Jump to content

Wikipedia:Village pump (miscellaneous)

From Wikipedia, the free encyclopedia
 Policy Technical Proposals Idea lab WMF Miscellaneous 
The miscellaneous section of the village pump is used to post messages that do not fit into any other category. Please post on the policy, technical, or proposals sections when appropriate, or at the help desk for assistance. For general knowledge questions, please use the reference desk.

Discussions are automatically archived after remaining inactive for a week.

« Archives, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78

Inexplicably popular article (by views)[edit]

Neatsville, Kentucky in April was the 2nd most viewed Kentucky-related article and has been similarly highly viewed for several months. I cannot make sense of this. This is a small unincorporated community in the middle of rural Kentucky. I cannot find any TV show or movie referencing it. It also doesn't make sense that anyone would be gaming this outcome for months (although I suppose this isn't impossible). Am I missing something? Stefen Towers among the rest! GabGruntwerk 21:00, 28 May 2024 (UTC)[reply]

Fascinating. Two-year pageviews are even higher on average, peaking in mid-2023. I see no news coverage or anything else that would drive this traffic. BD2412 T 21:28, 28 May 2024 (UTC)[reply]
The start of this climb in pageviews seems to have been on 24/25 August 2021 ([1]), when daily pageviews climbed from 2 to 410 to 1,717. Perhaps this may narrow the search for what is causing this. Curbon7 (talk) 22:39, 28 May 2024 (UTC)[reply]
Billy Joe in the same Kentucky county announced he saw a UFO on 8/24. LOL. Stefen Towers among the rest! GabGruntwerk 23:47, 28 May 2024 (UTC)[reply]
Also, nearly all of the traffic coming to the article is from unidentified external routes (which is highly unusual), and there is virtually no traffic from this article to other articles (also highly unusual). BD2412 T 22:02, 28 May 2024 (UTC)[reply]
Maybe there's a viral post or tweet somewhere with an easter egg? Schazjmd (talk) 22:07, 28 May 2024 (UTC)[reply]
Possibly. Although I've not heard it, I can easily imagine a meme in which "Neatsville" (a redirect to the article) becomes a trendy term of approval. (Compare Coolsville.) Alternatively, someone may be trying to get it into a most-viewed listing. It would be interesting to know how many different IPs have accessed the article (perhaps counting each IPv6 /64 as one), rather than just the number of hits. Certes (talk) 22:20, 28 May 2024 (UTC)[reply]
Redirects seem to be negligible in their impact. Unchecking "Include redirects" makes virtually no difference. Regarding someone gaming this, that's an awful lot of such to sustain. Of course, this could be a script disguising itself as a real person. Stefen Towers among the rest! GabGruntwerk 22:35, 28 May 2024 (UTC)[reply]
Thanks for the pointer on redirects: I hadn't spotted that. Yes, I assumed it was scripted. It does seem erratic and slightly seasonal, with peaks in spring 2023 and 2024, but does not vary much by day of week. Certes (talk) 22:49, 28 May 2024 (UTC)[reply]
That crossed my mind, but I think the incoming traffic would be more varied and identifiable for something like that, rather than a dark web monolith (speculation before further details). Stefen Towers among the rest! GabGruntwerk 23:23, 28 May 2024 (UTC)[reply]
This sounds like a repeat of Mount Takahe, which also has inexplicably high reader numbers. And like Takahe, Neatsville has fairly average reader numbers when only counting the Mobile App and only slightly elevated reader numbers with by spiders. FWIW, neither News nor Twitter/X show many if any mentions. Jo-Jo Eumerus (talk) 07:18, 29 May 2024 (UTC)[reply]
This is getting really ridiculous. It's skewing statistics, even to the point where new editors are noticing. I don't want make this into some huge problem, but I think "nipping it in the bud" is well called for now. Please admins block the access of this apparent script kiddie. Stefen Towers among the rest! GabGruntwerk 21:51, 3 June 2024 (UTC)[reply]
I have logged a case in WP:ANI. Stefen Towers among the rest! GabGruntwerk 22:10, 3 June 2024 (UTC)[reply]
Admins do not have the ability to block people from viewing articles, this would have to be handled by the system administrators. You would probably be best filing a ticket on Phabricator, though I'm not sure they'd take action. 86.23.109.101 (talk) 22:53, 3 June 2024 (UTC)[reply]
