This page contains discussions that have been archived from Village pump (technical). Please do not edit the contents of this page. If you wish to revive any of these discussions, either start a new thread or use the talk page associated with that topic.
I tried to add a link to a file under http://icecast.rte.ie to ga.wikipedia.org but it seems wikipedia replaces http with https in the href for hosts in rte.ie even though icecast.rte.ie only serves http. I have had to use the nowiki tag to stop wikipedia from generating the broken href, or any href for that matter: surely this is a step backwards from the web of hyperlinks! How can I get wikipedia to respect my urls, particularly http? 121.127.206.113 (talk) 00:28, 4 June 2022 (UTC)
121, this forum is for help with technical issues here, on the English Wikipedia. Did you make an edit here that you had a problem with? Can you provide a link to the specific page you edited and the time you edited it from its history (or a link to the Diff of the edit). — xaosfluxTalk00:37, 4 June 2022 (UTC)
If you are trying to add a link to a page here that isn't working, can you provide the exact entire link as an example? It is certainly possible to link to http only sites from Wikipedia, for example here is a link to http://www.gnu.org. — xaosfluxTalk00:43, 4 June 2022 (UTC)
You can see the same problem in my post here, so it is a problem here too. http://icecast.rte.ie is an https not http link. Following it leads to timeout, whereas http access succeeds. The fact that wikipedia does not break org and com addresses i think is good evidence that this is a problem that ought to be fixed. ga wikipedia code is mostly copied from en wikipedia, and almost everyone speaks english, and if I am asking for help in english, this seems to be the right place. 121.127.206.113 (talk) 00:47, 4 June 2022 (UTC)
Hi 121, that link is dead so something else may be at play, do you have an external http link that actually works? — xaosfluxTalk00:51, 4 June 2022 (UTC)
OK, I found the "problem" and it is sort of us, but because of the remote site. Wikipedia automatically overrides http to https on sites that have HSTS declared - such as icecast.rte.ie (per this check). In this case the "fix" would be for the external domain operator to un-declare HSTS (or just fix their https), as it will also break many browsers that use it. — xaosfluxTalk01:03, 4 June 2022 (UTC)
Basically meaning that even if we didn't do this, most common readers that use modern browsers are going to have the problem anyway, as their browsers will use HSTS and be unable to get there as well. For example I can't manually load that site in default Firefox or Chrome (but I could with a custom browser config - that we would never expect most readers to use). — xaosfluxTalk01:05, 4 June 2022 (UTC)
(A) Wikipedia's present behaviour actually violates RFC 6797 2.4.1.1.6 "UAs need to disallow security policy application to peer domains, and/or higher-level domains, by domains for which strict security policy is enabled."
icecast.rte.ie does not set Strict-Transport-Security
its parent rte.ie does not set Strict-Transport-Security
only www.rte.ie sets it: applying this to icecast (its peer) or rte.ie (its parent) clearly violates the RFC. (See also RFC 6797 11.4.2)
(B) I have had no trouble accessing the site with the packaged Firefox on my BSD machine, on my schools CS dept GNUbuntus, and with Chrome on the MS machines at my local public (state) library. Maybe these browsers actually adhere to RFC 6797?
I'm relatively confident that Chrome/Firefox wouldn't have allowed it onto the HSTS preload list at some point if it didn't meet the technical criteria. I suspect that rte.ie at some point had includeSubdomains set, when they didn't actually want that and it made it into the list. The FAQ entry about Removal is particularly important here: "Domains can be removed, but it takes months for a change to reach users with a Chrome update and we cannot make guarantees about other browsers."
I just checked, the domain is still in the Firefox HSTS preload list, which is where MediaWiki pulls from. So there's really not much we can do about it. Legoktm (talk) 05:04, 4 June 2022 (UTC)
Yup, and even if we did - browsers are still going to have the same problem. So what now 121? Well, there is nothing we, the editors on the English Wikipedia, can do about this. Feel free to open a bug report that there is something wrong with the handler, or a feature request for a format that editors of mediawiki could use to bypass this behavior. See WP:BUG for directions for both of those. This still won't ultimately fix the problem for most readers whose browsers are still going to follow the preload list and make the link useless. — xaosfluxTalk10:55, 4 June 2022 (UTC)
The way this HSTS list is used violates actual and proposed standards and degrades WP, so WP/MW shouldn't use it. If as you say, all these browsers obey this list, I do not see why WP should impose it on everyone with browsers that do not so grossly disobey actual and proposed. This reminds me of the issue with MS being unable to display web pages with an xml/html header. 121.127.206.113 (talk) 13:43, 4 June 2022 (UTC)
The fact that you thought http://icecast.rte.ie was dead because you were mistakenly accessing https://icecast.rte.ie after I reported that mistake in particular, and that you gave a link to a site you misunderstood as saying icecast.rte.ie has HSTS enabled without even bothering to read the HTTP headers icecast.rte.ie and rte.ie actually send, very strongly suggests to me that you have no idea what you are talking about, nor any desire to form an accurate idea about it. 121.127.206.113 (talk) 13:54, 4 June 2022 (UTC)
It gets worse. I just checked your example of www.gnu.org and found that host in particular does actually set Strict-Transport-Security, yet WP does not rewrite that href. 121.127.206.113 (talk) 13:59, 4 June 2022 (UTC)
I thought it was dead not because of anything on Wikipedia, but because multiple common, modern, browsers failed to follow that link even when manually entering it in the address inputs; I was only able to get a response on that link when I tried very hard - something that 99%+ of encyclopedia readers are not going to do. In any case, I've already referred you to file all the bugs and feature requests you want for Mediawiki software and/or WMF hosted wiki configurations - as this is not something specific to the English Wikipedia so there is nothing we can do about here on this forum. — xaosfluxTalk14:17, 4 June 2022 (UTC)
You could also open bugs with those browser manufactures (I had the same issue with Firefox 101.0 64bit; Chrome 102.0.x 64-bit) I was able to get it to load on an ancient copy of MSIE 11 and with wget. — xaosfluxTalk14:33, 4 June 2022 (UTC)
Thanks for reporting the issue with those browsers. I doubt that I can get them to really fix their broken-ass software, but if rte.ie can be removed and so make things easier on gaeilgeoirí, that'll do me. 121.127.206.113 (talk) 15:29, 6 June 2022 (UTC)
How many of those browsers actually support TLS? Do HSTS sites tend to have stricter TLS version requirements? How many of those browsers support those stricter requirements? Obviously sites like www.gnu.org that do support TLS (though maybe do not faithfully implement HSTS) do not benefit from this list. It seems to me that this is the wrong solution for the wrong problem. 121.127.206.113 (talk) 22:57, 4 June 2022 (UTC)
You're mixing numbers - it isn't about "what percent of browsers" that really matter, it is about % of GET's - that a small fraction of old browsers don't work is fine to know , but it is many orders of magnitudes less of actual page views because most of our pageviews are from modern browsers. Which goes back to the fact that even if we didn't rewrite that http-->https on sites with wonky HSTS directives, the readers browsers are already going to do it -- which ultimately means that that external link is going to be unreachable for most readers. — xaosfluxTalk14:58, 5 June 2022 (UTC)
It's not that icecast.rte.ie has wonky HSTS: it doesn't HSTS. What is wonky is those browsers. Just because some browsers are wonky does not justify breaking WP for everyone. Those browsers need to be fixed, just the same as WP.
The Chromium list and HSTS itself are dodgy AF, but according to Mozilla's description of its list, icecast.rte.ie should not be on it. I plan to look into that. BTW WP has some pretty wonky-ass HSTS. Try requesting https://en.wikipedia.org/ and see what headers you get.
Anyway I don't think we are getting anyway with this so let's give it a rest, shall we?
Adding parameters to an Infobox (specifically Motorcycle Rider Infobox)
Having found myself adding a bio for a motorcycle racer to Wikipedia, & as a software developer, I was initially not dismayed when I realised I needed to add a set of additional parameters to the Motorcycle Rider Infobox. It quickly became clear that it wasn't obvious to me what I needed to change, where I needed to change it, what I needed to be careful of not changing, & how I could play with all this safely in a sand box to find out without inadvertently breaking things that have a massive & widespread impact on every page that uses Motorcycle Rider Infobox. I rather cheekily hoped to request that somebody else might add the parameters I was requesting, piggy backed onto an earlier request that had gone unanswered because it was never flagged as a change request. The response has been that I should do it myself, which I would love to, but despite having had a look at the Wikipedia LUA page, & made numerous plausible looking attempts to add the parameters to the code, I've so far achieved nothing, or achieved nothing but also broken everything.
The problem may be my incompetence, but I can't help thinking that the complete lack of code comments or documentation concerning how the nested race result stuff works may be a factor. I had hoped to track the changes made by one of the experts in this domain to reverse engineer them & enhance my understanding so that I could make future changes easily.
How does the world of Wikipedia feel about code comments? Is somebody who does understand Infobox Motorcycle Rider & the nested race results child objects, & how to add them, prepared to go back & add in some useful code comments? — Preceding unsigned comment added by Khaylock (talk • contribs) 09:58, 6 June 2022 (UTC)
I have rarely seen comprehensive code comments or proper developer documentation for Wikipedia templates or modules (though there are prominent exceptions). Documentation for users (editors) is also of widely different quality. Documentation for readers is non-existent. This is to be expected in an all-volunteer platform where most development projects are small, and not expected to scale. Again, there are exceptions. My advice would be to get a good grasp of the syntax Wikipedia uses for templates and modules (native and Lua scripting) and then sandbox extensively until you get it right. Be advised that there is no proper software evaluation done in Wikipedia. To compensate, there is often drama. 50.75.226.250 (talk) 15:02, 6 June 2022 (UTC)
@Khaylock: Template talk pages often have very low activity. You used {{edit semi-protected}}[2] to draw attention but that template is meant for cases where you post the wanted code or making it would be easy, but you are unable to edit the page due to protection. Wikipedia:Requested templates can be used to ask for template help. You can also try {{Reply to}} to alert users who have edited a given template and are still active. PrimeHunter (talk) 16:29, 6 June 2022 (UTC)
Ah, thank you! Yes, I piggy backed onto an earlier request that looked similar but was apparently also in the wrong place.
A very helpful user has stepped up to implement a working prototype the change I was hoping to make, & I may try to pay this forward by adding the code comments I would have wanted when I was trying to work out how it worked so I could make the changes myself... Ken Haylock (talk) 17:03, 6 June 2022 (UTC)
Cannot edit watchlist
Hello,
I cannot edit my watchlist. Everytime I click on the "Edit your list of watched pages" button I get a server time-out.
This is probably caused by the fact that I have added every page I visited to it, like the bread crumbs in Hansel and Gretel.
It has 23,984 articles now.
Would it be possible for an administrator to empty/purge my watchlist?
Then I can use it the way intended.
to isolate Presidential Election tables. However, sometimes this crashes, and I'm sure there's a better way to do this than run a huge regex on the whole page. Solutions? ― Qwerfjkltalk18:28, 5 June 2022 (UTC)
Fair enough. I wonder whether .match(...)[1] would be quicker than .replace(..., "$1"), but an efficient implementation might be equally fast for both. Certes (talk) 12:59, 6 June 2022 (UTC)
To clarify that: an expression starting with [\\s\\S]*, obeyed naively, starts at the character 1, greedily matches the whole page but fails to match headerRegexp.source at end of page, backtracks so [\\s\\S]* matches the whole page except the last character, fails to match headerRegexp.source against the last character, and so on back to the first character; then repeats all that starting at character 2, 3... On a page with n characters, it tests for headerRegexp.source n2/2 times. With match, the initial (and final) [\\s\\S]* can be removed, so it only tests for headerRegexp.source n times. On a 10 kB page, that's 5000 times faster. Certes (talk) 21:41, 6 June 2022 (UTC)
What links here functionality
Recently we've apparently lost the ability to filter redirects and tranclusions from this report. Not too long ago we had an external tool that gave a little help fixing broken section links from redirects. What's with the declining functionality in this area? Being able to manage links between articles is a key feature for editors and readers of an online encyclopedia. ~Kvng (talk) 17:26, 6 June 2022 (UTC)
Looks like the special page got an OOUI treatment. Click on "˅ What Links Here" and tick the checkboxes as necessary. Nardog (talk) 18:01, 6 June 2022 (UTC)
I thought that the links were gone too. It turns out that they have been placed in a large collapsible box that needs to be expanded. You then need to click the options you want and click "Go". Is there a nice CSS way to ask this box to be expanded by default? That would save me one click, at least. – Jonesey95 (talk) 18:09, 6 June 2022 (UTC)
Thanks. Sorry I missed that. The interface has been "improved" from 1-click to 3 clicks with discoverability issues :( Why? (The phab page referenced doesn't offer a clue) ~Kvng (talk) 18:15, 6 June 2022 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 7 June. It will be on non-Wikipedia wikis and some Wikipedias from 8 June. It will be on all wikis from 9 June (calendar).
I'm currently at work but without any WiFi, so I'm using my mobile data and rightly so, the IP range is blocked but anon only. However, I'm also an WP:ACC member, and because the IP range is also 'Account Creation Blocked' I am prevented from creating an account requested though ACC. I want to get the community's thoughts on either:
1. Apply 'Anon Only to 'Block Account Creation' when both are set, so registered users (read: Extended Confirmed users) can register a new account when logged in
2. Allow those with 'Account Creator' permission to bypass 'Account Creation Blocked'
Since a couple of hours I have a weird issue when editing talk pages. A bunch of characters just don’t work anymore. I can’t type a colon, rather get some strange ː character instead, so have difficulties indenting my comments and the tildes, to add my signature, don’t work properly and I have to use the preset ones underneath the edit window instead. Anyone else having these issues. At first I thought it was a keyboard setting on my computer, but I’m getting the issue on both my computer and any mobile device I use to edit Wikipedia. It thus appears to be caused by the software.Tvx1 21:47, 6 June 2022 (UTC)
Check boxes are blank if overridden by user's own colours
{Note: I (still) have not gotten myself an access to "Phabricator", so I can't report this problem there.}
When I switch to "Edit (source)" mode, and scroll down, both, the "Minor change" and the "Watch this page", check boxes are blank/empty; they only show my background colour.
This is in-dependent of whether I actually edit or not.
And when I click into a "Minor change" check box, the frame of this check box becomes thicker. So far, so good.
Problem:
But the space in this check box remains blank;
there appears no (white) "check" (on blue ground) in it.
And when I return, e.g. from a preview,:
this check box is still blank, plus
the frame of this check box is thin (again), as it was at the opening of this editing page,
so I can't see whether I have already checked this box or not;
I can only know from my memory - if this tells me :) .
This phenomenon is only in the English WP, not in the German WP.
In the German WP these check boxes are smaller than in the English, but show the check on blue ground.
I use: Windows 8.1 and Firefox (latest edition).
In Firefox / tab [General] / sub heading "Language and Appearance" /
I have set:
"Colours" to my own colours and
"Override the colours specified by the page with your selections above" to "Always".
When I change my setting of "Override..." to "Never", then the content of the check box(es) is visible, but for this I would have to sacrifice my colours.
In 1923 clicking on a month in the Birth section, it displays month from the Deaths section. Clicking on a month in the Death section, it does not jump to that section. Template is "Template:Births and deaths TOC". Reporting here and asking for help to fix. JoeNMLC (talk) 16:21, 8 June 2022 (UTC)
I have worked around this problem by reading the somewhat muddled template documentation. A more straightforward fix would have been to break up the "Events" section into individual months instead of grouping them, but I didn't want to mess with an established article. – Jonesey95 (talk) 17:00, 8 June 2022 (UTC)
@ESanders (WMF): Your reply is tagged as having used the "reply" button + "mobile edit". I always have reply link on desktop site, but now on mobile, on this page, I don't see the reply link. How did you use it? —CX Zoom[he/him](let's talk • {C•X})12:59, 8 June 2022 (UTC)
Hi all, I don't know if this has been requested in the past but I think this idea will help the project in certain areas, so I'm proposing this here first. Sometimes, we have have page titles separated by slashes that are clearly NOT subpages of the part of title that comes before the slash, although MediaWiki thinks so. For example, MediaWiki thinks that Template:He/she (pronoun temp) is a subpage of Template:He (Hebrew), or that Talk:R/place (subreddit) is a subpage of Talk:R (Latin character). Because of this, if we someday decide to move R to "R (character)" and check the move subpage box, which is useful when a article talkpage has too many archives, we would be left with Talk:R/place moving to "Talk:R (character)/place". Similarly if we ever move Template:He to Template:Hebrew, and check the move subpage box because doc, sandbox & testcases, the subpage Template:He/she goes to Template:Hebrew/she. But more importantly, the really annoying and stupid backlink we currently have to Template:He that is shown just below the Template title of Template:He/she. Now if we use __NOTSUBPAGE__ on Template:He/she, MW stops thinking that it is a subpage of Template:He. Checking the move subpage box won't move it now, neither does that stupid backlink shown. Thoughts? —CX Zoom[he/him](let's talk • {C•X})11:27, 2 June 2022 (UTC)
Not sure how needed this would be. We can already disable subapges on namesapces - and do so for content pages where the title can be important (for example Dev/null is not considered a subpage) - for other namespaces we can pretty much pick any titles we want to use and can simply avoid control characters in the page name. — xaosfluxTalk13:28, 2 June 2022 (UTC)
The problem is that although we can disable it per namespace, we generally want it enabled for all talk namespaces. The question is: how many non-talk namespace pages would we want this to apply to? Otherwise, we could have a different rule: if Foo/bar exists, then Talk:Foo/bar should not be a subpage of Talk:Foo, since Foo/bar is not a subpage of Foo. Animal lover |666| (talk) 16:12, 2 June 2022 (UTC)
...and since there is no articles (apart from some redirects) that end with /GA1, all GAN talk pages should not be considered subpages? __NOTSUBPAGE__ seems to be the better choice. NguoiDungKhongDinhDanh16:17, 2 June 2022 (UTC)
It got WONTFIXed on the basis of how is the software supposed to know that? Either subpages are enabled for a namespace, or they're not. Since a magic word is understood by MW, I think this solves the "how is the software supposed to know that?" part. —CX Zoom[he/him](let's talk • {C•X})18:04, 2 June 2022 (UTC)
Using a NOTSUBPAGE switch was discussed in the merged task, phab:T24597. This certainly doesn't preclude trying again (reopen the old ticket, challenging the WONTFIX). — xaosfluxTalk18:11, 2 June 2022 (UTC)
The software starts at the rightmost part and goes to the left until it encounters the first NOTSUBPAGE, which serves as the breaking point. So the Archive page becomes a subpage of Talk:AC/DC but because it already has NOTSUBPAGE, neither Talk:AC/DC nor the Archive page is a Subpage of Talk:AC. —CX Zoom[he/him](let's talk • {C•X})09:45, 3 June 2022 (UTC)
When logged in to Wikipedia, latex code is being displayed rather than the parsed formulae. This does not happen when I am logged out of Wikipedia. It happens for both Firefox and Safari. Burrobert (talk) 06:47, 11 June 2022 (UTC)
Under "Math" the radio button for "Latex Source (for text browsers)" is filled in. Regarding examples, afaict this happens on all pages that contain math formulae, e.g. quadratic residue. Burrobert (talk) 14:32, 11 June 2022 (UTC)
It seems to be doing what you asked it to do. If you want to see images, try "MathML with SVG or PNG fallback". – Jonesey95 (talk) 14:39, 11 June 2022 (UTC)
Please fix the encoding of the hosting page name created by {{PAGENAME}} in {{Userpageblue&round}} - when inserted into a page of a user with spaces in a name, only the first word of the name gets included in the resulting URL. Example: Tidjani Saleh. --CiaPan (talk) 05:52, 11 June 2022 (UTC)
Fixed: {{PAGENAME}} is now replaced by {{PAGENAMEE}}. This will produce underscores instead of spaces, providing a continuity to the URL. With {{PAGENAME}}, the spaces in user's name were produced as it is which resulted in the URL being broken at the first space. —CX Zoom[he/him](let's talk • {C•X})06:03, 11 June 2022 (UTC)
After some updates, the Wikipedia Adventure can no longer progress after clicking on the edit button. While I don't know about the hooks of the guided tour, I know that we could easily add buttons that say "Clicked!" to have the tour progress itself instead of waiting for an event that never comes. Aaron Liu (talk) 06:27, 4 June 2022 (UTC)
Pinging Ragesoss, Mattflascen. I wish I had time or know-how to fix and update, but I really don't anymore. Do either of you know a code-fix for the on-edit step advance? Best, Ocaasit | c05:31, 8 June 2022 (UTC)
Both "Start your adventure" buttons work for me. I am using legacy Vector, FWIW. What does Aaron Liu mean by can no longer progress after clicking on the edit button? Can you provide step by step instructions for what does not work? – Jonesey95 (talk) 11:07, 8 June 2022 (UTC)
It doesn't work for me either. If I'm not mistaken it worked a month or two ago when someone on my watchlist was playing it and I too went for fun. Using Vector legacy, beta features on. —CX Zoom[he/him](let's talk • {C•X})13:24, 8 June 2022 (UTC)
What I meant was not about the start your adventure buttons(these work for me), it was about how the guided tour was stuck on the "click the edit source button to..." step even after you clicked the button. I would like to add some buttons that say "Clicked!" in these steps to have the tour progress. Aaron Liu (talk) 01:57, 9 June 2022 (UTC)
Are you using the 2010 Wikieditor? (check if the wikieditor in the beta features is off) If not, then that is the issue. It depends on which one you are using what can be done.--Snævar (talk) 12:54, 8 June 2022 (UTC)
Hmm, on current firefox I'm having the problem - but only in monobook so far. On an old version of firefox it works still. — xaosfluxTalk13:35, 8 June 2022 (UTC)
For Wikieditor 2017 support, shouldSkip and attachTo in the TWA scripts need to be changed, whenever the user is in editing mode. The styles used for Wikieditor 2017 are different than 2010 wikieditor, which is why attachto needs to support both. I have not tested the code, hence not making a edit request right now. (the code below can probably be compacted and improved)
Hi, in my opinion table of contents in the new 2022 vector (which exists on the left menu bar) should have an scroll bar. By adding this capability, first, a user can see he/she is in what location of table of contents (near the top or near the bottom of TOC) and second, he can scroll the TOC more easily. I really recommend to add an scroll bar to the TOC by these reasons. Additionally, the word "Contents" should be outside of this scroll bar, and should not be scrolled at all, i.e., it (the word "Contents") should be displayed unrelevant to this fact that we are in what location of an article. Hooman Mallahzadeh (talk) 15:17, 11 June 2022 (UTC)
Unbulleted lists inside a {{br separated entries}} in an infobox seem to generate one extra line. For instance, Special:Diff/1092557588 and Special:Diff/1077039645 (two different places, same infobox, different articles). Is there a reason for this and how can it be avoided?
That was not a decision I made, one of the few items left over from the old Infobox television station structure. Should those be converted to ubls, Jonesey95? Sammi Brie (she/her • t • c) 05:17, 11 June 2022 (UTC)
I'm not familiar enough with the accessibility recommendations to know, but it seems those are what is being cited as the reason for replacing br tags with {{ubl}}. It seems like a strange mixture to have ubl wrapped in a template that outputs br tags. – Jonesey95 (talk) 05:28, 11 June 2022 (UTC)
Yeah, that's definitely a concern. In any event, migrating the br templates to ubl fixed the underlying "extra line space" issue in the template. Thank you, Jonesey95. Sammi Brie (she/her • t • c) 06:31, 11 June 2022 (UTC)
The diff and wikitext of 1092557588 is a mess. 1) Don't use lists for things which are not lists (namely, the location). 2) Don't use lists for things which are not at 'the same level' of list (namely, wikitext like {{ubl|'''Analog:'''|9 ([[Very high frequency|VHF]], 1949–2009)|'''Digital:'''|38 (UHF, 2002–2018)}}). 3) Don't use lists for start dates ever. Izno (talk) 16:55, 11 June 2022 (UTC)
As for 1077039645, this is not in the infobox. But I expect my cleaning edit for the page owning 1092557588 may/not have fixed that one. Izno (talk) 16:59, 11 June 2022 (UTC)
The line break / extra white space was there; the infobox has been changed since then. You can replicate it by putting this in ExpandTemplates: {{br separated entries|{{ubl|[[Secaucus, New Jersey]]|[[New York City|New York, New York]]}}|United States}} - Jonesey95 (talk) 17:20, 11 June 2022 (UTC)
In the context of expandtemplates at least, the parser is more or less correctly inserting a paragraph to wrap the United States line, which has top margin for its CSS. Plus the br and you get right around an extra line break. Izno (talk) 19:42, 11 June 2022 (UTC)
Recovering sources archived at WebCite (webcitation.org)
Though WebCite supposedly still make archives readable, none I've tried are in fact readable. Often, date-specific archives at the Wayback Machine are not available either, since in early days I saved the sources only to WebCite. The result is that sometimes archives are not available, which of course is bad if the original page is now gone.
Request/Question: Is there any way of recovering what was saved at webcitation.org? Were the WebCite archives themselves archived somewhere else? Or are they lost forever?
Please {{reply|RCraig09}}, thanks. —RCraig09 (talk)04:33, 12 June 2022 (UTC)
@RCraig09: if I understand [4], it appears likely a fair amount of the content archived at WebCite was once made available and retrieved by ArchiveTeam in 2019 to be made available at the Internet Archive. However it's not clear to me the current status of this effort other than it being not yet publicly available. [5]. It's possible the demise of Webcite will push efforts forward, but who knows, especially since I don't know the reasons it's still incomplete (perhaps the complexity of integrating it). Until this happens, unless Webcite comes back, your only option is to look for other archives. Unfortunately the content ArchiveTeam retrieved which was formerly available is no longer, so even if someone else wanted to make an effort to revive it, it might be difficult. (Although possibly ArchiveTeam would be willing to assist if it's someone who they feel it's worth their while.) Nil Einne (talk) 08:52, 12 June 2022 (UTC)
Using the automatic citation function of the VisualEditor for this site results in an URL with accidental doubling of some part (e.g. double www). The output is:
<ref>{{Cite web |title=World Record Alert {{!}} USA’s Hunter Armstrong sets new global 50m backstroke standard |url=https://www.fina.org/news/2596440/www.fina.org/news/2596440/world-record-alert-usas-hunter-armstrong-sets-new-global-50m-backstroke-standard |access-date=2022-06-11 |website=FINA - Fédération Internationale De Natation |language=en}}</ref>
Is this something for Phabricator? Searching for "automatic citation of web sources" gives many results there. --Kallichore (talk) 19:24, 11 June 2022 (UTC)
It was completely down 5 minutes ago, I confirmed with isitdown. Now it is running very slow as of 08:31, 12 June 2022 (UTC) Anybody else experiencing this? —usernamekiran (talk)08:31, 12 June 2022 (UTC)
wikimediastatus.net reports an outtage at 8:14 which was resolved at 8:44 today. That page is an user readable website of the status of WMF wikis. Can not find an phab ticket behind it, might be hidden.--Snævar (talk) 09:07, 12 June 2022 (UTC)
I've had a recurrence of phab:T301505, it lasted for 15 mins or more. Sometimes "upstream connect error", sometimes a HTTP 503 Service Unavailable. This time, the problem also affected Phabricator itself. --Redrose64 🌹 (talk) 08:33, 12 June 2022 (UTC)
Well, yes, it must be OK now since we can post here. The thing that I posted is an after-the-event report, but also noting that phab was also unavailable, something that hasn't happened before. --Redrose64 🌹 (talk) 08:47, 12 June 2022 (UTC)
I had limited access to the site in "Desktop mode" on mobile (Chrome/Android), though I didn't check it until a minute or so before the site came back, so I don't know if it was available through the entire outage. – PhilosopherLet us reason together.08:46, 12 June 2022 (UTC)
I was on desktop mode on Android too. I faced this issue too. The last Wikipedia page to open correctly was at 08:11 per browser history and the next opened at around 08:27 UTC. In the meantime there were errors. This is a different timeframe from what Snævar notes above. In fact, I registered an edit within the outtage period per wikimediastatus.net. —CX Zoom[he/him](let's talk • {C•X})09:12, 12 June 2022 (UTC)
to be exact, the error was upstream connect error or disconnect/reset before headers. reset reason: overflow The cryptic nature of the error message leads me to believe that their septic/flush tank had some issues, and overflowed. —usernamekiran (talk)19:44, 12 June 2022 (UTC)
Hello from german wikipedia! For the german wikipedia I wrote some script to identify dead interwiki links, now after 2 month almost all those "redlinks" are gone. So maybe the english wikipedia (or others) are interested in such a list. Currently I find 3,706,692 interwiki links to 401 wikis and there are 129,018 "redlinks" in 39,420 articles (I did not check namespace draft). So 3.5 % of those interwiki links are not working, in german wikipedia we had about 3 % such dead links, so there is no big difference. If you are interested, I can create such a list. --Wurgl (talk) 09:11, 12 June 2022 (UTC)
I find a surprising amount of hostility on en-wiki to inter-languguage links. I do not understand why, machine translation is so easy and the imperfections are fairly minor. Of course, you must apply some common sense in how you use them. But most of the time, it's more than adequate.
