1 October 2022 (UTC)
- Thanks for your comments.
- MOS:DATES suggests that abbreviated dates like 2 Sep 2001 or Sep 2, 2001 are acceptable in tables. I think the full unabbreviated dates would take up too much space in most tables.
- Do you have any suggestions to make the edit links less obtrusive? They are necessary to allow editors to change the values, but are only currently visible to logged in users.
- The maximum number of references is set at 3 and could be reduced further.
- — Martin (MSGJ · talk) 14:32, 1 October 2022 (UTC)
- Thanks for your comments.
- "Sep" as an abbreviation for September looks very odd in UK English, which would appertain to a list of Welsh people. The problem is the lack of easy-to-understand customisation. If I have five sources for a dob, I can choose to include only say no 3 because it is reliable & accessible, or put in two sources because one is highly reliable but not readily accessible, and another less reliable but accessible, &c&c. I have absolutely no idea how to do that within Wikidata, and no desire to have to learn a new and cumbersome editing interface. Also using things like n/a for the death date of a living person feels disrespectful. No clue how to make the edit links less prominent; I dislike them in infoboxes, but there's usually plenty of whitespace there to absorb them. Perhaps if they only showed in edit mode, somehow? And how are logged-out editors supposed to amend things? Espresso Addict (talk) 23:44, 1 October 2022 (UTC)
- @Espresso Addict: absolutely. If you are curating a list/table to such a degree then you will almost certainly want to forgo the convenience of the template and gain the greater flexibility. But 99% of lists are not like this, many of which are a bare list of links without additional information or references. For these, the template can produce a nice looking table with more detail, and is more likely to stay up to date. PS I use "Sep" all the time as an abbreviation for September, and it doesn't look odd to me! — Martin (MSGJ · talk) 15:51, 3 October 2022 (UTC)
- "Sep" as an abbreviation for September looks very odd in UK English, which would appertain to a list of Welsh people. The problem is the lack of easy-to-understand customisation. If I have five sources for a dob, I can choose to include only say no 3 because it is reliable & accessible, or put in two sources because one is highly reliable but not readily accessible, and another less reliable but accessible, &c&c. I have absolutely no idea how to do that within Wikidata, and no desire to have to learn a new and cumbersome editing interface. Also using things like n/a for the death date of a living person feels disrespectful. No clue how to make the edit links less prominent; I dislike them in infoboxes, but there's usually plenty of whitespace there to absorb them. Perhaps if they only showed in edit mode, somehow? And how are logged-out editors supposed to amend things? Espresso Addict (talk) 23:44, 1 October 2022 (UTC)
- Unless it can be altered on ENWP without going to Wikidata, I am in line with the previous consensus that wikidata imported content should not be used in article space except under the very limited exceptions. If it can be amended on ENWP and draw through to Wikidata, all my concerns disappear. Keep in mind those exceptions exist precisely because of the issues with Wikidata, chiefly its another project with its own rules and policies, its own admins, far less active users to combat deliberate vandalism, BLP violations etc. Importing lists that include living people is most BLP-watchers nightmare when you have to go to other projects to rectify it. (The same issues exist with imported commons content but at least thats relatively simple to fix). Being able to alter the content as it displays on ENWP, while on ENWP, without having to go to another project would seem to be something the WMF's tech tech could spend some time & cash doing (hint, its not difficult as anyone who has worked with a data warehouse and multiple databases knows) instead of whatever waste of time they are concentrating on. Only in death does duty end (talk) 16:22, 1 October 2022 (UTC)
- Frankly, I read the spirit of the various RFCs on wikidata as being pretty clearly against their use in making articles - so the claim that a table isn't "article text" and so it is fine to make tables from wikidata ... strikes me as very much a rules-lawyering type of statement. I wish that the proponents of wikidata would quit pushing it in such ways - it doesn't help their "cause". Ealdgyth (talk) 13:29, 3 October 2022 (UTC)
- The past RfCs have explicitly considered Wikidata use in tables and found no consensus on the question, so I'm not sure where you're getting that it goes against their "spirit" other than that you're reading your preferred opinion into them. Your idea of "pushing" can just as easily be flipped: I wish that those who fail to see Wikidata's potential would quit resisting it in such ways. {{u|Sdkb}} talk 15:19, 3 October 2022 (UTC)
- Its more that we do see the potential ... for abuse. And there needs to be either a strong mitigation or exceptional reason in place to ignore that risk. Which when it comes to wikidata, there rarely is. Only in death does duty end (talk) 16:47, 3 October 2022 (UTC)