I'm not sure what action can or should be taken. This doesn't seem to be a denial-of-service attack (or, if it is, it's an incredibly lame one). Wikipedia's terms of service don't prevent anyone from viewing pages, even multiple times; in fact it's encouraged. I don't know whether the hosting system can, or should, rate-limit a particular IP address or range, even assuming that most of the unusual traffic comes from one IP or a small range. Certes (talk) 23:13, 3 June 2024 (UTC)[reply]
Indeed. I wouldn't be reporting this as a performance or security issue, but rather a data corruption issue. And I sense this might not be taken very seriously, but I have a thing against the presentation of false data and that in that presentation, the person doing it is getting away with it, possibly encouraging more of this kind of corruption by others. I think it is in our long-run interests to stop it or put some kind of brakes on it. Stefen Towers among the rest! GabGruntwerk 23:52, 3 June 2024 (UTC)[reply]
If this is due to a malicious botnet, shouldn't you have WMF report this to law enforcement? –LaundryPizza03 (d) 01:20, 4 June 2024 (UTC)[reply]
I don't know if it's malicious. It's just skewing our cumulative views data on a single article. I might rather have an ISP notified if that could be pinned down. Stefen Towers among the rest! GabGruntwerk 02:10, 4 June 2024 (UTC)[reply]
The internet can be a bit of a wild west sometimes. I don't think calling the police to report a DDOS attack would result in anything. DDOS attacks are usually carried out by hacked zombie computers, and are often transnational. So it's a bit hard to police. –Novem Linguae (talk) 07:43, 4 June 2024 (UTC)[reply]
An inexplicable steady increase in readership to an article happened one time before, and the explanation was that it had been included as an example/default link somewhere. Will see if I can find the details. –Novem Linguae (talk) 23:20, 3 June 2024 (UTC)[reply]
That's a possibility if it's not a link from English Wikipedia but another project or website. I had already reviewed EN pages linking to the article and didn't see anything. Thanks for checking. Stefen Towers among the rest! GabGruntwerk 23:46, 3 June 2024 (UTC)[reply]
It's tempting to put a banner on the top of the article: "Please tell us what brought you to this article" with a link to the talk page, see if any of the 17,000+ readers answer. Schazjmd (talk) 23:49, 3 June 2024 (UTC)[reply]
Many years ago I found – guess how – that the address anton@pobox.com was used as an example in what appeared to be a guide to email for new users (in Russian, but hosted in Israel). —Tamfang (talk) 22:17, 12 June 2024 (UTC)[reply]
Found this through some searching, not really sure where it came from: urlscan1: Kepler's Supernova article, urlscan2: Neatsville, Kentucky article. The scan was for a different url, which redirected to those Wikipedia pages with some (ad tracking?) parameters. – 2804:F1...99:B28F (talk) 05:48, *edited:06:42, 4 June 2024 (UTC)[reply]
Mind you, the interesting thing would have been to know where that original link was from (possibly emails? unsure) - both were scanned on the 17th of last month and both articles have an increase in views, but without knowing where that's from and if it always redirects there, it doesn't really mean it's even related with the view count unfortunately. – 2804:F1...99:B28F (talk) 06:42, 4 June 2024 (UTC)[reply]
Thanks for bringing this here. Is it fair to say that Kepler's Supernova is also getting the same kind of fake views? Or could its extra recent views have a legitimate reason behind it? Stefen Towers among the rest! GabGruntwerk 07:03, 4 June 2024 (UTC)[reply]
Not that I could find, both noticeably grew in views since April: Kepler's Supernova, Neatsville, Kentucky
According to wikitech:Analytics/AQS/Pageviews#Most viewed articles the most viewed list (same data as the graphs) tries to only count page request from "human users", so it's not clear if the views are fake, though a reason is also not obvious. Do you know why the Neatsville article had similar numbers in from March to June of last year? – 2804:F1...99:B28F (talk) 08:21, 4 June 2024 (UTC)[reply]
I have no idea, and I'm in Kentucky. This place really is "in the sticks". Stefen Towers among the rest! GabGruntwerk 08:34, 4 June 2024 (UTC)[reply]
Talk page for Kepler's Supernova says Publishers Clearing House for some reason included a link to [the page] in email (promoting daily contests) for awhile. Page view patterns are the same as with Neatsville. Not sure if this IP is relevant either 107.128.181.22 (talk) 08:31, 4 June 2024 (UTC)[reply]
Publishers Clearing House for some reason included a link to [the page] in email (promoting daily contests) for awhile. This seems like the most plausible explanation so far. –Novem Linguae (talk) 12:10, 4 June 2024 (UTC)[reply]
