Jump to content

Wikipedia:Village pump (technical): Difference between revisions

From Wikipedia, the free encyclopedia
Browse history interactively
← Previous editNext edit →
Content deleted Content added
VisualWikitext
M.Sakhaie (talk | contribs)
517 edits
No edit summary
Line 522: Line 522:
::::{{ping|Gadget850}} Those stats are quite fine, but quite dated: "User preferences statistics; data as of 23:12, 19 June 2014 (UTC)." Is there a way to update them, or get a month-by-month graph? [[User:LawrencePrincipe|LawrencePrincipe]] ([[User talk:LawrencePrincipe|talk]]) 02:01, 14 February 2015 (UTC)
::::{{ping|Gadget850}} Those stats are quite fine, but quite dated: "User preferences statistics; data as of 23:12, 19 June 2014 (UTC)." Is there a way to update them, or get a month-by-month graph? [[User:LawrencePrincipe|LawrencePrincipe]] ([[User talk:LawrencePrincipe|talk]]) 02:01, 14 February 2015 (UTC)
:::::Looks like this was run on the old Toolserver. I don't know of a replacement. --<span style="color:Turquoise">''''' &nbsp;[[User:Gadget850|Gadget850]]'''''<sup>[[User talk:Gadget850|&nbsp;''talk'']]</sup></span> 02:23, 14 February 2015 (UTC)
:::::Looks like this was run on the old Toolserver. I don't know of a replacement. --<span style="color:Turquoise">''''' &nbsp;[[User:Gadget850|Gadget850]]'''''<sup>[[User talk:Gadget850|&nbsp;''talk'']]</sup></span> 02:23, 14 February 2015 (UTC)

== signature ==
Why I can not sign correctly? does anybody know? M.Sakhaie 11:29, 14 February 2015 (UTC)

Revision as of 11:29, 14 February 2015

 Policy Technical Proposals Idea lab WMF Miscellaneous 
Shortcuts
The technical section of the village pump is used to discuss technical issues about Wikipedia. Bugs and feature requests should be made in Phabricator (see how to report a bug). Bugs with security implications should be reported differently (see how to report security bugs).

Newcomers to the technical village pump are encouraged to read these guidelines prior to posting here. Questions about MediaWiki in general should be posted at the MediaWiki support desk.


Lua module for date handling

Could someone please make (or point me to it, of one exists already) a Lua module which takes a date in a variety of unambiguous formats, and outputs it in one of a few a standard formats, as selected by a switch parameter?

For example, it should accept dates in formats such as:

  • 01 February 2014
  • 1 Feb 2015
  • 1 Feb. 2015
  • 23/4/2014
  • 23-04-2014
  • 2014-04-23
  • 2014-04-01 <- note, not ambiguous, as ISO compliant
  • 1FEB2014
  • others tbc?

(all case insensitive, and allowing other permutations of "1" and "01", "4" and "04"); and output dates in these formats:

  • 1 February 2015
  • February 1st, 2015
  • 2015-02-01
  • [as three parameters, YYYY, MM and DD]

It should reject dates in the format:

  • 01-02-2015
  • 02-01-2015
  • 1 Feb 15

as ambiguous, and should reject invalid dates such as:

  • 29 February 2015
  • 31 September 2015

I would like this module to be available for use in a number of other templates. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:50, 3 February 2015 (UTC)[reply]

mw:Help:Extension:ParserFunctions##time is pretty advanced and uses an existing well-known PHP function. I'm not sure it's a good idea to make an alternative. PrimeHunter (talk) 14:20, 3 February 2015 (UTC)[reply]
{{#time:}} seems to accept almost all of these and output what you want.
  • 01 February 2014 → 2014-02-01 = 1 February 2014
  • 1 Feb 2015 → 2015-02-01
  • 1 Feb. 2015 → 2015-02-01
  • 23/4/2014 → Error: Invalid time. (This one fails as "Invalid time", presumably this is parsed as M/D/Y, which makes it invalid. Clearly, the format is ambiguous.)
  • 23-04-2014 → 2014-04-23
  • 2014-04-23 → 2014-04-23
  • 2014-04-01 → 2014-04-01
  • 1FEB2014 → 2014-02-01
It will, however, also accept your ambiguous and invalid examples, and try to do its best to parse them.
  • 01-02-2015 → 2015-02-01
  • 02-01-2015 → 2015-01-02
  • 1 Feb 15 → 2015-02-01
  • 29 February 2015 → 2015-03-01
  • 31 September 2015 → 2015-10-01 (these two are really fun, heh)
I agree that it would be better to stick to it rather than create an alternative. As the one invalid "good" example you've given shows, it's really hard to do this well, and #time does an okay job. Matma Rex talk 15:09, 3 February 2015 (UTC)[reply]
The extension parser function #time can silently "correct" instances of dates stated in the Julian calendar. For example, 29 February 1900 (which was observed in Greece and Russia) would be rendered as 1 March 1900. This doesn't seem like the kind of function you would want to hide under the covers for use by editors who are potentially unaware of the subtleties of calendars. I would think a function that screams bloody murder when fed a date it considers invalid would be a better choice. Jc3s5h (talk) 15:38, 3 February 2015 (UTC)[reply]
Quite. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:39, 3 February 2015 (UTC)[reply]
How would we get the multi-parameter output? (Assuming that 01-02-2015 is 2015-02-01 is very, very dumb.) Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:39, 3 February 2015 (UTC)[reply]
@Matma Rex: Did you see the above question? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:48, 10 February 2015 (UTC)[reply]
Module:Citation/CS1 should have most of this functionality. --  Gadget850 talk 06:13, 4 February 2015 (UTC)[reply]
@Gadget850: I asked there, but there was no indication that that was the case, and little inetrest in implementing such ficntionality. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:48, 10 February 2015 (UTC)[reply]
@Pigsonthewing: Um, what is "multi-parameter output"? Matma Rex talk 16:03, 10 February 2015 (UTC)[reply]
From my initial post: "[as three parameters, YYYY, MM and DD]". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:21, 10 February 2015 (UTC)[reply]
@Pigsonthewing: Ah. You can just produce each one separately, like this (look at source wikicode): year=2014, month=04, day=01. Matma Rex talk 15:11, 11 February 2015 (UTC)[reply]
Thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:40, 11 February 2015 (UTC)[reply]

Past discussion

Something similar has come up before, {{Cite web/auto}}, The point of that was to fix various errors in citation templates. But there was no support for it; other than the author people thought it better to have the template report problems rather than fix them silently, otherwise problems would persist for much longer. Although these aren't strictly errors I think the same applies: better to flag them and have editors fix them, checking sources for e.g. ambiguous or clearly wrong (31 September etc.) dates, rather than have software guess, possibly get it wrong, and hide the error so other editors have less chance of fixing it.--JohnBlackburnewordsdeeds 16:22, 10 February 2015 (UTC)[reply]

The above request isn't about citations, and it's not about guessing I specifically referred to unambiguous dates). Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 23:52, 10 February 2015 (UTC)[reply]
I was bringing it up as something similar that has been considered, in particular for citations templates which you enquired about yourself, though it is unclear which templates this is for so how relevant. I still think the same applies; flag it as incorrect and get the editor to fix it, ideally as they preview it so before they hit save. Don't silently 'fix' it its in the correct format but the source is still non-standard.--JohnBlackburnewordsdeeds 00:39, 11 February 2015 (UTC)[reply]
My two-penn'orth: I'd like to see {{date}} extended so that it emits date in user preference. I.e. a render-time. N.B. 'preference' to include user language, so {{date|13 feb 2015}} would give 13 fév 2015 for a French speaker, etc. -- Unbuttered parsnip (talk) mytime= Fri 08:47, wikitime= 00:47, 13 February 2015 (UTC)[reply]
I think French is a bit too far – the French WP is over there – but even within English there are different preferences, and differences between British and American practices, The problem though is this could cause problems, from the subtle to the serious, as e.g. one editor adds dates to a table or infobox, say, with one preference and it's viewed by a user with different preferences. It might just look untidy, it might be too wide and end up wrapping, making rows of the table twice as high perhaps. The second user might then fix it, but it might then look wrong, even broken, to the first or another editor who reverts or repairs it. And so on. Such disputes would be hard to resolve as different editors would be seeing different things, even with the same browser and OS.--JohnBlackburnewordsdeeds 01:53, 13 February 2015 (UTC)[reply]