The most mundane approach is just to use an inter-wiki link (e.g.. [[:language-code:articlee_name]]).
In preference to this is {{ill|article_name|language-code...}}. This is generally acceptable, but raises a lot of discussion as to the most languages you should ever specify. (I would peronally not go crazy about this, but it seems to generate more heat than light.)
However, I think that is greatly inferior to speicfying the wikidata parameter with {{ill}}. This is the most maintainable approach. People seem to complain about the Wikidata interface, and I really have no idea why. I don't doubt the Wikidata interface can be improved, but it's just not that onerous. Of course, it's confusing until you understand what's going on, people who care will figure it out, there are many stumbling blocks in using Wikipedia.
Anyway, I have no understanding of WP auomation or scripting, so I couldn't use it, but it sounds like a great offer and something that somebody in en-wiki should be taking you up on. Fabrickator (talk) 20:26, 12 June 2022 (UTC)
(not directly related, but I will ask anyway) What about interlanguage links that link to redirects, and that redirect goes to a page that is present in the same Wikidata item as the original article (where the interlanguage link came from)?--Snævar (talk) 07:39, 13 June 2022 (UTC)
Hello. A few minutes ago, I realised I was logged out of my this (usernamekiran) account automatically. I login to Wikipedia/media/tech using four devices. One computer, and three mobiles. All of them are password protected, and only I have access to them. In case somebody wants to use my computer for any purpose, or mobile browser, I always log out. Logging out from on device logs me out from everywhere. For using other accounts, I always use private browsing, and it doesnt have any effect on my main window's login session. It certainly hasn't been 365 days since my last login (I had logged out recently). So it was surprising when I found out I was logged out of the account. And it is not even Thursday. I have a super-duper strong password, which I changed a few minutes ago. I also checked my global contribs, no suspicious activity found. Has this happened with someone else as well? —usernamekiran (talk)08:57, 13 June 2022 (UTC)
@Usernamekiran: If you are logged in on two or more devices, logging out of any one of them will always log you out on all of the others, regardless of whether it's a private session or not. It's because the action of logging out sends a message to all servers to invalidate all login cookies with your name on. This is not new, it's been the case for at least as long as I've been around (13 years). --Redrose64 🌹 (talk) 09:08, 13 June 2022 (UTC)
@Redrose64: yes, I am aware of that. I don't click on "log out" during the private session. I directly close the window, that doesn't log me out from the other sessions. While editing English, and Marathi wikipedia, I regularly do the account switching/hopping from private window for KiranBOT, and KiranBOT II; it doesn't log me out from main window's "usernamekiran" account. —usernamekiran (talk)09:36, 13 June 2022 (UTC)
Yes, I've had issues too. Ignore and carry on. I suspect that sometimes when someone's doing some IP 'forensic' work that it is possible to get knocked off. Neils51 (talk) 11:49, 13 June 2022 (UTC)
Tech News: 2022-24
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
The new version of MediaWiki will be on test wikis and MediaWiki.org from 14 June. It will be on non-Wikipedia wikis and some Wikipedias from 15 June. It will be on all wikis from 16 June (calendar).
Some wikis will be in read-only for a few minutes because of a switch of their main database. It will be performed on 14 June at 06:00 UTC (targeted wikis). [8]
Starting on Wednesday, a new set of Wikipedias will get "Add a link" (Abkhazian Wikipedia, Achinese Wikipedia, Adyghe Wikipedia, Afrikaans Wikipedia, Akan Wikipedia, Alemannisch Wikipedia, Amharic Wikipedia, Aragonese Wikipedia, Old English Wikipedia, Syriac Wikipedia, Egyptian Arabic Wikipedia, Asturian Wikipedia, Atikamekw Wikipedia, Avaric Wikipedia, Aymara Wikipedia, Azerbaijani Wikipedia, South Azerbaijani Wikipedia). This is part of the progressive deployment of this tool to more Wikipedias. The communities can configure how this feature works locally. [9]
The New Topic Tool will be deployed for all editors at Commons, Wikidata, and some other wikis soon. You will be able to opt out from within the tool and in Preferences. [10][11]
Future meetings
The next open meeting with the Web team about Vector (2022) will take place today (13 June). The following meetings will take place on: 28 June, 12 July, 26 July.
Future changes
By the end of July, the Vector 2022 skin should be ready to become the default across all wikis. Discussions on how to adjust it to the communities' needs will begin in the next weeks. It will always be possible to revert to the previous version on an individual basis. Learn more.
Warning when trying to update a file which prompts that your updated file cannot be accepted.
Upload warning, SVG does not match
I'd like help with an error that pops up every time I try to update files, it doesn't matter if I match the size of the new image with the old one or the file type.. it will simply give you warnings over and over again, making it impossible to progress, it shouldn't be this hard to update files when you use the correct file type which is shown. Hogyncymru (talk) 15:27, 13 June 2022 (UTC)
Even if it is an oversight, why is it so sensitive?, I shouldn't have to figure out why my file isn't accepted. I'll give you a test if you'd like, it's set up here , if you download the file in question and use that to update the old file see what happens (the file needs updating anyway as it's inaccurate)Hogyncymru (talk) 15:37, 13 June 2022 (UTC)
So did you download an SVG file, edit it in to a PNG formatted file, then try to upload it to replace the old SVG file? If so, after editing you should render it as SVG first. There are many benefits of different file types that those wanting to use or reuse files may consider, so having the name reflect the content is generally useful. commons:Help:Converting#Converting_images and commons:Help:SVG#Converting_to_SVG have more information on this as well. — xaosfluxTalk15:44, 13 June 2022 (UTC)
"It's so sensitive" precisely because editors who don't understand what they're doing can ask editors who do and actually receive help with the issue.
I assume what you did was that you downloaded one of the rendered .svg files which are rendered to PNG and saved it without changing the file name. And/or you made edits to the PNG subsequently. You are now trying to update the onwiki version, but because you have a PNG and not an SVG, the system is correctly warning you that you have a PNG. To correct the issue, you will need to get the original SVG and then edit that. Izno (talk) 15:52, 13 June 2022 (UTC)
Thanks for the information, it does help me understand better, but why does it matter if it changes file type? I'm a little confused, why can't wiki deal with the problem Iv'e ended up with? is there a desire to update it to make it easier to change the file no matter the file type?Hogyncymru (talk) 16:22, 13 June 2022 (UTC)
why can't wiki deal with the problem Iv'e ended up with Software is not magical. Given the problem you've encountered, no software can fix the issue (at least to the fidelity of the original SVG). This won't be changed in the software. Izno (talk) 16:52, 13 June 2022 (UTC)
And the change you are trying to do will replace a relatively high-quality image with a lower quality one (svg-->png). You certainly can fork the image and upload your PNG file, then change pages to use your new image - but replacing it with a high quality SVG would be better for most everyone else. — xaosfluxTalk17:27, 13 June 2022 (UTC)
@Hogyncymru There are two main ways of storing images. The first is called raster graphics, you make a list of every pixel in the image and you note down what colour they are, a hypothetical raster graphics file might look something like "row 1: pixel 1 = red, pixel 2 = light red, pixel 3 = pink ...". The second is called vector graphics, you store the image as a list of instructions that are required to draw it, a hypothetical vector file might look like "start with a green background 10 by 10 cm, draw a red line from (2,3) to (5,6) ...". Vector files have a number of advantages over raster ones, the main one being that you can zoom in indefinitely and the image will never get blurry because it can keep on rendering more detail. SVG is a vector file format. PNG is a raster file format. "wiki" cannot automatically change the file format here because it is very very difficult (bordering on impossible) for a computer to take a finished image and figure out a sensible set of brushstrokes it would need to reproduce it - software isn't magic. Even for cases where is is possible to do an automatic conversion (like converting between different types of raster graphics) it has the potential to be problematic and may involve making editorial decisions that affect how the image looks - in the conversion process you can loose detail and create artefacts, you can mess up colour scales and brightness, you can mess up transparent layers, you can end up needing to pad files with borders or trim off the edges, etc. If you want to edit an image the best thing to do is to keep it in the same file format throughout. 192.76.8.94 (talk) 20:17, 13 June 2022 (UTC)
It's Thursday and I think there might be a system problem but I don't see any messages here about it. It goes from minor things, like my edit count page not updating, to curious things, toolforge says that an AnomieBOT III report was issued at 21:56 UTC when no report has been issued since this morning, to Wikipedia:Database reports/Empty categories, which was just updated a few minutes ago, including a category that was deleted 15 hours ago.
Since AnomieBOT III wasn't working, I used a Quarry query to find broken redirects in article space and, yes, it did find one...one that existed early this morning and was later fixed a few hours later so it wasn't broken any longer! It's like the system is working with a version of Wikipedia that existed over 12 hours ago, not its current state with all of the changes that have happened since early this morning. Any ideas on what's up? Thanks for any insight your collective minds can offer. LizRead!Talk!01:17, 10 June 2022 (UTC)
Toolforge bots and tools like Quarry rely on copies of the Wikipedia database, occasionally the process that keeps them up to date breaks and they fall behind. According to the replag tool, it's currently 16 hours behind. I filed phab:T310325 to let the wiki replicas maintainers know about it. Legoktm (talk) 01:43, 10 June 2022 (UTC)
I love when people who know how this MediaWiki system works have immediate answers to my questions! Thank you, Legoktm. I guess the only part that surprises me is that it's just now being noticed. Thank goodness we have regularly scheduled bot reports, sometimes the only time I know something has gone wrong is when they don't operate as they always do so dependably. LizRead!Talk!02:16, 10 June 2022 (UTC)
The latest message, linked from the phab ticket, says that it should be recovered by Monday. Enjoy the weekend, everyone! – Jonesey95 (talk) 05:01, 11 June 2022 (UTC)
Monday, huh? Those bot reports are going to be loooong when they finally get updated. That's 4 days worth of data on things that need fixing. Monday will be a busy day! LizRead!Talk!05:44, 11 June 2022 (UTC)
The latest info on that phab ticket is that Monday is the latest it would take to recover from this system lag, they expect some parts to be back to normal tomorrow, on Sunday. LizRead!Talk!01:12, 12 June 2022 (UTC)
The lag maxed out at about 75 hours and then took multiple hours to recover, so services that depend on an up-to-date database were down for more than three days. Ideally, we will receive some advance warning when this will happen again, which it apparently will, since this outage was caused by normal (but infrequent) updates, not a system failure. – Jonesey95 (talk) 20:47, 12 June 2022 (UTC)
Important reports
@Liz and Jonesey95:: could you share which reports you find the most important to have updated on a regular basis? We should investigate moving those into MediaWiki special pages so they are supported at a production level. Legoktm (talk) 19:41, 11 June 2022 (UTC)
Here are a few that I look at daily, or more often:
Also see this thread. It occurs to me that converting reports to Special pages may not be desirable, since Special pages are not typically watchable or customizable and have no talk pages or history to generate diffs from. For example, looking at a daily diff of Wikipedia:Database reports/Unused templates/1 is very useful, since it shows any new entries in a straightforward way. – Jonesey95 (talk) 20:35, 11 June 2022 (UTC)
Oh, there are so many. It is easier to name the bots whose work I rely on. They include AnomieBOT III, SDZeroBot, BernsteinBot, DumbBot, JJMC89 bot, these are the main ones, they all issue reports daily or throughout the day, that include pages that need tending to. I can track down the exact pages if you need those. So far, SDZeroBot's and DumbBot's reports don't seem to be as affected maybe because they don't rely as much on up-to-date information database information since PRODs were tagged 7 days ago and expiring drafts can be predicted far ahead. Other bots that report broken redirects or empty categories rely on edits that have happened recently, over the past 24 hours. I also rely on Special Page reports but those, so far, have been unaffected by this system lag.
The phab ticket has some updated information on when this problem will be resolved but I think I saw you post there, Legoktm, so you've probably seen it. I think the most frustrating aspect of this is seeing the tech folks say that this lag was "unavoidable" which implied that it was expected. There used to be notices I'd see when Wikipedia was going to be worked on and would go into read-only mode, so they could have post some kind of notification, if not to the general pool of registered editors than here at WP:VPT or on the bot noticeboard. I think it's an issue of managing expectations and when some aspect of Wikipedia doesn't work as it reliably has for days, weeks or months, it can cause confusion and alarm. Knowing that the bots will be down for 3 or 4 days would eliminate much of this frustration. As Jonesey95 and you state, they also implied that some bot reports should be moved over to Special Pages which I guess is on MediaWiki but I don't know how realistic or feasible that suggestion is. The special page talk pages are unwatched and the ones I rely upon about categories can only be updated every 3 days which makes them less useful than bot reports which are typically more frequent. You can also come to the bot operator when there are problems and, in my experience, they have been very responsive. LizRead!Talk!00:57, 12 June 2022 (UTC)
If you don't mind a dumb question, what is this distinction that the developers make on the phab ticket between a task or job being "production" or it not being production? Does this define what problems they work on and ones that are someone else's responsibility? LizRead!Talk!01:05, 12 June 2022 (UTC)
Good question, it's really jargon these days. Broadly speaking, Wikimedia infrastructure are divided into two main categories, "production", which consists of the servers, databases, and associated services that run all the wikis (anything on the primary wikipedia/wikimedia/etc. domains), and "cloud", which consists of Toolforge, Quarry, Beta Cluster, and more (usually on wmflabs/wmcloud/toolforge domains). production is ultimately maintained by the Wikimedia SRE team (~45 people), who are on-call 24/7 in case of outages/emergencies. In contrast, cloud is maintained by the Cloud Services team (5 people), which doesn't have amount of staffing for on-call. (I haven't been on the Cloud team since late 2020, so I don't have a good sense of that the current threshold for outages that would page people on a weekend/late night).
Complicating matters is that the wiki replica databases straddle the production/cloud boundary (the sanitization step occurs in production, while the views that allow tools to make queries are maintained on the cloud side), leading to shared responsibility, which both is helpful but can also make it more difficult in diagnosing what is wrong and who is responsible for that part (when I was trying to escalate the issue on IRC I pinged the wrong group of people initially). I hope that answers your question on what "production" is.
In this case, the lag is caused by a schema change on the production databases that is taking quite a while to apply to the wiki replica database. Up until recently, schema changes were a really manual and rare process. I think we're seeing some of the growing pains of being able to do schema changes faster - and I expect that we'll get better at notifying editors or avoiding the lag entirely over time. Legoktm (talk) 04:19, 12 June 2022 (UTC)
Wow, that is really helpful, Legoktm. You are clearly very familiar with the backend of the system we all use. I appreciate the explanation and how some tasks might straddle the line between work groups. Do you expect that schema changes will happen more frequently and we will run into these lags more often? Again, it's about expectations, in reality, I think we all appreciate improvements to the system even if they take some time to do. LizRead!Talk!04:37, 12 June 2022 (UTC)
I'll defer to what Ladsgroup said at phab:T310325#7996931. The impression I get is that we have a large backlog of database cleanup, so in the short term we might have some more lag incidents but my hope is that this isn't really an issue long term.
P.S. back in the Toolserver era we had to deal with replag on the scale of months. I think it's really great that we are now complaining about a few days of replag, it shows how far we've come (and that we still have room to improve!). Legoktm (talk) 05:39, 13 June 2022 (UTC)
Thanks both for the input. In my opinion, things that are essential to the operation of the wiki should be part of MediaWiki so they get that "production" level quality of support, both from SREs as explained above, but also from the shared maintenance MediaWiki core gets from all the various devs that work on it versus the one or two botops that operate current reports. MW has a pretty robust system of special pages that can effectively serve as database reports, but as always need more love. I've started working on existing issues like T307314 and T310456 before looking into implementing more reports as special pages.
I am intrigued by the idea of highlighting new entries somehow, I agree that makes the special pages somewhat inferior to wiki pages. Legoktm (talk) 05:56, 13 June 2022 (UTC)
Moving to special pages might have advantages but could cause problems. As one small use case, I follow WP:Database reports/Linked misspellings (many are typos even when not wikilinked) but deal only with its new entries. I want a diff with the previous day, and I want to get that in arrears after a wikibreak. With a report, that comes free with the page history; with a special page it would be difficult. Enhancing or fixing a report can be done locally; modifying a special page would have to join a long queue of tasks awaiting scarce developer time. Certes (talk) 11:56, 13 June 2022 (UTC)
Are contributions from the last two days going to be eventually counted on XTools?
Not sure if this is the right place to ask, but currently, if you look at an edit counter like mine, there is a notice at the top that says "Caution: Replication lag is high, changes newer than 2 days may not be shown." What does this mean, and will the changes from the last 2 days be eventually counted when the lag ends? weeklyd3 (message me | my contributions) 02:15, 12 June 2022 (UTC)
I've noticed that this user has contributions and a live talk page and does not generate the nonexistent user message, but the Xtools tools fail to find this user. They have 3 edits to date, all of which regarded a BLP draft that has been speedy deleted per G10. –LaundryPizza03 (dc̄) 11:51, 12 June 2022 (UTC)
I have no doubt that I'm right here, but that's between me and the reverting user. Thanks for showing me how to do it and for informing me of the revert. Animal lover|666|06:21, 14 June 2022 (UTC)
Display of numbers goes wrong in this list in mobile view only
Please see this mobile version of List of vehicle speed records. Evenon a wide screen, it geos terribly wrong in some cases (especially in the section "Rail vehicles"), but on a smaller mobile screen, nearly all the numbers get a confused jumble across multipe lines, making it impossible to correctly read this. I have no idea what specific element causes this (perhaps the Template:decimal-align?), I guess the people at VPT are more qualified to answer this. That mobile editing has numerous serious issues is well-known, but mobile viewing should work... Fram (talk) 08:35, 13 June 2022 (UTC)
@Fram does the first table there look better for you in this revision? The page authors have applied some very specific width allocations on those tables and their columns - in the revision I linked above I let the first table just fill naturally without that specification (left the other tables for comparison). — xaosfluxTalk09:47, 13 June 2022 (UTC)
Thanks, but the improvement is minor. On a widescreen, it is better, but when reducing the width e.g. the first cell still reads "1,2.98" (first line) "275" (second line)" instead of "1,227.985" (or e.g. "1,227" (first line) ".985" (second line)). For some reason he is line-wrapping the number in this very strange way: while the wrap itself may be caused by the widths, the wrapping in two parts should never happen and seems to be due to either a local template (convert or decimal-align) or a more global technical issue. Fram (talk) 09:54, 13 June 2022 (UTC)
Yup. There is a proposal to do this aligning properly in CSS, based on the character, but that's not supported in any browser yet. The best way for now to do this is simply have an equal amount of numbers after the decimal point and right align. These kinds of hacks that decimal-align applies are really shaky. —TheDJ (talk • contribs) 12:55, 13 June 2022 (UTC)
May need both of these to fix it for more screens, add trailing zeros to make the decimals the same length, and get rid of the fixed width allocations. — xaosfluxTalk13:40, 13 June 2022 (UTC)
Yeah, I'll post an RfC at the article if required but...
I don't know how to construct these particular files/maps so have posted elsewhere around Wikipedia asking for help - Wikipedia:Graphics Lab/Map workshop, Talk:COVID-19 pandemic in Tennessee, Wikipedia talk:WikiProject COVID-19, a user talk page - trying to get them updated but have been unsuccessful so far. The information on these data-maps dates from February 2021 and March 2021...I have found sources to update the info but graphics-coding is not my strong suit. I am at the point that I think the maps should be excised from the infobox and am wondering what the Village Pump:Technical community thinks.
If anyone around here can update the 4 maps, YAY! Here are the sources I have found. Some of the initial pandemic data sources have gone dark so getting at June 2022 info isn't as easy as it might have been early on.
These are the 4 2021 data-maps, they are hosted on Commons:
Right place... Well, bless your heart. Did you see that I had mentioned above I had already asked at both those places? (Plus some other WP-venues before I posted here?) Virtually no replies. Oh well, at this point I'm going to just delete the maps - they are so far out of date they are useless. But thanks - Shearonink (talk) 21:49, 14 June 2022 (UTC)
Honestly, Wikipedia editors seem to have lost interest in regularly maintaining COVID-19 stats. If no one updates them, be bold and remove them from the infobox. A much better position for them will be the February 2021 section in te timeline. —CX Zoom[he/him](let's talk • {C•X})22:04, 14 June 2022 (UTC)
Yeah, I know the data is overwhelming and we are all overwhelmed - I get that, there's a g-d- pandemic going on. I just wish I could get someone to do the updated maps to sometime this year.... I'll remove them sometime next week when I can sit down and figure out where to stash them in the timeline of the article. Thanks for your input CX Zoom, that's a good idea. I do hate to lose visible good work off of an article. Shearonink (talk) 23:31, 14 June 2022 (UTC)
Wrong message when restoring an old version of a page.
For some reason the text of MediaWiki:Postedit-confirmation-restored has been changed from "The page has been restored." to "The page has been undeleted.". When you do a manual revert on a page you therefore get a message about the page being undeleted, rather than being restored to an old version. Is this an intentional change or a bug? I don't think "undeleted" makes much sense in this context. Could someone override the new message locally until this is resolved? 192.76.8.94 (talk) 11:56, 15 June 2022 (UTC)
192: can you be more specific, I haven't run in to this yet. What do you mean when you say "a manual revert"? When I go to a page history, edit an old version, and publish it ("a manual revert", such as in testwiki:Special:Diff/514665) I'm not seeing this message. What are the exact steps you are doing that is leading you to this? — xaosfluxTalk13:22, 15 June 2022 (UTC)
@Xaosflux This morning I did a manual revert on a file description page. I opened the version of the page I wanted to restore, Entered an edit summary about reverting sock edits, then saved the old version of the page without changing anything. The green "edit successful" popup was displayed, but instead of saying "The page has been restored." it said "The page has been undeleted.", which doesn't make sense in this context. 163.1.15.238 (talk) 13:48, 15 June 2022 (UTC)
@Xaosflux Thank you for opening a phab ticket. Sorry that my original message was unclear, I'll give more details when reporting problems in the future. 163.1.15.238 (talk) 15:03, 15 June 2022 (UTC)
No worries, it may have been more obvious to someone else, that you were watching and followed up was most helpful! The dev's seem to agree this is in error and are looking at an upstream fix. — xaosfluxTalk15:06, 15 June 2022 (UTC)
If you check it right now, you'll see what I've seen in the past, a category for non-empty categories that is actually full of empty redirect categories! In the past, I have made a minor edit to the redirect category, like adding a blank line or space, and that has removed it from this category. I don't know why basically a null edit would do this but it does. But I really don't want to spend time tonight making pointless edits to 59 redirect categories so I came here looking for a longer term solution. Do you know why some redirect categories would randomly show up in an incorrect category when they, in fact, are empty? Thanks for any help you can provide. LizRead!Talk!02:28, 16 June 2022 (UTC)
@Liz The English wikipedia is too big and gets too many page views to update things in real time. Rendering a page (i.g. turning the wikitext into HTML that can be sent to readers and calculating all the metadata about the results) can take a significant amount of time, 20-30 seconds or more for big or complex pages, so to save computing time the results are calculated and stored in a cache for later use. Mediawiki uses things called "links tables" to keep a record of all the links in a page and what types they are, which are used to fill up categories and pages like Special:WhatLinksHere. To save processing time this table is not updated when the page is viewed or purged, it is only updated when a page is edited (on the assumption that that is the only time a page will have new content added that needs to be scanned for links), hence why you can only update the categories on a page by null editing them.
When it comes to categories there's another complication to consider - it is possible to add or remove categories from millions of pages at once by editing templates. To prevent the servers getting locked up trying to add categories to loads of pages at once mediawiki uses what is called a "job queue" - instructions that pages need to be added to or removed from a category via a template are stored up and processed when the servers are less busy - this is why it can take weeks/months for some template populated categories to empty.
The best solution here is probably a bot. The API has a special option you can set when you can set when purging a page called forcelinkupdate which forces the page to update the links table, something the normal Special:purge page doesn't do. A bot could scan over the category and purge any pages in it a couple of times a day. 192.76.8.94 (talk) 03:20, 16 June 2022 (UTC)
I don't know if the matter described at Wikipedia:Village pump (technical)/Archive 196#G12 category populated by ghosts was ever fully cleared up - if not, it is probably caused by the same problem. Generally speaking, whenever you have a page (which we shall call "Foo") that is showing a category at the bottom (which we shall call "Bar"), but the Category:Bar page isn't listing Foo, the fix is to perform a WP:NULLEDIT on Foo. Similarly, if the Category:Bar page lists Foo, but Foo doesn't show Bar among its categories, again perform a null edit on Foo. Of course, you should have enabled Preferences → Appearance → Show hidden categories first. --Redrose64 🌹 (talk) 07:41, 16 June 2022 (UTC)
Malformed entries derived from preloads - is there a better way to do this?
Sometimes, a link opens an edit page where users can fill in content. Sometimes, the new section already has some text in it. Many times, I've seen users write their request/contested speedy deletion/something else under the signature inserted by the preloaded text.
For example:
== Contested speedy deletion ==
This page should not be speedily deleted, because (your reason here) [[User:Example|Example]] ([[User talk:Example|talk]]) 19:08, 15 June 2022 (UTC)
(a reason why the page should not be speedily deleted)
oder
== Edit request ==
{{request edit}}
[[User:Example|Example]] ([[User talk:Example|talk]]) 19:08, 15 June 2022 (UTC)
(description of the edit request)
Is there a way for preloads to use the DiscussionTools, which automatically inserts signatures, instead of the edit screen? So that when users write their request at the bottom, it is autosigned.
I'm talking about something like this:
User clicks 'Request edit' button
'New section' form appears with a preload like this:
{{request edit}}
<!-- State UNAMBIGUOUSLY your suggested changes below this line, preferably in a "change X to Y" format. Other editors need to know what to add or remove. Blank edit requests will be declined. -->
User fills in edit request below comment and saves the page.
DiscussionTools fills in the signature automatically.
We've considered this when working on the DiscussionTools new topic tool, you can read the discussions at T269310. (I'm one of the developers.)
It would be possible to add, and we had a mostly working implementation, but we've discovered that this would probably have to be enabled individually for each use of preload, because the existing ones wouldn't always work well with it (there are some that don't expect a signature at the bottom or a heading at the top, and some that don't work well in visual mode – particularly those using <!-- --> comments and subst: templates).
@Matma Rex It would definitely be nice to have this as an option to opt into using the new topic tool with preloads, there would just need to be a way to specify a fallback. --Ahecht (TALK PAGE) 13:10, 16 June 2022 (UTC)
How to get page views for all fungi articles for a partner organisation
Hi all
I've recently started a project with Kew Gardens to try to improve coverage of plants and fungi. I'd really like to give them a number of page views to describe the audience for fungi articles. I've tried using Category:Fungi but this includes people and products eg Marmite and Beatrix Potter. The is a Wikiproject box which should capture all the relevant articles but I don't understand how to get this into a pagepile or similar. If someone could do this for me it would be really really helpful.