I have reported this as a security issue (re: data integrity) to Phabricator. Stefen Towers among the rest! GabGruntwerk 06:54, 4 June 2024 (UTC)[reply]
It might be very helpful to know how many different IP addresses access the page a lot (say >100 times a day) and whether they're in a single range. Obviously this requires access to non-public information, but it should be safe to pass on a digest with the actual IPs removed. Certes (talk) 11:04, 4 June 2024 (UTC)[reply]
Update: Neatsville, Kentucky in May was the top most viewed Kentucky-related article. This effectively trashes the point of having a Popular pages list. There are bigger things to be outraged about in this world, but as far as Wikipedia goes, this really honks me off. Stefen Towers among the rest! GabGruntwerk 17:00, 10 June 2024 (UTC)[reply]
The number of views 26k is so low it could easily be explained by a default link somewhere. The Publishers Clearing House explanation given above sounds reasonable, or something like it. These kinds of things are not uncommon. If the popular pages list is important, you could modify the list with another bot. -- GreenC 17:20, 10 June 2024 (UTC)[reply]
This isn't a very recent phenomenon. The views have been skewed off and on since over a year ago (see "Two year pageviews..." link above). Also, the explanation as such doesn't absolve this as not being a problem. There is no excuse for PCH or any entity for sending non-purposeful (junk) links to people. Whether or not it affects our system performance, it is abusive. As far as modifying Popular pages results, if there was a straightforward way to asterisk, strikethrough, hide or shade an entry based on particular criteria, that would suffice, but writing a new bot seems overwrought. I could temporarily strikethrough, hide or shade the top or nth entry via CSS but then that would require monthly maintenance. I think I'll just write a nasty letter to PCH - that may be our real solution (half-joke, half-serious). Stefen Towers among the rest! GabGruntwerk 20:47, 10 June 2024 (UTC)[reply]
Meh, somebody put a link in an email or newsletter or something. That doesn't strike me as abusive; if people are clicking the links and reading our article that's really no different than anyone who sees one of our articles through a link in a tweet or Discord, that page was popular. It doesn't seem like there's anything to be done. The WordsmithTalk to me 21:24, 10 June 2024 (UTC)[reply]
We'll just have to disagree on this. They had no business skewing views to these articles. What on earth is the purpose? These are not legitimate views. Stefen Towers among the rest! GabGruntwerk 22:37, 10 June 2024 (UTC)[reply]
Wait one min-u-ette here. If these are all genuine human visits off an e-mail or promotion, how come I'm the only one to edit the article (once) since September? With the huge amount of visits, that seems to defy reason. For a small rural town, it has a kind of interesting story, having been relocated twice – so it's weird that edits wouldn't have happened. These are highly likely bot hits disguised as human hits. That's not a problem?? Stefen Towers among the rest! GabGruntwerk 22:57, 10 June 2024 (UTC)[reply]
Is it possible for Wikipedia articles to be embedded into a webpage, and if so, is it possible these collect pageview data without people clicking through? Curbon7 (talk) 23:43, 10 June 2024 (UTC)[reply]
Yes (<iframe>) and yes. Probably uncommon though. –Novem Linguae (talk) 23:53, 10 June 2024 (UTC)[reply]
No. it's not a problem. Who cares why any of our articles are read and who by? Phil Bridger (talk) 06:36, 11 June 2024 (UTC)[reply]
Thanks for your opinion, but it's not as simple as that. This is systems data used beyond the superficial aspect that you imagine. Note that if views data wasn't important, it wouldn't be collected and stored in the first place. It can be used for various purposes, like for instance, project prioritization. Corruption of data is a real problem. I am not suggesting this specific issue reported here is a huge problem but one that should be addressed lest it really get out of hand. Stefen Towers among the rest! GabGruntwerk 06:53, 11 June 2024 (UTC)[reply]
I agree with Phil. Usually website backlinks are a good thing, for search engine optimization and brand awareness reasons. If it causes one aberrant data point in one report, that's fairly minor. –Novem Linguae (talk) 07:13, 11 June 2024 (UTC)[reply]