Automatically substituting Template:Unsigned and friends

Would anyone object to automatic substitution of Template:Unsigned and other similar signature templates? There are almost 30,000 transclusions of them, so it would be better to make sure there is a consensus to do it before letting a bot loose on them. Please see the discussion at Template talk:Unsigned and the corresponding edit request here. — Mr. Stradivarius ♪ talk ♪ 02:47, 4 February 2015 (UTC)[reply]

  • Yes, I object. This request not only assumes bad faith on the part of an editor who may have simply forgot or been unable to sign ( the ~ isn't available on all mobile devices and copy and past if you have edittools or charinsert isn't always easy ) it is also a change to the content of a discussion that isn't visibly apparent in the rendering and fails COSMETICBOT. I've commented as such and closed the edit request until there is a consensus which should've been obtained first. — {{U|Technical 13}} (e • t • c) 03:21, 4 February 2015 (UTC)[reply]
    How does it assume bad faith? The templates are already there, so I don't see how substituting them makes any kind of statement about the editors who made the posts. ekips39 05:59, 4 February 2015 (UTC)[reply]

I thought there was a consensus for this. I put a whole bunch of unsigned templates on a talk page, then tried to get bots to archive, and Clubot refused to archive the transcluded unsigned templates (thus, this is a real problem and not just cosmetic). When I posted about it at the bot's talk page, I was told that the bots behavior is intentional since the unsigned template should always be substituted, and the user then went the the talk page and substituted them all. So, to fix this, I put it in this bot's category, but the bot doesn't do it if there's over 100 transclusions. Unsigned has over 20,000, so it would be way too much work for a human. Oiyarbepsy (talk) 03:20, 4 February 2015 (UTC)[reply]

Does anyone know the rationale for (a) us wanting all these occurrences to be substituted, (b) the bot not wanting to perform too many substitutions at once? Is the latter due to fear of temporarily overloading the system, though the end result is still desirable? EdJohnston (talk) 04:25, 4 February 2015 (UTC)[reply]
I'm not aware of any special rationale for (a), although with this kind of template the reason for substing is usually to make sure that users are guaranteed to see the message as it was originally sent. (If the message is not substed, and someone changes the template after the message is sent, it could look completely different to what the sender intended.) As for (b), this is to prevent vandals from maliciously substing templates with thousands of transclusions, as reverting all the edits would take a disproportionate amount of time and effort. The edits themselves don't put much strain on the system; they are (I think) done at the rate of 6-10 a minute, a similar rate to other bots. This means that while the system load wouldn't be very high, a run of 30,000 substitutions would take a few days to complete. — Mr. Stradivarius ♪ talk ♪ 06:31, 4 February 2015 (UTC)[reply]
SmackBot, which was operated by Rich Farmbrough, used to do this work; here's an example. Graham87 08:32, 4 February 2015 (UTC)[reply]
As far as (a) goes, I assume this to be the dating; a transcluded template containing the current date will always contain the (now) current time (it doesn't, off course, because of bugs/reasons, but it does after an action=purge), while the substituted template will have the time of substitution. Martijn Hoekstra (talk) 13:16, 4 February 2015 (UTC)[reply]
I'd thought (a) in this case was mainly to make pages render faster for the server for templates that would be transcluded probably hundreds of thousands of times without subst:, especially if User:SineBot didn't add them substituted. These templates were almost added to Anomie's auto subst list in 2012; only a lack of interested admins at the time prevented it. ‑‑xensyriaT 15:13, 4 February 2015 (UTC)[reply]
It's a pretty minor issue either way. But I would be perfectly willing to restart this task, should circumstances permit. All the best: Rich Farmbrough, 18:34, 4 February 2015 (UTC).

Should unsigned templates be substituted

Sounds like we need an answer to the basic question here. I see three options:

  • Substitute All unsigned templates should be substituted. Bots should substitute any transcluded versions, so these templates need to be added to Anomiebot's list.
  • Transclude All unsigned templates should be transcluded. Archiving bots will needs to be updated to ensure they can read a transcluded unsigned template
  • Who cares? It doesn't matter. We still would need to change the archive bots, though.

Hopefully we can get this resolved. Oiyarbepsy (talk) 01:43, 5 February 2015 (UTC)[reply]

  • Substitute, SineBot substitute them, Anomie's unsigned script substitute them, and there is no difference on substituting and transcluding. I don't think making archivebots to read {{Unsigned}} is good idea. — Revi 04:38, 5 February 2015 (UTC)[reply]
  • Substitute, no point in transcluding as autosigs never need updating. Transcluding only adds to transclusion depth thus slows down rendering, especially on archived talk pages. (also fixed wrong word.) -- [[User:Edokter]] {{talk}} 12:13, 5 February 2015 (UTC)[reply]
  • Transclude at least for 7 days before substituting. Immediate substituting implies bad faith on the part of the OP and they should have a chance to go back and remove the template adding their own three tilde signature. Good faith new users may see the template being substituted as a permanent red mark on their editing. This is different than how sinebot does it because sinebot leaves a notice on the user's talk page explaining the situation and teaching the user how to fix the situation. Alternatively this is another case of the perennial Wikipedia talk:Signatures#Automatically sign talk posts which has been rejected by the community multiple times. — {{U|Technical 13}} (e • t • c) 14:20, 5 February 2015 (UTC)[reply]
    Technical, I'm just not buying the bad faith argument. If there is an element of bad faith, it's not in the bot that substitutes the template, but in the human user that places it in the first place (since our signing bots substitute it). And having one bot transclude it and another bot substitute 7 days later seem entirely ridiculous to me. Now, on the other hand, if you want the signing bot to wait before signing, that addresses your argument, but we're not talking about signing posts, we're talking about modifying posts already signed. Oiyarbepsy (talk) 15:30, 5 February 2015 (UTC)[reply]
    If you're really worried about making the editor unhappy, then I think the more important point would be to change the bot signature to say something like <!-- auto signed --> [[User:Example|Regular signature goes here]] rather than the current display, which says Ha, ha! User:Example doesn't know how to sign comments!. WhatamIdoing (talk) 00:22, 12 February 2015 (UTC)[reply]
  • Substitute. I don't see any need to include a delay as suggested immediately above; for one bot actions aren't instant, except for those bots reversing vandalism. Mostly it should be substituted when first added anyway. Getting a bot to do it gives the OP more time to correct it. Not that they can't easily correct the final substituted version, by editing the HTML or undoing (e.g.) AnomieBot's action before replacing the template.--JohnBlackburnewordsdeeds 14:40, 5 February 2015 (UTC)[reply]
    It's a lot harder for new users to figure out how to replace substituted HTML then a template. I've seen it done and it has resulted in half the page being left with an open <small> tag but no close which breaks half the page. — {{U|Technical 13}} (e • t • c) 15:03, 5 February 2015 (UTC)[reply]
Then someone will fix it. Saying something should not be changed because a subsequent edit might break things is a recipe for making no changes at all, or at least no changes that introduce relatively complex HTML/Wikicode. So no adding CSS to tables for example.--JohnBlackburnewordsdeeds 15:25, 5 February 2015 (UTC)[reply]
  • Substitute: the exception being if parameters are left out because whoever added it hasn't bothered to look up the timestamp or whatnot (like {{Unsigned-unk}}; these should all be added to a hidden category if they aren't already). Would be simpler for bot design and debugging, and should lower server-load. Temporary transclusion for a fixed time before this isn't a bad idea though, and wouldn't be a problem for archive-bots or server-load. ‑‑xensyriaT 15:33, 5 February 2015 (UTC)[reply]
  • Substitute — I don't feel there's a truly compelling reason to transclude at this point. The template is unlikely to ever be significantly changed (and really hasn't been, historically), doesn't need to blend in with generalized UI changes, and doesn't need to dynamically add or remove categories (or update text based on dynamic conditions). On top of all of this, every bot, script, and tool that makes use of it would need to then account for the new format as well as the old, plus enforcing transclusion (i.e., transforming a substitution into a transclusion) in an automated way is a giant pain in the ass inevitably fraught with bugs. :P Overall, seems to be an unnecessary hassle for more people than it's hypothetically—and I'm not sure how, hypothetically—it's supposed to help in any significant way. --slakr\ talk / 04:25, 6 February 2015 (UTC)[reply]
  • Transclude A transclusion creates shorter wikicode, making it easier to read the source. --Stefan2 (talk) 11:36, 6 February 2015 (UTC)[reply]
  • Substitute: no legitimate reason to change the signature after signing; reduces server load and prevents any future tampering. BethNaught (talk) 11:41, 6 February 2015 (UTC)[reply]
    • Exactly, it prevents tampering in a way that suggests to new editors that they are not allowed fix it to what their actual signature is and in doing so assumes bad faith that editors are all of a sudden intentionally not posting their signature and fully automated signatures for all posts need to be done to prevent increased disruption to the encyclopedia. — {{U|Technical 13}} (e • t • c) 11:49, 6 February 2015 (UTC)[reply]
You know very well that I meant it prevented tampering with the template as opposed to the signature. If a new editor knows how to sign, I would have thought they would know how to resign. The template is not about assuming bad faith, it's about being helpful when people accidentally forget. I really can't understand why anyone would take offence at a bot signing when they had forgotten. Your deduction that editors are all of a sudden intentionally not posting their signature and fully automated signatures for all posts need to be done to prevent increased disruption to the encyclopedia simply doesn't logically follow. Also, my name is not Beth. BethNaught (talk) 11:54, 6 February 2015 (UTC)[reply]