One option is to use SPARQL on Wikidata. This query works just for Amanita. (Click the triangle in a square, bottom left.) In theory, you can replace Amanita's id (Q213938) by that of fungus (Q764). In practice, there so many results that it times out. You may be able to fix it by working through all the taxa at some suitable level one at a time. Certes (talk) 13:20, 16 June 2022 (UTC)
Installing AutoWikiBrowser in Linux
Hi,
I'm having trouble installing AutoWikiBrowser on Linux. I tried installing it with Wine, but it doesn't work. Can you walk me through how to correctly install it on Linux? Thanks, Interstellarity (talk) 12:34, 16 June 2022 (UTC)
Good luck: it can be a pig. User:Certes/AWB on Ubuntu may be of help. Some of what I did there is unnecessary, but I don't know which bits. It will also need changes if you use a different Linux distribution. I now use wine-5.0.3, which works at least as well as the 4.0.2 I originally had, so any version between should be fine. AWB now works for me, though not smoothly. I usually use JWB, which does simple things just as well, and reserve AWB for the tasks JWB can't handle. Certes (talk) 13:01, 16 June 2022 (UTC)
Preview image for Charles V, Holy Roman Emperor uses signature instead of portrait
I originally brought this issue up in the Teahouse, and was directed to here.
The preview image for Charles V, Holy Roman Emperor's article is the second image—his signature—even though the first image has a full portrait. I was told that maybe the dimensions of the image (1771×3000 px) makes it less preferable to the extension that picks the preview image, but it seems to me that the ratio (0.59 or 1.69, depending on how it's calculated) is well within the allowed ratio by the extension. Is there any way to force the extension to either pick the portrait or not pick the signature?
(By way of suggestion as well, could it help if the extension's criteria be modified to score colored photos higher than pure black-and-white?) — Preceding unsigned comment added by Yo.dazo (talk • contribs) 23:19, 16 June 2022 (UTC)
I've added notpageimage to the signature block of the infobox. This will remove signatures from being page images on all pages for all signatures. I think that's probably reasonable for signatures. Izno (talk) 23:42, 16 June 2022 (UTC)
Adiutor (new gadget proposal)
Hi everyone, I would like to propose a new gadget I coded this gadget with OOUI from scratch, this gadget helps various user processes, I first coded Adiutor for Turkish Wikipedia, then I thought about adapting it here, I don't know how the application process works here. I am still developing this gadget.
Hi Vikipolimer, we don't normally jump right to "gadget" level for new scripts until they have some uptake. Feel free to list at Wikipedia:User scripts/List and advertise it gently in places it could be relevant. I suggest you put a backlink comment in your directions so that you can track the 'what links here' adoptions. If you script starts getting quite popular, then yes this is the venue to ask for it to be moved from a WP:USERSCRIPT to a site gadget. — xaosfluxTalk00:23, 17 June 2022 (UTC)
Twinkle already does speedy deletion and proposed deletion. I don't think you'd get approval for activity status for a gadget, but I think there is already a script for that. Recent diffs can be obtained from (ancient) popups gadget. (I don't know if there are others.) Copyright violations might reasonably be scripted, but OTOH I know that our copyright violations processes already suffer from insufficient processing access.
And any way, I'm not really sure all those different functions should be in one script.
I don't mind the idea that there should be competitors to our existing gadgets and scripts, but it might be more useful to work on those directly instead. (I understand that's maybe difficult to think about since you have already started and it was originally made for Turkish WP.)
As an aside, there's a general expectation that OOUIjs will at some point in the nearish future be deprecated in favor of Codex, which is also designed by the WMF and is based on Vuejs, so it uses modern Javascript. Izno (talk) 00:30, 17 June 2022 (UTC)
Thanks for your review and advice @Izno, I plan to add some key features to this script that will distinguish the new tool from Twinkle, of course, I'm not asking for new ideas here, but it's a good start to create a roadmap, thank you again. Also, I had no idea about Wikipedia will pass the VueJS. 𝗩𝗶𝗸𝗶𝗽𝗼𝗹𝗶𝗺𝗲𝗿 ℣00:54, 17 June 2022 (UTC)
:@SD0001: The article I created on Alexander S. Wolcott passed GAN and was promoted to Good Article yesterday, however Legobot notified me it had failed. Legobot got confused. The GA Counter missed this one also. I really have 218 Good Articles, but the GA Counter shows only 216. Seems to be a pattern here. Technically beyond me to know how to fix. Thanks for help. = This one solved.--Doug Coldwell (talk) 10:04, 15 June 2022 (UTC)
@SD0001: The article I created on John Johnson (inventor) passed GAN and was promoted to Good Article 01:15, 17 June 2022, however Legobot notified me it had failed. Legobot got confused. The GA Counter missed this one also. I really have 219 Good Articles, but the GA Counter shows only 217. Seems to be a pattern here. Technically beyond me to know how to fix. Thanks for help.--Doug Coldwell (talk) 09:21, 17 June 2022 (UTC)
Give me some time. I'll sort it out. For the GA counter to work, Legobot or someone else needs to add {{good article}} to the article. John Johnson now has it, and so it's indeed being counted. As for Lambert Gas, it just got missed. I can add the entry manually in the database, but I am hoping to do some analysis and figure which other articles got missed randomly, so that the issue doesn't arise in the future. – SD0001 (talk) 13:33, 17 June 2022 (UTC)
Enter an interwiki or namespace prefix into the left hand search box that appears, e.g. "c:" or "phab:" or "Help:"
Select the text in the search box, and copy it by dragging to the right hand search box
Single letter interwiki prefixes are replaced by the letter followed by an underscore, e.g "c:" becomes "C_". Multi-letter prefixes are replaced with the word "download".
There seems to be some kind of javascript (?) involved which is stripping prefixes from page titles for some reason, e.g. "Help:Getting started" is replaced with "Getting started" when copied across in this manner. Is this an intended feature? I don't see how it is useful to remove namespace prefixes on copying, it just seems like it would end up sending people to the wrong place? Normal words can be copied across in this manner with no issues. Tested on chrome. 192.76.8.88 (talk) 21:26, 18 June 2022 (UTC)
(1) Vector is the default desktop skin so if someone doesn't mention the skin it's safe to assume it's Vector. (2) There indeed are two boxes even in Monobook, just on the left side instead of top right. Nardog (talk) 00:07, 19 June 2022 (UTC)
I wish it was safe to assume Vector and other defaults but many users fail to say they have changed preferences and installed scripts or browser extensions when they report issues. In this case it was an IP so Vector seemed relatively safe to assume. It could still be a registered user who was logged in when testing but not when posting. It would be a lot easier to answer issue reports if all reporters were sensible. PrimeHunter (talk) 00:21, 19 June 2022 (UTC)
(edit conflict) All IP's have Vector. It's the default skin. Other options are only for registered users. MonoBook also has two search boxes at your link: The normal in the left pane and the larger one at top of Special:Search. There are also Chrome issues in MonoBook no matter which search box you try to drag text from. PrimeHunter (talk) 00:13, 19 June 2022 (UTC)
It appears to be a Chrome feature to download something by dragging it from a browser window. Colons are usually disallowed in file names so if the feature thinks that text is meant to be a file name, it may replace colons by underscores. I don't know whether a site can ask Chrome to ignore the feature in some circumstances. PrimeHunter (talk) 23:00, 18 June 2022 (UTC)
Vector border (i.e. under article titles) appears lighter
I've noticed that the Vector theme's border color has changed to a lighter shade of gray. Noticed it when I was logged out a few minutes ago (logged in I have monobook). Figured I'd bring it up here in case it was mistaken, not sure where all the master CSS is or where changes to it are discussed so I don't know where to verify whether it was intentional myself. If it was intentional, of course, disregard this. Thanks. --Duonaut(talk | contribs)04:27, 19 June 2022 (UTC)
Hi all, to me, the right-part of the navigation links above a page (Read/Edit source/.../More/TW) appear to have shifted a lot towards the left and now slightly overlapping over the left-part of the navigation links (Article/Talk) on Vector 2022. I'm on mobile using desktop version but this is a brand new issue on it. Anyone else facing this? Something to do with today being Thursday? —CX Zoom[he/him](let's talk • {C•X})20:15, 16 June 2022 (UTC)
Cleaning up some dusty corners of WP, I found a page using the citation Template:Appletons' which is throwing a redtext error as the cite encyclopedia template needs a title (or it could be some deeper bug). I have no idea how to code templates and maybe one day I'll want to learn, but today's not the day. It seems like a quick fix for those in the know, so if someone has a few minutes to look at it it'd probably be helpful for a lot of pages still using this template. Thanks all. SamuelRiv (talk) 21:58, 18 June 2022 (UTC)
Not a WP:VPT issue.
{{Appletons'}} uses {{cite encyclopedia}}. The error message is telling you that the citation needs the name of the article in Appletons' that supplied text for the Diego Delgadillo en.wiki article; this because surely the whole of Appletons' is not the source of that borrowed text. Use either of |wstitle= (if the Appletons' article is available at Wikisource) or |title=.
I have thought about implementing something like that – I've done it for some templates that use {{lang}} → Module:Lang – but haven't gotten farther than that.
Thanks guys, didn't know that about the template. Nothing on WikiSource for my guy, so it'll stay redtexted for now, but we'll see if it's posted elsewhere. SamuelRiv (talk) 06:34, 20 June 2022 (UTC)
Radio buttons to select version on Page history
Regarding WP:PAGEHISTORY, From past 4-5 or so days, I am seeing that on the page history the latest and the oldest versions are auto selected. Earlier it was the first two versions that were auto selected. Not sure who or how this is changed, but I would like to go back to the default option where the top two versions are auto selected. Please ping when you respond. Venkat TL (talk) 11:31, 20 June 2022 (UTC)
@PrimeHunter top 2 versions selected when I clicked the safe mode link (in that link I can see myself logged in) Skin is default one. Browser firefox (Vector legacy 2010). Disabling the user scripts did not help. Venkat TL (talk) 12:22, 20 June 2022 (UTC)
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
The Wikipedia App for Android now has an option for editing the whole page at once, located in the overflow menu (three-dots menu ). [14]
Some recent database changes may affect queries using the Quarry tool. Queries for site_stats at English Wikipedia, Commons, and Wikidata will need to be updated. Mehr lesen.
A new user_global_editcount variable can be used in abuse filters to avoid affecting globally active users. [15]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 21 June. It will be on non-Wikipedia wikis and some Wikipedias from 22 June. It will be on all wikis from 23 June (calendar).
Users of non-responsive skins (e.g. MonoBook or Vector) on mobile devices may notice a slight change in the default zoom level. This is intended to optimize zooming and ensure all interface elements are present on the page (for example the table of contents on Vector 2022). In the unlikely event this causes any problems with how you use the site, we'd love to understand better, please ping Jon (WMF) to any on-wiki conversations. [16]
Parsoid's HTML output will soon stop annotating file links with different typeof attribute values, and instead use mw:File for all types. Tool authors should adjust any code that expects: mw:Image, mw:Audio, or mw:Video. [17]
recent version update mangled the display (vector-2022), and the top menu shows now "TalkRead". the "talk" part links to the talpage, and the "read" part does the same as the leftmost menu item ("Project" here, "Article" in mainspace, etc.).
arguably, the "read" part is superfluous, and in any case, as it is now, it's harmful, and mangles the display.
no doubt this will be fixed soon by the software, but in the meantime, it can be "patched" locally by editing medaiwiki:Skin-view-view, and saving it as empty page.
It looks like this is by design, one is an "indicator" that you are on the main page or the talk page, the other is an indicator that you are in 'read mode' or 'edit mode' while providing nav back to the other modes. I'm not seeing this "mangling" the display or otherwise making it ever have the words "TalkRead" actually attached to each other? Similar to Wikipedia:Village_pump_(technical)#Thursday_issues_in_Vector-2022? above, if you are seeing overlapping text would you please open a WP:BUG, include exact steps to reproduce, include screenshots, and then let us know the task number for follow up? I'm guessing you are possibly doing something like trying to use this skin with some sort of extreme resolution? — xaosfluxTalk10:24, 20 June 2022 (UTC)
@Kipod, I don't see any issue in the English Wikipedia. I tried it with ?useskin=vector-2022, logged-in and logged-out, Firefox and Chrome. There may be a problem somewhere, but I suspect it's not universal. Amir E. Aharoni (talk) 12:00, 20 June 2022 (UTC)
guess it's something with my account. i should have tested it in "safe mode" before bothering the community. apologies.
issue still exists, and is not user-account dependent, and also visible out-of account and in safemode.
_however_, it only visible when using chrome, and so far only detected on linux (specifically, "mint" distribution, with current chrome showing "Version 87.0.4280.66 (Official Build) (64-bit)".
i guess "issue with specific version of chrome, visible on linux only" is not worth dealing with. FF on same OS does not show the issue.
MORE PRECISE DESCRIPTION:
the issue has nothing to do with "read", but rather with the 2nd half of the top men, which is split in two: left part has "Project page" (or "Article" or whatnot, depending on namespace) and "Talk", and right part has "Read", "Edit source" and so forth.
when the issue materializes, the right half justifies leftward, such that "Read" is concatenated to Talk in the left part (hence "TalkRead").
again, this was only viewed with chrome browser, and only on linux. i tested chrome on chromebook and windows, and both display correctly, as does FF on linux.
@קיפודנחש the Chrome version you reported above is from Nov 2020, the current stable release is 102.0.5005.125 - can you verify if your browser-specific problem resolves if using the current version of Chrome? — xaosfluxTalk18:49, 20 June 2022 (UTC)
in the meantime, i dug a little more, and found that it's mw bug.
other browsers (and apparently, newer version of chrome) are more lenient, but it will be nice if mw can fix it anyway.
the issue has to do with "justify-content" attribute of the 2nd half of the menu (element id is #right-navigation).
mw set it up as "end" (i.e., justify-content: end;). this is illegal, even if most browsers respect it anyway: it's a "flex" container, and w3 says you should use "flex-end", not "end". see https://www.w3.org/TR/css-flexbox-1/#justify-content-property. when i manually changed the value on my browser to "flex-end", the display straightened up and showed correctly.
admittedly, this is a small bug, maybe not worth fixin', especially since most browsers are nice enough to interpret "end" as "flex-end". i happen to use linux mint, and apparently they are using way behind upstream top version of chrome. i can live with it, but it will be nice to fix the bug and use correct value for this attribute.
needless to say, the "solution" i suggested in 1st post is complete junk.
@קיפודנחש thanks for the udpate, please see WP:BUG for how to report the bug on phabricator where it could be fixed for all mediawiki users, placeing a local language-specific message hack isn't the right way to solve this. Please do let us know the task id so we can note it here for anyone that comes across this. — xaosfluxTalk19:11, 20 June 2022 (UTC)
thanks. i'll pass on the suggestion to create a task on phab:. i added the "update" here for completeness...
I think the W3 spec is outdated (at least compared to what browsers implement). The CSSWG's CSS Box Alignment Model Level 3 draft knows about end, though MDN's compatibility table does say it's only been supported in Chrome since version 93. So I'd say this isn't a bug per se, though it may still warrant changing, since both IE and Safari notably don't appear to support the feature. Rummskartoffel19:29, 20 June 2022 (UTC)
Don'r rely on anything that isn't a full W3C Recommendation. CSS Flexible Box Layout Module Level 1 is a W3C Candidate Recommendation, and has been since March 2016 - it's been revised four times since then, so might not yet be stable enough to progress to a Proposed Recommendation. Indeed, back in 2014 it was busted from Candidate Recommendation back down to Last Call Working Draft, so this may well happen again. The fact that some browser vendors have chosen to implement these new features demonstrates that they are providing users with the ability to conduct tests. So what you could do is provide Chrome's authors with feedback on your experiences with this property. --Redrose64 🌹 (talk) 10:38, 21 June 2022 (UTC)
Well, yeah using a newer browser resolved the issue I noted in the above section. Noting that (even new version) Edge on Android probably doesn't respect the justify-content: end; thing. Chrome does though. —CX Zoom[he/him](let's talk • {C•X})19:41, 20 June 2022 (UTC)
Image request for UK people or Ireland
Is this possible to request images? Is anyone going to Restricted Forest Festival, Back to Festival, Falkirk Club, Edniburgh City 7S/Dundee Club, Glasgow Club, Drogheda Club, Colour Clash Festival/Club Cardiff, Clubland in the Beach, Doncaster Club or Guildfor Club in UK or Festival Punchestown in Ireland and could take photo of event and Basshunter? List of locations: Facebook, Instagram or Twitter. Eurohunter (talk) 16:17, 21 June 2022 (UTC)
@Xaosflux: Good idea but there could be option to reach and more or less ask directly Wkipedians who live exactly in these places or if they are just going there for ocassion than listing them in more general category. Eurohunter (talk) 17:14, 21 June 2022 (UTC)
A long-time editor I know (since '02!) who took a break for many years, now needs a password reset. They didn't set an email in user prefs, but list their email on their userpage. Is emailing ca@ still the right way to ask a dev to set the email for the account? – SJ +20:41, 17 June 2022 (UTC)
@Sj: yes, emailing ca@ is the best way, especially if you can vouch for them being the legitimate account owner. Legoktm (talk) 05:10, 18 June 2022 (UTC)
The devs will require absolute evidence that the long-time editor concerned really is the same person that was using the login ID concerned. As noted by Xaosflux, it's much easier for the long-time editor to just create a new account. My advice is that an early (if not the first) edit of that new account should be to create a user page stating something like "This user has previously edited under the name of (insert previous login ID) but lost their password". That makes it plain that sockpuppetry is not involved. Also, get them to set that email address. --Redrose64 🌹 (talk) 21:11, 18 June 2022 (UTC)
I would think the edit history of their user page showing that they listed their email on their user page should be evidence enough that that editor owns that email address. --Ahecht (TALK PAGE) 14:14, 21 June 2022 (UTC)
Minor one-time glitch, probably: exception encountered when moving page
Just wanted to put this somewhere before I lose it. Tried moving Lisa Ellis (fighter) to Lisa Ellis (martial artist) on Safari on a mobile device on mobile Wikipedia and received the following:
[b92c7ff8-12fe-47d0-9498-dd4c2e77e0cc] 2022-06-22 08:14:36: Fatal exception of type "MWException"
As a person who regularly works with categorization cleanup, I've noticed a recurring problem that I wanted to ask if anybody can solve.
What happens, over and over again, is that somebody adds a {{citation needed}} tag for the population of a community that's using {{Infobox settlement}} without explicitly referencing the population there — but because they frequently pop the tag directly onto the population number in population_total = instead of placing it in the population_footnotes = field where it really belongs, the automatic comma-delimiter function in population_total automatically also comma-delimits the year in the citation tag's transcluded maintenance category, resulting in Special:WantedCategories constantly getting cluttered up with nonsense categories like Category:Articles with unsourced statements from June 2,022 (note the comma in the year) where it really should have been Category:Articles with unsourced statements from June 2022.
But this obviously isn't a thing that those of us who work on category cleanup should be expected to just put up with repeatedly mopping up over and over again — it's a thing that by rights we should never be seeing at all, because it should be flatly impossible for the template to even be able to cause such a nonsense category to get generated in the first place. WP:TEMPLATECAT even explicitly states that "When templates are used to populate administration categories, ensure that the code cannot generate nonsensical or non-existent categories, particularly when the category name depends on a parameter".
Obviously we don't want to completely shut off comma-delimiting in that field, because it's necessary in the population numbers themselves, but is there any way that some sort of "firewall" can be coded so that it can continue to comma-delimit the population numbers while simultaneously being prevented from fucking up the maintenance category if somebody places the citation tag in the wrong field? Bearcat (talk) 18:46, 17 June 2022 (UTC)
So you're giving me broken code that doesn't work in the first place, and even if it did work I still wouldn't have the foggiest notion what I was actually supposed to do with it anyway? I asked for somebody who knows how to fix this to just fix it, not for half-baked advice on fixing it myself. Bearcat (talk) 12:09, 22 June 2022 (UTC)
{{#invoke:String|replace|source={{{population_total|}}}|pattern=%[%[Category:All articles with unsourced statements%]%]%[%[Category:Articles with unsourced statements from %a+ %d+%]%]%<sup class="noprint Inline%-Template Template%-Fact" style="white%-space:nowrap;"%>%[%<i%>%[%[Wikipedia:Citation needed%{{!}}%<span title="This claim needs references to reliable sources%.% %(%a+ %d+%)"%>citation needed%<%/span%>%]%]%<%/i%>%]%<%/sup%>|plain=false|replace=}}
This should return the parameter without any citation needed tags. (It only took me an hour, too.) I would fix it myself, but I'm not a template editor, nor do I have the energy to make an edit request. Sorry. ― Qwerfjkltalk15:18, 22 June 2022 (UTC)
I had this happen one time before where it was more obvious and I was able to remove the offending source. In this case, I don't know if the filter warning was referring to citations present that I had removed which were marked as self published sources, and among the sources I added and read they all seemed reliable to me for the scope of the article. (I could be wrong about that though.) My point is I am getting a warning, which I am not even sure is relevant to my edit, which I can't fully read due to a non-resizable window, and does not clearly tell me which citation/s I added are from a predatory journal. Appreciate any help with this. Darcyisverycute (talk) 11:50, 21 June 2022 (UTC)
This was phab:T184379. I guess they've made the dialog as big as they're going to. It should be scrollable, but also I believe it's on us to use TemplateStyles or something else to conditionally show smaller messages based on the viewport size. This isn't a bad idea anyway as mobile users can face the same problem. — MusikAnimaltalk21:56, 21 June 2022 (UTC)
I am pretty sure the primary problem here is that the caption is set to nowrap. The secondary problem would indeed be related to the message box having certain expectations in which it is displayed, which VisualEditor has more or less so far failed to accommodate since edit filter messages were first supported, and which certainly cannot be fixed before container media queries are well supported (and even in the current proposal, mw-parser-output would also need to be set in CSS to be a valid container for queries, which I've been thinking about filing a task for a bit now and just haven't done so because they aren't implemented yet). Izno (talk) 02:30, 22 June 2022 (UTC)
Why is there an entire table here though.. Just link to a a page with the details...
Another easy improvement we can make. If .oo-ui-popupWidget-popup .*mbox then kill margins, decorative images etc. Is it good, is it how it should be done ? no, but it fixes an immediate problem. —TheDJ (talk • contribs) 14:26, 22 June 2022 (UTC)
Yes, that would fix it, but also split the styles from their soon-TemplateStyled home. (I literally just need to fix Module:Listen and then deploy a new version of Module:Side box before we can do that. We're almost there!)
Yes, but on mobile we're at a resolution where you could use media queries to affect the presentation. (That we don't yet is simple technical debt that at least TheDJ has played around with resolving but which hasn't been merged into the relevant styles.) In other words, VE continues to have no reason why this flyout is as small as it is. Izno (talk) 16:06, 22 June 2022 (UTC)
How to print a specific part source code of an article?
Hi, is there any template, tool, query or anything else to print a specific part of an article source wikitext given the article title and regex pattern? For example, I want to get the entire coord template from Ab Bar article. Thank you! ⇒ AramTalk14:30, 20 June 2022 (UTC)
@Certes: Thank you for your reply! That’s a great template and solves a lot of my problems, but in my question I meant the wikitext, not the actual value. For example, I want to get this: {{coord|36|55|21|N|48|58|28|E|region:IR|display=inline,title}} Thanks! ⇒ AramTalk20:48, 20 June 2022 (UTC)
The nearest I can get is something like {{#invoke:string|match|{{:Ab Bar}}|%{%{coord.%}%}|}}, which does not work because {{coord}} has already been expanded before #invoke:string sees the wikitext. A modified copy of Module:Template parameter value should be able to do the job, or there may be some other way to get the actual wikitext without its templates getting expanded. (That shouldn't be hard to write, and I can't believe no one has done it, it's just a question of guessing what they might have called it.) Certes (talk) 21:50, 20 June 2022 (UTC)
@Certes: Thank you very much for your answers! I hope someone makes such a template because it is really important to have one. ⇒ AramTalk11:33, 21 June 2022 (UTC)
Thanks: Page is exactly the module I was looking for. The above solution preserves a typo from my failed attempt. I left out a hyphen (non-greedy quantifier); it should read {{#invoke:string|match|{{#invoke:Page|getContent|Ab Bar|as=raw}}|%{%{coord.-%}%}|}}. Certes (talk) 21:36, 22 June 2022 (UTC)
@Andrybak: Actually, we are working on creating all Iranian cities (about 1,000 articles on ckbwiki) using a bot. Our biggest problem was getting the coordinates. The coordinates are on Wikidata, but some pages (on Wikidata) have more than one value and we should use regex to shape them as a wikitext. However, I worked hard on them after your reply, but didn't get very accurate results. I was very pleased with your answer and I am sure I can use it elsewhere. Thank you very much for your reply!
@Qwerfjkl: and @Certes: It's incredible! Thanks for improving the code above to give us exactly what we want. I was working on the code above a little while ago, but I put a plus sign (+) after the dot (.) and brought it up to the stub template. :( After User:Certes replied, I was surprised that you made it. :) That would be great if you can make it into a new template for those who want to get Wikitext in the future. I still can't believe it! Thank you all so much! ⇒ AramTalk22:26, 22 June 2022 (UTC)
Main Page Popup
Hi, I also asked this question on the main page talk page. Upon using an internal wikilink to the main page (like this), the page preview shows an error like the one in the screenshot.