Through my background in database development and 20 years as a Wikipedian, I insist it's a real (though not currently huge) problem by what I've already stated. Also, there seems to be an insistent assumption these are true views. Based on information that's been made available, the strong suggestion is that these are effectively bot hits. Also, I highly doubt we are getting SEO benefits from distributed junk hits, and who doesn't already know our brand? The bottom line is this has a potential to really bollocks up various processes that use this data if it isn't nipped in the bud. "Fairly minor" is today. But tomorrow? Yeah, let 'em increasingly tarnish our data. Cool, man, cool. Stefen Towers among the rest! GabGruntwerk 07:32, 11 June 2024 (UTC)[reply]
We really need to ask someone with access to private logs whether these views come predominantly from one IP (or a small range) or are widespread. If the latter then they may also be able to tell us (perhaps from the referrer) whether they are predominantly from one webpage, perhaps via an iframe embedded in HTML bulk e-mail. Certes (talk) 09:53, 11 June 2024 (UTC)[reply]
I concur. That's a part of why I logged the issue in Phabricator, so that an investigation can be conducted. Stefen Towers among the rest! GabGruntwerk 19:37, 11 June 2024 (UTC)[reply]
Also, I realize that when I said "distributed", I was buying into an assumption but yes, it's possible this comes from one IP or a small range. Stefen Towers among the rest! GabGruntwerk 19:39, 11 June 2024 (UTC)[reply]
Even that's not absolute proof. A significant portion of our page views from mainland China come to use through just two (2) IP addresses (used by a VPN service). If you find that most of the traffic comes from a single IP, that does not mean that a single person is reloading the same page every few seconds round the clock. It could mean that a lot of people are using a VPN or other shared service.
You might also be interested in https://theconversation.com/2022-wasnt-the-year-of-cleopatra-so-why-was-she-the-most-viewed-page-on-wikipedia-197350 and similar reports. WhatamIdoing (talk) 15:41, 12 June 2024 (UTC)[reply]
I guess my view doesn't count because I have only been editing Wikipedia for 17 years and my background is in systems programming, but I'll state it anyway. It is that the only problem here is with people who place too much faith in reports. Measure what you actually want to measure, not what's easiest to measure, and don't try to change what you're measuring to make it easier. Phil Bridger (talk) 20:47, 11 June 2024 (UTC)[reply]
Some reports do weed out automated views, sometimes by limiting their scope to articles which have between 5% and 95% of their views from mobiles. (Example: Signpost.) This technique is helpful but not foolproof, especially if someone who reads the report is trying to appear on it in some sort of SEO game. Certes (talk) 21:51, 11 June 2024 (UTC)[reply]
As someone with a similar background, these are the kinds of arguments you find in IT departments, I suppose. The report isn't the problem but rather the report is indicative of a data problem, and it's the data problem that should be solved, because that problem could increase and cause other issues. And yes, we should change what we're measuring, rather, prevent bad data input (the case here), because you don't want "garbage in". Spending time to assure clean data going into further processing in other systems was a significant part of my IT work. Stefen Towers among the rest! GabGruntwerk 17:01, 17 June 2024 (UTC)[reply]
I am inclined to concur that we aren't looking at genuine readers here - few people seem to go from Neatsville to other articles. Compare Donald Trump, where almost all readers then go on to read other articles. That might be an iframe deal or a bot, but not people directly reading the article. Jo-Jo Eumerus (talk) 07:45, 11 June 2024 (UTC)[reply]
Of course we aren't. But what does any of this have to do with Wikipedia? Phil Bridger (talk) 20:47, 11 June 2024 (UTC)[reply]
There are lots of people who are interested in how widely shared information on a given Wikipedia page is. That tells us something about which topics are important, which ones need to be taken care of etc. Distributing information is the purpose of a Wikipedia page after all. Jo-Jo Eumerus (talk) 06:01, 12 June 2024 (UTC)[reply]
This may be related (or unrelated), but my talk page received an unusual number of page views each day from late March to early April: see here. Besides a couple of messages from the bots, there weren't any other activity on my talk during that time [2]. I doubt those page views (at least on my talk page) are genuine. Some1 (talk) 03:40, 18 June 2024 (UTC)[reply]