Echo notificiations

Tracked in Phabricator
Task T59684

I've got a red "1" next to my name, but I don't have any new notifications. I clicked on all of the ones listed just to try to clear it but it won't go away.--v/r - TP 18:04, 5 February 2015 (UTC)[reply]

Even if it doesn't go after a WP:BYPASS, it should go away in a little while. Just be patient. I think we've all had this problem at some point. Jared Preston (talk) 18:28, 5 February 2015 (UTC)[reply]
@TParis: Note there may be two tabs. When you click the 1, you should see either just "Alerts" or both "Alerts" and "Messages" (the latter is used for Flow). If you see "Messages", make sure you click that to check any notifications in that tab (in the "Messages" tab, there is also an X to "Mark as read" (if visiting doesn't work for some reason) and a "Mark all as read". Mattflaschen (WMF) (talk) 00:56, 11 February 2015 (UTC)[reply]
I look at both and everything was read. No matter, though. Someone else pinged me, I got a "2", read the new one, and it cleared.--v/r - TP 02:18, 11 February 2015 (UTC)[reply]

I seem to be getting selective echo "mentions" on user talk pages, perhaps 50 percent of the time, with the others disappearing into the ether. Does anyone know what's going on? Viriditas (talk) 02:22, 12 February 2015 (UTC)[reply]

@Viriditas: See recent posts by Quiddity (WMF) (talk · contribs) at WT:Echo (also at Template talk:Reply to and Template talk:User link). Have these selective mentions been happening for some time, or did they begin yesterday? --Redrose64 (talk) 12:17, 12 February 2015 (UTC)[reply]
@Redrose64: I just looked at the last two pings I didn't receive, and in both cases the editor did not sign their posts. Viriditas (talk) 19:31, 12 February 2015 (UTC)[reply]
Well, that would certainly explain it. A valid signature must be added with the same edit that the link to your user page is added. --Redrose64 (talk) 19:33, 12 February 2015 (UTC)[reply]
@Redrose64: Thanks for helping me isolate the problem. Since there are still editors that are still not up to speed on Echo, would you be in favor of advertising the simple fact that one must sign and ping in the same edit to the wider community, possibly through a sitewide watchlist message or banner notice? Viriditas (talk) 02:52, 13 February 2015 (UTC)[reply]
Or can that be fixed? Not being able to fix a typo isn't ideal. Sarah (SV) (talk) 04:41, 13 February 2015 (UTC)[reply]
The requirement to sign on the same post as the user link is already described, at WP:Echo, also {{replyto}} (other places too). I can't force people to read documentation.
Not being able to fix a typo is a known problem with Notifications. --Redrose64 (talk) 09:25, 13 February 2015 (UTC)[reply]
Question: should we expect our users to read documentation on every feature? Or should we make them more intuitive with some kind of error checking? Viriditas (talk) 10:32, 13 February 2015 (UTC)[reply]

@Redrose64: It looks like the problem is still unresolved. Here's another recent ping I never got: [1] And, another one: [2] Can you look at those diffs to see if anything went wrong? Viriditas (talk) 03:13, 13 February 2015 (UTC)[reply]

The first one possibly failed because it has {{ping|User:Viriditas}} instead of {{ping|Viriditas}}. The second one possibly failed because a subheading was added. --Redrose64 (talk) 09:25, 13 February 2015 (UTC)[reply]
Thanks! Viriditas (talk) 10:32, 13 February 2015 (UTC)[reply]

Visible markup on category pages

Category pages are now showing visible markup <a name="Pages_in_category" id="Pages_in_category"></a> immediately before the section heading 'Pages in category "..."'. This comes from MediaWiki:Category header, which hasn't changed in over eight years. Do other people see the problem? If so, it must be a recent bug in the MediaWiki software, since it wasn't doing that this morning. --Redrose64 (talk) 20:59, 5 February 2015 (UTC)[reply]

Yes, it looks like the HTML is being escaped now. Alakzi (talk) 21:03, 5 February 2015 (UTC)[reply]
I see it too. HandsomeFella (talk) 21:09, 5 February 2015 (UTC)[reply]
But <a> is not whitelisted through Sanitizer.php so it should always show. Deleting the page would return it to the default Pages in category "$1". --  Gadget850 talk 21:10, 5 February 2015 (UTC)[reply]
The real question is why it never showed before? --  Gadget850 talk 21:25, 5 February 2015 (UTC)[reply]
@Gadget850: There are several pages in MediaWiki space that contain HTML elements that are not whitelisted, including <a>...</a>. The markup doesn't show up (normally). --Redrose64 (talk) 23:34, 5 February 2015 (UTC)[reply]
I'm guessing that was added for a specific gadget or JavaScript. It's from far before my time here, does anyone know why it was added exactly or what script makes use of it? I'm guessing it is used as a marker or container for such a script to add something to the page. I'm guessing it was never seen before because the script hid it or it was ignored since it is an empty link element. — {{U|Technical 13}} (e • t • c) 21:14, 5 February 2015 (UTC)[reply]
Pinging Mets501 as the editor that added it. — {{U|Technical 13}} (e • t • c) 21:16, 5 February 2015 (UTC)[reply]
@Technical 13: It is valid HTML to leave the <a>...</a> element empty, provided that it has a name= attribute. This attribute (introduced right at the start) is used as the anchor for a URL that has a fragment. Since HTML 4 introduced the id= attribute, which may be applied to any element and can also serve as an anchor, the need for empty <a>...</a> elements has dropped significantly, except of course for old browsers that don't recognise anything later than HTML 3.2. The name= attribute is marked as "obsolete but conforming" in HTML5. --Redrose64 (talk) 23:34, 5 February 2015 (UTC)[reply]
  • I wasn't saying it was an invalid anchor or link, I was just saying that it appeared to be something that was injected as a way to find that specific spot by a script and I was wondering what snippet of code made use of that anchor. — {{U|Technical 13}} (e • t • c) 23:41, 5 February 2015 (UTC)[reply]

Today's outage may have changed something. There was a brief period that unvisited pages were bolded in my watchlist (rather than the hard-to-see green bullet). --NE2 21:28, 5 February 2015 (UTC)[reply]