- Particularly for lists that include BLPs, there is potential for BLP violation going unnoticed; a particularly common form of vandalism is to state a living person is dead, or less malevolently, to believe unreliable sources and assume incorrectly that death has occurred. This happens time and time again, and requires careful oversight. Espresso Addict (talk) 22:13, 3 October 2022 (UTC)
- Its more that we do see the potential ... for abuse. And there needs to be either a strong mitigation or exceptional reason in place to ignore that risk. Which when it comes to wikidata, there rarely is. Only in death does duty end (talk) 16:47, 3 October 2022 (UTC)
- No, Wikidata has been repeatedly banned from the body of the article. Every time a Wikidata-activist tries to shove Wikidata into the body of the article there is always consensus against it, they can't even get consensus for infoboxes. That's stuck in no-consensus. It appears that the only way to stop the recurring and disruptive creeping rollout-contrary-to-consensus by Wikidata-enthusiasts is to entirely shut off the calls to Wikidata in Wikitext itself. As long as it's available they just keep cooking up new ways to shove it out unilaterally. Alsee (talk) 06:50, 7 October 2022 (UTC)
- Additional issue: editing of tables like this in Visual Editor is nearly impossible and gives, for the few things it can do, very poor results. I tested this with List of dams in Tochigi Prefecture. Compared to "normal", non-Wikidata lists, here I can only add a new line at the top, not anywhere else in the list, and the result is badly formatted. I don't know if the WD list template can be changed to solve these issues, but if not, I don't think it is acceptable to introduce new things into Wikipedia which are incompatible with Visual Editing (even though I loathe it, it is used by a fair percentage of people and many new editors, and making it impossible for them to edit a type of articles is not anything that should be tolerated). Fram (talk) 10:04, 13 October 2022 (UTC)
- That could be partially mitigated through proper use of TemplateData. {{u|Sdkb}} talk 14:06, 13 October 2022 (UTC)
- How so? My remark about the template is not that an editor won't know that they need to add the wikidata template or add a Qnumber, but that even with that information, you can't produce a good result in VE. Perhaps I am missing something, but I don't think Templatedata can solve or even mitigate the underlying technical issue. Fram (talk) 14:26, 13 October 2022 (UTC)
- VisualEditor at this point is capable of doing anything with templates that source editor can do, I believe. Having good TemplateData makes the interface easier for editors. {{u|Sdkb}} talk 17:56, 13 October 2022 (UTC)
- Have you actually tested this, e.g. with List of dams in Tochigi Prefecture? Try to add e.g. a new dam in the middle of the list, or to move one of the existing entries up or down. TemplateData won't change anything about this functionality. Fram (talk) 18:32, 13 October 2022 (UTC)
- VisualEditor at this point is capable of doing anything with templates that source editor can do, I believe. Having good TemplateData makes the interface easier for editors. {{u|Sdkb}} talk 17:56, 13 October 2022 (UTC)
- How so? My remark about the template is not that an editor won't know that they need to add the wikidata template or add a Qnumber, but that even with that information, you can't produce a good result in VE. Perhaps I am missing something, but I don't think Templatedata can solve or even mitigate the underlying technical issue. Fram (talk) 14:26, 13 October 2022 (UTC)
- This is a valid point, and I have made a start on creating TemplateData for this template, which can be observed on List of dams in Tochigi Prefecture. I do not know if rows can be added in different places, but I will seek advice on this. — Martin (MSGJ · talk) 12:08, 14 October 2022 (UTC)
- Thanks. Adding entries anywhere in the list is just one of the issues, you e.g. also can't delete the existing ones, and when you add a new line at the top (using the Wikidata template) it only fills the first cell, instead of filling the table as it should (no idea if this list is exhaustive, I stopped testing after this). Templatedata helps at editing the already existing lines (though not removing them or moving them inside the table), but does nothing to make the editing of the table possible. Just tested again at List of dams in Toyama Prefecture, and the new Templatedata is good (e.g. overriding an existing row label), but no solution to the fundamental issue. Fram (talk) 12:47, 14 October 2022 (UTC)
- That could be partially mitigated through proper use of TemplateData. {{u|Sdkb}} talk 14:06, 13 October 2022 (UTC)
