Wikipedia:Village pump (miscellaneous)
| 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, no archives yet (create) |
Whether to put common name first or full official name first when it comes to country or political-entity articles
 |
|
We currently have United States starts with the official name, as the United Kingdom, while having various countries or political-entities start with their common name first. I presume the rationale for full official name would be that the main article title at the top is already the common name, therefore new information takes precedency. However, I'd like to request community's opinion on this since it seems rather incoherent throughout the site. Lolitart (talk) 10:06, 14 November 2021 (UTC)
- I don’t think it matters as long as both are clearly mentioned in the opening sentence. Blueboar (talk) 12:47, 14 November 2021 (UTC)
- Not sure what examples are but I don't think there are any issues with that. You probably want to read WP:COMMONNAME. There is consensus on that already. It takes precedence. AXONOV (talk) ⚑ 15:55, 14 November 2021 (UTC)
- Arguable not a case for common name which only concerns the article title. The question is since the common name is already at the top of the page, it seems more reasonable, as in the case of United States and United Kingdom, to use the full name to open the paragraph as that provides more information, but I'm uncertain if there is a consensus about that as many other articles begin the paragraph with common name as well. Lolitart (talk) 00:10, 15 November 2021 (UTC)
- Per Blueboar, does it really matter so long as the opening sentence is clear and easy to read. That might well dictate the order used. Regards, Cinderella157 (talk) 00:30, 15 November 2021 (UTC)
- MOS:FIRST suggests that
the page title should be the subject of the first sentence
butit may appear in a slightly different form
. There are exceptions, such as MOS:FULLNAME for biographies, but I don't see one which applies here. Certes (talk) 00:35, 15 November 2021 (UTC)
IP's = trolls?
From my experience, iP's making certain edits are a lot more likely to be reverted than a well-established registered user in good standing making the same edits. This is unfair. Just because you're an IP doesn't mean drastic reformulations are bad, worse, incorrect or incomplete. --2A02:AA1:1009:3306:BDA3:4FD1:9B42:F7FD (talk) 18:49, 26 November 2021 (UTC)
- Your general premise seems spurious and difficult to prove. I'd be interested to see examples of those reverted good-faith edits that you are mentioning. Generally if an edit is reverted, it is assessed based on the quality of the edit and not of the editor itself. You are also more likely to be reverted if you are editing in a contentious field or a rapidly developing current event. WaltCip-(talk) 17:17, 28 November 2021 (UTC)
- @WaltCip: I think it's a significantly more complex then that. On the one hand IP edits, and more generally those by non-ac (and given our increasing complexity often many non-ec) accounts are just flatly more likely to have problems, broken formatting, missing citations, misunderstanding our basic purpose, etc. On the other hand there are edits made by IPs that would not be reverted if an established user made them.This occurs for a few reasons, first all non-ac edits are more likely to be reviewed, indeed ec editors are essentially invisible to most edit filters, and the more people that see your edit the more likely it is someone will disagree with the change. Second is heuristics, if I'm unsure if an edit is good or bad in an area I'm unfamiliar with I may make the decision of whether or not to revert based solely on the username, because otherwise the time needed to assess will increase sharply. Finally there's the issue of social capital, Wugapodes has written on the topic extensively and can explain it better than I, but in short reverts have a social cost, and that cost depends to a considerable degree on who you are reverting. The social cost of reverting IPs is usually minimal; the user probably doesn't know how to use bring the dispute to the concern of a larger audience, hence you can often make a series of objectively bad reverts without anyone noticing. With someone who's been around a bit the considerations shift, you may have to engage in a talk page conversation, it's more likely the revert will be brought to the attention of a wider audience, and a fair chance the revert will be remembered. With a truly established user you now not only have to concern yourself with how they will respond in the long-term, but also their friends, and if some of those people have +sysop the cost considerations grow further still.Anyway, I don't know if anyone has tried to tease all these details apart, but it's possible that Whatamidoing (WMF) has some data the WMF has collected on this issue. Regards, 91.221.17.220 (talk) 19:51, 30 November 2021 (UTC)
- Largely agree with anon. If you want a sense of good faith IPs getting reverted checkout recent changes filtered to include likely good edits in mainspace made by IPs that were reverted. Includes, for example, this edit adding links to other articles which was reverted with a citation to WP:OVERLINK despite that section having very little to do with what actually got linked. — Wug·a·po·des 23:06, 30 November 2021 (UTC)