It's an unintended side effect of the weekly software update. I'm taking care of it. Jackmcbarn (talk) 21:33, 5 February 2015 (UTC)[reply]
Another example (from an old HTML dog).
::<a name="Pages_in_category" id="Pages_in_category"></a>Pages in category "Roman Catholic Archdiocese of Anchorage" ::
Seeing an extra > before the closing </a> .
Regards, JoeHebda (talk) 21:39, 5 February 2015 (UTC)[reply]
Only partially fixed. See Category:Women's studies journal stubs. Vegaswikian (talk) 21:46, 5 February 2015 (UTC)[reply]
We all knew that this house of cards PHP jenga was gonna come tumbling down one day. ;-) Alakzi (talk) 21:54, 5 February 2015 (UTC)[reply]
I knew. PrimeHunter (talk) 22:08, 5 February 2015 (UTC)[reply]

This should be completely fixed now. Jackmcbarn (talk) 00:16, 6 February 2015 (UTC)[reply]

The real question is why was this not showing before? <a> is not whitelisted through Sanitizer.php so it should have always been exposed. We don't even have to nowiki the markup in question: <a name="Pages_in_category" id="Pages_in_category"></a>. --  Gadget850 talk 11:51, 7 February 2015 (UTC)[reply]

It showed on the MediaWiki page itself [3] but many MediaWiki messages allow <a> when they are used. A search on href for example gives MediaWiki:Wikimedia-copyright which is displayed at the bottom of all pages and has working links there. PrimeHunter (talk) 13:48, 7 February 2015 (UTC)[reply]
My guess is that not all MediaWiki: pages are transcluded at the same time. If this transclusion happens at a late stage, it might be that the sanitiser has already run, and possibly the Wikimarkup parser too, which leaves no choice but to use raw HTML. I count something like 40 system messages with one or more <a>...</a> elements. Of customised messages, we have MediaWiki:Centralauth-login-progress, MediaWiki:Googlesearch, MediaWiki:History copyright, MediaWiki:Readonly lag, and as noted above, MediaWiki:Wikimedia-copyright. --Redrose64 (talk) 15:44, 7 February 2015 (UTC)[reply]
Not a timing issue; some system messages simply never go through the sanytizer because they do not go through the parser. -- [[User:Edokter]] {{talk}} 16:43, 7 February 2015 (UTC)[reply]
Starting to make sense. @Edokter: what does "some" mean here? How would I know if an interface page went through the parser? --  Gadget850 talk 20:57, 7 February 2015 (UTC)[reply]
It depends on what part of MediaWiki is actually using the messages. Messages used in the user interface, like MediaWiki:Copyright, are usually not parsed, because wiki markup has no relevance outside content space. Such messages let HTML pass unobstructed. The same goes for messages used by special pages. Some people see this as a security risk (because admins can use raw HTML there). There is a bug at phab:T85864 listing those special pages. -- [[User:Edokter]] {{talk}} 21:51, 7 February 2015 (UTC)[reply]
They all should go through the parser (or just be HTML-escaped, but that obviously limits customization). If any don't, that's a bug. The bug for MediaWiki:Copyright is T45646. Matma Rex talk 16:21, 9 February 2015 (UTC)[reply]

http://stats.grok.se partial data

Both January 27 and February 4 only show partial data for pageviews. Both dates compiled only articles up to some point in the alphabet between Royce White and Tim Hardaway, Jr.. All articles after this point in the alphabet have no pageview data for these dates. Recall that December 31, 2013 stopped between TNZ and TO and was not compiled completely for several months thereafter.--TonyTheTiger (T / C / WP:FOUR / WP:CHICAGO / WP:WAWARD) 21:31, 5 February 2015 (UTC)[reply]

Currently, February 5 has compiled to somewhere between Emily Ratajkowski and Frank Underwood (House of Cards).--TonyTheTiger (T / C / WP:FOUR / WP:CHICAGO / WP:WAWARD) 05:40, 6 February 2015 (UTC)[reply]
No Page view statistics at all are available for February 6, and Henrik's talkpage is pretty much deserted Ottawahitech (talk) 14:32, 7 February 2015 (UTC)[reply]
2/7 did almost none of the pageviews. It stopped somewhere between Aardvark and Anthony Davis (basketball).--TonyTheTiger (T / C / WP:FOUR / WP:CHICAGO / WP:WAWARD) 09:07, 8 February 2015 (UTC)[reply]
It appears that the program follows the alphabet not only in the articles but also in the Wikipedias it serves. According to this posting the Arab Wikipedia's Page view statistics are still working while the English Wikipedia's are not. Ottawahitech (talk) 14:37, 9 February 2015 (UTC)[reply]
Is the source data from Wikimedia complete? It's best to narrow down the offender. It'd be nice if the Foundation could follow through and actually make their own page view statistics in a readable format rather than relying on a website that frequently breaks. Killiondude (talk) 00:27, 10 February 2015 (UTC)[reply]
You may be thinking of WMF Labs "Wikiviewstats", which has been in this condition since last October: Noyster (talk), 16:41, 10 February 2015 (UTC)[reply]
The source data from [4] appears to be complete. Mr.Z-man 05:06, 11 February 2015 (UTC)[reply]

The following is a better summary of the problem dates. The following shows the last page I quickly found with seemingly full stats and the first page that I did not see stats for.

Jan 27: Royce White and Tim Hardaway, Jr.
Feb 4: Royce White and Tim Hardaway, Jr.
Feb 5: Emily Ratajkowski and Frank Underwood (House of Cards)
Feb 6: None (before Aardvark)
Feb 7: Aardvark and Anthony Davis (basketball)
Feb 8: None (before Aardvark)
Feb 9: Campbell's Soup Cans and Cloud Gate
Feb 10: None (before Aardvark)
Feb 12: None (before Aardvark)
Feb 13: None (before Aardvark)

I continue to assume the English WP pageview stats are compiled alphabetically and base the above information on that fact.--TonyTheTiger (T / C / WP:FOUR / WP:CHICAGO / WP:WAWARD) 06:42, 10 February 2015 (UTC)[reply]

In Polish Wikipedia stats are down from 2015-02-02. --Swd (talk) 07:47, 10 February 2015 (UTC)[reply]

It's up again for 11 February now. While I saw a graph yesterday where only 6th, 8th and 10th February were missing, the graph for the same page today shows a gap from 5 thru 11 February... Does anyone know if the missing data will be restored? --78.53.66.106 (talk) 21:29, 12 February 2015 (UTC) (PS ping @Jackson665: as you asked there :)[reply]

No data?

I've been checking both the English and Japanese Wikipedias, and essentially, data for February 4 and 5 is missing. What happened? Narutolovehinata5 tccsdnew 13:49, 6 February 2015 (UTC)[reply]

I use the service regularly for it.wiki and I've noticed some lack of data on January, the 27th, but in the previous months it has been performing quite well. It did show some interruptions in the past, but not recently.--Alexmar983 (talk) 21:26, 10 February 2015 (UTC)[reply]
I reported this earlier after seeing Stand Up, Stand Up for Jesus apparently not have any views for the 27th despite being on DYK. But when I reported it, no-one appeared interested. I would like to know why there is no data and why WikiViewStats are down too. The C of E God Save the Queen! (talk) 09:54, 11 February 2015 (UTC)[reply]

Anyone interested?

In collaborating on a new, stable, pageview system? grok.se and wikiviewstats are both maintained by single users who haven't been active in months. I don't have the time or interest to do it all myself (plus, we obviously want to avoid having a single person running important systems like this), but I have some experience (and already-written code) from popularpages. So if 2 or 3 other people are interested, I'd also be in. Mr.Z-man 05:06, 11 February 2015 (UTC)[reply]

I'm not sure if this is still a plan, but last October, the WMF said that they were trying to build a new page view system: "Negrin told [the Signpost] that they are aware of the problem [with grok.se] and are currently working to replace the current apparatus with a "modern, scalable system," which will come out in a preliminary form next quarter." Now, my memory could be wrong and the WMF was referring to something else ... but that's one reason we have pings. TNegrin (WMF) Ed [talk] [majestic titan] 18:09, 11 February 2015 (UTC)[reply]
What about https://stats.wikimedia.org/ ? Anyway there could be a replacement needed, as @Henrik: seems to be on the leave? (Unfortunately, I am no programmer, even not when I log in ;) PS. stats.grok.se says: "This is very much a beta service and may disappear or change at any time." :( --78.53.66.106 (talk) 21:54, 12 February 2015 (UTC)[reply]

Working again?