Rename the article?[edit]

This constant pinging of our article could easily be disrupted by renaming the article without leaving a redirect, if only for a day or two. Of course that might still count as vandalism, and make Skynet very angry. NebY (talk) 19:02, 16 June 2024 (UTC)[reply]
  • Note: I have also temporarily changed all incoming links from reader-facing spaces. In case anyone is concerned that the above move will impede legitimate searches for this title, it is now the first search result that comes up when searching for Neatsville, Kentucky. Cheers! BD2412 T 16:25, 17 June 2024 (UTC)[reply]
    Neat! And nicely done. I'm looking forward to seeing how it turns out; already I've been surprised to find that the iOS app is still giving me the article at Neatsville, Kentucky (but a blank talk page and an "unexpected response from the server" for its history) as well as at Neatsville, Adair County, Kentucky. Maybe I'm the last to learn the iOS app uses a different database, which lags. NebY (talk) 20:38, 17 June 2024 (UTC)[reply]
Thank you both for the concept and implementation. I imagine at the very least the results will add to our body of knowledge. Stefen Towers among the rest! GabGruntwerk 17:22, 17 June 2024 (UTC)[reply]
Seems a bit odd to let off-site pressures dictate the titles of our articles. Also if the Publisher's Clearing House explanation is accurate, we have now broken this link for regular users. Also may be a violation of WP:PMRC. –Novem Linguae (talk) 01:14, 18 June 2024 (UTC)[reply]
This is effectively an experiment to determine whether moving the article — for one week — resolves the issue that has been reported. It may well be that these views are the result of an internal glitch rather than on off-site one, and this resolves that all the same. It may be that when the article is moved back, the issue will resume. The only way to find out is to perform the experiment and gather the data for analysis. As noted, the correct article is still the number one article that comes up when using the search function, and given the page views prior to this situation arising, actual inconvenience to regular users should be nominal. BD2412 T 02:50, 18 June 2024 (UTC)[reply]
I would oppose a permanent rename for a flimsy reason like that, but all along, this was set up as a one-week test, and I don't see a big problem there. Anyway, I saw that Neatsville, Kentucky was redirected after this test was started, so I wonder if that defeats the point of the test. Stefen Towers among the rest! GabGruntwerk 02:58, 18 June 2024 (UTC)[reply]
I don't think it will defeat the point, since pageviews of redirects are tracked separately from pageviews of their targets. But then I could be misunderstanding. * Pppery * it has begun... 03:04, 18 June 2024 (UTC)[reply]
Yes that's true, but now these miscreant/fake hits will be hitting a live mainspace page that happens to be the same page they were targeting before. So, they won't be getting any indication they are hitting a nonexistent page like they would have when the test was first set up. Stefen Towers among the rest! GabGruntwerk 03:10, 18 June 2024 (UTC)[reply]
(edit conflict) The point of moving without the redirect was to see whether the absence of anything at this target would "break" whatever is causing the excessive page views. Perhaps the few hours during which there was no redirect was enough to do that. The test does not have to run for a week, that was an arbitrary time set figuring that whatever process was involved might itself be on a week-long clock. Maybe a few days would do. BD2412 T 03:11, 18 June 2024 (UTC)[reply]