- Thanks for the interesting discussion. I came here after participating in Wikipedia:Articles for deletion/List of CryEngine games and Category:Video games by game engine, and I was pointing people to here to see if a bigger discussion could be had. However this discussion is interesting with respect to the deletion discussion, as one suggested solution to prevent information being deleted was actually using WikiData to store the information, and not need categories or lists, and for users who are interested they could pull the data out of Wikidata. I've read with interest everyone's positions, and I agree with user User:Rhododendrites and User:CaptainEek. We should absolutely not force people to use them, but if editors use them appropriately that is also fine. I feel that inboxes and tables are good uses. Perhaps *not* inline text *for the moment* - we need to still make Wikipedia useful to casual editors and this is likely a step too far. However a table in the text is fine. With the original RfC being done in 2013 (as noted above), it is likely a good time to revisit the topic. I think we should use structured data in tables to help WP articles, and the benefit is that using it consistently means that updating information in WikiData updates it on all language Wikipedias (?potentially even other wikis such as WikiVoyage). The downside is single point of failure; however also single point of fixing - this may need other policies e.g. only logged in edits on WikiData. However may be better for consistency than some pages in WP where information on the same topic is markedly different as pages have not all be updated simultaneously. The application to lists for data is something I'm equivocal about. I hate a lot of list pages and think they should be better categorised, however there are supporters and detractors to this method of organisation as well. I do not like deleting referenced information, and if there was a way of archiving and easily searching it via WikiData I would be all for that. It would be nice if in future a "WikiData Explorer" page type was created where we could just pass it a search string to autogenerate data so we could get rid of lists. This is one for the future. However in the meantime I think we need to at least try improving our use of structured data - not ban but also not force its use. And if it doesn't work go back to normal tables. I quite like the WikiData sourced list of dams provided by the OP. Out of interest for User:Fram - I looked at the example you gave at List of learned societies in Australia. It was interesting as I think the table is generated by passing individual IDs - my futurist hat btw would be looking at this and saying in future it should be "table = learned society + based in australia -> autogenerate table with these fields". Why is this not a case of taking the ID code pointing to the wikidata entry out of there? This kind of error would also be made if manually creating the table, and isn't a commentary on why not to use wikidata for information. It's more a comment on how if you're not careful incorrect information can be put in any type of table, and what I get out of this is that if we truly passed a query to a database (if we made sure we had properly structured data) the presented information would be "better". Or have I missed the point here? - Master Of Ninja (talk) 07:42, 14 October 2022 (UTC)
- Thanks for sharing your comments.
- No one is suggesting using data for inline text, and I can't think of a situation when that could be appropriate.
- The reason that each row of the table is produced seperately is to allow human editors to override the data with locally defined content.
- I agree completely with your comments on keeping tables updated.
- There are all sorts of ways to browse Wikidata already, and Wikipedia is not really needed for that. See wikidata:Wikidata:Tools/Visualize data for ideas.
- I can see the possible benefit, if someone is searching for a list which we do not yet have on Wikipedia, of linking to an automatically generated list. But there would be a lot of technical and policy issues to navigate.
- — Martin (MSGJ · talk) 12:21, 14 October 2022 (UTC)
- Thanks for sharing your comments.
- Not only does Wikidata have questionable sourcing and verifiability policies not fully compatible with Wikipedia, but I wonder if folks would consider whether it's worth the trouble to make editing list pages even more complicated than they already are? I hate having to jump back and forth between Wikipedia and Wikidata just to manage interwiki links, and it would be an absolute nightmare if it was allowed to make up significant chunks of actual list content. Just say no to more complex articles, so we can focus more on encyclopedic content and less on formatting or template garbage. As for the idea that Wikidata is easier to edit than Wikipedia... I am highly skeptical of this claim, given that I think the average person couldn't even define what a knowledge graph is, much less find out how to add a statement to a Wikidata page. For new and anonymous editors in particular, it is likely extremely confusing to click edit on a cell in a Wikipedia list and then be taken to the read mode of an entirely different website that 99% of our readers have probably never heard of at all. Steven Walling • talk 20:41, 17 October 2022 (UTC)
- On a procedural note: it seems clear from previous RFCs that existing consensus is toward disallowing Wikidata use in articles. It should almost certainly be removed from any existing lists unless and until a new consensus supporting it develops. Steven Walling • talk 21:01, 17 October 2022 (UTC)
Is Wikipedia a genealogy site, or isn't it, or is it partly, or perhaps a teensy bit?