Ha. Looks like it may be working again. Pkeets (talk) 02:35, 14 February 2015 (UTC)[reply]

All I'm getting is "Internal server error". Squinge (talk) 10:38, 14 February 2015 (UTC)[reply]

Wikipedia Library Minimum Criteria

Hi Folks: The Wikipedia Library has some pretty stringent requirements for access: a 1-year old account and 1000 edits.

I want to learn many potential users we are 'losing' with this scheme. Could someone create a query that looks for the number of editors active in the last year with at least a 1 year-old account and at least:

  • a) 250 edits
  • b) 500 edits
  • c) 1000 edits?

This would ultimately help us get more accounts to more editors, if we can make the case that we are missing a substantial number of potentially qualified recipients! Cheers, Jake Ocaasi t | c 11:46, 6 February 2015 (UTC)[reply]

Hi Jake. First pass - 37940 users on enwiki registered before 2014 and over 1000 edits; 60302 with over 500; 98153 with over 250. However (big caveat) not all of these are still active - the first one I checked was indefinitely blocked, and many more will have made their thousand edits and vanished by 2007. I'll try and figure out the best way to check for # of reasonably active users - has edited in the last month, say? Andrew Gray (talk) 23:35, 6 February 2015 (UTC)[reply]
Hey Andrew. Thanks! Yeah, I think active in last month, or even last year would give a pretty good sense of proportion either way. So if you can do last month, awesome... cheers, Jake Ocaasi t | c 19:36, 7 February 2015 (UTC)[reply]
Not been having much luck with this, I'm afraid. There's a "last touched" field in the user database, which would be exactly what we need, but (presumably for privacy reasons) it's either blank or restricted to internal WMF visibility only. Andrew Gray (talk) 21:12, 9 February 2015 (UTC)[reply]

Change in blocking reasons

I've just blocked someone spamublock, and found the blocking reasons have been changed in order. That's no real problem, but the reason shown on the red banner of blocking (and in the drop-down list to select from) is now " ({{uw-spamublock}} <!-- Promotional username, bad faith -->) " instead of "({{spamusernameblock}}) ". Is this supposed to be an advance? These spamublocks aren't always 'bad faith' in the sense that vandalism is. They are often through ignorance and not through deliberate attack on Wikipedia. For a now example, see User:Sumtrix. and a before at User:Mkn000. Peridon (talk) 11:48, 7 February 2015 (UTC)[reply]

I've posted a message at AN as well to send people here. Peridon (talk) 11:53, 7 February 2015 (UTC)[reply]
If there isn't sign of bad faith, then {{spamusernameblock}} isn't appropriate. You'd want {{softerblock}}, {{uw-softestblock}}, or perhaps {{causeblock}} instead. —Cryptic 12:30, 7 February 2015 (UTC)[reply]
Soft block is for an organisational name, not for advertising, as I see it. I'm talking about the ones who genuinely haven't realised that they can't advertise here like they can on social media, or haven't realised how promotional their post was - and have an inappropriate name. This change is accusing all the spamublock ones of bad faith. Has there been a discussion on this, and if so where? Or has an intended non-displaying comment <!-- Promotional username, bad faith --> become displayed on the red banner? That looks like comment coding, but it's displaying on the banner. Peridon (talk) 12:46, 7 February 2015 (UTC)[reply]
You have a better AGFometer than me, I suppose.

I've tested the new block message; the comments appear in the block log, but not in the message that shows up when you try to edit, so I don't see how it's any worse than the template name including the word "spam". —Cryptic 13:00, 7 February 2015 (UTC)[reply]

If you look in the Sumtrix link above, it's showing there - for me, at least, in Monobook and signed out in Vector. Peridon (talk) 13:57, 7 February 2015 (UTC)[reply]
I'd prefer a move to AN rather than MediaWiki - there'll be more participation over here and it's an en-wiki problem not an MW one. I still think the comment might not be supposed to display outside the list, and that's probably technical. Peridon (talk) 20:30, 7 February 2015 (UTC)[reply]
@Peridon: MediaWiki talk:Ipbreason-dropdown is a page on en-wiki. It's the talk page for MediaWiki:Ipbreason-dropdown, where all those block reasons are listed. Pages on MW begin mw:, as in mw:MediaWiki:Ipbreason-dropdown. --Redrose64 (talk) 20:53, 7 February 2015 (UTC)[reply]
Thanks for that. I hate going to other places that don't show up on my contribs list. I lose track of them... Peridon (talk) 21:22, 7 February 2015 (UTC)[reply]

Well since the conversation is continuing here, I guess I'll reply here. I suggested this change on January 26 at this very village pump, now archived here. There didn't seem to be any great uproar at the idea. I took the "additional verbiage" as Ponyo describes it directly from the descriptions at WP:UTN#Blocks. For example, {{uw-spamublock}} is the appropriate message for a bad-faith promotional username block and has been since 2008 when that template was created. (Ironically, that was the template I used in my original suggestion as an example of what would be changed.) If the majority agrees that the additional verbiage is not required, that's great; I only included those to assist users that are somehow not familiar with the correct template names, so they can find the appropriate one to use. I'm happy to hear that somebody agrees that people should know the correct templates and their usage by now. --Geniac (talk) 02:48, 8 February 2015 (UTC) Addendum: I'm happy to see other users successfully choosing the updated block reasons without too much of a problem: [5], [6], [7], [8], [9] --Geniac (talk) 03:01, 8 February 2015 (UTC)[reply]

If the template were sanely named, it wouldn't need the additional verbiage. "uw-spamublock" is almost completely meaningless, like most of what WP:UW has come up with; the redirect from {{spamusernameblock}} is actually readable by humans. —Cryptic 08:37, 8 February 2015 (UTC)[reply]
What the templates are named are what the templates are named. If you wish to for a change in that regard, I'd suggest proposing that at WT:UTM --Geniac (talk) 16:10, 8 February 2015 (UTC)[reply]
  • It's the wording appearing on the red banner that worries me, not what it says in the dropdown. Softerblock is OK for very minor promotionality, but for a full-blown ad there's only spamublock no matter whether they posted it as a deliberate attempt to flout our rules, or whether they were either under the impression that it was OK (as it is at Facebook) or thought that that was how it should be worded, being indoctrinated by PR wording seen everywhere. (An example in the UK in particular is a sign on a shop saying 'Car park to rear' because estate agents seem incapable of using the word 'at' even for things that are very immobile.) Until a revision of these classes is made (with possibly a reduction of size of spamublock), I don't thing the words 'bad faith' should be on the banner. They're fine as a comment on the dropdown because only admins see that. Peridon (talk) 12:51, 8 February 2015 (UTC)[reply]
BTW I for one of the admins who delete lots of stuff only come on this board when there's a problem. And when I do come here, I don't see the others I see frequently in CSD etc, unless they've also experienced the problem. As it's an admin thing rather than a technical thing (apart, of course, from making the change), I'd have thought AN a better place for the suggestion. Peridon (talk) 12:57, 8 February 2015 (UTC)[reply]
If I'd seen that heading here, I wouldn't have known what it was about. I also notice that only one of the two that commented was an admin. But if I had seen this, and looked in, I would have taken <-- Promotional username - Bad faith --> to be a comment line for admins to see. Peridon (talk) 13:08, 8 February 2015 (UTC)[reply]
How about this for a solution to both issues (1: users only familiar with the redirect template names and 2: the additional verbiage showing up in the block banner not just the dropdown)... the entries are changed to say something like {{uw-spamublock}} <!-- spamusernameblock -->. So issue 1 is solved; the redirect name some may be more familiar with is listed right there. And issue 2 is solved: no "bad faith" wording showing up in the block banner message. --Geniac (talk) 16:10, 8 February 2015 (UTC)[reply]
Looks fine to me. Thanks. @Ponyo: Peridon (talk) 18:30, 8 February 2015 (UTC)[reply]
I've made some changes along those lines ([10]). I'd probably still like it a little better without the comments, or at least if the comments somehow didn't show up in the actual block log, but I don't mind it too much now. --Bongwarrior (talk) 07:29, 9 February 2015 (UTC)[reply]
That looks OK too. Thanks. Peridon (talk) 09:17, 9 February 2015 (UTC)[reply]
I'm happy with those changes as well.--Jezebel's Ponyobons mots 16:31, 9 February 2015 (UTC)[reply]
I'm also happy with the "hard block/soft block" verbiage. Everybody fine? Things are cool? Ok, time for a nice hot cup of tea. --Geniac (talk) 01:59, 10 February 2015 (UTC)[reply]
Sorry to delay teatime, but could I suggest one more small change? For {{uw-softerblock}} and {{uw-spamublock}}, can we use a more generic term like "organization name" instead of "company" because these block templates can apply to other types of organizations (for example a band, musical group, etc.) rather than just a company. -- Ed (Edgar181) 13:15, 11 February 2015 (UTC)[reply]