New data: This is remarkable. Two days after moving the article, Neatsville, Kentucky continues to average close to 20,000 pageviews per day, but Neatsville, Adair County, Kentucky is averaging 50 pageviews per day. Anyone actually navigating to the Neatsville, Kentucky link would be redirected to Neatsville, Adair County, Kentucky, which should therefore also have those tens of thousands of views. This definitively means that visits to Neatsville, Kentucky are not organic views from regular readers, but are queries of the URL itself that therefore do not get redirected. BD2412 T 17:15, 19 June 2024 (UTC)[reply]

That's maybe not so clear; I find that if I click "include redirects" then Neatsville, Adair County, Kentucky is receiving 20,000+ pageviews a day[3]. On the other hand, toying with the Agents setting gives me another puzzle. Over the last 90 days, the ratio of "User"[4] to "Automated"[5] views of Neatsville, Kentucky varied from 1:1 to 8:1 and more, but both peaked on 01 June 2024. Even assuming some views misidentify themselves, I can't even start to explain both the variation and the coincidence. NebY (talk) 18:26, 19 June 2024 (UTC)[reply]
My understanding (and this may be incorrect) is that including redirects merely adds the number of views to the page and the number of views to the redirect. I do not believe it is possible to have a view of the redirect that results in the viewer being redirected to the page, but does not also lead to a view to the page itself, such that pageviews alone should always be higher than redirect views alone (compare pageviews of "FBI" versus "Federal Bureau of Investigation"). BD2412 T 19:09, 19 June 2024 (UTC)[reply]
Ah, that it's simple addition does make sense. Still, the FAQ does say If a user browses to a redirect, a pageview is registered for the redirect but not for the target page. That suggests to me that it's technically feasible that ~20,000 human readers went to Neatsville, Kentucky, were redirected, and did read Neatsville, Adair County, Kentucky - but I've little experience of this tool, could be very wrong. NebY (talk) 19:42, 19 June 2024 (UTC)[reply]
I have yet to find a working redirect that has more pageviews than the page to which it redirects. BD2412 T 04:12, 20 June 2024 (UTC)[reply]

Hatnote styling?[edit]

Did the hatnote stylesheet change? It seems to have become harder to read and a different font color in the last few days. -- 64.229.90.32 (talk) 06:40, 14 June 2024 (UTC)[reply]

See discussion at Wikipedia:Village pump (technical)#Hatnotes have Minerva-style background color? PamD 07:09, 14 June 2024 (UTC)[reply]

Kurds and the rule of crowd[edit]

— Preceding unsigned comment added by Novem Linguae (talkcontribs) 15:00, 14 June 2024 (UTC)[reply]

Wikipedia in Fiction[edit]

Nature magazine has been running a series of Science Fiction short stories called "Futures". The latest one -- "Plastic-eating fungus caused doomsday:[2][3] A collaborative effort" -- is told as a series of entries to a Wikipedia talk page. Never thought of Wikipedia as a genre. AFAIK, this is the first Wikipedia fiction -- not counting hoax articles, of course.