- I don't have data myself, but I remember being told that mw:ORES research evaluated this, and they found that, for whatever details they were measuring, not being logged into a registered account was the strongest predictor of whether an edit would be reverted. @EpochFail would be able to tell you much more. Whatamidoing (WMF) (talk) 19:15, 6 December 2021 (UTC)
- We did some analysis around this. See https://dl.acm.org/doi/abs/10.1145/3449130 for a paper discussing the effects of algorithmic flagging on revert rates. There's a breakdown for IP editors. I don't think we have a good general assessment of the quality of edits by anonymous editors. But using the dataset that is used to train ORES, we can get a rough approximation. See https://labels.wmflabs.org/campaigns/enwiki/4/?tasks= for the raw data. Edits are labeled as "damaging": true if they caused some harm and "goodfaith": false if the harm appears to be intentional. You'd need to write a script to look up the details of each revision to find out if the editor was anon and the data is from 2015 so it is a little old, but it should be informative. I'll try to find some time to dig around, but someone else please feel free to beat me to it. --EpochFail (talk • contribs) 00:27, 7 December 2021 (UTC)
- One other thing worth mentioning is m:Research:Anonymous_phenomena -- a project I proposed but never got the time/resources to complete. Still, there's a good review of some of the discussion of the nature and qualities of anonymous editors in there. --EpochFail (talk • contribs) 00:28, 7 December 2021 (UTC)
- I don't have data myself, but I remember being told that mw:ORES research evaluated this, and they found that, for whatever details they were measuring, not being logged into a registered account was the strongest predictor of whether an edit would be reverted. @EpochFail would be able to tell you much more. Whatamidoing (WMF) (talk) 19:15, 6 December 2021 (UTC)
- Largely agree with anon. If you want a sense of good faith IPs getting reverted checkout recent changes filtered to include likely good edits in mainspace made by IPs that were reverted. Includes, for example, this edit adding links to other articles which was reverted with a citation to WP:OVERLINK despite that section having very little to do with what actually got linked. — Wug·a·po·des 23:06, 30 November 2021 (UTC)
- @WaltCip: I think it's a significantly more complex then that. On the one hand IP edits, and more generally those by non-ac (and given our increasing complexity often many non-ec) accounts are just flatly more likely to have problems, broken formatting, missing citations, misunderstanding our basic purpose, etc. On the other hand there are edits made by IPs that would not be reverted if an established user made them.This occurs for a few reasons, first all non-ac edits are more likely to be reviewed, indeed ec editors are essentially invisible to most edit filters, and the more people that see your edit the more likely it is someone will disagree with the change. Second is heuristics, if I'm unsure if an edit is good or bad in an area I'm unfamiliar with I may make the decision of whether or not to revert based solely on the username, because otherwise the time needed to assess will increase sharply. Finally there's the issue of social capital, Wugapodes has written on the topic extensively and can explain it better than I, but in short reverts have a social cost, and that cost depends to a considerable degree on who you are reverting. The social cost of reverting IPs is usually minimal; the user probably doesn't know how to use bring the dispute to the concern of a larger audience, hence you can often make a series of objectively bad reverts without anyone noticing. With someone who's been around a bit the considerations shift, you may have to engage in a talk page conversation, it's more likely the revert will be brought to the attention of a wider audience, and a fair chance the revert will be remembered. With a truly established user you now not only have to concern yourself with how they will respond in the long-term, but also their friends, and if some of those people have +sysop the cost considerations grow further still.Anyway, I don't know if anyone has tried to tease all these details apart, but it's possible that Whatamidoing (WMF) has some data the WMF has collected on this issue. Regards, 91.221.17.220 (talk) 19:51, 30 November 2021 (UTC)
Film article with outdated plot vs new plot with flaws
Im came by here and want to ask about a question relating about questioning or objecting if it is correct or wrong on the edit that is once proper but then revert due to copyediting. 122.11.214.200 (talk) 08:16, 27 November 2021 (UTC)
- It would help if you asked the question - your statement is vague and unclear as to what exactly you're trying to say. Popperman99 (talk)
Would there be interest in a bot that makes a "watchlist" just for recently-edited pages?
I feel like I've seen requests for this over the years, but the issue is always that it would be prohibitively expensive, API-wise, to have a bot emulating the effects of Special:Watchlist on user subpages. It occurred to me recently, though, that that wouldn't be the case if you just had a bot periodically update a given user subpage with links to pages they've edited recently; Special:RecentChangesLinked for that page would then function identically to a watchlist for those pages, meaning the bot wouldn't have to actually update on each edit. I got this idea after realizing I could use Special:RecentChangesLinked/User:Tamzin/spihelper log to track SPIs I've been involved in. This would be the same principle.