Wikipedia Main Page preview
Since the main page is mostly templates and the like, this is expected, but isn't there a way to change the text to something more friendly, like "The Wikipedia Main Page"? Thanks. UrbanVersis32KB ⚡ (talk | contribs)02:45, 23 June 2022 (UTC)
Lag in edit window with Microsoft Edge Toucan browser extension
I am not sure if this is not obsolete, but it is still used by some comic infoboxes. I just noticed that Czech and Hungarian are not supported. Can they be added (needed for Funky Koval, FYI)? And ideally, this needs to be automatic, without people like me (and you) wasting our time on adding some languages... Piotr Konieczny aka Prokonsul Piotrus| reply here15:32, 19 June 2022 (UTC)
Commons being too wide since yesterdayMeta being too wide since yesterdayWikidata being too wideEnglish Wikipedia being fine as of yet
Hi all, as many of you know I almost always use mobile device in desktop mode. Suddenly, since yesterday the global projects (commons, meta, wikidata) became way too much wide, and thus, the text became too small. This issue has never happened in the past. Meanwhile, the Wikipedias (en, bn, pi, etc.) are fine as of yet. Screenshots of main pages of commons, meta and enWikipedia are attached. Thanks! —CX Zoom[he/him](let's talk • {C•X})07:34, 23 June 2022 (UTC)
This is a WP:ITSTHURSDAY thing, and will probably hit us too. The Devs™ have decided that people can't read long lines of text and they have seen fit to artificially constrain the width for us. Now, speaking personally, I bought a wide monitor because I want long lines of text - those big white stripes up the left and right are a complete waste of space, and have all the appearance of being set aside ready for the day that Wikipedia carries advertising. On those occasions that I want shorter lines, I unmaximise the window and drag the left or right window borders until I get the width that I want. --Redrose64 🌹 (talk) 07:51, 23 June 2022 (UTC)
I wish they'd listen to that advice. I do see advantages to short lines. By not making the browser window full screen, I can make the lines any (shorter) length I like, and recycle the freed space for another window containing a source document, another wiki page or my shopping list. A compromise might be to make the text a main element (it's still the most important item, isn't it?) and the right column an inner element within it which can be suppressed with JavaScript or ad blockers, but I don't see that happening. Certes (talk) 08:18, 23 June 2022 (UTC)
> The Devs™ have decided that people can't read long lines of text and they have seen fit to artificially constrain the width for us. "
The train was stopped last night, which in plain English means that this change (as-is) will never come to the English wikipedia. It was solely stopped because of the issue in this task (source: phab:T308070). An less aggressive change along the same lines will still be done, however. phab:T306910 caused this change.--Snævar (talk) 11:52, 23 June 2022 (UTC)
This was fixed in the last hour. You may need to purge your page. Please let me know if you are still experiencing this issue ASAP. Jon (WMF) (talk) 14:46, 23 June 2022 (UTC)
@Locke Cole yes, it requires the (reupload-shared) permission - that only admins have. This is normally a bad idea and should really only be used for some special cases where we have extremely high-used images locally. — xaosfluxTalk15:09, 21 June 2022 (UTC)
It seems odd to allow a file, unused anywhere I might add, on Commons to hold a name on this project for a file that is clearly fair-use and will likely never qualify to be uploaded to Commons. Regardless, the upload wizard text should be changed to reflect that it can't actually be overridden unless the uploader is an administrator. The current text simply leaves the impression that you have to confirm your action, not that it can't be done. —Locke Cole • t • c15:42, 21 June 2022 (UTC)
I'm still trying to suss out why a local project is being forced to abide by the name of a repository? I get having a warning to discourage people from overwriting files there, but there's no reason to restrict this for users who are experienced and show no history of vandalism. I seem to recall overwriting files stored on Commons locally in the past, so clearly this was a change that was made, I'm just curious about the reasoning, the policy/guideline, and when/where the discussion was. —Locke Cole • t • c16:57, 21 June 2022 (UTC)
Name shadowing is the worst, period, and multiple people here have had to deal with the mess it leaves in the general case. As Ahecht has said, the name on Commons sucks, so the image there should be moved. Izno (talk) 17:46, 21 June 2022 (UTC)
It's been at least 10 years - I'm not feeling like an archeologist right now to go through config files - but it would be out there in the codebase somewhere; "reupload-shared" seems to have been created in 2005 - not sure when enforcement started here. — xaosfluxTalk17:56, 21 June 2022 (UTC)
@Ahecht: Sadly, no. The page could not be moved, for the following reason: The filename chosen is already in use on a shared repository. Please choose another name. I suspect because while the file was moved, a redirect exists. —Locke Cole • t • c15:12, 22 June 2022 (UTC)
Hi all, is there a way to disable the big pop-up that appears on the right when you preview an edit with a new link to to a dab page? I already have "Display links to disambiguation pages in orange" in Preferences/Gadgets selected, which is quite sufficient for me. Cheers, MinorProphet (talk) 21:42, 23 June 2022 (UTC)
@Xaosflux: Great, that seems to work. Thanks very much. On the same subject, is it possible to disable the pop-up when you hover over an orange link? (Except for your name, of course) Cheers, MinorProphet (talk) 00:23, 24 June 2022 (UTC)
@MinorProphet I'm not sure which one that is? Are you getting the notification area popup there too, or is this the normal popup you get for all links (should appears near where you mouse is). — xaosfluxTalk09:18, 24 June 2022 (UTC)
@Xaosflux: Sorry, I've realised it's the usual popup, saying "This title relates to more than one page", with a blue link to "View similar pages", eg Buc and Buc, Yvelines. I've changed my browser preferences to "Always open links in a new tab", which I think solves the problem for me. Thanks again for your kind assistance. MinorProphet (talk) 09:56, 24 June 2022 (UTC)
Some of these taxobox support templates have been little changed for a decade. There is also some overlap between several templates. I don't remember noticing {{Long fossil range}}. I'll try and find some time to have a look and see if we can automate or add a maintance category. — Jts1882 | talk17:13, 24 June 2022 (UTC)
Proposal to change citation templates which hurt articles' Google ranking
Many archived links used in citation templates have a dead link which says "the original". This hurts ranking in Google, because while the reader can pass over it, the crawler picks it up and Google's algorithm processes it as if the article is full of dead links. This can make articles appear lower on Google search results than they would otherwise. How could this be fixed?
Which of these ideas do you like best, or do you have other ideas?
dead links could have a non-hyperlinked url text following "the original"
or "the original" could have a tooltip text with the url instead of a hyperlink
the link status parameter could be given three possibilities: "dead", "permanent dead", and "live", with "permanent dead" showing a non-hyperlinked url instead of a dead link to "the original".
a "permanent dead" link parameter could show a non-hyperlinked url as a tooltip text instead of a dead link to "the original".
I believe the first option (non-linked URL) is best. Tooltips raise accessibility and compatibility concerns. Adding a "more dead than dead" parameter may be convoluted, and overly complex for developers, users and readers. 50.75.226.250 (talk) 15:24, 21 June 2022 (UTC)
Assuming for the sake of the argument that dead links do hurt PageRank (I have no idea if that is the case), I would prefer that we optimize for page layout rather than for search engine ranking. That being said, if there is a way to tell a "permadead" link from a "tempdead" one, why not unlink the former. TigraanClick here for my talk page ("private" contact)15:29, 21 June 2022 (UTC)
External links also have nofollow. I imagine that damages their page rank significantly more than the text in the link itself. Izno (talk) 16:02, 21 June 2022 (UTC)
Either way, I see no reason to help Google here. Dead links in "the original" are either temporarily or permanently dead, so their accessibility by any old user is also probably worse for their page rank than anything we can do to the text of the link. There is no reason to help Google find them in any sense. Izno (talk) 16:03, 21 June 2022 (UTC)
Tigraan, yes, unlinking would be fine with me, that advice was also given on stackoverflow. There is uncertain information online about how dead links affect search rankings. It is possible that some (low-quality) links don't really matter one way or the other whether they are alive or dead. If you are wondering about this, you can search for advice like, "do dead links hurt SEO". Generally it is recommended to fix them for one reason or another. I experimented with one article that was sagging in search rankings by removing the dead link url and putting the archival url into the url parameter and adding (Archived...) just outside of the citation template. It lasted awhile before it got changed back, but while it lasted it seemed to help. Search engines don't just give out their "secret sauce" algorithm; people have to guess it and test for it. More testing could be done about this on wikipedia too.
50.75..., yes, some dead links are anyone's guess about living again. Yet there are some categories of dead links which are unlikely to ever live again, such as links to long-defunct websites and multiple links to government sites or pdfs which all follow the exact same url format, and which die all at once.
Izno, Google does not strictly obey nofollow code on links anymore. This official policy change from March 2020 had previously been suspected by some to be their practice. I wonder if the nofollow code was supposed to protect Wikipedia from the bad SEO effects of dead outbound links. We don't need to avoid the dead links for Google's benefit. Yet some articles about popular topics may not be on the first page of search results. They get buried instead; I am concerned about that. I agree that there is some benefit to any old user in not hyperlinking dead links known and expected to stay dead.
Jonesey95, thank you for the talk page. Especially with having it show non-hyperlinked text instead of a tooltip, the coding changes at hand deal with only the template. I will link to this discussion there, and it might end up getting continued there too. Changing the template will also affect some bots.--Epiphyllumlover (talk) 17:46, 21 June 2022 (UTC)
Re "I wonder if the nofollow code was supposed to protect Wikipedia from the bad SEO effects of dead outbound links" - nofollow was turned on for external links in mainspace in Jan 2007 explicitly to deter spam by removing the SEO incentive (announcement). I don't think the effects on our own search engine rankings were something anyone considered at that point. Andrew Gray (talk) 06:32, 22 June 2022 (UTC)
Unlinking is a bad idea. It makes things much harder to verify (e.g. is the link truly dead or not?). What Google does is irrelevant. Headbomb {t · c · p · b}12:20, 22 June 2022 (UTC)
What Headbomb says has traditionally been the accepted rationale. In light of this new information, it carries much more weight to remove the hyperlinks. It's a tradeoff. The benefits of increased page views can't be overstated. -- GreenC19:35, 22 June 2022 (UTC)
@Epiphyllumlover, Also, verifying if the dead links is an editor task, not a reader task. It doesn't seem to0 much to ask of the small subset of editors that actually verify these links to copy & paste a URl (though maybe it's harder than one would think). ― Qwerfjkltalk13:09, 23 June 2022 (UTC)
Maybe a bot could be written to verify that urls labeled as permanent dead links are still dead. One problem would be permanent dead links which were recreated by a domain parker or content farm; I have seen that happen. So maybe humans would be required to do the work.--Epiphyllumlover (talk) 16:11, 23 June 2022 (UTC)
This problem is probably infrequent; a workable solution would be a bot which verifies the links are still dead, and makes a list for human editors of links which aren't. Animal lover|666|08:20, 24 June 2022 (UTC)
User:InternetArchiveBot had a bad spate of thinking websites were down when they weren't, so it's probably more common than you would expect.
It might also be a good idea to restrict IABot to adding archive links within ref tags. WP:ELDEAD only rarely wants archive links in the ==External links== section. WhatamIdoing (talk) 17:39, 24 June 2022 (UTC)
Do nothing Since when does Wikipedia care about an article subject's SEO? This is an issue between the subject and Google, it's not our problem. We're an encyclopedia, not a marketing vehicle. Roger (Dodger67) (talk) 12:31, 22 June 2022 (UTC)
There are two graphs, (recent and older) at stats.wikimedia.org. The graphs show rather consistent growth prior to October 2013, a period of flat viewership, and then there is a peak in April 2020. Since then it has overall gone down. April 2020 is just after the March 2020 announcement by Google that their policy changed. It is suggestive that the policy change hurt viewership. If your "Do nothing" option is what happens, how much lower would the views need to go for you to reconsider? (10% loss? 20%? etc.)--Epiphyllumlover (talk) 18:12, 22 June 2022 (UTC)
I read that page view graph as "flat since 2016, with a COVID spike in 2020". I see zero evidence for your suggestion. —Kusma (talk) 18:24, 22 June 2022 (UTC)
Yes, there is a Covid spike, and also seasonal factors. But look at the graphs again, trying to account for that. Google Knowledge graphs corresponds with the major hit on the older graph, although it didn't set in all at once. There is also a smaller hit to viewership on the recent graph. Changing it to monthly can help with comparing specific months with the same months in prior years.--Epiphyllumlover (talk) 18:29, 22 June 2022 (UTC)
We should pay zero attention to the Google rankings of our articles, and we should not try to optimise them every time Google changes their algorithm. Improve the article quality, not search rankings. —Kusma (talk) 18:29, 22 June 2022 (UTC)
If readership drops by a quarter, I will start to be interested why. Do they not click through to our page because Google's snippet is good enough? Do they read our articles at Wikiwand instead because our layout sucks? The only reason to prefer to have readers here instead of at a mirror is the small chance of converting them into editors. Where do you think the potential readers we might be losing because of the dead link issue are going? —Kusma (talk) 18:53, 22 June 2022 (UTC)
Active readers is the pool active editors derive from. Reduce one the other will go down. Search engines are the main driver of active readers. -- GreenC19:25, 22 June 2022 (UTC)
Still, if we can choose between expanding effort on improving the conversion rate reader->editor by 10% or increase total readership by 5% I would choose the conversion rate. —Kusma (talk) 20:38, 22 June 2022 (UTC)
IMO anecdotal opinion a prime reason Wikipedia is so successful is Google. Without that, readership would probably 50% or less, the site would have less social impact, less importance, fewer people who cared, less money etc... Good or bad, how it is. Wikipedia is not unique in that position. Big Tech sucks but also real. -- GreenC19:10, 22 June 2022 (UTC)
When we had fewer readers, we didn't have the terrible amount of responsibility to get it right that we have now. —Kusma (talk) 19:16, 22 June 2022 (UTC)
The SEO thing is a death by a thousand cuts. There are other things on Wikipedia which could be changed to improve SEO, and might have a larger impact, but this seems to be one of the easiest possible changes.
The Knowledge Graph is part of the issue, both the snippet and the sample questions and answers. The questions and answers may appear above the article. And the attractiveness of the sample questions and answers is compounded by having the wikipedia article's placement being lower down, maybe under even irrelevant placements. The usual winners are businesses and organizations who practice SEO. For example, if I search for "Door Peninsula" in Google, the top result is the "Door Peninsula Winery". Number two is the wikipedia article. So the Door Peninsula Winery is the winner in that case. But so is Google, because if they can keep viewers from seeing the wikipedia article so easily, they'll spend more time on Google. And the more time they spend on Google, the more click-throughs Google gets on their ads. In the early days of Knowledge Graph it seemed like they were more likely to be derived from Wikipedia, but now most of them aren't. The content is aggregated from all over.
I wonder what the viewership is on wikiwand, I couldn't find it. The figures, if they are high enough, have the possibility of changing my perspective.
A redeeming factor to this is that none of the larger wiki encyclopedia forks are exploiting this. If one of them decided to get serious about SEO, that could change, but I don't see that happening. Recently I saw one of the smaller wiki content sources (a niche topic site rather than a clone site) beat Wikipedia for awhile on a specific search, but now Wikipedia is ahead again.--Epiphyllumlover (talk) 19:36, 22 June 2022 (UTC)
For me (in the UK), the Wikipedia article is the top result, followed by Britannica and Expedia. —Kusma (talk) 20:36, 22 June 2022 (UTC)
When I search for "door peninsula" on Google, I get the WP article, FWIW. Wait a minute, Google displays ads? I haven't seen an ad on Google for 15+ years. – Jonesey95 (talk) 20:57, 22 June 2022 (UTC)
I get the Wikipedia article at the top in both searches, but I can replicate the behaviour you see by using a VPN that geolocates to the United States. —Kusma (talk) 08:44, 23 June 2022 (UTC)
Some time back I submitted a complaint to Google about listing the winery first. That might be why one search is different than the other.--Epiphyllumlover (talk) 16:06, 23 June 2022 (UTC)
Is there a way to make it harder to post death threats and other threats of harm on Wikipedia?
Looking in my inbox, I see that I have e-mailed User:Emergency twice this month: Once on Friday, and once on the 14th. On the 14th, a death threat was made in an edit summary to kill the subject of an article. This Friday, there was a threat on an admin's user subpage of killing babies (I'm not going to explain this more). Two threats this close by seems a bit dangerous. Maybe we should make an edit filter that warns users when adding something like /kill\s(him|her|the|other nouns here)/img? Maybe that'll prevent some people from adding it. weeklyd3 (message me | my contributions) 05:52, 26 June 2022 (UTC)
Yeah, insource:"kills him" in the search bar returns 5,256 results in article space. This is discounting the pronoun variants or the variants of the word kill. Discussions about such content may very well exist in any form of talk pages as well. So there probably would be huge false positives, if edit filter is enacted. —CX Zoom[he/him](let's talk • {C•X})07:40, 26 June 2022 (UTC)
The NewPP limit report (view HTML source to see) shows the post‐expand include size being exceeded. Replacing {{sock list}} by something simpler like {{Bulleted list}} to show the four big lists as plain text would reduce that size by 85% and fix the problems. Certes (talk) 18:52, 25 June 2022 (UTC)
I don't know if the following workaround would exactly reproduce what {{sock list}} does. Working that out would require examining the parameters in the wikitext and deciding whether they work when used as follows. At any rate, the page renders without error after replacing each
@RoySmith:{{socklist}} would have probably not been that expensive had {{checkuser}} and {{checkip}} been Lua-ified, which I just did. A quick test showed that using a module helps reducing post‐expand include size by 1,619,512 (current revision – call Module:Sock list directly) - 773,702 (replace frame:expandTemplate with its Lua equivalent) ~ 850 thousand bytes. NguoiDungKhongDinhDanh03:51, 26 June 2022 (UTC)
Watchlisting a category does give you a list over new entries in it, the entries are valid for 30 days or until your watchlist fills up, depending on which happens first.--Snævar (talk) 16:10, 26 June 2022 (UTC)
Hi, this good faith edit appears to have broken the reflist for the page. But, for the life of me, I can't figure out how. I've tried tweaking parts of the references and even deleting the original citation and re-inserting it, but no dice. I am, however, getting the following warning:
Script warning: One or more {{cite web}} templates have maintenance messages; messages may be hidden (help).
Script warning: One or more {{cite book}} templates have maintenance messages; messages may be hidden (help).
Script warning: One or more {{cite news}} templates have maintenance messages; messages may be hidden (help).
I have no idea what this means. Any of you more technical folks know what's going wrong here? — Czello17:25, 23 June 2022 (UTC)
Ok, so step 1 worked. Anyway, to further fix the uses of templates that come after the reference list you will need to reduce the uses of {{music}}. Whether that's some sort of page split or simply raw removal is up to you. Izno (talk) 17:43, 23 June 2022 (UTC)
{Time signature} is already using Module:Su via Template:Su; are you suggesting removing that one layer will be a substatial improvement? Wouldn't replacing {Music} calls with (Time Signature} calls be better? (I don't know if template size is an issue or just number of templates.) - R. S. Shaw (talk) 00:34, 24 June 2022 (UTC)
Size presents the primary issue, but that is compounded by the number of templates called by a template of use because those are counted twice.
To answer the question directly, yes, removing layers tends to induce a substantial improvement.
Regardless of further template adjustments, I have removed a number of signatures on that page from section 12 to section 59+ such that all templates on the page now render fully. Further removal of uses of the template will be a stopgap if the issue occurs again. Bypassing the use of {{music}} with {{time signature}} would be a second step to help, as would bypassing the use template:Su. Izno (talk) 16:37, 26 June 2022 (UTC)
Before applying any fixes to any templates or modules, perhaps the correct thing to do for this article is to reduce the unnecessary redundancy. For example, §1 8 has these (the first three; other entries similar; references removed):
The Four Quarters by Thomas Adès has a measure of 1 8 just before rehearsal mark 3 in the first movement.
Mädchentotenlieder, by Bo Nilsson. Bar 11 is in 1 16 time, bar 53 is in 1 8 time.
Since that section is the 1 8 section, it seems pointless to me to repeat the 1 8 time signature in every entry of the section. Additionally, since this is §1 8 and Mädchentotenlieder is also mentioned in §1 16, there is no need to mention the 1 16 time signature in Mädchentotenlieder's entry in §1 8. Removing the redundant {{music}} templates would, it seems to me, go a long way towards fixing that list article.
Display items from Wikidata under the title of a page
I am trying to find in Preferences the option to show "Display items from Wikidata under the title of a page" or something like this. Where is it? Philocypros (talk) 15:43, 26 June 2022 (UTC)
I was recently trying to contact an editor who had engaged in some poor behavior. Upon clicking on the user name in their contribution history, I was taken to the Main Page (with no "(Redirected from User:Xxx)" notice to get me back to the offending user page). The same thing also happened with the offending users talk page. This was because they had created redirects to Main Page from bothpages prior to the behavior. I've requested a Bot to monitor for future attempts to create these types of tricky redirects, but was wondering if there was any blacklist or feature in MediaWiki that could be used to prohibit creating a redirect from a root User/User talk page to Main Page? Or if not, if this is something that might be worth seeing if it can be addressed in MediaWiki? From that Bot request discussion, there is a query that will at least make finding such pages easier (and I already nuked one other page that had a similar trick-redirect). From a policy/guideline standpoint WP:UP already disallows such redirects as a guideline (the subection at WP:SMI would also apply). —Locke Cole • t • c18:02, 26 June 2022 (UTC)
@Locke Cole:[I] was wondering if there was any blacklist or feature in MediaWiki that could be used to prohibit creating a redirect from a root User/User talk page to Main Page? Yes, there is; abuse filters can do so quite easily – there's no need for a bot. Try requesting one at WP:EF/REQ. NguoiDungKhongDinhDanh18:16, 26 June 2022 (UTC)
Since yesterday (25th of June, 2022), if I open a link in Wikipedia from my mobile phone, I get to the desktop version instead of the mobile version (https://en.m.wikipedia.org). Am I the only one experiencing this? is it long term change or just a bug? I really prefer the user experience using the mobile Web interface Minerva. PAC2 (talk) 07:06, 26 June 2022 (UTC)
I'm in the process of building a tutorial for the Template:Adjacent stations system. For creating the module, I tried the following code:
<inputbox>
type=create
buttonlabel=Submit
editintro=
preload=Module:Adjacent stations/blank
prefix=Module:Adjacent stations/
placeholder=Name of system
summary=Create module for transportation system
break=no
minor=no
nosummary=no
</inputbox>
Unfortunately, I'm not getting the preload. I did try to replace the edit page to the Template: namespace, and there the preload worked. Is it possible to do this in the Module: namespace? Animal lover|666|20:41, 26 June 2022 (UTC)
The preload feature has been disabled for non-wikitext talk pages since it posed a potential security issue in some cases (see T297725). There is a proposal with patches waiting for review that would re-enable it for modules, since it is not a security issue there: T300644. Matma Rextalk20:51, 26 June 2022 (UTC)
So. This Template:Editor of the Week topicon that I created... On my user page - User:Shearonink - it now appears first, before all the other topicons there (FA, FL, GA, Protection, and, lastly, the 10 Year Society top icons). I would like for the EotW top icon to appear last, or at least next to the Ten Year Society laurel leaves, after any and all article or list top icons (like GA/FA/FL). I think the EotW is appearing first because, yeah, "Editor of the week" comes first alphabetically. To make it appear last or next to the Ten Year Society the solution would seem to be to move it to a new title, perhaps re-titling it as "Wikipedia:Editor of the Week"? I want to make sure (before possibly changing it) that such a title change/page move is WP-correct. Thanks, Shearonink (talk) 21:25, 26 June 2022 (UTC)
Xaosflux I thought about adding a sortkey but now that there's a sortkey parameter...when I put a number in to the Template itself, won't my sortkey number override other people's possible sortkeys? Couldn't I just change the title of the Template? Shearonink (talk) 21:58, 26 June 2022 (UTC)
@Shearonink no no, you leave that blank, anyone that wants to use this sorted, feeds the sortkey to it themselves so it sorts where they want it to sort - you shouldn't try to force sorting on to others. See example of calling these with sortkeys here on my topicons: User:Xaosflux/Topicons. — xaosfluxTalk22:16, 26 June 2022 (UTC)
Ok, I think I understand it. Sort of lol... Last question though...If I changed the Template name to "Wikipedia:Editor of the Week" would it then be automatically sorted in alphabetical order and automatically appear behind the GA/FA/FL top icons? Thanks for all your help, appreciate it so much - Shearonink (talk) 00:57, 27 June 2022 (UTC)
@Shearonink in looking at Template:Top icon it appears they sort "alphabetically" by sortkey parameter, then by "name" (as a parameter), then by id. Look at Template:GA user topicon for an example, it uses a "name" parameter - so that would sort in a group apart from your new template that only has an "id" parameter; sorting of these has historically been a big mess on non-articles - you could try to add a "name" parameter and make it start with a W and see though. — xaosfluxTalk01:34, 27 June 2022 (UTC)
Thanks Anomie! Can you (or anyone else) please raise an issue in the task? In flags, even the slightest detail matters, in this case the majority of the stars have vanished. This needs to be fixed soon. Thanks! —CX Zoom[he/him](let's talk • {C•X})12:39, 26 June 2022 (UTC)
Your best bet in getting it fixed any time soon would seem to be editing the SVG to avoid the bug. It looks like the problem comes down to when a <use xlink:href="..." /> refers to another <use />. I did the necessary edits for this image and uploaded a copy at testwiki:File:Flag of Arkansas fixed.svg to see if it worked (looks to me like it did); now someone would have to get Commons admins to upload it there as the image there is fully-protected. Anomie⚔15:37, 26 June 2022 (UTC)
I found a couple of weird ways that editing on mobile seems to be activating a crippled version of the reply tool, where some of the most problematic things happening are being forced to publish a comment without an edit summary, and being forced to publish a comment only at the bottom of the page and nowhere else. I have documented the results of my testing, as well as instructions on how to replicate the issue over on the talk page of my sandbox. Thanks. `Huggums537 (talk) 09:04, 26 June 2022 (UTC)
@Huggums537 (I'm one of the developers) I wanted to clarify that the current mobile talk page experience is unrelated to the reply tool, and also that we're planning to replace it with the reply tool (slightly adapted to the mobile version) in the near future. I think we should have done it a while ago and I honestly don't understand why it's taking us so long to make this decision, but I hope we'll do it soon. Matma Rextalk11:42, 26 June 2022 (UTC)
Ok. Thanks for the reply. The current mobile "talk page experience" is very similar to the reply tool in the places where I was experiencing these issues, so that's why I thought it was related. I don't know why nobody else has reported on this yet. Not being able to leave an edit summary or comment anywhere else but the bottom of the page seems like a pretty big deal. Huggums537 (talk) 22:13, 26 June 2022 (UTC)
The actual [reply] tool and its matching ==New Topic== tool will be made available to everyone at jawiki, arwiki, frwiki, kowiki, viwiki, hewiki, bnwiki, and zhwiki (i.e., not most places, and not here) later this week. If it turns out (as I suspect will be the case) to be better than the existing system, then you could request that it be turned on at enwiki as well. Whatamidoing (WMF) (talk) 21:53, 27 June 2022 (UTC)
Tech News: 2022-26
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
The new version of MediaWiki will be on test wikis and MediaWiki.org from 28 June. It will be on non-Wikipedia wikis and some Wikipedias from 29 June. It will be on all wikis from 30 June (calendar).
Some wikis will be in read-only for a few minutes because of a switch of their main database. It will be performed on 28 June at 06:00 UTC (targeted wikis). [20]
Some global and cross-wiki services will be in read-only for a few minutes because of a switch of their main database. It will be performed on 30 June at 06:00 UTC. This will impact ContentTranslation, Echo, StructuredDiscussions, Growth experiments and a few more services. [21]
Users will be able to sort columns within sortable tables in the mobile skin. [22]
Future meetings
The next open meeting with the Web team about Vector (2022) will take place tomorrow (28 June). The following meetings will take place on 12 July and 26 July.
Some wikis will be in read-only for a few minutes because of a switch – Remember when a m:server switch was a big deal, and could easily stop editing for half an hour, even if everything went well? Whatamidoing (WMF) (talk) 22:00, 27 June 2022 (UTC)
Wikilinks from italic titles
When editing the article Baltic Sea cruiseferries, I have had for numerous times to type [[MS So-and-so|MS ''So-and-so'']]. Would it be possible for the wiki software to simply ignore the italic signs and interpret [[MS ''So-and-so'']] as a direct wikilink to [[MS So-and-so]]? JIP | Talk01:56, 22 June 2022 (UTC)
{{Ship}} (and {{Chem2}}) do the opposite, they require the editor to specify html-parts by /documented prescriptions:
{{ship|HMCS|Kootenay|H75}} → HMCS Kootenay (H75) (works for topic-experienced users I guess)
OP question is like: {{html'ed title in - labeled wl out|HMCS7Kootenay H75}} → [[HMCS7 Kootenay H75|HMCS<sup>7</sup> ''Kootenay'' H75]]. Requires knowledge of required formatting of the editor all right ({{DISPLAYTITLE}}/topical like ship/chemical naming area). I'd suggest a dedicated module (-option), strictly defined, for this; having to find out wikitext-editing modules for a {{NewShips}} template is not easy enough is my experience. -DePiep (talk) 16:59, 26 June 2022 (UTC)
We have the same situation in chemical names, where there are all sorts of random italic syllables and letters scattered throughout, and superscripts/subscripts as well. Do we have a {{strip-all-that-html-shit|(−)-''trans''-Δ<sup>9</sup>-tetrahydrocannabinol}} to give (−)-trans-Δ9-tetrahydrocannabinol (link to "(−)-trans-Δ9-tetrahydrocannabinol")? DMacks (talk) 15:30, 26 June 2022 (UTC)
First attempt is at {{wl}}, which is based on {{plain text}}. It works for italics (original use-case that starting this discussion). But that reveals {{tl:module:plain text}} removes <sup>...</sup> and <sub>...</sub> chunks altogether rather than simply un-tagging:
Topical formatting (like {{Ship}} see above, {{Chem2}}), could ose or even require a switch? In that case, round trip formatting can be tested. -DePiep (talk) 06:56, 28 June 2022 (UTC)
If a template does not have [[..#Template data]], start it in the /doc and the template will have a TPU after beginning of next month (TPU is updated per 1st of each month). -DePiep (talk) 09:02, 28 June 2022 (UTC)
Unless things have changed, there’s a 5,000 limit. What happens when you hit it? Very active editors may get there this year. It’s hard to remember to go back and unsubscribe. Doug Wellertalk18:33, 26 June 2022 (UTC)
@Xaosflux Thanks. I think ideally they’d be deleted after a certain time. I can’t think of many if any good reasons to want to keep year old ones, probably not six months. Doug Wellertalk18:53, 26 June 2022 (UTC)
It would be technically possible to make them expire and get deleted automatically, we have a task for it here: T278190. We also have one about making it easier to clear out your subscriptions manually: T292035. I don't know if we'll get around to working on these, though. Matma Rextalk20:47, 26 June 2022 (UTC)
And the answer to what would have happened, if they hadn't turned it off, is: You wouldn't be able to subscribe to any new threads at that wiki until you had manually removed some.
oA==, how do I find out what it is? It looks like a space, but evidently isn't a regular space as atob('oA==').match(/ /) doesn't match it. MediaWiki appears to insert this thing when a section title has an exclamation mark in it. — Alexis Jazz (talk or ping me) 14:53, 28 June 2022 (UTC)
Xaosflux, I figured it out: atob('oA==').charCodeAt(0) returns 160, making this a no-break space and it only exists in the HTML, not in wikitext. On [24] you'll get 160 when you run $('H2')[4].innerText.slice(3,4).charCodeAt(0). — Alexis Jazz (talk or ping me) 15:08, 28 June 2022 (UTC)
When prompted to enter a CAPTCHA on page creation, the original section heading (subject) is mangled to be an edit summary with asterisk comments and "new section" text
This isn't the right place because this is not at all specific to enwiki (should be reported to MediaWiki Phabricator directly), but in due time I'll report this here (doubtful I'll report it myself to Phabricator, without an account). When I created a new page (Special:Diff/1095451549) and got prompted for a CAPTCHA, the CAPTCHA prompt changed the subject line from Original subject to /* Original subject */ new section. This would be expected if this was the edit summary, but in this case it mangled the original section heading (h2) to something else than the intended input. 84.250.14.116 (talk) 13:02, 28 June 2022 (UTC)
When you created that page which editor were you using? (e.g. wikitext, visualeditor, discussion tools, mobile, etc). To open a phab we need step by step directions to reproduce the problem, including environment information. — xaosfluxTalk13:52, 28 June 2022 (UTC)
New section (top right corner, whatever the default Vector skin is)
Enter an external URL such as https://example.com/ as body text, "Test" as subject
Preview, everything should still be fine
Press "Publish page", the CAPTCHA prompt is shown and the subject line gets mangled. The changes will not be published.