I don't know how long this link will be good, so I downloaded a pdf copy of this story in case it goes away. -- llywrch (talk) 02:44, 17 June 2024 (UTC)[reply]

Good find, thanks! Gråbergs Gråa Sång (talk) 07:59, 17 June 2024 (UTC)[reply]
@Llywrch see lena by qntm Mach61 23:20, 17 June 2024 (UTC)[reply]
I have to do this: [1]

References

  1. ^ Burnett, Emma (12 June 2024). "Plastic-eating fungus caused doomsday[2][3]". Nature. doi:10.1038/d41586-024-01723-z.

Gråbergs Gråa Sång (talk) 08:11, 17 June 2024 (UTC)[reply]

AFAIK, this is the first Wikipedia fiction Nope! Pre-dated by works like Neurocracy (2021), Missing Links and Secret Histories (2013), and I'm sure there's many more examples. – Teratix 10:02, 17 June 2024 (UTC)[reply]
@Gråbergs Gråa Sång, Llywrch, and Teratix: I have redirected Wikipedia in fiction to Wikipedia in culture. If the use of the above mentioned works is discussed in sources, it would be worth adding mention of them to that article. BD2412 T 19:49, 17 June 2024 (UTC)[reply]
I'm surprised that I missed Missing Links and Secret Histories, since I've read every issue of the Signpost since its creation years ago. But my oldest daughter was 6 at the time & having children that young limits every activity outside of work, eating & sleeping -- & sometimes the last two are also affected. -- llywrch (talk) 06:57, 18 June 2024 (UTC)[reply]

If there's any doubt that we need more sysops...[edit]

...I've made a graph comparing stats across English projects. Active users per admin is in blue. Cremastra (talk) 20:21, 18 June 2024 (UTC)[reply]

A better comparison would be the other major Wikipedia's (German, French, Japanese, etc). BilledMammal (talk) 20:27, 18 June 2024 (UTC)[reply]
Given that its users/admin, as opposed to straight-out number of admins, I'm not sure I see much of a difference. Cremastra (talk) 20:29, 18 June 2024 (UTC)[reply]
I think there are a lot of different variables to account for when comparing non-Wikipedia projects to Wikipedia projects that makes a direct comparison tricky. Ultimately, though, English Wikipedia has a lot of editors paying attention to it, and so out of necessity its community has tried to develop processes to handle large numbers of edits. (As I've stated previously, I do think that more administrators are needed in order to feed the pipeline of experienced users.) isaacl (talk) 02:12, 19 June 2024 (UTC)[reply]
That doesn't demonstrate anything much. It could just as easily be used to argue that (a) other projects need more sysops per user, or (b) other projects don't need more sysops per user, they just make it available to more people as a proportion. AndyTheGrump (talk) 20:46, 18 June 2024 (UTC)[reply]
Indeed. It might also indicate that we have well-developed dispute resolution systems, effective patrolling by non-admins, a strong culture of mutually advising and warning editors, and/or more efficient sysops, due in part to those other factors. NebY (talk) 21:47, 18 June 2024 (UTC)[reply]
Why is users-per-sysop a valuable metric? Admins are not responsible for users in the same way a schoolteacher or prison guard is responsible for their charges. What you should be measuring is perhaps troubled users (vandals, etc) and required admin tasks (page protection, blocks) per admin. If you have 1 million users who behave and 3 admins, who cares? RudolfRed (talk) 01:31, 19 June 2024 (UTC)[reply]
More users, in general = more admin tasks. Cremastra (talk) 01:33, 19 June 2024 (UTC)[reply]
It would be interesting to compare the average admin workload (i.e. number of logged actions per admin) on different projects. RoySmith (talk) 02:10, 19 June 2024 (UTC)[reply]

Help for article creation[edit]