I propose: Special:EditWatchlist can have a button that says & does: "Remove redlinks from watchlist". -DePiep (talk) — Preceding undated comment added 21:47, 7 February 2015 (UTC)[reply]

And while I am here: why not a link in my watchlist: "unwatch this page"? -DePiep (talk) 22:00, 7 February 2015 (UTC)[reply]
@DePiep: You can install this snippet of JavaScript in your common.js: mw:Snippets/Unwatch from watchlist. Helder 22:10, 7 February 2015 (UTC)[reply]
Should be wiki. Enough of these hacks I must maintain and that bug my screen ten years after. btw Do you know how I can get rid of that 0-subs "More" tab I inherited? What is more logic that those two button-links in a wikipage? -DePiep (talk) 22:16, 7 February 2015 (UTC)[reply]
Watch/unwatch is available in popups, more general solution than just unwatch from just watchlist. DMacks (talk) 09:11, 8 February 2015 (UTC)[reply]
OK. Time for me to accept modernity. Thx. -DePiep (talk) 16:09, 8 February 2015 (UTC)[reply]
There is an shorter bookmarklet here: c:User:Rillke/Code#Remove_red_links_and_Deletion_Requests_from_watchlist (you can also simply paste it in your console to execute).
PS: The script from Krinkle get hooked on every page load?? This feature is needed only all 2 months (he is good but I saw already some strange scripts from him) → User: Perhelion  12:08, 9 February 2015 (UTC)[reply]

API namespace issue / page reported existing in two namespaces

Tracked in Phabricator
Task T87645

In AWB we are doing an API query to list pages in Category:Book-Class Wikipedia-Books articles. Something odd is happening, we are getting results indicating that pages such as Book talk:Math are in mainspace. Digging into this, we run this query (seems you won't 'continue' to the right page, but you should get the idea). In the results returned we get:

<cm pageid="45227281" ns="0" title="Book talk:Math" />

and then in a later 'continue' query we get:

<cm pageid="43653252" ns="109" title="Book talk:Math" />

So can anybody please explain why the API says the same page is in two namspaces? Thanks Rjwilmsi 09:56, 8 February 2015 (UTC)[reply]

Looks like exactly the same thing that happened at Wikipedia:Help desk/Archives/2015 January 31#User:ClueBot III/Indices/Wikipedia talk:WikiProject Encyclopaedia Britannica. —Cryptic 10:50, 8 February 2015 (UTC)[reply]
Yes. See also phab:T87645, User talk:ClueBot Commons#Puzzle, User talk:Redrose64#Please help again, User talk:MZMcBride#Links to the userspace and User talk:ClueBot Commons#Indices listed at Special:ShortPages. --Redrose64 (talk) 13:15, 8 February 2015 (UTC)[reply]
Yeah, this is definitely part of a larger problem — I've listed, at User talk:JaGa#New toolserver problem, 37 titles in Template:, User:, Wikipedia: or User talk: space which, despite being located in internal project spaces, are being erroneously picked up as uncategorized articles by the Untagged Uncategorized Articles toolserver — which should never be picking up anything outside of mainspace. (The WikiProject Encyclopaedia Britannica example that Cryptic raised above is one of the titles in question.) JaGa did a database search, and found that many of the affected pages do have metadata files with the datum "page_namespace: 0" (meaning mainspace) in them — although he only checked the "Wikipedia:" titles and didn't specifically check the others, this is almost certainly the explanation for the others as well. But if he looks at the metadata files that are directly attached to the pages themselves, they're coded correctly — so in other words, there appears to have been a database corruption issue of some kind, in which some pages in project namespaces somehow ended up having duplicate metadata files ghosted into mainspace for some reason. That's how a page can be "in" two namespaces at once: there are two separate metadata files for the same page, with different namespace codes in them. This certainly needs to be corrected somehow — it's interfering with critical maintenance tools, and absolutely cannot be tolerated as a permanent speedbump that users of those tools should be forced to work around — but I'm not sure how to repair the problem directly. Bearcat (talk) 21:09, 9 February 2015 (UTC)[reply]
I should add as well that the list I gave to JaGa provides one very valuable clue that may help pinpoint the problem: one of the ghost titles is User talk:108.69.192.159, a user talk page which has somehow ended up with a metadata file even though it's not just a page that got deleted, but one which has never actually existed at all. The user's sole contribution was on January 27, so whatever caused this blip most likely happened on or soon after that date. Bearcat (talk) 21:27, 9 February 2015 (UTC)[reply]
To confirm (I guess) what you are saying, the API seems convinced that both of the above pages exist, the first in namespace 0, and the second in 109: API info. Johnuniq (talk) 22:25, 9 February 2015 (UTC)[reply]
Most (if not all) of this is already known, see that Phab ticket which I linked, also the Help desk page. From that, we can see that the first edit in error was at 08:15, 27 January 2015 and the last was at 12:08, 2 February 2015. --Redrose64 (talk) 22:54, 9 February 2015 (UTC)[reply]

Deprecating HTML code breaks

In a template, I changed align="center" into style="text-align:center" , being deprecated HTML code. However, in mobile view the new code does not center images any more. Can someone take a look? A demo is in User:DePiep/sandbox1, /sandbox4's are open for your edits. (Warning: old wikitable infobox construct ahead). -DePiep (talk) 16:07, 8 February 2015 (UTC)[reply]

Images aren't text. To center an image, use the |center option, see WP:EIS#Location --Redrose64 (talk) 16:57, 8 February 2015 (UTC)[reply]
Used |center=yes in module:InfoboxImage. Is that disadvised in a wikitable? -DePiep (talk) 18:00, 8 February 2015 (UTC)[reply]
No, should work. -- [[User:Edokter]] {{talk}} 11:47, 9 February 2015 (UTC)[reply]
Which template? Center is not that easy (as you can see on the deWP is an extra help-page for this) In De there is also an class centered, but default on Mediawiki is class="center". You can try it (it is a bit more than text-align:center). → User: Perhelion  11:58, 9 February 2015 (UTC)[reply]
class="center" didn't do the trick for me elsewhere, I trust that tidy will handle align="center" before the validator sees it. –Be..anyone (talk) 22:13, 9 February 2015 (UTC)[reply]
Tidy doesn't care about any attributes. -- [[User:Edokter]] {{talk}} 22:21, 9 February 2015 (UTC)[reply]
Tidy will not fix align as it is whitelisted through Sanitizer.php, thus it is still valid for Wikipedia. The center class should work, but it sets width: 100% so it will expands tables to full width which is usually not desired. See Wikipedia:HTML5#center for help. If I can get a specific example I am sure I can help. --  Gadget850 talk 22:24, 9 February 2015 (UTC)[reply]

Wikipedia pages getting hung up pretty frequently lately

I've been getting the following message pretty frequently on Wikipedia pages in the last week or so:

A script on this page may be busy, or it may have stopped responding. You can stop the script now, open the script in the debugger, or let the script continue.

Script: https://bits.wikimedia.org/en.wikipedia.org/load.php?debug=false&lang=en&modules=jquery%2Cmediawiki&only=scripts&skin=monobook&version=20150205T202925Z:4

Happens probably every 5th page I load, more or less. No other site besides WP. I'm 95% clueless on how all this works. From this message, is this enough info to tell if this is a problem on my end, or on WP's end? Is this something in my .js or .css file, which I can remove? Or do I have to live with it because it's not under my control?