I've been around now for almost 15 years, logged in, and one thing bothers me especially. Info boxes are supposed, I believe, to give our readers quick, accurate facts about article subjects. But an article about a person born into royalty can only cause rather severe info-box confusion for any reader who is not a genealogist or at least aware of Wikipedia's stubborn and absolute policy of acting like Ancestry, MyHeritage, Geni and other genealogy sites where mothers traditionally always are listed by their unmarried names. The amount of our new/uninitiated readers who make corrections such as Queen Silvia, not Silvia Sommerlath, being her name when her children were born, and who then get reversed due to our stringent maiden-name pseudo-genealogy-site policy, are considerable indeed. It is an absolute fact (such as are sought by readers in our info boxes) that a woman named Silvia Sommerlath did not give birth to the royal couple's 3 children. People who see that and know no better have the intelligent right to assume that the king had an affair with a Ms Sommerlath. That's absurd! Genealogy sites are cleverly formatted so that they inform about couples who had offspring but never, or rarely, allege specifically that Princess Daughter is the daughter of Mother Maidenname. Wikipedia always does that, in all our pertinent info boxes.
Proposal In the info-boxes of people whose mothers were royal, Wikipedia policy should allow for actual facts, not genealogically-formatted idiosyncrasies, in info boxes regarding the names of all mothers when their children were born and it should not automatically be reversed when such facts are given in info-boxes, --SergeWoodzing (talk) 18:02, 3 October 2022 (UTC)
- Oppose Not being a genealogy site means we don't present genealogical information on non-notable people. Royalty is quite a different matter, especially given that titles are inherited, and prosopography is important. Maiden name is useful information; it is far more useful to know that Silvia Sommerlath married King Carl XVI Gustaf that the Queen married the King. She's still Silvia Sommerlath of course; she just uses the title of Queen Silvia. And people who know no better are by definition not intelligent. Hawkeye7 (discuss) 21:10, 3 October 2022 (UTC)
- Or uninformed. Facts: Queen Silvia's surname, legally, is * (that's right an asterisk), as used by the Swedish Tax Authority's official census for members of the royal family who do not use surnames. Were one to apply a surname to her anyway, it would be Bernadotte. --SergeWoodzing (talk) 20:23, 5 October 2022 (UTC)
- Comment: This proposal does not concern who married whom or what their names were then, but what name the mother had when her children were born. That's how we confuse a vast amount of readers with false info-box naming. --SergeWoodzing (talk) 20:23, 5 October 2022 (UTC)
- I'm inclined to agree with SergeWoodzing. The names in infoboxes should be either (a) the name the person was known by at the relevant time [whether legal or not] or (b) the name the person is known by now. In the case of notable people, the current name can be determined by looking at the article title for that person.
- Also, we already make exceptions to this rule; look at the infobox in Kylie Jenner to find an infobox that doesn't use the pre-marriage names of either parent.. WhatamIdoing (talk) 23:31, 5 October 2022 (UTC)
- Oppose I agree with Hawkeye7 (discuss) that "Silvia Sommerlath married King Carl XVI Gustaf...[however] she's still Silvia Sommerlath of course; she just uses the title of Queen Silvia." Pregnancy has nothing to do with it. I have been to many genealogical sites and multiple royalty listing sites and have never seen the claim that a woman should be recorded by "...what name the mother had when her children were born". I vote against female fade-out. Thank you. Wordreader (talk) 06:19, 8 October 2022 (UTC)
- I can't see how it would be to fade a female out who was queen when her children were born and no longer just an olympic hostess whose surname used to be Sommerlath. Are women to be defined by their maiden names rather than their actual statuses? --SergeWoodzing (talk) 19:38, 8 October 2022 (UTC)
Section tag creation
The tag {primary sources|date=} informs that "This article relies excessively on references to primary sources". Can someone create the same tag by replacing "article" with "section"? Thank you, Manamaris (talk) 10:08, 7 October 2022 (UTC)
- @Manamaris You can use {{Primary sources section}} for this. the wub "?!" 10:17, 7 October 2022 (UTC)