I want to create article about Genetic studies on Bengali. I created many historical, political or other simple articles so I know how to find sources for these topics. But I am clueless about source indexing of genetic studies in search engines and I am having difficulties finding sources about genetic studies of Bengali people. Do you people have any idea how to find genetics-related sources effectively? Or can anyone redirect me to anyone who can really help me about the sources? Mehedi Abedin (talk) 04:33, 19 June 2024 (UTC)[reply]

Mehedi Abedin, you might have better luck asking at WP:RD/S. The people there will likely know how to find research materials about genetic studies. Folly Mox (talk) 05:37, 19 June 2024 (UTC)[reply]

Is this a Wikipedia error now appearing as an RS?[edit]

Until this edit[6], Austronesian peoples stated that Madagascar was reached by Austronesians c. 50–500 CE ("In the Indian Ocean, they sailed west from Maritime Southeast Asia; the Austronesian peoples reached Madagascar by ca. 50–500 CE"). Looking at that version[7], there are three references with failed verification tags. The last two references have been associated with this part of the article for some time. Precisely what is wrong with the references is discussed on the talk page[8] (It's the third of three issues under this heading.) Note these last two references, The Culture History of Madagascar and A chronology for late prehistoric Madagascar.

In a recent version[9] of the article, we have two reintroductions of early dates for Austronesian settlement of Madagascar. The one that says Estimates for when this occurred vary from the 1st century CE,[153]... has a new ref, Herrera et al. If you read that paper, you find, in the introduction:

  • "[Austronesians] ...colonized Madagascar during ca AD 50–500 [1,2]."

Checkout those references – they are The Culture History of Madagascar and A chronology for late prehistoric Madagascar. Since neither of them support this date range, where did the authors of the paper get this from? There are plenty of other potential references in the literature (giving later dates). So why use these? The most likely explanation is that they were lifted straight from the Wikipedia article that existed at the time.

So, do we have a Wikipedia error (since corrected) now being recycled back into the article to perpetuate the error? What do you think? ThoughtIdRetired TIR 22:42, 19 June 2024 (UTC)[reply]

Link: Herrera et al. 2017.
It does look like citeogenesis. As far as I can tell the A chronology for late prehistoric Madagascar source is unrelated to the statement made. CMD (talk) 04:43, 20 June 2024 (UTC)[reply]

Wikipedia on visionOS[edit]

I neither have nor can afford an Apple Vision Pro at this time, and I don't think most of us do or can as well. I'd like to know, from those using the device, how Wikipedia on Safari functions as well as the Wikipedia mobile app there. George Ho (talk) 21:33, 20 June 2024 (UTC)[reply]

@SGrabarczuk (WMF), is this something that the Web team has tested? WhatamIdoing (talk) 22:46, 20 June 2024 (UTC)[reply]
Hey everyone, thanks for the ping @WhatamIdoing.
Not yet! Compared to our total traffic, few people use this device. We had a talk about it soon after it was launched, but I guess we're still far from the decision to buy one. SGrabarczuk (WMF) (talk) 00:37, 22 June 2024 (UTC)[reply]
They should make the interface shaped like a puppy and rotating on a few axes. Like three. Zanahary 07:42, 21 June 2024 (UTC)[reply]

Getting a list/count of values of a parameter.[edit]

(not real example). Template Infobox Guinea Pig has a parameter called "Tail color". I'd like to be able to see what values this parameter has over all articles that use the infobox. (I asked on the help desk and got no response)Naraht (talk) 21:46, 20 June 2024 (UTC)[reply]

I think the data only flows in one direction. You can double check at WP:VPT if you want. That's probably the best board for this question. –Novem Linguae (talk) 22:05, 20 June 2024 (UTC)[reply]
I am sad that there is no {{Infobox guinea pig}} template. Not sure it is quite what you want, but Bambots tracks parameters for all sorts of templates. It can list the different values used for a param. See the documentation and an example (click on "Click to show"). The template needs to have TemplateData set up — GhostInTheMachine talk to me 11:06, 23 June 2024 (UTC)[reply]