I tested with both Firefox 101.0.1 and Firefox ESR 91.10.0esr with same results, doubtful it's my environment. 84.250.14.116 (talk) 15:41, 28 June 2022 (UTC); edited 16:01, 28 June 2022 (UTC)
It also gets worse everytime I enter the wrong CAPTCHA. After three CAPTCHA prompts, the subject line looks like /* /* /* Test */ new section */ new section */ new section. 84.250.14.116 (talk) 15:47, 28 June 2022 (UTC)
This does not occur on pages which already exist (i.e. my talk page which exists, using steps 2-6). Step 1 is not optional (the page must not exist), in order to reproduce this. 84.250.14.116 (talk) 15:59, 28 June 2022 (UTC)
I was able to reproduce this on test2wiki, I suspect most use cases of this will go away once dtenable is the default for logged out users, but yes seems like a WP:BUG and phab is the right place. "When CAPTCHA is triggered while creating a new page with the new section parameter, the new section title is unexpectedly changed" or the like. — xaosfluxTalk16:01, 28 June 2022 (UTC)
Hey, if you don't watch the miscellaneous section you may ignore a message you perhaps would prefer not to ignore, so this is like a redirect within the VP to increase the visibility.
In a nutshell:
We would love to see Vector 2022 become the default for readers and editors across all wikis. In the coming weeks, we will begin conversations on English Wikipedia.
It will always be possible to revert to the previous version on an individual basis. Monobook or Timeless users will not notice any changes.
Join an online meeting with our team. It will take place on 28 June 2022 at 12:00 UTC and 19:00 UTC on Zoom. Click here to join. Meeting ID: 5304280674. Dial by your location. The following events will take place on 12 July and 26 July.
I acquired an iPad Pro a few weeks ago, which I've been using to read & edit Wikipedia at home -- amongst other things. Today when I opened a new page on Wikipedia, I was surprised to find the font about 50% larger than it had been last night. According to the documentation (which Apple likes to hide because "if you need to read the manual the user interface has failed"), I can adjust the font size by pressing the Cmd & + (or the cmd & -) keys at the same time. This didn't work. Nor did going to another Wikimedia project: the font was 50% larger there with no way to reduce its size. (I might have eye problems, but being forced to use a larger font doesn't make any difference.) I know this is Wikimedia-specific because none of the fonts on other websites I visited this evening have suddenly grown in size.
Rather than rant about the habits of Wikimedia developers & their disinterest in what editors & users want -- or bore everyone by recalling how the web browser was originally intended to let end users control how they wanted content to look (which is one of the things about Wiki Wiki software I like) -- I'd like to know if there is are parameters in CSS I can apply to fix this.
Or even better, is there a resource page somewhere that documents the parameters of CSS as implemented by Wikimedia? That way I'm not uselessly repeating Frequently Asked Questions. -- llywrch (talk) 03:08, 24 June 2022 (UTC)
This seems like a (at least to me personally) previously unknown behavior of Safari/iPad wrt to viewport management. This is definetly unintentional and reproducible. I've left a note on the task that dealt with the other viewport issue earlier this week. I suspect it'll be fixed by monday. —TheDJ (talk • contribs) 06:21, 24 June 2022 (UTC)
Vector legacy (2010). I just updated the iPad OS to the latest release (15.5.1, IIRC) in an unsuccessful attempt to get Safari to handle captchas in a usable manner.FWIW, I had a look at the latest version of Vector, & I'm not happy with replacing the options of "Talk", "Sandbox", "Preferences", etc., with icons. I just want a simple interface similar to what existed way back when without all of these "user-friendly" enhancements. (To offer my unsolicited opinion.) -- llywrch (talk) 18:10, 24 June 2022 (UTC)
@Whatamidoing (WMF):, I'm at work at the moment, but I remember testing that skin the other night & seeing icons when I did a Preview; I assume what I see there is what I'd see once I make the change. Once I'm at home I can check it again on my iPad Pro.BTW, I'd still like to be directed to any resource page about CSS client settings for Wikimedia. -- llywrch (talk) 22:02, 27 June 2022 (UTC)
I just tried that link on my work computer (which I don't manage) & I see what I consider are icons in the top right corner: glyphs or symbols for "Alerts", "Your notices", "The list of pages you are monitoring for changes" (can't we just label this "List of watched apges"?), & the hamburger symbol/icon which opens a pull-down menu with more symbols/icons in it. Maybe my UX terminology is different from yours? -- llywrch (talk) 22:14, 27 June 2022 (UTC)
Oh, not at all. I was thinking of the icons that the mobile site uses for the Edit button. Yes, there are a variety of icons at the very top of the screen.
I'm sure there are people who prefer icons. I happen not to, & unless I can change that I won't be using Vector 2022. -- llywrch (talk) 03:55, 28 June 2022 (UTC)
I am thinking, why cannot we save edits using AWB like we do in Huggle. Let me be clear. In huggle, we perform certain operations one by one, and all of them are left in a queue, and huggle performs them one by one. We don't have to wait for Huggle to perform the edits. But in AWB we have to wait until AWB saves the edit and then a new page appears. What are the problems that can happen if AWB was configured like Huggle and it also had a edit queue? ItcouldbepossibleTalk04:31, 19 June 2022 (UTC)
@Itcouldbepossible, changing the behaviour may mean the loss of the 'last accessed item' window which can be useful. Multiple sessions can be used and if 'in the zone' on a list then one session can work from the list top and the other the bottom. Swapping between sessions, sure it's extra work but not onerous. Neils51 (talk) 03:35, 22 June 2022 (UTC)
@Neils51 Can you please elaborate, like how can we run two sessions? Then we will have to click save on the sessions, then again, and again like this simultaneously. Won't that be troublesome? ItcouldbepossibleTalk08:06, 24 June 2022 (UTC)
Multiple sessions is a case of multiple startups/logins in the usual fashion on your chosen device. Works fine on Windows though I don't know with Linux, YMMV. I rotate via selecting the taskbar thumbnails and I think the most sessions I have rotated through is five. By the time you get back to the first it has definitely finished its writeback (it may be possible to rotate sessions via script?). The time taken to read and process a new item will be dependent on a number of factors, including size of pages, additional regex in your normal/advanced settings, function option selections, etc. I suppose it depends on the way you work and what you are endeavoring to achieve. For instance some editors will look at a list of 1500 items and want to process them in one hit whereas I may decide to take two weeks. As Kusma has mentioned an AWB editor must also be checking all edits and that can take time. A responsible AWB editor will sometimes come across rule exception conditions and have to decide as whether it's a one-off or perhaps there may need to be exception modifications made to a rule, or at least make a post to advise others and get their input. In an ideal world AWB would have an option for sync or async writeback operation. However, I'm reminded of a previous life when a piece of software was issued that met the original design requirements and then the change requests start coming in and a colleague comments that "apparently we need to rewrite the operating system". :-) It's possible that someone will take this on as a project one day, creating a separate thread, however today there are ways of being quite efficient with AWB and there are bots for the big lists! Neils51 (talk) 10:29, 24 June 2022 (UTC)
You can also run AWB in pre-parse mode, which will then only present you with articles that AWB could edit according to the rules you've setup. Headbomb {t · c · p · b}12:22, 22 June 2022 (UTC)
@Headbomb You have possibly misunderstood me. I am talking about saving edits one after the other without actually waiting for AWB to save it. There should be a save queue. We can perform whatever actions we like, like saving or skipping. AWB will save them one by one like as in a queue. For example, we have processed what we will do in 50 pages, and a queue has been formed, and AWB has saved, say, 42 pages. Then it will likewise save the rest of the pages, which the user has already mentioned. If you use Huggle, you will understand better about the queuing. Won't this be more effective? ItcouldbepossibleTalk08:09, 24 June 2022 (UTC)
I think your suggested feature is too likely to be abused by people who won't review all of their edits. (Anyone who uses AWB without reviewing all of their edits should get a bot account or have their access removed). —Kusma (talk) 08:16, 24 June 2022 (UTC)
@Kusma There are many pros and cons of a given feature. For example, the mass rollback script, it can be dangerous and useful too. So like this everything has its good and bad side. And moreover I don't think untrusted users would have their AWB access request accepted. Users with good track of editing and those who are reliable are given AWB access. Still if someone misuses something, well, then admins can use this feature upon them. But it shouldn't stop any development from taking place. ItcouldbepossibleTalk08:33, 24 June 2022 (UTC)
How will your proposed change make sure you can only save edits after you have manually reviewed them? I have seen too many "trusted users" who believe that AWB can fix typos and then introduce one typo that changes the meaning of things (the typo fix "new york"->"New York" can produce things like "New York Mall" when something is about the "new York Mall", a new mall in York, making it sound like it is on a different continent) per a couple dozen small fixes that don't affect meaning (say, changing it´s to it's). People are not careful enough with AWB as is, and too often not willing to take responsibility for their edits, blaming them on the tool. —Kusma (talk) 08:48, 24 June 2022 (UTC)
@Kusma Actually it is not that. There are types of edits which always doesn't need to be reviewed. For example a regex that removes a parameter from a given infobox and only from that infobox and from nowhere else. Those type of edits doesn't need to be always reviewed since Regex wouldn't possibly remove something wrong. And people don't close there eyes and hit save, if in a hurry they make a wrong edit, they can always revert it. It has happened many times with me. I revert if I made a wrong edit. Typo fixing and fixing errors in new pages is what needs strict review. Things like I ones I mentioned needs only speed to save time and clear maintenance categories. You can also think of AWB bots in a way. They are running day and night. Who is reviewing them. But at the end of the day we see that it has done a lot of job and there are zero mistakes because of a fixed regex. ItcouldbepossibleTalk08:17, 25 June 2022 (UTC)
For what it's worth, as a botop this would be useful. Currently I'm checking around 200,000 pages and editing around half. At AWB's slow rate I expect this to take me around a month, or a week using multiple AWB sessions, even when running AWB for around 5 hours a day. Checking the edits obviously doesn't apply here, so this feature could easily be restricted to bots. ― Qwerfjkltalk13:14, 24 June 2022 (UTC)
@IAmChaos, that's part of one of Awesome Aasim's scripts, and probably doesn't apply to a discussion about AWB. My personal opinion on this script is that it's probably fine to have no delay, and if the API calls are asynchronous then it could be done pretty quickly (i.e. around 90 pages in under 10 seconds). However, page creation can't be reverted, and there's no editing requirement for that script, so it's probably worth erring on the side of safety. ― Qwerfjkltalk20:12, 28 June 2022 (UTC)
I was aware it is a script, (although had forgotten the creator). I was wondering if that was what was meant, to queue up edits and then let them work themselves through eventually. Happy Editing--IAmChaos21:22, 28 June 2022 (UTC)
Question - is Template / Columns-list not working?
Today, at article Federally Administered Tribal Areas, section for "Towns and villages", I added many more articles & the Columns-list template. It's not displaying into columns. Is that because I'm on the new skin Vector (2022)? or some other issue? Before writing here, I checked several other articles & they have the same issue. JoeNMLC (talk) 13:59, 29 June 2022 (UTC)
Works for me. You had |colwidth=35em and the content pane of Vector2022 is hardcoded to have a maximum width regardless of window-size that does not accommodate the 70+em needed for two columns. I set |colwidth=20em as a test and it works fine for me. Feel free to set whatever width you think is best and pile on wherever you can find a bug-report about this new max-width "feature". DMacks (talk) 14:19, 29 June 2022 (UTC)
@DMacks: - Thanks. Mostly I use "cmn" shortcut instead of full template name. Changed to cmn & that works also. Will add this "trick" to my wikicode "copy-and-paste" file. Should this change be noted in the Template documentation regarding Vector2022? JoeNMLC (talk) 16:00, 29 June 2022 (UTC)
I would assume...feel free to do it. No idea how "final" this design is, but as always things can be changed whenever necessary. DMacks (talk) 16:05, 29 June 2022 (UTC)
Because it is a non-standard property. Regardless, you should not use column-count except in specific circumstances; always prefer column-width or set both parts of the columns property. Izno (talk) 20:31, 29 June 2022 (UTC)
Is there a bot that removes maintenance tags that do not apply?
@Weeklyd3 That seems like a bad idea that would run afoul of WP:CONTEXTBOT. You can't just say "the article contains something between <ref> tags, the clean-up tag can be deleted". How will you tell the difference between <ref>'s used for sources and those used for other purposes, like notes? How will the bot know that the added source is appropriate - if a reference is added that is blatantly unreliable, doesn't support the content, is spam etc it would be better to delete the ref than the tag. How will the bot decide whether it's appropriate to delete the tag or replace it with something like {{more citations needed}} - this is a decision that needs to be made based on whether the added sources support the content in the article, not how many there are. Addition of these kinds of tags is forbidden by bots because they are an editorial decision, I can't see that we'd treat deleting them any differently. 192.76.8.85 (talk) 22:02, 29 June 2022 (UTC)
I wrote a bot to add{{Unreferenced}} see User:GreenC_bot/Job_11. It worked fine received zero complaints. Added about 10,000 increasing Category:Articles_lacking_sources from about 180k to 190k at the time - hardly much at all. The trick is par down everything to which you can be 100% confident about, and skip everything else. You can get a lot this way, but not all, not even most; and it's a big headache of edge cases and testing. In the end barely worth it functionally, and not worth dealing with nay sayers. -- GreenC00:08, 30 June 2022 (UTC)
Hear ye, hear ye – Enabling the New Topic Tool by default
If you don't want the New Topic tool, then go to Special:Preferences#mw-prefsection-editing-discussion and turn it off. (There's also a built-in button for switching to the old editing window, but I assume that anyone who reaches this page will have already dismissed that.)
I want to add: This tool is a bit better for less experienced editors, and it really does cut down on certain problems, like unsigned comments. The Teahouse has been using it for a couple of months (in the link under their big blue button). Lots of editors like it. But: the average editor isn't every editor. As with any significant change, it takes time to figure out a new tool, which you may or may not feel is worth your energy right now. Also, it is, by design, a simplified tool, which isn't a good match for everyone's editing. If you don't want it, then please just turn it off. The goal here is to get a match for your editing, and you are the best person to decide that for yourself. Whatamidoing (WMF) (talk) 22:54, 27 June 2022 (UTC)
Discussion tools enabled on legacy Vector Is this tool supposed to work with legacy Vector? I looked at the info page and the discussion page and did not see any mention of it one way or another. The "subscribe" link does not align horizontally with the "edit" link, and I still see "New section" instead of "New Topic" or "Add topic" (the documentation says "Add topic", but the post above says "New Topic"). Screen shot at the right. It is quite possible that one or more CSS hacks I have put in place to fix various display problems are conflicting with the tool. Others using legacy Vector may end up with similar problems. – Jonesey95 (talk) 00:50, 28 June 2022 (UTC)
@Jonesey95 Looks like an incompatibility with the gadget "Move section [edit] links to the right side of the screen". By default the [edit] link appears immediately after the heading text. the wub"?!"09:45, 28 June 2022 (UTC)
I use it in Vector 2010 all the time, and it works. I don't know what it would take to get the gadget working, but as the Editing team is planning some visual changes to talk pages (e.g., adding a light-gray note about how many people/comments are in each thread), it might be worth waiting a month or so, in the hope that the fix will remain fixed. Whatamidoing (WMF) (talk) 15:42, 28 June 2022 (UTC)
@Jonesey95 The tab reads "Add topic" on almost all English-language projects, but English Wikipedia overrides it to "New section" for reasons beyond my comprehension (MediaWiki:Vector-action-addsection). The documentation on mediawiki.org is intended to apply to all wikis, so it's using the default wording. Matma Rextalk18:10, 28 June 2022 (UTC)
Will the new tool cut down on meaningless headers to new threads? Or not mentioning what a thread is about until several lines in? DuncanHill (talk) 15:24, 28 June 2022 (UTC)
DuncanHill, Bawl warns about new sections with a title that already exists on the current page. So at least if there's already a section titled "Question" the next person who tries to enter that gets warned. With some invisible sections at the top such titles always produce a warning. But that's not DT. As for not mentioning what a thread is about until several lines in, that would require an AI that probably doesn't belong in either DT or Bawl. — Alexis Jazz (talk or ping me) 19:36, 28 June 2022 (UTC)
@Alexis Jazz: Meaningful headers and making it clear what a thread is about early on are certainly beyond some human intelligences. A cynic might assume someone wanted to "announce" something without anyone noticing. DuncanHill (talk) 19:46, 28 June 2022 (UTC)
Are these new "reply" links supposed to be the same size as the "edit" links at the top of each section? For me, the edit links are the usual 13px (93% of normal), the subscribe links are also 13px, and the reply links are 14px (100% of normal). It seems odd that two UI elements that are trying to look the same are not quite the same. At least for me, with my skin and preferences. – Jonesey95 (talk) 16:53, 29 June 2022 (UTC)
They are supposed to be similar, but you're right that they're not exactly the same. The reply links have the same font size as the surrounding text, which depends on the skin – e.g. it's 14px on Vector, 12.7px on MonoBook, and 16px on Minerva (mobile). If I were king they'd be the same as the section edit links, but making changes here would certainly annoy some people and would have little benefit. Matma Rextalk17:21, 29 June 2022 (UTC)
(Also they're not really new, at least not in this deployment – the reply links have been around for a while, today I only enabled the new topic tool and did not change anything about the reply tool.) Matma Rextalk17:22, 29 June 2022 (UTC)
OK, someone clue me in...
I'm sorry but I don't like the new/improved Editing/"new section" thingy that apparently just came onboard. I personally find it hard to follow and it also doesn't seem to have a Preview? If it does have "Preview" I couldn't find it. Anyway, I need to know how I can permanently switch back to the so-called "legacy experience". Lol couldn't find that either. Thanks in advance, Shearonink (talk) 18:35, 29 June 2022 (UTC)
@Shearonink, if you're in the wikitext source mode (which you probably were), there is an automatic live preview underneath the editing window. Above the editing window, there should be a big gray bar with a link to switch back to the normal editing window, and when you get there, it has a link for making that change permanent. You can also reach the settings directly by going to Special:Preferences#mw-prefsection-editing-discussion and turning off the item about "quick topic adding". Whatamidoing (WMF) (talk) 18:44, 29 June 2022 (UTC)
(edit conflict)Xaosflux - Oh? Well, I didn't see Preview...I guess it's there somewhere but I didn't see it. Thanks as always for your help Xaosflux & thanks for chiming in Whatamidoing (WMF). I saw the link to switch back to the legacy editing, & did so, but there was a button or linkage in that window to make the change to legacy permanent? Missed that too,
I tried this New section thingy before when it was in Beta and it just drove me nuts even though everyone else just seems to LOVE it. Ummm...the Powers That Be...they're not going to ever get rid of the legacy editing/optioning-out ARE THEY?!?Shearonink (talk) 19:01, 29 June 2022 (UTC)
There are several use cases where the "new section" tool isn't the best option - especially related to creating a brand new page - so this is an important one to be able to opt-out of (thus it is). For most novice users/unregistered users - this should be a big benefit though. — xaosfluxTalk19:08, 29 June 2022 (UTC)
Thank you for that...I'm an editor and writer of words who has taught myself to edit on WP through lots and LOTs of trial and error...every little time something gets jiggered around here it can send me for a loop. Shearonink (talk) 00:32, 30 June 2022 (UTC)
Above the editing window, there should be a big gray bar with a link to switch back to the normal editing window – I can't find this. Is there a screenshot of it somewhere? Maybe I'm just missing something obvious... DanCherek (talk) 19:13, 29 June 2022 (UTC)
If you've already used it (e.g., months ago), then it may not show for you. If today's the first time that you remember seeing/testing this, can you please tell me which browser and skin you're using? Whatamidoing (WMF) (talk) 19:30, 29 June 2022 (UTC)
I've seen it in use at the Teahouse but don't think I did any other previous testing. I'm using legacy Vector in Google Chrome. I guess my other question is whether the switch-back banner is meant to be temporary while people are getting used to the new tool. I was hoping for a long-term way to easily switch between the two. Thanks, DanCherek (talk) 19:38, 29 June 2022 (UTC)
Possible bug (clicking New section and then legacy mode from diff view)
I was just looking at the latest diffs of this page via my watchlist, then clicked the "New section" link at top of the page, which brought up the new interface, then clicked "switch back to the legacy experience". That brought up the following:
Section editing not supported
Section editing is not supported in this page or is disabled for this view.
That's pretty weird, thanks for the bug report. This is related to the fact that you can't use section editing while editing an old version of the page, and adding a new section after following a diff is… kind of similar. We'll fix it. Matma Rextalk19:40, 29 June 2022 (UTC)
@Simeon: Are you saying it appeared while it was a draft? That shouldn't happen. I guess you saw the search result after it became an article and it's the article you see. Otherwise the page heading would say "Draft:Kendra Dacher - Wikipedia". The noindexing of the draft namespace does not happen in our robots.txt at https://en.wikipedia.org/robots.txt so search engines are allowed to view pages there. Drafts are noindexed by placing noindex on the draft page itself. But redirects don't have noindex if the target allows indexing so Google can index https://en.wikipedia.org/wiki/Draft:Kendra_Dacher after it was redirected. MediaWiki redirects are not real URL redirects where the browser is forwarded to another URL. Instead the content of the redirect target is shown on the redirecting URL, and MediaWiki uses JavaScript to rewrite the URL in the address bar without actually going to the target URL. This can be seen by disabling JavaScript in the browser (real redirects work without JavaScript). So search engines like Google can index the article at the draft URL if they ignore the fake MediaWiki redirect. I think it's correct behaviour for them to ignore it when their indexed content wasn't actually retrieved from the rewritten URL. It's MediaWiki which plays tricks, not the search engine. PrimeHunter (talk) 13:57, 30 June 2022 (UTC)
Thanks for this detailed explanation of what's happening! I think I indeed searched after moving it to be an article. When clicking on the link in the search results, I did visit Draft:Kendra Dacher first and then got redirected to the article. But, I mistakenly thought that this meant it got indexed as draft and the redirect now took place because the namespace within Wikipedia changed. I think the behaviour is then indeed correct because it should be ok to index Draft:Kendra_Dacher when it's a redirect and not a draft article. I also learned something about how MediaWiki works so thanks for that :) Simeon (talk) 14:41, 30 June 2022 (UTC)
Citation format
I am sort of new here, recently I'm translating article Oboe Concerto (Strauss) to Chinese Wikipedia. I see there is lots of citations not using template "cite web" or "cite book" but just contain wikitext, which is kind of hard for me to transfer, but is that a typical way to cite them? QiuLiming1 (talk) 05:18, 30 June 2022 (UTC)
Just using <ref> whatever you know about the reference </ref> is all it takes to include a reference. All those templates are just a fancy ways to format the middle part, they certainly can be useful - but are not required by most projects. — xaosfluxTalk14:51, 30 June 2022 (UTC)
Show hidden categories globally
Hi all, on Special:Preferences#mw-prefsection-rendering we get the option "Show hidden categories" but on the global prefs equivalent of it, we don't have this option. But I know 5 languages and sometimes edit there, further sometimes I do a little global work too. I would like to set "Show hidden categories" because that'd be very helpful. Is there a way to do that? Thanks! —CX Zoom[he/him](let's talk • {C•X})11:14, 30 June 2022 (UTC)
Oh no! I'm sorry, it was there but it somehow missed it. Thanks for pushing me to look out there once again, so I found it. —CX Zoom[he/him](let's talk • {C•X})12:18, 30 June 2022 (UTC)
Hi, in the new vector 2022 and for articles like Tehran, Tokyo, and other cities, and on top right part of article, the texts "Read Edit source View history TW" and coordinates data like "Coordinates: 35°41′21″N 51°23′20″E" are overlapping on each other. Please make them unoverlapping. Thanks, Hooman Mallahzadeh (talk) 06:48, 1 July 2022 (UTC)
Anyone has any idea why was the header (.mw-body-header) and article toolbar (.vector-article-toolbar) were swap around recently? Previously the article toolbar was displayed before the header but now it is header first then article toolbar. —Paper9oll(🔔 • 📝)01:58, 1 July 2022 (UTC)
That's a weird change in my opinion with the rationale they gave in the tracked task ... maybe I gotten use to the old layout which has been the same placement for ages. —Paper9oll(🔔 • 📝)05:45, 1 July 2022 (UTC)
It's an odd change, and it'll take some getting used to, but I think I like it. For some reason, it looks better to me. I'm reminded of the advice I got from the developer of a forum I used to frequent ten years ago: "give it a week". Take a week to get used to it. If you still don't like it, then it's time to look into what specifically you don't like and what can be improved. --rchard2scout (talk) 08:14, 1 July 2022 (UTC)
defining favorite website in Wikipedia:Book sources
On Wikipedia, ISBN links point to Wikipedia:Book_sources, where users can select various libraries and bookstores where the book in question may be found. Back in 2005, Lunchboxhero shared a script which allowed users to be taken directly to a predefined library or bookstore upon following an ISBN link, obviating the need to select it manually from the "Book sources" page every time. Unfortunately, the script does not appear to work anymore. Looking at the associated talk page, it seems like this is because it is using deprecated parameters. As I'm not a programmer, I don't know how to fix the script, but perhaps others are willing to take a look? Pablo (talk) 11:35, 1 July 2022 (UTC)
@Pablo Those issues were both resolved last year, and the script works fine for me. It looks like you have both Lunchboxhero's script and independent code both trying to do the re-write in your monobook.js, so they might be interfering with each other (and while Lunchboxhero's script will work with 13-digit ISBNs or ISBNs with dashes, the code you have at the top of your monobook.js won't). Also, because you put this in your monobook.js and not your common.js, it will only work when you're using the monobook theme. --Ahecht (TALK PAGE) 14:22, 1 July 2022 (UTC)
Recovering deleted redirects
Hey, tech folks,
I've run into this question before, both on my talk page and at WP:REFUND, and since I have a little extra time this morning, I thought I'd bring it to WP:VPT. Is there any way to locate redirects that once existed but were deleted when a page has been deleted? I know that if you know the exact name and namespace of a page, you can go to Special:Undelete and search for it there. But it won't tell you what pages redirected to that page.