I'm bringing this up here at VPM, rather than VPT, because it's not really a question of how to make it. I already have an idea of how to code the bot in a manner that would pass BRFA, and can explain those details to anyone interested in doing it themself (since I won't have time to for a month or two, probably). What I'd like to know, though, is if this is something that the editor-base at large would be interested in. I know I'd use it, because I'm the kind of person who often checks their contribs and looks for ones that aren't "current", rather than using my watchlist, but I don't know how common a tendency that is. -- Tamzin[cetacean needed] (she/they) 23:07, 1 December 2021 (UTC)
- I achieve something similar by watchlisting pages I edit for a week or a month. Sadly, there is no way to make this the default; it is necessary to change the duration from "permanent" to "1 week" manually for each edit. Certes (talk) 23:30, 1 December 2021 (UTC)
- That, and also the issue that in practice, you can't have a lot of long-term watched pages without it drowning out those short-term ones. I enjoy having a few thousand permanently-watched pages. When I'm more active, it's nice to check in on them. For instance, I watch some pages that tend to get vandalized every few months in subtle ways that RCPers often miss—low-profile trans and non-binary people's biographies, for instance. But if I'm editing a bit less, and I really just want to keep an eye on any plates I already have spinning, I turn to just monitoring my own contribs, but that system is far from perfect. Hence my interest in the above. Should I consider you someone who'd be interested in this, Certes? -- Tamzin[cetacean needed] (she/they) 23:43, 1 December 2021 (UTC)
- Like you, I effectively have two watchlists: one transient (did someone improve or undo my recent contribution? did the vandal I just reverted return?) and one permanent (topics that interest me or are vandal magnets). They're merged on Special:Watchlist, with transients marked by a clock. I keep meaning to add JavaScript to change my watch duration to 1 week by default, but would be interested in alternatives. Certes (talk) 23:53, 1 December 2021 (UTC)
- @Certes: follow phab:T25014 for development request for auto-watching edited pages. — xaosflux Talk 00:19, 2 December 2021 (UTC)
- That, and also the issue that in practice, you can't have a lot of long-term watched pages without it drowning out those short-term ones. I enjoy having a few thousand permanently-watched pages. When I'm more active, it's nice to check in on them. For instance, I watch some pages that tend to get vandalized every few months in subtle ways that RCPers often miss—low-profile trans and non-binary people's biographies, for instance. But if I'm editing a bit less, and I really just want to keep an eye on any plates I already have spinning, I turn to just monitoring my own contribs, but that system is far from perfect. Hence my interest in the above. Should I consider you someone who'd be interested in this, Certes? -- Tamzin[cetacean needed] (she/they) 23:43, 1 December 2021 (UTC)
Low views - recent record
I notice that the overall daily pageviews have not exceeded the 250 million mark since 12 September. This is by far the longest period below that mark since the current stats began in 2015. They are still falling - 222 M for the 10th. Johnbod (talk) 02:36, 12 November 2021 (UTC)
- Increased indoor activity likely predicts increased online activity. It seems to roughly peak and trough based on weather temps in the northern hemisphere. Reaching a low around now. This year has been unusually warm on average. As it will go nearly every year, winter/fall/spring are getting shorter from global warming. -- GreenC 04:37, 12 November 2021 (UTC)
- This makes no sense at all, even without the obvious contradiction between your first bit and the rest. November is usually one of the busier periods, with a short drop for Thanksgiving. You will find the academic calendar has a bigger effect than temperatures. Johnbod (talk) 00:15, 13 November 2021 (UTC)
- @GreenC and Johnbod: Try a zero-base for the graph to put this in proportion. The results are (right). The English Wikipedia back to 2015 seems fairly constant around 250 million. There's a bit of seasonality with a slight drop in the summer. This year it doesn't seem to be bouncing back so much but it doesn't seem to be a big deal, given the historical volatility. If there's a reason, the pandemic is most likely to be a factor.
- Some other languages are shown for comparison. They seem even flatter.