I haven't changed my .js or .css files in months, and haven't (intentionally) changed or upgraded browsers in the last month or so. --Floquenbeam (talk) 16:45, 8 February 2015 (UTC)[reply]

Is your browser Firefox? --Redrose64 (talk) 16:58, 8 February 2015 (UTC)[reply]
Sorry, I know I'm supposed to say that. Yes, Firefox 35.0.1. --Floquenbeam (talk) 17:19, 8 February 2015 (UTC)[reply]
Me too, but I get the problem on lots of web pages; rarely Wikipedia. Newspaper websites are worst. It mostly seems to happen when I have lots of applications running, a slow connection and am trying to load a web page that is heavy on scripts or video content. It appears to trigger if the page fails to complete inside 60 seconds. Usually, if it's an advert I Stop script, but when waiting for text I Continue.
You could try clearing caches, and temporary internet files. --Redrose64 (talk) 18:02, 8 February 2015 (UTC)[reply]
You can try turning off the userscripts you are using to try to determine if one of those is causing the issue (you can find them on User:Floquenbeam/common.js and User:Floquenbeam/monobook.js (that is assuming you use monobook, which seems to be the case). If you remove them all, and turn them back on one by one, and the problem starts when you activate one, that's your issue. Martijn Hoekstra (talk) 20:31, 8 February 2015 (UTC)[reply]
Yes, I've been gettig this recently (Firefox 35.0.1) not just on WP but I've found just clicking "stop script" seems to leave the page properly displayed. Thincat (talk) 17:47, 9 February 2015 (UTC)[reply]
Thanks all. For me, it's only happening on WP, but I don't frequent other sites with a lot of scripts (well, that I know of). I get the page served up eventually, I'm just annoyed at the 30 second wait until I get the error message, when everything locks up and I can't do anything. I've followed Martijn's advice; if this solved the problem, I'll slowly add them back. Thanks again. --Floquenbeam (talk) 16:32, 10 February 2015 (UTC)[reply]
I've been getting the same thing for a few weeks now. It's obnoxious. I hope someone figures out a fix. Curly Turkey ¡gobble! 07:43, 14 February 2015 (UTC)[reply]

Table at Teahouse appears broken

The table that is at Wikipedia:Teahouse/Questions/Archive_Index appears broken to me. It stretches out way past the boundaries of the right side of the screen by maybe an extra 30% or so. Funny thing is it loads at the right size at first but then expands to mega-sized about one second later. Is anyone else getting this? I'm running Chrome on Windows 8.1 for what it's worth.  DiscantX 06:39, 9 February 2015 (UTC)[reply]

I can confirm what you're seeing, and it looks like a broken item is causing this. I have no idea how that archive index works, but I'll see if I can solve it. Martijn Hoekstra (talk) 10:00, 9 February 2015 (UTC)[reply]
I've added a </nowiki> which seems to fix the table; hopefully the bot won't undo it next time it updates the table. DH85868993 (talk) 10:19, 9 February 2015 (UTC)[reply]
(edit conflict) Solved it; there was a question with a nowiki tag in the title; this caused the next load of questions in the archive to be nowiki'd. Sam Walton (talk) 10:19, 9 February 2015 (UTC)[reply]
And I changed] the original section that caused the breakage, hopefully making sure the bot will do the right thing on the next run. Go go team fix-it. Martijn Hoekstra (talk) 10:28, 9 February 2015 (UTC)[reply]
Yay. Thanks for looking at it.  DiscantX 10:58, 9 February 2015 (UTC)[reply]

Tech News: 2015-07

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 software changes

Problems

  • MediaWiki was reverted to the previous version on February 4. It was because of a performance issue. It was restored later. [12]
  • UploadWizard was broken on February 4 because of the revert of MediaWiki. [13]
  • All sites were broken for 30 minutes on February 5. It was due to a network problem. [14]

Software changes this week

  • The new version of MediaWiki has been on test wikis and MediaWiki.org since February 5. It will be on non-Wikipedia wikis from February 10. It will be on all Wikipedias from February 11 (calendar).
  • You can have one user page for all wikis. Your Meta user page will show if you don't have a user page on a wiki. You can test this tool on test wikis. [15] [16] [17]
  • You can search for media files in VisualEditor more easily. Images are bigger and you see the size and license. [18] [19]
  • It is easier to review your changes when you save the page in VisualEditor. The window is wider. [20]
  • You can read the latest news about VisualEditor. You can now join weekly meetings with developers. During the meetings you can tell developers which bugs are the most important. The first meeting is on February 11 at 20:00 (UTC).

Future changes

  • Administrators will soon be able to delete change tags used fewer than 5,000 times. [21]
  • In the future you will be able to have personal lists of articles on the mobile site. [22]

Tech news prepared by tech ambassadors and posted by bot • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

16:27, 9 February 2015 (UTC)

Hello all, I'm having a bit of an issue with viewing pictures: I have a script/gadget enabled that opens pictures in a lightbox rather than MediaViewer, and until recently, I was able to close out by just clicking outside the picture. However, I'm now having to hit the back arrow and then re-searching whatever article I was at before (Splatoon, in this case). Any idea why it's now acting up? I've been consistently using the most updated version of Firefox (I've had it for a while, so no specific numbers), so it shouldn't be a browser issue. Supernerd11 Firemind ^_^ Pokedex 17:43, 9 February 2015 (UTC)[reply]

Bits on the fritz?

About half the time, I don't get any CSS/JavaScript served. What's up? -- [[User:Edokter]] {{talk}} 22:23, 9 February 2015 (UTC)[reply]

Anything in the Network tab of your browser's developer tools? Wondering if this problem happens with everything served from bits or just some parts. --AKlapper (WMF) (talk) 23:51, 9 February 2015 (UTC)[reply]
FWIW - not sure if its related to this or not but I'm getting reports like "... after 10 minutes or so of editing, when scrolling the original, the page rendering stops for about 10-15 seconds after each touch of the scroll bar, and I am unable to produce much work." among other "new" behaviors over on Wikisource (pretty much since the [re]deployment of 1.25wmf15 on 2/5).

So far - only FireFox users seem to be speaking to these new quirks. -- George Orwell III (talk) 00:05, 10 February 2015 (UTC)[reply]

Error: Invalid time.

I was finishing up a Good Article review, and putting the template in to signify the successful promotion on this page. However, when I saved it, in red text was the error "Error: Invalid time." I Googled it and found the same error elsewhere. I'm using Firefox 35.0.1 on OS X 10.10.1. Thanks, Acalycine(talk/contribs) 05:38, 10 February 2015 (UTC)[reply]

You did 4 ~ instead of 5 ~. I fixed it for you. ChrisGualtieri (talk) 05:54, 10 February 2015 (UTC)[reply]
Doh! Stupid me, that was the problem. Thanks Chris! Acalycine(talk/contribs) 05:59, 10 February 2015 (UTC)[reply]
As for the second one, the template m:Template:Archive header expects the last part of the page name to resemble a date. In the case of m:Help Forum/Archive 2, the last part is "Archive 2", which is not a date. --Redrose64 (talk) 12:07, 10 February 2015 (UTC)[reply]

Cite menu on toolbar

Hello. Is there a way to put my own templates in cite menu? Or to have in toolbar some of my own preferences? Xaris333 (talk) 10:09, 10 February 2015 (UTC)[reply]

I haven't tried it here, but I added the Redirect button to my toolbar on a wiki that didn't include it for everyone. (Ignore the unrelated top three changes, and look at the intermediate revisions if you want to add more than one button.) Mark Hurd (talk) 11:23, 11 February 2015 (UTC)[reply]

But how I know which are the buttons I can add? Xaris333 (talk) 21:20, 12 February 2015 (UTC)[reply]

Archives template

The {{Archives}} template cannot cope with more than 48 pages. Please can someone fix this? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:46, 10 February 2015 (UTC)[reply]