Some admins, like myself, use Twinkle to do all of our page deletions and Twinkle will delete redirects (but not redirect talk pages) when you delete a page so if you know what administrator deleted Article X, you can search for Article X on their Deletion log and see what redirects were deleted at the same time. Alternatively, broken redirects can appear at User:AnomieBOT III/Broken redirects, so if you know the rough period of time that Article X was deleted, you can go see if any broken redirects appear on this list at that time. But both of these methods can take a lot of time and don't cover broken redirects that are tagged for deletion by editors patrolling and tagging pages for speedy deletion.
This situation comes up more than you would expect and here's one scenario. Suppose Vandal A moves Article X to Article Wacky Page title. Typically, when move vandalism is addressed, the page mover doesn't leave a redirect behind. But Wikipedia's bots are very efficient and as soon as Article X was moved to Article Wacky Page title, all of Article X's redirects were changed to point to Article Wacky Page title. Without a redirect left when Article X gets returned to the proper page, all of those adjusted redirects are now broken and, if not spotted and fixed, they can be deleted either by a bot or admin as broken redirects. Or, and I just saw this happen, all of the articles created by Editor Q were deleted because they were suspected of being a sockpuppet. But, it turns out they weren't a sockpuppet! So, the articles were restored but what about any redirects that were deleted at the same time? If Editor Q created them, they can be found in their Deleted Contributions but not those that might have been created by other editors. In the event of these situations happening, it would be nice if there was a feature that captured what redirects to Article X existed before they were deleted as broken or as redirects to a deleted page.
I don't think these "snapshots" exist on Wikipedia but I just thought I'd check in in case there is some resource that exists that I don't know about. Thanks! LizRead!Talk!18:36, 1 July 2022 (UTC)
In case it wasn't clear, when you repair a page move that was vandalism, always leave a redirect behind or check What Links Here to see if there are redirects that were altered and change them back to point to the correct target! The redirect from Article Wacky Page title can always be deleted once the redirects have been changed back. My PSA for the day. LizRead!Talk!18:42, 1 July 2022 (UTC)
Botched edit-conflict resolution lost page history
When I began copyediting 2020–2021 Slovenian protests, the infobox indicated that the protests continued into 2022. I moved the page to 2020–2022 Slovenian protests before stupidly continuing to edit the lead, edit-conflicting with myself. I resolved the edit conflict as best I could (should've hit "Cancel", mistake #2), but somehow lost the previous page history. When I didn't see any sourced info that the protests extended into 2022, I moved it back to the original title; the previous history, however, is still gone. My attempted resolution of the edit conflict may have been a de facto cut-and-paste move, for which I apologize; I can't find the previous page history to fix it. Any help appreciated. All the best, Miniapolis01:45, 2 July 2022 (UTC)
Yes, but I don't know how it happened or how to get it back. I just undid a double redirect I fixed after the page move. Miniapolis02:11, 2 July 2022 (UTC)
What happened was that you moved the page, then overwrote the redirect left behind from that move, then moved it again and deleted the initial article with all the history to make way for the first move. I've listed this at Wikipedia:Requests for history merge since that's probably easier than trying to explain the complicated process involved in fixing it. * Pppery *it has begun...02:17, 2 July 2022 (UTC)
Thanks for fixing this, Izno; the full history is back. I clicked on too many radio buttons in the the paragraph-based edit conflict interface, which (I think) caused the problem. All the best, Miniapolis13:27, 2 July 2022 (UTC)
@Qwerfjkl Talk pages were enabled for logged out mobile web users in October 2021, Phab:T293946. I think logged in mobile web users have always had access to talk pages. I'm not sure how the apps handle talk pages. 192.76.8.85 (talk) 15:48, 2 July 2022 (UTC)
Change password option on special:preferences did not work
I searched for my special:preferences to change my password, and I clicked the change password option. Then I changed my password. But it didn't work. Why? After that, I have only changed my password by logging out and clicking "Forgot password" since I enabled my email address on Wikipedia, so I successfully changed my password. My question is, why does the "change password" option in my preferences not work? —Princess Faye(my talk)15:18, 2 July 2022 (UTC)
@Xaosflux: I used mobile web view, I never used the app. I will tell you what I have experienced. So I went to my preferences and clicked the "Change password" option. It redirects to another page, of course, but before you change your password, you need to type in your current password, so I did that. So after that, it redirects to another page again. Of course, I now need to type in my new password. And last, I clicked Change Credentials. So to be sure, I logged out of Wikipedia and tried logging in again with my new password, but it didn't work. So to log in, I typed my former password and that worked. —Princess Faye(my talk)15:40, 2 July 2022 (UTC)
@Princess Faye OK, so you eventually got to the "Change Credentials" page and tried to change your password, entered it and confirmed it. After you hit save did it give you an error or just go back to settings? Those steps worked when I just tried them. When you say you were on mobile web, was your URI "en.m.wikipedia", or were you just using a browser on a mobile device? Did you click on the "desktop" view either at the bottom of the screen, or in your browser settings? Were you on a "big" mobile screen, like a tablet, or a small one like a phone? — xaosfluxTalk16:49, 2 July 2022 (UTC)
@Xaosflux: Yes, your summarization is right. I'm using "en.m.wikipedia" using Google Chrome. I did not click "desktop view" nor the desktop view settings on my browser. I'm using a big-screen mobile device. Well, this might just be an isolated technical error that happened on my device, but I still successfully changed my password using the "Forgot password" technique, and not in the "Change password" option in my preference. I hope this is okay. —Princess Faye(my talk)17:07, 2 July 2022 (UTC)
Oh it's certainly fine - just trying to find out if something is broken for everyone so we can get it fixed. — xaosfluxTalk17:53, 2 July 2022 (UTC)
Re-ordering a large table, possibly off-wiki
One-party state has a huge table which was created in Dec 2019 when 4 separate tables were combined. It makes sense to have them in one table, but the previous listings weren't merged, just added end-to-end, so it is now in 4 A-Z sequences. I thought I had a technique to sort the table by copying it out into csv and then into Excel, sorting, and replacing, but that technique lost all the links, flag images, and refs.
Obviously the table can be sorted in real time when looking at it, but it would be better if the code of the table, and thus the default display, were in a logical order.
So: Can someone please either:
Point me in the direction of simple tools to sort a large table, or
Do some clever editing on this table yourself, to sort it A-Z by country.
I've raised it on the article talk page, where I mentioned another couple of possible sort orders, but as the original tables were mostly sorted A-Z by country I don't think this is controversial. The article had a sudden peak in pageviews last week, as it was the target of "Redactle" puzzle.
I attempted this but it's not easy to automate. Presumably it should be sorted by country, so "Tinoquista Costa Rica" appears under C, but there's no easily extracted clue that in this case the second word determines the sort order. Certes (talk) 10:51, 3 July 2022 (UTC)
@Certes: If you could do the heavy lifting of sorting them by whatever is there, I could go through and upgrade the minority of countries which need a {{sort}} added to fix their alphabetisation. Ah, thinking about it, a better plan: if I go through and add those {{sort}}s anyway - they'd always be needed for re-sorting back to A-Z after sorting on any of the other fields - would you then be able to re-order the code? PamD16:10, 3 July 2022 (UTC)
The more I look, the more puzzling it gets: I can't see what it's sorting on at the moment. Possibly the name of the "flagicon"? Puzzling.
If you sort the first column,
{{flagicon|Romania}}{{flagicon image|Flag of the Legionary Movement.png}} [[National Legionary State]] seems to sort as "Romania" (so presumably being picked up from the "flagicon" field, as the only mention)
{{flagicon image|Flag of Cameroon (1961-1975).svg}}{{flagicon image|Flag of Cameroon.svg}} [[Cameroon|Republic of Cameroon]] sorts as "Republic of Cameroon"
{{flagicon|Upper Volta}} [[Republic of Upper Volta|Upper Volta]] sorts as Republic of Upper Volta" (so UV and Cameroon differ: one filed by displayed title, and one by target article title)
[[Communist-controlled China (1927–1949)|Yan'an]] sorts as "Yan'an" despite not having a flagicon..
Can anyone explain how this sorting is working? I wonder whether it uses the country with which the flagicon is associated, and the displayed text if there is no flagicon? PamD16:33, 3 July 2022 (UTC)
I tried sorting the table using some find-and-replace magic, but the columns do not have consistent sort keys. If someone adds consistently formatted |data-sort-value= entries to the cells in the first column of the table, I should be able to sort it manually. – Jonesey95 (talk) 16:38, 3 July 2022 (UTC)
I've had a go at adding data-sort-value using Wikidata and some manual fiddling. Anomalies like how to sort Benin/Dahomey and various Chinas remain. Certes (talk) 21:23, 3 July 2022 (UTC)
Stray Brackett
I just created an article titled Battle of Punished Woman's Fork. (The article is still a stub, but I'm expanding it.) At the end of the summary paragraph is a stray bracket. In trying to edit the article, I can't find the bracket to delete it. Can somebody delete it? Thanks. Sorry to be so incompetent. Smallchief (talk) 11:06, 4 July 2022 (UTC)
Ignore the request. The stray bracket seems to have disappeared when I added more text. Another unexplained riddle of technology? Smallchief (talk) 11:13, 4 July 2022 (UTC)
It didn't disappear but remained at the end of the article which only consisted of the summary paragraph when you posted. Your added text just meant that the article ended later. I have removed the bracket.[25]PrimeHunter (talk) 11:46, 4 July 2022 (UTC)
True. EasyTimeline will be removed eventually. Use either Extension:Graph or HTML based graphs. The largest and most feature rich templates are Module:Graph for Extension:Graph and Module:Chart for HTML. There are also smaller templates. It depends on your usecase which one is more suitable. Snævar (talk) 12:03, 4 July 2022 (UTC)
Tech News: 2022-27
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 5 July. It will be on non-Wikipedia wikis and some Wikipedias from 6 July. It will be on all wikis from 7 July (calendar).
Some wikis will be in read-only for a few minutes because of a switch of their main database. It will be performed on 5 July at 07:00 UTC (targeted wikis) and on 7 July at 7:00 UTC (targeted wikis).
This change only affects pages in the main namespace in Wikisource. The Javascript config variable proofreadpage_source_href will be removed from mw.config and be replaced with the variable prpSourceIndexPage. [26]
Could somebody who has the ability to work with past dumps run a scan and provide me with the contents of Category:Austro-Hungarian films as of the last available dump date before June 27, 2022? It's been erroneously emptied out, so I need to reconstruct it. The output can be saved to my sandbox at User:Bearcat/Random.
(I tried processing a dump myself this afternoon, and failed hard.)
@Bearcat, I'm planning on mass reverting any edits by the bot that removed more than one category, and handling them separately, probably with a pywikibot script. It would be easier for me to handle these along with all the other pages, so could you hold off on reconstructing the category (unless it's too late)? ― Qwerfjkltalk20:01, 4 July 2022 (UTC)
How do I disable the new drop down box for Add Topic
This has to be one of the worst ideas ever. It keeps adding "no wiki" in the middle of sentences, and makes it a lot harder to just add a simple topic, because you have to go in and delete what it added that you didn't want. It's horrible. How do I disable it completely? — Maile (talk) 12:40, 4 July 2022 (UTC)
Is there a way to switch default to "Source editor" in the Add topic/reply system? The "visual editor" is non-intuitive for anyone who understands the markup. Source editing also shows a live preview below, so new editors won't find it hard to use. And it's better for them to learn by doing too, as many templates all over Wikipedia still do not support visual editing. —CX Zoom[he/him](let's talk • {C•X})13:46, 4 July 2022 (UTC)
@CX Zoom "default" - as in for all non-logged in users as well? I think you are overestimating the subclass of "new editors" that are brand-new editors. — xaosfluxTalk21:30, 4 July 2022 (UTC)
@Maile66, it looks like you made your first-ever edit in the Reply tool here at VPT, and were put into the wikitext source mode by default. Your next edit was also here, and it looks like you switched to the visual mode (maybe trying out all the buttons?). It looks like you never switched back. The Reply and New Topic tools use the same settings, so if you switch to the visual mode for Reply, you'll be in the visual mode for New Topics.
Fortunately, if you don't like it (or if you want to add templates to your comments), the switch is easy and sticky: Just click the "Source" tab, and it will automatically remember your choice until you change it again. Whatamidoing (WMF) (talk) 01:39, 5 July 2022 (UTC)
Is it clear why this edit on a pending changes protected article was automatically accepted? The user is not autoconfirmed and has no flags. Thanks. Ymblanter (talk) 20:19, 4 July 2022 (UTC)
MySun is autoconfirmed, the edit prior to theirs was accepted - so yes, this appears to have worked properly this time. @Ymblanter: why do you think they are not AC? — xaosfluxTalk21:51, 4 July 2022 (UTC)
Note, that article is also SPP, so everyone editing it will be autoconfirmed, making the PC1 mostly moot. — xaosfluxTalk21:53, 4 July 2022 (UTC)
They have only made 14 edits in total, I thought they need more (30 or 50?) for an autoconfirmed. Thanks anyway, it is clear that PC1 is not really appropriate for this article, but this is not a technical question.--Ymblanter (talk) 05:28, 5 July 2022 (UTC)
Could the big blue button at {{Submit an edit request}}, which is used in (I think) MediaWiki:Protectedpagetext, acquire a little extra text? I'd like it to say "Submit an edit request for changes to {{FULLPAGENAME}}". My hope is that this would discourage people from submitting requests to the wrong page. Or even something like "Submit an edit request (not for changes to articles)" for requests outside the mainspace. WhatamIdoing (talk) 00:23, 4 July 2022 (UTC)
There are very few misplaced edit requests, and generally the people making requests aren't reading any of the instructions anyway. That said, it doesn't hurt to add even more text. ScottishFinnishRadish (talk) 00:50, 4 July 2022 (UTC)
Yeah, 8 out tens of thousands, and they all look to have been been appropriately handled. I don't think adding even more text will help the thousands of blank requests that are created despite the instruction text in the edit window and the pre edit request explanatory page. ScottishFinnishRadish (talk) 08:41, 4 July 2022 (UTC)
I don't think it will add anything since the page title will be shown at the top of the page, and it states "This page is..." quite clearly, implying that the edit request that would be submitted is specific to the open page. Longer text only adds more text for users not to read and could end up with a multi-line button on width limited smaller screens. Clear and concise is normally the best approach. There is no "Read {$pagename} article | Edit the {$pagename} article | View the history of the page {$pagename}" as the three main buttons at the top of a page for a reason. Terasail[✉️]00:58, 4 July 2022 (UTC)
If someone does incorrectly place an edit request, a responder only has to notify the user of the error, either with {{EP|mis}} or a personalised reply. I am not sure that the problem is common enough to require this change where a responder can just guide the user to the correct place. Terasail[✉️]01:05, 4 July 2022 (UTC)
My understanding was that only empty edit requests or clear policy violations were blanked rather than simply misplaced requests. Terasail[✉️]13:03, 4 July 2022 (UTC)
I'm sure that's the usual practice for articles, but consider, e.g., the first one I linked above, which was reverted three minutes after it was posted. This might not be the best practice, but it does not seem to be unusual, either. WhatamIdoing (talk) 18:31, 4 July 2022 (UTC)
Wikipedia:Autobiography does not say that you should revert a newbie who is trying (however misguidedly) to write an autobiography. I'd rather that this newbie, and all the others as well, had known up front that clicking this particular blue button would do them no good at all. WhatamIdoing (talk) 22:59, 4 July 2022 (UTC)
I know its annoying to deal with others' mistakes, but this expectation that we often put out there that people should not make mistakes (and that adding text will help them not to make mistakes) seems unachievable to me. Mistakes are how people learn. We need to give people the opportunity to learn, or they won't even be here at all. This is after all Wikipedia, not the paper version of Encyclopedia Brittanica. —TheDJ (talk • contribs) 11:02, 4 July 2022 (UTC)
As a general rule, I'd say that substantially reducing text would help. Most people probably look right past the oblong gray blur of text there, but if they read anything, they read the text on the button itself.
Another good approach would be offering options. Perhaps we need one button that says "Request a change to the {{FULBASEPAGENAME}} page" plus another button that says "Ask for help with another page or something else" (maybe pointing to the Teahouse). If you only give people one button, they'll use it for everything. WhatamIdoing (talk) 18:38, 4 July 2022 (UTC)
This is a good point. We also see that with uploading. Ppl primarily care about uploading and not about licenses etc. So they will choose whatever allows them to achieve their goal. —TheDJ (talk • contribs) 08:14, 5 July 2022 (UTC)
WP:V brought me here, but the problem is more diffuse than that. Also, Template:Editnotices/Page/Wikipedia:Verifiability already exists. WP:Nobody reads the directions, so it probably doesn't do much good. It particularly won't do much good when the one bit that the person sees says "Submit an edit request", with no indication that this means "an edit request for this specific page, which is probably not the one that you want, and not for a generic queue that will ultimately get routed to the right place". WhatamIdoing (talk) 15:55, 5 July 2022 (UTC)
Just got the message "Error contacting the Parsoid/RESTBase server (HTTP 504)" when using reply
This also includes the visual editor on desktop for me. When I tries to save the page, it takes forever to save and would timeout, saying "Error contacting the Parsoid/RESTBase server (HTTP 504)". It doesn't work even when I purged the cache (both "?action=purge" and browser), log out and back. CactiStaccingCrane (talk) 12:29, 3 July 2022 (UTC)
Possibility to condense notifications of mass deletion discussions?
I was recently given no less than thirty notifications of deletion discussions at once, all related to a mass deletion discussion (see Wikipedia:Templates for discussion/Log/2022 July 2#Template:VR-K color) by User:Gonnym (ping intentionally left out as Gonnym is not really at fault here). This cluttered up my talk page and caused me to accidentally overlook a discussion I had not answered yet.
Might it somehow be possible that when such a mass deletion discussion is started, each editor involved in the discussion is only given one notice, possibly containing links to each of the pages (in this case, templates) nominated for discussion? JIP | Talk00:19, 3 July 2022 (UTC)
There's a wider issue here: we need a tool to create a nomination for many related pages at once. Twinkle does an excellent job of creating each nomination individually but combining them is a pain, often hampered by multiple edit conflicts with early responders as they paste their response into each nomination in turn. Certes (talk) 00:36, 3 July 2022 (UTC)
I know I've personally asked for Twinkle to be updated to handle that. Doing mass template or category nominations is just a pain atm. AWB is an option, but can work but only if you know the same editor is the creator, however that requires you to manually check each page which makes AWB moot. Gonnym (talk) 08:42, 3 July 2022 (UTC)
It would be fairly easy to write a script. It would be better if it could be integrated into Twinkle or as an AWB plugin. ― Qwerfjkltalk19:30, 3 July 2022 (UTC)
The website xinhuanet.com has thousands of links that a redirecting to the home page. For example http://www.xinhuanet.com/english/2018-09/11/c_137460450.htm .. viewed in a browser it first redirects to http://www.xinhuanet.com/webSkipping.htm then to the home page. The webSkipping page contains an error message in Chinese The content you viewed has expired and been archived; thank you for your attention to Xinhuanet. This is simple enough. However if you view the page with JavaScript disabled (or with curl), the underlying content is still there - it has not been "archived". Further, looking at the JavaScript code there appears to be some kind of logic decision involved in determining when the webSkipping page is served versus the actual content, something is triggering the redirect that is yes or no. I'm not too familiar with JavaScript can anyone make out what might be happening? Maybe it works from Chinese IPs? Or the redirect expires at some point? Ultimately would like to determine if these pages should be converted to archive URLs or left alone. -- GreenC16:20, 5 July 2022 (UTC)
@GreenC: Here's the script they use: pageCore.js. According to this file, if you access a page whose URL doesn't contain content_, you will be redirected to webSkipping.htm (and then the home page). Comments are mostly encoded so I'm not sure why this was used. NguoiDungKhongDinhDanh03:05, 6 July 2022 (UTC)
@PrimeHunter thought I'd solved it, but a preview of my common.js has the same problem. Ditto vector.js It's erratic, I just tried to edit 3 articles, 2 no problem, one the same issue. Doug Wellertalk15:20, 3 July 2022 (UTC)
@PrimeHunter Switching to Vector 2022 gets rid of the problem but I hate the wasted space. This is a new problem, something's changed somewhere, and I did nothing in the past few days. Doug Wellertalk16:25, 3 July 2022 (UTC)
And I found the problem. I'd added BitDefender's anti-tracking extension to Chrome. Remove that and no problem. Maybe they updated it recently. Doug Wellertalk15:16, 3 July 2022 (UTC)
And Mono breaks at least one script, the one that gives me details about an editor at the top of their talk page. It's too useful to give up. I really prefer the legacy version of Vector, does that make me a Luddite? Doug Wellertalk16:04, 5 July 2022 (UTC)
The signature of editor Lambiam present on these 800+ pages contains #ifeq code that I think was supposed to be substed at the time of its insertion, but it apparently was not. It is IMO messy, and it is somehow generating a transclusion of the editor's User talk page on each page in which it is used. It looks like it needs careful replacement by someone with AWB and the knowledge of the right thing to replace it with. – Jonesey95 (talk) — Preceding undated comment added 23:41, 30 June 2022 (UTC)
Please do not replace the broken signature with anything fancy such as the above. Just replace it with a standard signature. "Lambian ([[User talk:Lambian|talk]])" would be fine. Johnuniq (talk) 08:07, 1 July 2022 (UTC)
Because of the substitutions, I believe that would result in only [[User talk:Lambiam|Lambiam]] being actually saved to the page, but at that point, one might as well insert that directly. Rummskartoffel19:17, 1 July 2022 (UTC)
I noticed that the titleprotected message no longer displays when creating pages that you can't create because they are protected. Was this message deprecated or what? Aasim - Herrscher of Wikis18:06, 6 July 2022 (UTC)
@Xaosflux Maybe I should have been clearer - for a user with the inability to create a protected page, such as by creating This page does not exist, there is just the top message that says only admins can create that page, but there is no bottom message giving the explanation for titleprotected. Aasim - Herrscher of Wikis18:26, 6 July 2022 (UTC)
I am wondering if this is a bug or what. I was going to suggest that "submit a page creation request" be added to the message to allow the user to create a protected edit request. However, the titleprotected message which gives instructions on what to do is not at all being shown.
What is being displayed is something similar to this:
Interface message displayed
You do not have permission to create this page, for the following reason:
Some templates and site interface pages are permanently protected due to visibility. Occasionally, articles are temporarily protected because of editing disputes. Most articles can be edited by anyone.
The reason for protection can be found in the protection log. If there are no relevant entries in the protection log, the page may have been moved after being protected.
If you have noticed an error or have a suggestion for a simple, non-controversial change, you can submit an edit request by clicking the button below and following the instructions. An administrator may then make the change on your behalf. Please check the talk page first in case the issue is already being discussed.
It looks like an attempt to create a page that's restricted to intadmins (e.g. User:192.0.2.16/common.js) is also not showing MediaWiki:Titleprotected (although the message that does show is different). Pages that are blocked by the title blacklist (e.g. Wikipedia:∀∀∀∀∀) also don't show it. So maybe MediaWiki:Titleprotected just isn't functional at all at the moment – I can't find any way to trigger it (at least not by using ?action=edit URLs). --ais523 22:14, 6 July 2022 (UTC)
@Awesome Aasim so long as any configurable message is displaying, I'm not seeing what the problem is? Is the default text for the message that displays for a specific use case inaccurate? Especially if it is in usecases that have a normal workflow. In your example page:
@Awesome Aasim I have a feeling that the messages outputs have just been updated to use new messages and some documentation is missing, but sure open a bug and someone will check in to that and verify. — xaosfluxTalk01:18, 7 July 2022 (UTC)
Looking at the source code, the message is supposed to appear on an attempt to edit a protected title, i.e. a page that appears on Special:ProtectedTitles, and not in any other context. I tried to edit one of those pages, and it didn't appear, so I'm guessing it's just broken at the moment. (Some of the relevant source code is marked as "deprecated", so maybe it doesn't work or is being moved to a new system.) --ais523 01:05, 7 July 2022 (UTC)
I have a feeling that MediaWiki:Titleprotected is the low-level error for a permission failure, but most if not all code paths that would display it are checking for the error condition and displaying a different message first. The low-level message still needs to exist in case there's still a code path somewhere that doesn't do a higher-level check first. Anomie⚔11:26, 7 July 2022 (UTC)
Hidden category "Use dmy dates" truly is hidden on Ventricle (heart)
I do a lot of date format auditing and the majority of the time, the templates that create that hidden tracking category are plainly visible in the article. One in a great while though, the source of the hidden category is a mystery. In other words, the template isn't actually in the article, but is usually transcluded from another (navigation) template. With this article though, I haven't found the {{Use dmy dates}} template on any of the templates in the article. Is the issue me? Dawnseeker200010:19, 6 July 2022 (UTC)
If "Use dmy dates" should not be used in the template namespace, then the remaining cases of where it is used without noinclude should be removed too. Searching for it shows several more.--46.182.184.86 (talk) 10:05, 7 July 2022 (UTC)
I have checked and modified all of the transclusions of this template in template space so that they are either removed or noincluded. I have also updated the documentation. There was a 2021 discussion on the template's talk page, but nothing had been done about it. – Jonesey95 (talk) 15:58, 7 July 2022 (UTC)
I finally gave up and switched to Vector 2022 (see my post above) as it seems to be becoming the standard. But I can't preview from Chrome but can from Firefox. Doug Wellertalk16:44, 7 July 2022 (UTC)
At the risk of sounding like a broken record - have you disabled your personal user scripts and any special browser extensions and retested? — xaosfluxTalk17:33, 7 July 2022 (UTC)
I probably should have done that first, I was hoping there might be an easier answer. I’m finding everything a bit difficult with this damn chemo but I’ll do that tomorrow. Scripts are easier to disable now and that helps, but I can start by just blanking the relevant pages, easy to revert if it makes no difference. Doug Wellertalk17:41, 7 July 2022 (UTC)
Sort of: it currently loads both of those until phab:T301212 happens - so right now you should not put the same scripts in to both of those at once. If you are done with classic vector and just want to use -2022, then you can remove from the old and move to the new. — xaosfluxTalk10:20, 8 July 2022 (UTC)
When I tried to subst Template:Uw-ew onto someone's talkpage, any text I put in the additional text field disappeared, and I had to add it manually afterwards Nigel Ish (talk) 18:13, 7 July 2022 (UTC)
It should work find as long as you add the template in the format {{subst:uw-ew|pagenamehere|additionaltexthere}}. @Nigel Ish Do you have an example of this failing to subst additional text? Terasail[✉️]18:21, 7 July 2022 (UTC)
It didn't here - the text in the additional text field completely disappeared - this was using the new and improved talk page editing system.Nigel Ish (talk) 18:25, 7 July 2022 (UTC)
Yes it did - there was a diff in the text - diffs are perfectly sensible things to include in things like edit warring templates, so this ideally needs to be fixed, or at least explicit warnings rolled out to all the user warning templates.Nigel Ish (talk) 07:23, 8 July 2022 (UTC)
It applies to all templates with unnamed parameters. User warnings are often given with tools like Twinkle which automatically handle it. The documentation of {{Uw-ew}} does mention it. If "explicit warning" means something more prominent then it could quickly get annoying to see on thousands of templates. Some template documentations include it in example code. {{subst:Uw-ew|Article|Additional text}} could be changed to {{subst:Uw-ew|Article|2=Additional text}} (articles rarely have equals signs so 1=Article is rarely needed). I think that would be helpful but subtle enough to not get annoying to see everywhere. PrimeHunter (talk) 14:16, 8 July 2022 (UTC)
Hello again
Hello! This is about password observation. If a user wants to log in to Wikipedia but forgets the password, then the user clicks the "forgot password" option to change the password. He gets a temporary password from Wikipedia through an email. The email states that if the user remembers the old password, he can now ignore the temporary password given to him by Wikipedia. Now that user has remembered the old password and successfully logged in to Wikipedia. What will happen next to the temporary password sent to him by Wikipedia through email? Is it still working even though you are now logged in using the old password? I'm confused. —Princess Faye(my talk)12:54, 8 July 2022 (UTC)
@Princess Faye Temporary passwords are invalidated either after 7 days or when you reset your password with it. There is no way for you to force invalidate an unused temporary password, this is requested at Phab:T65837. 163.1.15.238 (talk) 14:27, 8 July 2022 (UTC)
SPA account name with no user or talk page appears blue instead of red
Recently an SPA made its one-and-only post to a talk page under DS sanctions. When I saw their post, their name was blue. Aren't user names with no history or userspace pages in red? Still unaware this was an SPA with no other contribs, I was surprised they had no talk page, but I tried to open one to post the DS alerts. That's when the user name turned red. After I closed my browser (which is configured to dump all history and other browser stuff) and log back in, the name is blue again. Can anyone teach me what is going on? Thanks here is the thread. The SPA is the OP NewsAndEventsGuy (talk) 18:53, 8 July 2022 (UTC)
@NewsAndEventsGuy links to userpages and usertalk pages are just like any other, they are blue if the page exists and red if they do not. There is nothing special that changes if it is a user that does or does not have local contributions or subpages. That OP doesn't have a userpage or usertalk page currently, so this is normal. Depending on your browser and any scripts/gadgets/feature you have installed the links could be recolored - most browsers will also treat "visited" and "unvisited" links differently. I don't see that link as blue at all, and hovering it shows that it has the &redlink=1 argument on the link as well - so it should be red. — xaosfluxTalk20:24, 8 July 2022 (UTC)
Ever since my Firefox upgraded itself recently (I'm now on FF 102), SUL is failing. I am logged in on English Wikipedia, but when I go to any other WMF site, such as meta or commons, I'm shown as logged out. I've tried explicitly logging out on en.wp, and logging in again to get a fresh login cookie, but no joy. Is this a Firefox bug, or MediaWiki? Is there a phab: ticket? --Redrose64 🌹 (talk) 19:00, 3 July 2022 (UTC)
You might be able to get it working again if you disable "Enhanced Tracking Protection" for SUL sites. To do that, click the shield icon at the left end of the URL bar while on the site, then click on the switch next to where it says "Enhanced Tracking Protection is ON for this site". The page should then reload and you should be logged in. Rummskartoffel21:40, 3 July 2022 (UTC)
"Create new section" tab and new editing interface
There has been a new editing interface added for the "Create new section" tab. There was an option to revert to the classic interface but this has disappeared. Why/how do I get it back. Regards, Cinderella157 (talk) 03:24, 6 July 2022 (UTC)
@Cinderella157 If you click the "Source" link on the top right of the new tool to switch to source mode, substing templates works just fine. You can also add dtenable=0 to the URL if there are specific cases where it should never be used. --Ahecht (TALK PAGE) 14:06, 6 July 2022 (UTC)
I just ran the Prosesize gadget on two different articles and I am seeing the results at least three times on the page and sometimes four. For instance, I ran it on George Washington and on Andrew Jackson. On GW I am seeing four different results - three partial results:
The page Quantum biology has the wrong tab. The tab bears the title Biology instead of Quantum biology (in Safari browser). I am not able to adjust that. Could someone else take care of that? Thank you very much Phacelias (talk) 07:59, 10 July 2022 (UTC)
Whoever adjusted, thanks a lot Phacelias (talk) 10:07, 10 July 2022 (UTC)
Decompress stuff that was compressed with mediawiki.deflate
The decompression seems to currently only be available server side. I tried Pako as that's what mediawiki.deflate is according to phab:T236210 but I can't figure out how decompress something that was compressed with mediawiki.deflate. pako works with Uint8Arrays and isn't accepting what mediawiki.deflate produces. — Alexis Jazz (talk or ping me) 16:04, 10 July 2022 (UTC)
Alexis Jazz, from looking at gerrit, it looks like they used deflateRaw, so the following works, as tested locally:
I'm working in Wikipedia in breton, the continental celtic language. I need some help and I would like to know if somebody might be willing to give a hand or to give me the name of somebody who could. I dont know anybody who could help here. We are a very small community and the holidays begun this month. The language itself should not give any problem. I linked the pages in breton (brezhoneg or br) to the pages in english and the code is mostly in english
I've been trying to adapt the template {{Uses Wikidata}} to our wiki to help to spread the use of infoboxes using Wikidata properties. I've created the template and module pages and also some of the other templates needed. Unfortunately it doesn't work as expected. I get the double bullets problem (in front of the properties names) you had in january and wich is mentioned in Module talk:Uses Wikidata. The box is also to large and on the left side of the page, the links are writen in bold.