- Thank you!--Manamaris (talk) 10:37, 7 October 2022 (UTC)
- @Manamaris and The wub: You can also just do
{{Primary sources|section}}
, as with most such templates. I've never really understood why people insist on making separate templates for the purpose. Anomie⚔ 11:45, 7 October 2022 (UTC)- @Anomie they prob got build in parallel; someone could merge them and make a redirect. Template:Primary sources/doc doesn't even mention the
|section
directive, and actually tells editors to use {{Primary sources section}}. As long as it is seamless for editors via redirects, merging should be uncontroversial. — xaosflux Talk 13:53, 7 October 2022 (UTC)- I'd suggest converting it into wrapper instead of redirecting, so that existing transclusions and existing format works post-transclusion. —CX Zoom[he/him] (let's talk • {C•X}) 14:19, 7 October 2022 (UTC)
- Yes, wasn't getting in to the details but if a straight redirect won't work a wrapper - then future updates only need to happen in one place. — xaosflux Talk 16:45, 7 October 2022 (UTC)
- It already is a wrapper. The "section" version was created in 2015, as a wrapper even then. The doc page at Template:Primary sources/doc mentioned "section" until July of this year when User:Mathglot rewrote it. Anomie⚔ 20:33, 7 October 2022 (UTC)
- Still does (twice), in sections §§ Parameters and Details. Mathglot (talk) 21:47, 7 October 2022 (UTC)
- @Manamaris:, if the current mentions of "section" (the parameter value is actually any string, section, paragraph, table, whatever you want) iin the template doc aren't prominent enough, you can fix it; everyone has access to the doc page. Perhaps all you need, is another line under section § Usage that includes it, or a new example in the § Examples section. Feel free to change it however you see fit. See the top of Template:Neologism for an approach which may be more to your liking. Mathglot (talk) 23:54, 9 October 2022 (UTC)
- It already is a wrapper. The "section" version was created in 2015, as a wrapper even then. The doc page at Template:Primary sources/doc mentioned "section" until July of this year when User:Mathglot rewrote it. Anomie⚔ 20:33, 7 October 2022 (UTC)
- Yes, wasn't getting in to the details but if a straight redirect won't work a wrapper - then future updates only need to happen in one place. — xaosflux Talk 16:45, 7 October 2022 (UTC)
- I'd suggest converting it into wrapper instead of redirecting, so that existing transclusions and existing format works post-transclusion. —CX Zoom[he/him] (let's talk • {C•X}) 14:19, 7 October 2022 (UTC)
- @Anomie they prob got build in parallel; someone could merge them and make a redirect. Template:Primary sources/doc doesn't even mention the
- @Manamaris and The wub: You can also just do
- Thank you!--Manamaris (talk) 10:37, 7 October 2022 (UTC)
- Adjusted the doc. Also trying to fix the bug with BLP and Section together. OK now? — GhostInTheMachine talk to me 11:30, 10 October 2022 (UTC)
Deletion discussions for lists and categories of games by game engine
Hello, there are two discussions going on whether we should keep or delete lists or categories of games by game engine, or both: categories, lists.
I notified about both of them in WikiProject Video games, but as there are a large number of lists and especially categories involved, it is a good idea to attract more people. Respiciens (talk) 11:27, 14 October 2022 (UTC)
Category for Summer Olympic host citiesh
Hello! Can you add a category for Summer Olympic host cities. The category should then be named Host cities for Summer Olympics and the same should count for Winter Olympics then it should be named Host cities for Winter Olympics. Yours sincerely, Sondre 88.88.4.178 (talk) 14:52, 19 October 2022 (UTC)
- The host cities display {{Olympic Summer Games Host Cities}} or {{Olympic Winter Games Host Cities}} at the bottom. I don't think a category is needed. It doesn't seem important enough for a large city to clutter up the list of categories. PrimeHunter (talk) 15:59, 19 October 2022 (UTC)
- Per WP:CLN, not every thing needs a category. Many major cities (basically all of the host cities are major cities) already suffer from overcategorization, and adding one more to the mix is a Bad Idea. People can find the list in many ways, such as at the navigation templates linked by PrimeHunter, or List of Olympic Games host cities. Categorization is not required, and not needed, here, as a navigational tool. --Jayron32 16:10, 19 October 2022 (UTC)