- Andrew🐉(talk) 20:04, 14 November 2021 (UTC)
- Thanks, but I'm not convinced. This format also makes consistent drops below 250 million easy to spot. Previously, they have been brief, and in high summer (northern). Not so this year, & I don't really see why the pandemic should hit now. The first lockdown wave took us to the highest views of these figures. We shall see. Johnbod (talk) 20:10, 14 November 2021 (UTC)
- The pandemic seems to have shaken people out of their rut as the lockdowns, furloughs and layoffs have caused them to change their habits and lifestyle. For example, I was reading today that it's hard to get a taxi now as many drivers have moved on to steadier work. This may have effects on work-related topics. For example, when I checked the readership for an article I wrote – Perfect is the enemy of good – I found that it halved at the weekend – down from 500/day to 250. For another article I wrote – The Great British Bake Off – the readership varies widely during the year, rising dramatically when a new season starts – ranging from 3,000 to 16,000 per day. So, the overall siteviews are a composite of many such different cycles. Andrew🐉(talk) 20:30, 14 November 2021 (UTC)
- Andrew, if you hadn't realized how day of the week affected views on work/homework subjects, try Logarithm. Johnbod (talk) 17:49, 20 November 2021 (UTC)
- I've uploaded another graph using the siteviews data, adjusted for small-period fluctuations. It does seem that there's been a slight dropoff in traffic compared to previous years, following the massive March 2020 boom due to the pandemic. On another note, I'm a little surprised that WP as one of Top 10 websites online doesn't even get a billion views a day -- are these unique views only? Daß Wölf 17:09, 15 November 2021 (UTC)
- Nice job! This entirely confirms my point that previous annual extended lows were in the northern summers. We are now below the 2021 summer low, & apparently heading downwards. Whether you call this a "slight dropoff" is a matter of taste I suppose; just reading off the graph we now seem to be some 20% lower than this time last year. Where next, I wonder? Johnbod (talk) 17:49, 20 November 2021 (UTC)
- Thanks. Calling it "slight" might've been an understatement, that's true. The numbers are currently a little lower than the trough of (northern) Summer 2020. I wonder if the causes are similar to the slump in Autumn 2017 that came roughly a year after Brexit and the Trump election, which were both followed by spikes. Was it general fatigue from the 24/7 news cycle, or growth and strengthening beliefs of the luddite crowd? Anti-vaccination protests are currently very popular in my area. Daß Wölf 19:22, 20 November 2021 (UTC)
- P.S. I wonder what was behind the October 2019 spike. Daß Wölf 19:22, 20 November 2021 (UTC)
- Nice job! This entirely confirms my point that previous annual extended lows were in the northern summers. We are now below the 2021 summer low, & apparently heading downwards. Whether you call this a "slight dropoff" is a matter of taste I suppose; just reading off the graph we now seem to be some 20% lower than this time last year. Where next, I wonder? Johnbod (talk) 17:49, 20 November 2021 (UTC)
- The pandemic seems to have shaken people out of their rut as the lockdowns, furloughs and layoffs have caused them to change their habits and lifestyle. For example, I was reading today that it's hard to get a taxi now as many drivers have moved on to steadier work. This may have effects on work-related topics. For example, when I checked the readership for an article I wrote – Perfect is the enemy of good – I found that it halved at the weekend – down from 500/day to 250. For another article I wrote – The Great British Bake Off – the readership varies widely during the year, rising dramatically when a new season starts – ranging from 3,000 to 16,000 per day. So, the overall siteviews are a composite of many such different cycles. Andrew🐉(talk) 20:30, 14 November 2021 (UTC)
- Thanks, but I'm not convinced. This format also makes consistent drops below 250 million easy to spot. Previously, they have been brief, and in high summer (northern). Not so this year, & I don't really see why the pandemic should hit now. The first lockdown wave took us to the highest views of these figures. We shall see. Johnbod (talk) 20:10, 14 November 2021 (UTC)
- @Johnbod, Andrew Davidson, and GreenC: Apparently, the Sept/Oct 2019 rise may have been caused by bot traffic, as I've been informed by MKampurath (WMF). There's an indepth analysis at the phabricator [1]. In that case the lower numbers this year are probably a little less worrying. Daß Wölf 18:16, 2 December 2021 (UTC)
- Just want to remind anyone looking at this that Correlation does not imply causation and that global warming, outdoor activity, academic calendars, and any other world events may or may not all have a effect on Wikipedia activity. Popperman99 (talk) — Preceding undated comment added 17:05, 3 December 2021 (UTC)
- It looks like a WMF analyst is working on a (so far non-public) task to "Investigate declining pageviews" on Wikimedia sites overall (see also), so perhaps there will be an update in the monthly audience metrics report for November.