It might be a problem in the template equivalent to {{Side box}} (where the sidebox is shown on the left side of the page, and in a size different from the english version) but I checked the code for the template and the module and it looks exactly the same as in english, as the breton module {{yesno}}. Kadwalan (talk) 11:31, 8 July 2022 (UTC)
Please link a page with the problem. You always have a better chance of getting help if people don't have to start by trying to work out where you want help. PrimeHunter (talk) 14:30, 8 July 2022 (UTC)
Thanks. I linked the pages to the similar english pages wich are linked in my explanation. Here are the direct links to breton wikipedia if it's easier :
You said "the double bullets problem". I don't see any bullets in any of your links and your version of {{Uses Wikidata}} doesn't have a single transclusion. PLEASE LINK A PAGE WITH THE PROBLEM. Why is it so bloody hard for posters to link a page with a problem they want help with? We shouldn't even have to ask for it. I'm out. Sorry if I sound rude but I have spent fifteen years searching for pages with a reported problem and asking posters for it. It's getting to me. PrimeHunter (talk) 16:13, 8 July 2022 (UTC)
Sorry for the lack of examples. You're right. Maybe because I didn't spend even one year searching for pages with a reported problem ? I tried not to use the template before because there was a problem. You can now have a look at the page br:Patrom:Infobox Skiantour/Teuliadur. Thank you. Kadwalan (talk) 17:02, 8 July 2022 (UTC)
Thank you very much. As I don't know much about code, and as I don't have the right to change this page anyway, I will let this for the time beeing. But would it be surprising if it is not the only problem ? The double bullets problem was mentionned on the english Module Uses Wikidata talk page in January and the the plainlist clas css was in use then, isn't it ? Kadwalan (talk) 17:57, 8 July 2022 (UTC)
I'm not going to attempt to figure out why there was an issue then. The issues are obviously unrelated at that point. Izno (talk) 19:18, 8 July 2022 (UTC)
As for "don't have the rights", you have the rights to use TemplateStyles instead of Common.css. Izno (talk) 19:18, 8 July 2022 (UTC)
Simplified example: Preview a page with {{foo|bar=1|bar=2}}, e.g. this. It says:
Warning: Wikipedia:Sandbox (edit) is calling Template:Foo with more than one value for the "bar" parameter. Only the last value provided will be used. (Help)
Looks like it got broken in I6a7c04. What happens now is that the message intends to replace $1 with $'"1, then parse the wikitext, then replace $'"1 with the page name. But the fullurl percent-encodes the $'"1 so it's not there to be replaced anymore. Anomie⚔12:37, 11 July 2022 (UTC)
Thanks. That also explains why #ifeq:$1|{{FULLPAGENAME}} is apparently always false. I have used Special:Edit/$1 instead.[41] This works but the edit link seems pointless if it's always displayed. It's meant to only display if it's not the previewed page itself but that sounds impossible to implement if the value of $1 is only inserted after the wikitext is parsed. PrimeHunter (talk) 15:08, 11 July 2022 (UTC)
Tech News: 2022-28
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Recent changes
In the Vector 2022 skin, the page title is now displayed above the tabs such as Discussion, Read, Edit, View history, or More. Learn more. [42]
The Anti-Harassment Tools team recently deployed the IP Info Feature as a Beta Feature at all wikis. This feature allows abuse fighters to access information about IP addresses. Please check our update on how to find and use the tool. Please share your feedback using a link you will be given within the tool itself.
Changes later this week
There is no new MediaWiki version this week.
Some wikis will be in read-only for a few minutes because of a switch of their main database. It will be performed on 12 July at 07:00 UTC (targeted wikis).
Missing edits from watchlist for personal talk page
Although I have the "Expand watchlist to show all changes, not just the most recent" option checked in my preferences, and have no issues with seeing all changes for other pages on my watchlist, today there were two edits to my own talk page, but only the most recent one appears in my watchlist. Does anyone know of anything that might cause this? isaacl (talk) 23:41, 11 July 2022 (UTC)
I am not entirely certain if this is an universal occurring problem however when inspecting images of svg's the preview breaks. I noticed this while viewing the svg of Los Alamos chess. I am using timeless but I checked with vector the issue still occurs, likewise I checked with Firefox, Chrome and Opera on all three browsers the bug was present. I highly doubt it's a hardware issue as my browser has standard API and nothing in my computer to my knowledge would cause this bug, the bug may only occur with chess related svg's as ther svg files such as this example displays correctly. As further analysis of board state on chess articles is useful it can be determental for people analyzing certain moves. I have brought an image of the bug for visual reference. Many thanks. Des Vallee (talk) 07:26, 10 July 2022 (UTC)
I don't see anything "broken". There are 36 images (6x6) in those two tables on the Los Alamos page. Iif you click one of the images, you'll see only that image. Malyacko (talk) 08:55, 10 July 2022 (UTC)
Malyacko It is likely an issue with my computer alone then upon inspect the preview turns into a zoomed in chess piece, on all svg's of chess irregardless of the browser. Des Vallee (talk) 12:53, 10 July 2022 (UTC)
The chess board you're seeing there is not one image. It's a grid of 36 images, one per square of the board. When you click on a square it shows the image for only that square. Anomie⚔14:15, 10 July 2022 (UTC)
Anomie Ah I see it is not a file on Wikimedia which explains why there is no inspect. Still this could easily be refined, under the current system someone studying the board must focus on the small blurry preview to the side, likewise it can be very difficult to transfer the svg to other wiki-projects (and more generally the internet) and drastically increases the byte size of an article for a simple svg, however I understand how this system could be more quick then creating an entirely new svg and uploading it to commons. With that mind this is not a bug, simply a misunderstanding thanks for explaining. Des Vallee (talk) 17:43, 10 July 2022 (UTC)
@Pppery: Would you be aware of any way to obtain this list of protection levels with JavaScript, either through the API or through something like mw.config.get()? I tried using the latter, and it returned null for this variable, and in my search to find this list through the former, I did not find any appropriate APIs. Joeytje50 (talk) 00:00, 12 July 2022 (UTC)
Also, the protection level names, as used in various contexts, can be accessed using action=query&meta=allmessages just like any other system messages. Example: [44]. The message keys starting with "action-", "protect-level-", "restriction-level-", and "right-" are the ones related to the user right (and protection level) rather than to the user group. PleaseStand (talk) 00:20, 12 July 2022 (UTC)
Hi, WPVPT!! I'm using Mozilla Firefox PC version. I uploaded an SVG image which is currently deleted unless sought via the suppressors to WP for use in articles, but it isn't showing on WP. It is an SVG converted from a PNG image, which I did through SVG converter websites obtained via Google Search. Before uploading it, it had the typical feel of an SVG, but it became blank on its upload to WP. Does this happen to such SVG converts from PNGs uploaded to WP? Is something wrong with MediaWiki or that new SVG converts from PNGs get uploaded on WP and don't show straight from upload? Intrisit (talk) 10:22, 12 July 2022 (UTC)
@Intrisit: The deleted File:2022 WAFCON.svg can also be viewed and undeleted by administrators but we cannot see how MediaWiki will treat it while it's deleted. MediaWiki converts SVG files to PNG files when they are displayed. This can fail for different reasons, e.g. an error in the SVG format which may not affect other SVG software (my browser can show the deleted SVG), compatibility issues with MediaWiki, or a temporary server delay. Can I undelete it? You will not be punished for uploading an image which doesn't display properly, but don't add it to articles while it fails. PrimeHunter (talk) 14:39, 12 July 2022 (UTC)
You can try using https://svgcheck.toolforge.org to get some feedback about possible issues. A lot of SVG converters are very bad at what they do. Also if you have a brand or something with a PNG, in MANY cases it is possible to find an SVG original (often as part of marketing materials or brand guidelines or whatever), which is simply much better because PNG -> SVG is always a lossy process, whereas if you have an SVG original any sizing is always lossless (which is why SVG is preferred in the first place). —TheDJ (talk • contribs) 14:44, 12 July 2022 (UTC)
In this case, the "conversion" was just to create an SVG that has an embedded PNG inside it with no actual vector content. Anomie⚔01:57, 13 July 2022 (UTC)
Glitches on Featured and good articles when using scripts
See here about an odd glitch in the duplicate link checker, that is only affecting Featured and good articles. There is a similar glitch, also only on Featured articles best I can tell, in the prose size script installed in tools. It is also generating duplicate info, only on FAs. So perhaps the glitch is not coming from the scripts, but from something else in the FA or GA template? The writer of the dup link script hasn't edited since Jan and hasn't responded; maybe someone else can locate the issue. SandyGeorgia (Talk) 21:56, 4 July 2022 (UTC)
I was hoping someone would find instead that the error was in one of the templates, in which case we could fix it. I guess we're stuck. Thanks for looking ! SandyGeorgia (Talk) 03:59, 5 July 2022 (UTC)
Interface admins will take bug fix edit protected requests. You should file one at the appropriate talk page. Izno (talk) 04:16, 5 July 2022 (UTC)
Thx, Izno, but it still makes no sense to me why both of these scripts are failing only on articles that have a GA or FA template? SandyGeorgia (Talk) 04:17, 5 July 2022 (UTC)
Basically, Evad was looking for a (supposed-to-be) single element that bears a class named .mw-parser-output (without the leading dot). This class marks the outputs of the wikitext parser, which is the content of the page, and, unfortunately, indicator templates (e.g. FA/GA star, {{pp}} etc.). NguoiDungKhongDinhDanh04:49, 5 July 2022 (UTC)
The change for indicators to output .mw-parser-output is also fairly recent, so that is why these scripts have had issues only lately. Izno (talk) 06:59, 5 July 2022 (UTC)
I have been unable to figure where to find the host of the prose size script that I have installed in tools on the left-hand side of screen ... sorry, techno-dummy ... but that needs similar attention. Over at WT:FAC, I am told any of these people can do the fixing:
$($(".mw-parser-output", "#mw-content-text")[0]) seems unnecessarily complicated. Wouldn't $("#mw-content-text .mw-parser-output").first() do the job? Nardog (talk) 07:03, 5 July 2022 (UTC)
Maybe it would do the job, I don't know. I do know that Prosesize is still throwing result-anomalies on GA & FA articles. I am getting 3 or 4 results instead of the previous one result. Shearonink (talk) 22:31, 10 July 2022 (UTC)
Hey, if you don't watch the proposals section you may ignore a message you perhaps would prefer not to ignore, so this is like a redirect within the VP to increase the visibility.
In a nutshell, in one to two weeks, we would like to begin the conversation to update the default design of Wikipedia to the Vector 2022. We are interested in deciding on the process we want for the conversation before discussing the change itself. This is because we want to have a good decision-making process that allows for a way to identify the needs of the community from the new skin.
Some some of pdf widget to select text in references and automatically create cites with page numbers (and other information)
Hope this is the right place for this post. Tell me if I should put it somewhere else. I've been thinking for a while about references in wikipedia. I think it might be nice if a reference could identify precisely which section of text supports a claim, and it might be nice if there were a GUI to do this.
This sort of GUI would:
Make it easier for people to cite, creating fewer errors, and speeding up the editing process.
Help ensure that all references had page numbers.
Make it easier for people to check citations.
Provide a more useful data set for machine learning tasks (can you see my real motivation?). Some machine learning tasks might be valuable for wikipedia, e.g. checking citations, suggesting to the reader which part of a page or document supports the claim, in a far flung universe automatically generating citaitons.
I imagine such data could be use of to wikipedia's internal team.
There is some prior work for this sort of stuff in the open source hypothes.is library associated with Jeff Atwood of stackoverflow fame. And meta have been messing around with citations for fact checking recently and open sourced their work. I'm not sure we should particular try to help meta, but we could make it easier for the community as a whole to build on their work. Talpedia (talk) 23:26, 12 July 2022 (UTC)
There are two aspects here, 1 creating an citation from pdf, 2 verifying pdf citation. For 1 there are a bunch of programs that help users to cite an pdf. There is no need for an wikipedia specific thing, because it is enough just to get an citation in an academic format. The requirement of an page number and section does filter these tools quite a bit. For 2, it is worth mentioning that IAarchiveBot started adding urls to pdfs to references with ISBNs. As for the problem itself, Wikipedia does require editors to reword things so they are not equal to the citation, so it would probably have to search for key words instead.
As for why this is not a thing in the three editors (VisualEditor, 2017 Wikitext editor and 2010 wikitext editor) much of the information you want is in the pdf itself. Getting the information needed is not the issue, but as I understand it, the developers are hesitant to temporarily store an pdf just to get information out of it. As for the page and section, you kind of need to get that from an PDF viewer, as then you can ask for the page the editor is currently viewing. The three editors can however tick most of the boxes if you give it an identifier, like an ISBN, if it has one.--Snævar (talk) 06:56, 13 July 2022 (UTC)
Thanks for the reply. I agree that all of this could be handled outside of wikipedia proper and through and external tool which provides citations. In fact, I imagine if this were to get into wikipedia proper it might have to be proven in such an external tool. I can see the "linuxy" argument for separate compatible tools. It's interesting to know that there is a tool that can give me page numbers from pdfs if I highlight a region, I'd like to give one of those a whirl if anyone new about them. The lazy version of me would love something to find the ISBN/doi from a pdf I'm viewing and include the page number.
I guess the main addition within wikipedia associated with is the idea of selecting which span of text within a document contains the information, as well maybe as providing urls that select the page and piece of text. I don't quite know how you would deal with multiple references to different spans on different pages.
I could also imagine people reviewing references selecting the bit of text that matches - assuming they are actually motivated to review what has been added. One thing I do sometimes when reviewing other people's claims is add page numbers.Talpedia (talk) 10:45, 13 July 2022 (UTC)
Anyway, this is phab:T136722. This will be added to mediawiki sometime, but it seems blocked on the backend software, Zotero, or adding another backend software to get the necessary information. Snævar (talk) 12:17, 13 July 2022 (UTC)
Hmm that doesn't seem quite the same. That bug is about pulling metadata out of a pdf.... while I want to pull out the page number and selected section from an open pdf (and potentially do something with it). Talpedia (talk) 15:02, 13 July 2022 (UTC)
Actually refs are having a lot of problem nowadays as we can't use <ref><ref></ref></ref> (nested refs) as such we can expand {{refn}} but after expansion will get problem. Thingofme (talk) 09:11, 13 July 2022 (UTC)
I stopped following the various tickets on this years ago, so I don't know. IIRC, this was a pretty basic problem, involving the way the parser handled recursion. 50.75.226.250 (talk) 15:09, 13 July 2022 (UTC)
Hello, I'm using Vector 2022 Skin and Dark mode Gadget. In the recent Vector 2022 skin, the title has moved up the tab. However, there is a problem here. There is no problem in light mode, but in dark mode, another line appears above the title (shown in red squares above). I don't know if this is a bug in a Vector or a bug in a Dark mode Gadget. Bluehill (talk) 11:17, 7 July 2022 (UTC)
It is because .mw-body has a 1px transparent border-top property which is causing the line to appear on the dark mode from what I understand. It is transparent with a white background, just not with dark mode, you could always disable this with custom css or change the color to the background color. Terasail[✉️]11:25, 7 July 2022 (UTC)
Thanks @Xaosflux. This is odd. I don't see the line when I'm logged in. Instead, the content is in a box, with all four borders. But in the private/incognito mode, I do see the line instead of the box. What's going on? I'm not sure what gadget/user script I'm using may change my experience! :D SGrabarczuk (WMF) (talk) 01:32, 12 July 2022 (UTC)
Not sure, I don't think this is rally a software "bug" - there is a box there, maybe it isn't needed for something else - maybe it is, but the local CSS is styling it. Since darkmode here is just a local css, if someone wants to identify and change the element in it they can just make an edit request over at Wikipedia talk:Dark mode (gadget) to change the styling I suppose. — xaosfluxTalk09:51, 12 July 2022 (UTC)
I agree, this isn't necessarily a software "bug". Our team should be aware of this, though, because we're working on the grid system, and there are just things that may make this change again. SGrabarczuk (WMF) (talk) 12:16, 12 July 2022 (UTC)
Hello -- I am User:CharlesGillingham, and I was locked out of my account by the password system. I think it was sending the "password reset" message to an old email, or something along those lines. The help page advised that, if I didn't receive the "password reset" message I should open a new account, so I did. However, I have lost the editing privileges I enjoyed as a long-time editor. (2006 or so, hundreds of thousands of edits.) Can anyone advice me on how to reset password of my old account? ---- CharlesTGillingham (talk) 00:45, 12 July 2022 (UTC)
@CharlesTGillingham:User:CharlesGillingham has not specified an email adress in the account. Special:CentralAuth/CharlesGillingham shows 21,415 edits here and a few to other projects but no advanced user rights. I don't know whether that is enough for somebody with the required access to verify your identity and set an email address for the account so you can get a password reset. It helps a lot that the user page states your identity (added by the user) so it can potentially be verified that the request is really from the account owner. Passwords never expire so you can keep trying. PrimeHunter (talk) 02:59, 12 July 2022 (UTC)
The only privilege I need is "auto confirmed", so that's not really a concern. (I remember being notified, around 2011 or so, that I had passed 100K edits ... but of course that's not an issue worth worrying about.)
The last password was Safari-generated, so there's no hope of my guessing the password -- literally intractable. It would still be nice to keep going in the same old account. Do you know how I might go about finding a person who might be able to set the email? ---- CharlesTGillingham (talk) 03:43, 12 July 2022 (UTC)
I lost a lot of my passwords when troubleshooting Firefox recently. (I clicked a button that said "You won’t lose essential information like bookmarks and passwords." Well, it was half right.) However, I still have both my previous laptop and a full backup, and I was able to look up the passwords there. In Safari, there's also the possibility of passwords being stored in Apple's iCloud or Keychain. I hope that you are able to find your passwords. Whatamidoing (WMF) (talk) 18:06, 13 July 2022 (UTC)
Considerations for edit notices on mobile
On mobile platforms like the Android app and likely soon https://en.m.wikipedia.org, edit notices are delivered as a popup that needs to be dismissed (and can be revisited with a button), as opposed to a passive box above the editing window. I've implemented the same in the editor script I'm using right now which causes edit notices to appear as a popup on desktop platforms too. On mobile, there often just isn't enough screen real estate to display warnings in a passive box. This is good for warnings like Template:Editnotices/Page/Talk:Goofy, it makes them harder to miss. But not all notices are used like this, Template:Editnotices/Namespace/File for example contains more generic information and pops up on every file description page the user hasn't edited before. Whether that's really needed is an open question, personally I lean towards no: I'd rather reserve edit notices for actual warnings specific to a particular page or subject.
To stop a particular notice from showing on mobile platforms it can be wrapped in an element with the nomobile class. We should consider this for notices that are primarily informative and not really warnings.
A wall of text is unlikely to impress a mobile user, so consider wrapping everything outside the core message in an element with the nomobile class.
Many edit notices use markup to draw attention: background colors, images of exclamation marks, red and/or bold text, borders, etc. In a passive box these make sense due to banner blindness, but they are largely inappropriate when already shoving a popup in the user's face. They can also cause issues at limited screen width. Mobile platforms try to strip at least some (if not all) of this stuff, but please keep this in mind when designing a new notice.
Shouldn't edit notices be popups on desktop too? It'll make them harder to miss, and do you really enjoy having that box sitting permanently above the edit window?
How should we handle more generic information for mobile platforms that is best delivered passively? One solution I'm thinking of is the introduction of a new class for editnotices that, when detected, suppresses the popup but still enables the button to revisit the notice so the content is accessible. Maybe call it "nopopupnotice" or something. (edit: Detection of elements with the "nopopupnotice" class will be available once EditNoticesOnMobile gets synced from my userspace. The actual classname may be changed in the future.)
Wow... this topic just made me realize how blind to edit notices I really am. Also, since I'm now thinking about it, we should probably revisit the idea of edit notices such as the one on Talk:Goofy entirely. The new reply tool skips the edit notice entirely, so if someone is using that they will never see the warning about Goofy not being a cow. ~ ONUnicorn(Talk|Contribs)problem solving18:30, 12 July 2022 (UTC)
DiscussionTools' ==New Topic== tool displays edit notices. The [reply] button doesn't. I'm not sure that it makes much difference for replying to someone else's comments. Talk:Goofy has only twoedits using the reply tool (and none with the ==New Topic== tool), and neither of them mention cows. Whatamidoing (WMF) (talk) 18:18, 13 July 2022 (UTC)
When viewed from an iPhone, without being logged in to Wikipedia, two of the blockquotes (but not the others) in Planet show in a font much too small to read:
@SandyGeorgia, when you say it is in font too small to read, which skin are you using? The default mobile skin ("Minerva") as is found at the mobile site? Or desktop ("Vector"), as are the links you provided? Just checking. Izno (talk) 04:51, 13 July 2022 (UTC)
Both of those quotes look larger than the surrounding text (and, weirdly, are rendered in serif font instead of the sans serif used in the body) on my iPhone using Safari, not logged in. – Jonesey95 (talk) 05:54, 13 July 2022 (UTC)
You are likely running into automatic font size inflation that phones provide. With this, they increase the size of elements like headers and paragraphs etc compared to items the device thinks are less important like block quotes and images. It's expected behaviour on a phone when using the desktop version of a website (which was not designed to be used on mobile) and is an attempt to make desktop websites more readable on phones. So actually the block quote is at the 'proper' text size, its the paragraphs around it that are 'too big'. Our normal paragraph font size is around 14px, but on mobile with text inflation, that becomes a font-size around 24px. —TheDJ (talk • contribs) 08:03, 13 July 2022 (UTC)
Thx, but ... Then why is it not happening on all blockquotes? Those two quotes are the only ones affected, and the reset of the text looks fine, as far as I can tell. I don't think the text around it is changed ?? Hard to tell ... SandyGeorgia (Talk) 16:22, 13 July 2022 (UTC)
@SandyGeorgia, yes, that's exactly why I asked the question. As TheDJ explains, skins which are not 'optimized' for mobile (like Vector today) are going to run into browsers attempting to display content sensibly. I have seen similar behavior depending on other content as you do with the different quotes.
I would highly advise against attempting to 'correct' this in the source page. The fundamental issue is Vector here, and not something worth changing because it may negatively impact other display, or as you seem to have discovered below, an awful lot of work that could compromise our guidelines on page layout. (I think some of your results are interesting, mind you.)
At some point in the near-ish future, the new version of Vector may do a better job supporting mobile, so the problem will become more-or-less irrelevant except for a tiny quantity of logged in editors viewing the website (which accordingly should not be optimized for). Izno (talk) 17:35, 13 July 2022 (UTC)
Izno OK, but I'm not sure if I've left this at the best place ... unclear what you are recommending, since I don't fully speak this language. Because this is a Featured article, it shouldn't compromise anything about images or quotes, and we should get it right (it could run WP:TFA once it passes FAR). Is User:SandyGeorgia/sandbox#2006 IAU definition of planet a good solution, compliant with everything ? SandyGeorgia (Talk) 19:36, 13 July 2022 (UTC)
I'm suggesting that trying to work around this issue on mobile (iPhone) using the desktop website (specifically the Vector skin) is doomed to failure right now and which should not at all be cared about at TFA or any other process. Make it look decent on the mobile website (en.m.wikipedia.org) while using mobile and on the desktop website (en.wikipedia.org) when using a desktop. That's it. Izno (talk) 20:15, 13 July 2022 (UTC)
No, you cannot select your skin while logged out besides using one of the two different websites (en.m. versus en.). Which you are doing when you click that link. :) Izno (talk) 17:37, 13 July 2022 (UTC)
I fiddled again with the blockquote, and re-checked from my iPhone, not logged it. The first quote in this section is tiny; the next blockquote in the same section is fine. This is from the desktop view on iPhone. When I click on "mobile view", it's fine. I am posting this from my iPad, which shows no problem. SandyGeorgia (Talk) 16:44, 13 July 2022 (UTC)
And, doing this solved the "sub-stellar mass" blockquote problem. So, it is related to not text, but image sizes around the quotes ?? So now, how do I get the first image up to a larger size if upright= is causing a problem? Have not tested switching the second to upright yet ... SandyGeorgia (Talk) 17:01, 13 July 2022 (UTC)
I know this is terribly vague, for which I apologize, but has anything changed in Parsoid in the past 24-48 hours? I'm working a heisenbug in a script I wrote which parses sock templates on user pages, calling $.get('/api/rest_v1/page/html/' + pageTitle); to parse out the template data.
Starting yesterday, this started to fail on seemingly random user pages. I've seen pairs of user pages with byte-for-byte identical content where one parsed correctly and the other not. And now it's failing on more and more of them. I even saw one of my test cases transition from "this one works" to "this one fails" in the middle of a debugging session. A plausible explanation would be that parsoid started returning a different parse tree sometime yesterday and what I'm seeing is the user page parse trees slowly transition as some HTTP cache times out.
I downloaded your repo from github and took a look.
So, the node you were trying to find has typeof="mw:Extension/templatestyles mw:Transclusion" . This change happened after @Izno updated the mbox template.
The fix is actually pretty simple in your case. Change $html.find('[typeof="mw:Transclusion"]') to $html.find('[typeof~="mw:Transclusion"]'). Note that ~ addition there. The typeof attribute can embed multiple applicable types for the node. We should perhaps clarify our docs.
{{FAQ}} does not have the more common v • t • e buttons in the corner. Instead, it has a big "Edit" button. Could someone make it have (at least) a "View" button as well? I don't always want to edit the page. WhatamIdoing (talk) 03:21, 14 July 2022 (UTC)
I'm attempting to use the "follback all" tab to revert these sock edits; although the edit summary box pops up, when I click "ok", nothing happens and the edits are not rolled back. Any other admins having this issue?-- Jezebel's Ponyobons mots16:44, 15 July 2022 (UTC)
Their particular way of being annoying is to revert their own edit right after they make it, so there is nothing to rollback. --Floquenbeam (talk) 16:49, 15 July 2022 (UTC)
Thanks for your comments - this may be well beyond my coding abilities at the moment - any specifics (ie, actual installation coding on an article page) welcome of course - iac - Stay Safe and Healthy !! - Drbogdan (talk) 15:10, 15 July 2022 (UTC)
@Drbogdan short answer is: we can't today, because we don't have software that can do that available. Long answer is above, we don't like to just say "nope" for no good reason, but extension development is a long process for sure. Some options you could look in now: use a gallery to just show the images side-by-side in static form; make an animated GIF that does the slider back and forth automatically. The former is traditional and should have little resistance, animated gif's in articles are slightly contentious - but have some support. — xaosfluxTalk15:39, 15 July 2022 (UTC)
@Xaosflux: (and others) - Thank You for your reply - and *Excellent* Comments and Suggestions - they're *Greatly* appreciated - the "GIF" option seems interesting - may try this at some opportunity - Thanks again for your reply - and - Stay Safe and Healthy !! - Drbogdan (talk) 18:35, 15 July 2022 (UTC)
CS1 maint.
Hi all, I've just fixed (I hope) some |url-status=bot: unknown params at Royal Ulster Rifles, for example [1]. This is the first time I've come across this message. I wonder if someone could explain for a non-techie, please: Why does this happen? ie What is it that the bot can't recognise about what appears to be a standard 404 error? And why does this ref[2] generate the same CS1 maint. message although the original url is still live? Cheers, MinorProphet (talk) 12:37, 15 July 2022 (UTC)
Have you asked the bot operator? |deadurl=bot: unknown was added at this edit (5×). In that edit it looks like IABot simply created |archiveurl= and |archivedate= from archive snapshot urls that were improperly assigned to |url=.
The St Patrick's Barracks citation was added to the article by Editor Dormskirk at this edit. Perhaps that editor can answer your question about that citation.
Thanks, both. I probably don't need to worry my head too much about it, I'll just fix things whenever I come across them. Cheers, MinorProphet (talk) 18:47, 15 July 2022 (UTC)
The showing Editnotices to mobile editors RfC passed today. This means Wikipedia:EditNoticesOnMobile will soon be enabled as a default gadget. To ensure no unexpected issues arise testers are wanted. While the gadget has been tested, there are endless combinations of browsers, devices and editnotices out there and they can't all be tested by one person. Testing is simple! Enable "EditNoticesOnMobile" in your preferences and do what you normally do. If you run into any issues, report them on Wikipedia talk:EditNoticesOnMobile. If you don't run into any issues, also leave a note! This way we get some idea of how many people have tested the gadget. Thank you in advance! — Alexis Jazz (talk or ping me) 23:59, 15 July 2022 (UTC)
Asked at Tea but no satisfactory response, so heading to where the techies hang out.
Is there a way to see how many articles are in a category inclusive of articles in subcategories (bonus points if dupes not counted but no biggie)
Specifically, I was trying to compare Category:People from Bangkok with Category:People from New York City. The former can be done by hand, but the later is not so easy. And if there is an on-wiki way, then I can expand my comparison to other cities/regions. Slywriter (talk) 18:53, 15 July 2022 (UTC)
Petscan may work for you. Here's a category search that goes three levels deep; the category tree goes deeper than that, but you have to be careful about ending up in subcategories that are something else entirely. I get 31,195 results (no duplicates) from that search, limited to 10,000 shown on the default output table; you have to change the output in the Output tab to get full results. You may also want to traverse the category tree to understand what you are getting, including a lot of fictional people. – Jonesey95 (talk) 21:30, 15 July 2022 (UTC)
It's not possible to set PetScan to go to unlimited subcategory depth: you need to specify depth as an integer (which effectively has no upper bound in either direction, it even permits negative values, even though those have no meaning). Don't set very high values - not only does it increase the processing time, it increases the possibility of hitting a category loop. --Redrose64 🌹 (talk) 08:39, 16 July 2022 (UTC)
"Tag: Reverted" showing up on edits that clearly were never reverted
I don't know how long this has been going on, or what the extent of the problem is, but, for instance, this edit and the one following it have never been reverted, evn though they were tagged as reverted: [46].
(Obviously I'm a bit miffed because one of the edits was mine, and it was reverting vandalism -- so when I check my own edit history and see that my rvv has been tagged as reverted, I have to spend a considerable amount of time trying to figure out what's going on and whether anyone in fact reverted me without me getting a notification.)
Yes, the current revision is identical to a revision less than 15 edits ago (mw:Manual:$wgRevertedTagMaxDepth), so all those revisions are tagged "reverted". It's a tag which can come and go depending on the current version. It doesn't mean that an edit was individually reverted. PrimeHunter (talk) 03:51, 16 July 2022 (UTC)
When I view that page, the top post has the ArchiverLink but the other three posts on that page do not. I do not know why...and I wanna know. Thanks, Shearonink (talk) 18:49, 17 July 2022 (UTC)
What does "upstream error" mean? It happened a few days ago, when I clicked on a Wikipedia page. I could pull up anything else on the internet, but not Wikipedia, Wikisource or Commons - not on any of my browsers. The error message just appeared again, and only lasted a minute or maybe less. But like last time, I could not pull up any wiki sites or Commons, but had no problem with any other sites. What is causing this? — Maile (talk) 01:55, 18 July 2022 (UTC)
@Maile66 It means that the PHP part of Mediawiki cannot connect to the database part of Mediawiki. It can be caused by all kinds of issues, overloaded servers, malfunctioning extensions, bots making too many edits, database maintenance, misuse of extensions etc. The only people who really have the tools to diagnose this are the developers and sysadmins. 192.76.8.85 (talk) 02:48, 18 July 2022 (UTC)
On what basis are interlanguage links regionally categorised?
Interlanguage links on English Wikipedia article Nakshatra as on 17 July 2022.
On what basis Russian (language), Marathi (language) & Malayalam are languages of the middle-east, and how is Spanish (language) African? It doesn't make much sense. Yes, these language show up in their respective continents (Europe, Asia) too but I still don't understand the basis. If languages of immigrants is included in deciding interlanguage link categorisation, we should have all Indian languages show up in America section but it doesn't. —CX Zoom[he/him](let's talk • {C•X})14:18, 17 July 2022 (UTC)
It's generally due to former colonies of those countries in those regions, where the language was or is an official language. For instance, there is Dutch in the americas due to their former Caribbean colonies. Spanish is in Africa because of for instance Equatorial Guinea. Malayalam is in the Middle East, probably because of Judeo-Malayalam ?? There are a lot of states which simply recognise a lot of languages for various reasons. The list with this mapping is jquery.uls.data.js, part of jquery.uls, part of the Universal Language Selector MediaWiki extension. —TheDJ (talk • contribs) 11:49, 18 July 2022 (UTC)
Problem showing UTC live clock gadget in Vector 2022 skin
Ah looks like you did. Are you only seeing this problem in vector-2022 on enwiki, or are you also seeing this problem on other projects with vector-2022? — xaosfluxTalk14:15, 18 July 2022 (UTC)
I just saw it on testwiki as well. Think it has to do with per-project CSS settings - if one of the devs has fixes we can import them. — xaosfluxTalk14:24, 18 July 2022 (UTC)
FYI, on eu.wikipedia.org and mediawiki.org it shows outside the menu altogether. When I tried to do this for en.wikipedia.org I got some complaints (I can't find the appropriate link right now), so it might benefit from some discussion about whether this should appear in the dropdown at all?
FWIW the clock is in the user dropdown but is not clipped for me on English Wikipedia. Not sure why. Perhaps there's another gadget interfering here? Can you try disabling a few gadgets to narrow this down? Jdlrobson (talk) 17:33, 18 July 2022 (UTC)
The "New section" tab at the top of talk pages (such as this talk page!) is not working when JavaScript is disabled: when you click on the tab, nothing happens. The tab used to work fine without JavaScript, but the underlying code must have changed recently and broke it for no-JavaScript editors. It would be great to restore this functionality for no-JavaScript editors. See mw:No-JavaScript notes for more reasons why this is important. Biogeographist (talk) 16:43, 18 July 2022 (UTC)
I turned off the "Enable quick topic adding" preference, and it works: the old behavior of the "New section" tab is restored. Thanks! I never would have figured that out by myself. Biogeographist (talk) 17:10, 18 July 2022 (UTC)
And what is your browser? Does it work if you log out? It works for me in Firefox and Edge, both logged in and out. Try to bypass your cache. Use Ctrl+F5 in most Windows browsers, not F5 alone. If you disable JavaScript for testing then bypass after disabling it. PrimeHunter (talk) 17:08, 18 July 2022 (UTC)
You should have said you use Noscript to "disable" JavaScript. I'm not suprised you get problems if JavaScript is enabled in the browser itself and you use a browser extension to control which JavaScript is allowed to run. I don't have Noscript. If I disable JavaScript in Firefox and "Enable quick topic adding" is enabled then it works for me. I don't get the quick topic feature which requires JavaScript but I do get the traditional new section window. Does it work if you really disable JavaScript? See e.g. https://www.lifewire.com/disable-javascript-in-firefox-446039. PrimeHunter (talk) 17:58, 18 July 2022 (UTC)
Sorry that wasn't clear at first. NoScript is a pretty common way to disable JavaScript in Firefox, so I don't think the scare quotes are needed around "disabled", but you're right that it's necessary to differentiate methods of disabling JS due to technical implications. I just tested disabling JS via about:config, and the new section tab works as expected with that method even with "Enable quick topic adding" enabled, although that's not the method I use to disable JS. Biogeographist (talk) 18:50, 18 July 2022 (UTC) & 19:05, 18 July 2022 (UTC)
NoScriptblocks JavaScript. I don't know the technical details but I guess your Firefox tells MediaWiki that it has JavaScript enabled and MediaWiki therefore chooses the JavaScript interface but NoScript blocks it. I don't know whether MediaWiki can make a workaround. PrimeHunter (talk) 19:44, 18 July 2022 (UTC)
I noticed that DLynch asked at phab:T313246 whether I have "New wikitext mode" enabled, and I don't, nor any of the other beta features. Also, I have NoScript configured to allow <noscript> content, so blocking of that content is not the issue. I just tested it again while logged out, and the first time I clicked on the "New section" tab, it didn't work. But then I clicked on it a second time, and the second time it worked—it redirected to the basic editor. And now it always works when I'm logged out (always redirects to the basic editor), so I can no longer reproduce the issue when logged out. I don't know what changed; I swear it didn't work when I tested it before when logged out. So it appears that phab:T313246 could be closed. I'm sorry if I unnecessarily caused a bug report to be opened, although I'm glad I mentioned the issue since Xaosflux taught me the preference fix when I'm logged in. Biogeographist (talk) 21:04, 18 July 2022 (UTC)
academia.edu/download
Most users report this URL returns 404. However some say it works. The works and not-works are both in the USA. One works user reports success on Windows 10 (Edge) and macOS (Safari). One not-works user reports failure on Windows 7 (Chrome and Firefox) and Linux (Chrome and Firefox). There are about 1,000 URLs with academia.edu/download that have the same problem. Does it work for you? -- GreenC19:51, 18 July 2022 (UTC)
If I go to the main page (academia.edu), it requires a login. Maybe the few "works" folks signed up for an account at some point and it kept them signed in? Schazjmd(talk)19:58, 18 July 2022 (UTC)
There's something very weird going on with that one. It redirects to https://d1wqtxts1xzle7.cloudfront.net/30869670/Turismo_y_Territorio_en_Salta-_Caceres_et_al-_CONICET-UBA_2012-with-cover-page-v2.pdf?[big-awful-hash]. Headbomb {t · c · p · b}20:14, 18 July 2022 (UTC)
Yes that's what it does when it works.. they host the content at AWS so it redirects there. When it doesn't work it looks like this. What browser did you use? -- GreenC20:41, 18 July 2022 (UTC)
The issue indeed seems to be of requiring an account. I'm on Firefox on Android, and have logged in account with Academia. the link works when I click in normally, but when I open it in incognito mode, it goes to a 404 page on Academia. TryKid[dubious – discuss]20:52, 18 July 2022 (UTC)
Thank you, User:Schazjmd for figuring this out and TryKid for verifying. I emailed the site to let them know about it. The 404 status will cause bots to mark it dead. -- GreenC21:32, 18 July 2022 (UTC)
Tech News: 2022-29
Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.
Problems
The feature on mobile web for Nearby Pages was missing last week. It will be fixed this week. [47]
Changes later this week
The new version of MediaWiki will be on test wikis and MediaWiki.org from 19 July. It will be on non-Wikipedia wikis and some Wikipedias from 20 July. It will be on all wikis from 21 July (calendar).
When adding {{Infobox television}} to a new page on desktop and clicking "Show preview", I see two Lua errors. Firstly, at the top of the prose part of the screen: "Lua error in Module:Infobox_television at line 106: bad argument #1 to 'find' (string expected, got nil)." Secondly, at the top of the infobox: "Lua error in Module:Infobox_television at line 284: bad argument #1 to 'match' (string expected, got nil)." I don't see this error once the page has been created, either in the article proper or the preview screen.
I believe the issue is that the Lua is failing to grab the name of the page that the infobox is on, as the title at the top of the infobox does not display in the preview. I'm using the editing preference "Show previews without reloading the page", in case that's relevant. I note that I'm not the only one with this issue. It's just the TV infobox that's causing the issue: the page title is shown correctly and there are no Lua errors with {{Infobox film}}, for instance. — Bilorv (talk) 11:02, 19 July 2022 (UTC)
I couldn't replicate it when testing it on Etetetetete. I however added nil checks to the lines the error message pointed to. In general though, post issues at the most relevent location. This isn't it. Gonnym (talk) 11:23, 19 July 2022 (UTC)
Search categories for both article and talk space conditions
Is there a way to search for articles matching an AND condition for categories amongst both the talk page and the main article page? eg. something like
$.ajax( { url: 'https://de.wikipedia.org'+mw.config.get('wgArticlePath').replace('$1','Special:GadgetUsage?useskin=vector'), cache: false, type: 'GET' } ); I'm not asking for account data, this request can be made without cookies as far as I'm concerned. It results in a CORS error. So I searched and found many solutions that don't work. In many questions about it it's unclear which options are for clients and which are for servers. I stumbled upon [48] which basically says "use this proxy". The stupid thing is: that actually works. It's obviously infinitely less secure and less reliable, but it works. *facepalm* There has got to be a better solution. — Alexis Jazz (talk or ping me) 21:11, 18 July 2022 (UTC)
which makes the request without complaining, but no obvious way to get the page content. — Alexis Jazz (talk or ping me) 22:10, 18 July 2022 (UTC)
Oh for fucks sake: https://stackoverflow.com/questions/43262121/trying-to-use-fetch-and-pass-in-mode-no-cors. Long story short: in this case, evil people can easily get around CORS by using a proxy and good actors are forced to use a proxy as well because this is pretty dumb. What's the bloody point in blocking public resources? The actual request is still made, so if my goal was to send sensitive data to evil-incorporated.com I still could. And if I need to retrieve evil-incorporated.com/mook.js I still could. *facepalm* — Alexis Jazz (talk or ping me) 22:52, 18 July 2022 (UTC)
There is no better solution. This limitation of the web platform isn't to block you from retrieving whatever, it's to block you from retrieving it using someone else's device. As a simple example, if anyone on the internet could fetch https://de.wikipedia.org/wiki/Spezial:GadgetUsage from your browser, then they could extract your username from the result and associate it with your IP address. There are omre complicated scenarios, e.g. someone could access an intranet website that's only available from specific IP addresses. That's why using a proxy is the accepted workaround – then it's the proxy doing the retrieving from their own device and with their credentials, not with yours. Matma Rextalk23:18, 18 July 2022 (UTC)
Matma Rex, thanks. Extraction of account details seems like it should be a non-issue: there should be a way to make the request without cookies. For the intranet, that would be a valid point, but an intranet site (which Wikipedia is not!) should specifically specify that its content should only be accessible by accepted domains. No, wait actually: if the data on the intranet site is sensitive in any way, it should be locked behind a login. Also, using ToolForge as a proxy may not be allowed in this case. ToolForge rules are confusing with that. — Alexis Jazz (talk or ping me) 23:40, 18 July 2022 (UTC)
Even revealing the existence of an intranet site by allowing you to retrieve the login page is a privacy leak. Anyway, you're of course right that this could be done in a better way, but that is the platform we have to develop for and it has become this way for historical reasons; I thought you'd appreciate the explanation. There are sites on the internet (and definitely on intranets) that pre-date the introduction of CORS, and it has been specced this way to be compatible with sites that do not know of it.
I don't see how that wouldn't be allowed on Toolforge, but what do I know.
(Maybe the real solution to your problem would be to write a patch for the Gadgets extension that would make the gadget usage data available in the API somewhere. It seems like it could be added to action=query&list=gadgets easily enough. I'd be happy to review it if you cared to write it.) Matma Rextalk02:00, 19 July 2022 (UTC)
Matma Rex, I do appreciate the explanation, but it's still oddly designed and when "use a proxy" is the correct answer something has gone horribly wrong somewhere along the way. I could appreciate if intranet sites would specify that browsers should pretend they are unreachable when trying to access them from a foreign domain. I could also appreciate browsers doing that when they detect an internet site is trying to access something in a local range. But neither should be standard. And say you know about some intranet site and you know how to hack it. Trick an employee into visiting your blog and you can make requests towards that intranet site! Never getting a response certainly complicates things, but the fact you can make requests at all should be worrying enough. Thinking about that, I'm starting to worry about routers. Getting the usage data through the API would certainly be very nice. I'm pretty much completely unfamiliar with MediaWiki though and I haven't done any PHP in ages. I do wonder: if the API can be accessed from other domains, why can't other pages be accessed? If security is an issue, isn't there some render-as-anon option that could make CORS more lax? — Alexis Jazz (talk or ping me) 02:51, 19 July 2022 (UTC)
Consider if you were logged into your banking site and then visited a site with some malicious Javascript that made a request to access private information from your bank web site. To prevent it from going through, the cross-origin request has to be blocked by default. A proxy would be unable to access your info, as it would not be logged in as you. isaacl (talk) 03:46, 19 July 2022 (UTC)
Isaacl, a request without cookies or other session data (compare opening a private window in your browser) would have the same effect for internet sites, wouldn't it? — Alexis Jazz (talk or ping me) 03:59, 19 July 2022 (UTC)
Sure, making sure you're always logged out of your accounts would provide protection (and it's why this is a general security recommendation). In practice, users don't always log out, and don't browse every site with separate sessions restarted from scratch. Third-party Javascript code can be injected by third-party advertisers/trackers/discussion widget providers and so forth. Private browsing is, in a sense, overkill as not all cookies and other session data are problematic, but it's a reasonable default to wall off interactions between sites. In a similar manner, it's reasonable to block cross-site Javascript requests by default. isaacl (talk) 06:31, 19 July 2022 (UTC)
I don't mean ask the user to log out, I mean more like {mode:'no-cors'} resulting in a request akin to how that request would be made if a private window was opened for it. That's what I initially expected from that option, but instead it just prevents you from getting any response at all. — Alexis Jazz (talk or ping me) 06:39, 19 July 2022 (UTC)
The native Javascript fetch API lets you specify that all credentials should be omitted, same origin credentials should be sent, or credentials should be sent in all requests (same origin or cross-origin). I don't know if the jQuery ajax API supports setting these options. isaacl (talk) 21:29, 19 July 2022 (UTC)
Yep, you definitely should worry about routers.
I do wonder: if the API can be accessed from other domains, why can't other pages be accessed? If security is an issue, isn't there some render-as-anon option that could make CORS more lax?
I think in principle this would be possible, I don't know for sure why it's not allowed. I could guess that a) folks who worked on that wanted to encourage using the API instead of parsing the HTML output and b) it would require reviewing all of MediaWiki and extensions to make sure everything respects that parameter, and doesn't directly access global state like cookies. (There is a render-as-anon option in the API: origin=*, intended especially for anonymous CORS, for example: https://en.wikipedia.org/w/api.php?action=query&meta=userinfo&origin=*.) Matma Rextalk10:47, 19 July 2022 (UTC)
Matma Rex, I don't see how that wouldn't be allowed on Toolforge, but what do I know.wikitech:Wikitech:Cloud Services Terms of use: "Using Wikimedia Cloud Services as a network proxy: Do not use Wikimedia Cloud Services servers or projects to proxy or relay traffic for other servers. Examples of such activities include running Tor nodes, peer-to-peer network services, or VPNs to other networks. In other words, all network connections must originate from or terminate at Wikimedia Cloud Services." Not so sure what I'd need would be allowed by that. Wikipedia isn't part of Wikimedia Cloud Services I think. And it certainly would be forbidden for any domain outside Wikimedia. I thought about some script to create citations or similar things for stuff not supported by Citoid. Like importing ISBNs or something, processed by the client. CORS generally prevents that and ToolForge prohibits running a proxy to get the data from www.isbnsite.example to the user. mw:Wikimedia Labs/Agreement to disclosure of personally identifiable information: "and agree your IP address will be made publicly available and not be treated as confidential". Okay, hard pass. — Alexis Jazz (talk or ping me) 05:09, 19 July 2022 (UTC)
A page that is labelled as a draft with talk page comments from 8 years ago? I don't think you have much to worry about there. Izno (talk) 05:32, 19 July 2022 (UTC)
Huh, okay, I guess that does technically prohibit doing this, although I don't think that's the intended spirit of the rule. I'd ask for a clarification and I'd expect that doing what I suggested would be allowed. But I guess this is a moot point given SD0001's better solution below. Matma Rextalk10:57, 19 July 2022 (UTC)
I agree with Izno that the random mw.o page you found is outdated, but to be clear, anyone who SSHs into Toolforge will have their IP made available to anyone else who is logged into the machine (run w -i). There's nothing stopping you from using a VPN, etc. to mask your real IP from Toolforge. Legoktm (talk) 19:16, 19 July 2022 (UTC)
This API is missing the information about whether the gadget is default or not though, so you have to guess whether "0" means no users or all users… But I realized you can cross-reference this with the data in https://en.wikipedia.org/w/api.php?action=query&list=gadgets, which marks default gadgets. Together with that I think this matches the Special:GadgetUsage functionality. Matma Rextalk11:03, 19 July 2022 (UTC)
I am little confused. What null edit where? I see image is now in the categories but I don't see any null edit in the category or image. What am I missing? Whpq (talk) 21:47, 19 July 2022 (UTC)
If either of these situations apply:
page P shows category C at the bottom, but P is not listed at the page for category C
the page for category C lists P, but C is not shown in the categories box at the bottom of page P
Thanks for the explanation. My confusion came from thinking that a null edit was the same thing as a dummy edit. Now I understand they are not the same thing. Whpq (talk) 23:18, 19 July 2022 (UTC)
Yeah, we have a lot of jargon here. In my experience fixing weird little problems, a null edit is always your best bet. Click Edit, even on a section, then click Publish without changing anything. If that doesn't fix the problem, then you really do have a problem that needs investigating. – Jonesey95 (talk) 05:29, 20 July 2022 (UTC)
Could a bot assist with table column manipulation?
Is there a bot that could assist with table column operations in dozens or hundreds of articles? I reported a SYNTH problem at this discussion involving what appeared to be a dozen articles, where the fix involves primarily dropping one column from a table on each article having the issue, plus a couple of other concomitant column tweaks. I can do that and have been plodding through, but what I thought was a dozen articles has ballooned to 165 now, and I don't think that's the end of it. It's too many to handle individually. Is there a batch operation that could assist with this? Thanks, Mathglot (talk) 07:17, 18 July 2022 (UTC)
@Qwerfjkl: Thanks; currently it's up to 165. I probably can't use AWB, as the regex often has to be tweaked because of all the variation in styling among table cell definitions in the various articles. My latest is s!(\|{1,2}\s*rowspan="\d"\s*)?(\|{1,2}){{(:?Flag\|)?[^|]+?}}\s*.(\|[\-}])!|-!gm, and each version gets a little better at handling variation until there's some new variant I haven't seen before, so it sort of keeps evolving. Which is easy enough for me to tweak on the fly, but I guess would be tougher in AWB (dunno; haven't used it). So I suppose I'm stuck with my hand-rolled, ever-expanding regex. Thanks again, Mathglot (talk) 10:08, 20 July 2022 (UTC)
Edit summary box - turn off enter to save changes?
If you cursor is in the edit summary box when you press enter, the edit finalises. I have, too often, caught the enter key when reaching for the backspace and saved a change I'm part way through. Is there a user setting to turn off this behaviour? (More than happy to be told it's a browser setting, rather than a WP setting, as long as I can turn it off) Little pob (talk) 13:47, 20 July 2022 (UTC)
I hadn't included the link above because that user already had a common.js file, but it's useful for others (now, and later for people looking through the archives), so I added it there too. MANdARAX XAЯAbИAM 14:10, 20 July 2022 (UTC)
Start a discussion on the talk page, and contact the editor who moved the article. It is not a technical problem, but rather a contested move which should be discussed. Donald Albury17:59, 19 July 2022 (UTC) Edited 18:01, 19 July 2022 (UTC)
@DanCherek: True. I never remember about it. @Donald Albury: Start a discussion? Contested move? His move is actually wrong (move without reason - look at article) so it should be reverted without question and then he can start discussion about it. I could understand if I was the editor who moved page. Eurohunter (talk) 20:25, 19 July 2022 (UTC)
- There I was told that I should report this here, as this is a CSS/configuration issue which must be solved on the specific WikiMedia server instance which is the Wikipedia (I use EN and DE).
- I hope that after the fix one can browse the Wikipedia with the Vector Skin's header on top and sticky headers stacking nicely below that. Thanks PutzfetzenORG (talk) 15:25, 20 July 2022 (UTC)
@PutzfetzenORG:, I'm assuming you are referring to the testing and development gadget "Make headers of tables display as long as the table is in view, i.e. "sticky" (requires Chrome v91+, Firefox v59+, or Safari)", correct? Gadgets in that section are not well supported. The original author for that one can be contacted at User talk:TheDJ. If this is preventing you from viewing or editing pages right now, please disable it. If anyone has specific fixes to propose for it, please drop an edit request at MediaWiki talk:Gadget-StickyTableHeaders.css. — xaosfluxTalk15:34, 20 July 2022 (UTC)
The text buildingterror.blogspot.com. The script runs in more places than just references to catch issues like in works cited lists. Izno (talk) 16:55, 20 July 2022 (UTC)