- As noted above, it can be tricky to visually distinguish seasonal ups and downs from more fundamental changes. Short of an actual statistical time series analysis (or using stats software to produce a seasonal decomposition plot [2]), a simple, useful option is to plot the data year-over-year for several years (example).
- Regards, HaeB (talk) 13:51, 4 December 2021 (UTC)
.png&action=edit&redlink=1){kind=link}
Collateral Search Gold
Non-contextual words in a document resulted in a search result from google:
- Marilyn Monroe comes to visit Einstein (Rutger_Hauer)
- the search: site:www.uscis.gov/sites/default/files/document/foia/
- may have things of interest, any idea which talk pages to mention this?
- 0mtwb9gd5wx (talk) 00:23, 4 December 2021 (UTC)
Problem with Wikipedia: Requested articles
If one goes to Wikipedia: Requested articles, and clicks on the link "Biography", one will see that most of the names are in red wikilinks. However, today I made a request for an article on Scott Erickson, the visual artist. The link was in blue, because there are already articles on people with that name in Wikipedia. None of these, however, were on the person for whom I am requesting an article. Seeing the link in blue may have confused Wikipedians, as it may have led to people thinking there was already an article on the visual artist called Scott Erickson. How can we stop the problem of unintentional blue wikilinks at Wikipedia: Requested articles?YTKJ (talk) 12:56, 4 December 2021 (UTC)
- Maybe qualify the new request as Scott Erickson (artist)? — GhostInTheMachine talk to me 16:37, 4 December 2021 (UTC)
- Thank you, User:GhostInTheMachine. I have done that, but my request seems to have been removed. YTKJ (talk) 22:10, 4 December 2021 (UTC)
- My request for an article on Scott Erickson (artist) appears to have been accepted now. I think my problem before was that I did not supply enough information, and my only external link was a primary source. YTKJ (talk) 19:29, 6 December 2021 (UTC)
MF DOOM
I’m not donating any more money until MF DOOM is spelled properly in all caps on his page. — Preceding unsigned comment added by 2600:8801:1B84:AE00:D0EF:76C6:2CBE:7308 (talk) 19:40, 4 December 2021 (UTC)
- Replied at Wikipedia:Teahouse#Question. PrimeHunter (talk) 20:38, 4 December 2021 (UTC)
English fundraising banner campaign - update
Dear all,
The first week of our English fundraising banner campaign is behind us and I would like to thank you all for your support and understanding. So far our campaign has run smoothly and we have been trying different messaging in all of our banners to see what resonates most with readers.
Several comments have been left both here and on the talk:fundraising page. We will post comments to those in the next few days.
Hear from the donors:
I am a guest lecturer on cruise ships and turn to Wikipedia for information on the ports and places so I can prepare my PowerPoints and presentations. Without access to Wikipedia my work would be so much harder. And you keep making it better, deeper and more intuitive. (donor from the US)
Yeah 2020 felt like the earths magnetic core finally flipped. I work in icu so I’ve seen more death than usual and you’d b surprised at how much we use Wikipedia for basic info to save lives. We know most stuff but once in awhile we like to refresh or just look up new meds that no one has heard of so in a way ur helping us help society. Thanks for all of it. (donor from the US)
Recently an article about my wife, [...], was put up. We are very, very grateful because she wanted to get a page. I had even started an account with the goal of writing the page when, lo and behold, it appeared one day, once more proving that if I just lay around the house like a no-good husband good things will still happen. We think this effort might be attributable to the Wiki Women in Red campaign." (donor from the US)
In the meanwhile:
- If you have specific ideas to share, please feel invited to add them to the fundraising meta talk page.
- If you need to report a bug or technical issue, please create a phabricator ticket.
- If you see a donor on a talk page, VRT or social media having difficulties in donating, please refer them to donate(at)wikimedia.org
Thank you all very much for your support and understanding during the campaign. I will be in touch again latest by the 22nd of December.
Best wishes, JBrungs (WMF) (talk) 07:57, 8 December 2021 (UTC)
- Thank you for thos messages showing people still think they are donating to Wikipedia / enwiki and that their money goes to make the contents better, when in reality it goes to the overbloated and quite rich enough WMF and the costly (and too often problematic) affiliates and grants. Fram (talk) 08:12, 8 December 2021 (UTC)