I've been working on an archive module for WP:CHU archives, and it's starting to turn into a more general-purpose archive module. I'll see if it can apply over there as well. — Mr. Stradivarius on tour ♪ talk ♪ 11:39, 10 February 2015 (UTC)[reply]
(ec) I'll have the setup for an alternative done in a bit, but getting all the parameters supported might take a bit. Martijn Hoekstra (talk) 12:32, 10 February 2015 (UTC)[reply]
As a proof of concept, User:Martijn Hoekstra/archivealt2 seems to work as an adequate replacement for {{archives}} (it replaces the call to the inner module with a call to a lua module Module:Martijn Hoekstra/archivecounter) apart from the root parameter. I can attempt to make this nice and all tonight CET. Review appreciated (I'm looking at you Mr. Stradivarius/Mr. Stradivarius on tour) Martijn Hoekstra (talk) 13:11, 10 February 2015 (UTC)[reply]
That should be done. See Template_talk:Archive_list_long#Luafication. Any advice on how to move forward? Martijn Hoekstra (talk) 20:28, 10 February 2015 (UTC)[reply]
Seems to be working well; thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 23:49, 10 February 2015 (UTC)[reply]

Reminder: Weekly VisualEditor triage meeting tomorrow at 20:00 UTC (12:00 PST)

As a reminder, on Wednesday, 11 February 2015 at 12:00 (noon) PST (20:00 UTC) there will be the first of a weekly series of open triage meetings about VisualEditor.

We will discuss the release criteria for VisualEditor, and jointly prioritise the work of the team, talking about the bugs and features which are most important to you. We particularly welcome the presence of volunteers who enjoy contributing MediaWiki code.

The joining instructions are available on MediaWiki.org. Hope to see many of you there.

Jdforrester (WMF) (talk) 19:29, 10 February 2015 (UTC)[reply]

Jdforrester (WMF) I missed this. Is there a transscript/log? Martijn Hoekstra (talk) 21:58, 12 February 2015 (UTC)[reply]
@Martijn Hoekstra: We're not recording the calls, no; however, I'm about to write-up the outcomes with an e-mail to wikitech-l which I can link to from here. Next week's meeting (at 08:00 PST == 16:00 UTC) will have a slot where people can nominate things to be de-listed, and new and re-nominations will be considered in the main session as this time. Jdforrester (WMF) (talk) 20:51, 13 February 2015 (UTC)[reply]

Lua

I will like to know where are the reasons or discussion for the integration of Lua in order to render templates/infoboxes. --Keysanger (talk) 22:08, 10 February 2015 (UTC)[reply]

A scripting engine was created for maintainability, ease of writing logic, and performance. Lua specifically was chosen for its sandboxing capabilities. Martijn Hoekstra (talk) 22:35, 10 February 2015 (UTC)[reply]
Do you know whether there is a paper about, or RfC? --Keysanger (talk) 23:19, 10 February 2015 (UTC)[reply]
There sure is a whole lot of mailinglist stuff about it. There is [23] which is one of the oldest documents about it I guess. I'm not 100% sure if there was an RFC, but the general sentiment was one of thank god. An RFC wasn't really needed either, as nothing was really changing, just additional options were provided. People started using it naturally, and there wasn't much, if any, opposition. Mediawiki templates are really not suitable for programming, and something like Scribunto, which is the name of the Lua extension was sorely needed. You could ask Tim Starling or Anome, who both worked on the implementation, they surely remember if there were any en.wiki discussions about the implementation. Martijn Hoekstra (talk) 23:35, 10 February 2015 (UTC)[reply]
While I don't particularly recall an RfC as such, there were quite a few discussions. I'd try checking out the archives of he various Village Pumps or the discussion pages for the particular template(s) that you are interested in. – Philosopher Let us reason together. 23:51, 10 February 2015 (UTC)[reply]
There's a decent summary of the history and reasons in File:Wikimedia Hackathon 2013 - Scribunto presentation slides.pdf. Anomie⚔ 01:08, 11 February 2015 (UTC)[reply]

That is what I was looking for, thank you, Martijn, Philosopher, Anomie. --Keysanger (talk) 10:06, 11 February 2015 (UTC)[reply]

New search not finding title matches

Special:Search/U of A Community College at Hope doesn't have University of Arkansas Community College at Hope on the first page, and List of colleges and universities in Arkansas is #19. --NE2 04:05, 11 February 2015 (UTC)[reply]

The "U" and "A" are being searched for literally, first at least, and other pages mention say "U.S.A." and so are matching more words completely.
I assume you're suggesting the old search did "better" with this. Could it be that shorter words matching produced a lesser relevance score than now?
Mark Hurd (talk) 10:23, 11 February 2015 (UTC)[reply]
If the search results aren't "fixed" to your satisfaction (or you really want a short term workaround), specify "country = U.S.A." in the Infobox (which should really be done anyway, seeing as this is not US-specific, but I haven't done it so as to not change this search result). Mark Hurd (talk) 11:08, 11 February 2015 (UTC)[reply]
Google, whose core business is web search, has some pretty nifty innovations, among which understanding that "U of A" in some contexts means University of Arkansas, and rather than treating all tokens separately additionally searches for the set phrase "University of Arkansas". It's some pretty cool stuff. Our search isn't nearly as sophisticated, and doesn't draw that same conclusion. I don't think our search will ever be as good as Googles. As an alternative, you could use google search rather than our own search. Martijn Hoekstra (talk) 11:57, 11 February 2015 (UTC)[reply]

Tool down, maintainer notified

The tool at http://tools.wmflabs.org/bibleversefinder, mantained by a veteran editor, has been down for almost (or "at least"?) five months now. That editor's talk page has three comments on it (including mine, a few days ago) from other users. What to do in that case? I would like to edit one page which uses that tool, but don't want to seem rude about it. I wrote the third note on the veteran editor's talk page. The other two notes were written by two other different editors, within months of each other. In wikipedia-editing time (if there is such a thing), is it the right time to do something about it? I'm not quite offering myself to mantaining it...but perhaps would like to know what it might entail to maintain a lab tool (I'm new to the registered Wikipedia "world," but not to editing articles). -- Capikiw (talk) 22:37, 11 February 2015 (UTC)[reply]

Problem with DS alerts template

There's a problem with one of the discretionary-sanctions templates that I can't see how to fix. Template:Ds/alert#Codes says that gg refers to gamergate and GGTF to gender gap task force, but currently both alerts – {{subst:alert|gg}} and {{subst:alert|ggtf}} – lead to GGTF. They have different messages: the alert produced by gg is correct, but it should be called ggtf. If someone could fix this it would be very helpful. I can't see how to get to it. Sarah (SV) (talk) 22:46, 11 February 2015 (UTC)[reply]

Sorry, cancel that. I just worked out how to do it. Sarah (SV) (talk) 22:54, 11 February 2015 (UTC)[reply]

Issue with fonts caused by Windows Vista update

Just a public announcement if anyone else is having an issue with fonts looking like shite: [24] --NE2 02:37, 12 February 2015 (UTC)[reply]

Wikipedia:Reference desk/Computing#Microsoft: the font of all disgruntlement. --  Gadget850 talk 10:45, 12 February 2015 (UTC)[reply]
The official line. --Redrose64 (talk) 12:26, 12 February 2015 (UTC)[reply]
Ha! I just wanted to post this one! Learned a lot today about fonts, while I was trying to figure out wat was causing the problem. See also Patch Tuesday: Font Corruption with KB3013455 and Update KB3013455 Messes Up Fonts. Joshua Jonathan -Let's talk! 20:06, 12 February 2015 (UTC)[reply]

nbsp

Is there a way to replace all "& nbsp ;" with {{nbsp}} throughout Wikipedia? The first one is hard on editors - hard to read and easy to get wrong. Bubba73 You talkin' to me? 20:09, 12 February 2015 (UTC)[reply]

Please, no. {{nbsp}} is two characters longer, and not all bots or scripts recognise it, whereas all recognise &nbsp; --Redrose64 (talk) 20:14, 12 February 2015 (UTC)[reply]
And it creates <span class="nowrap">&nbsp;</span> which can break things when used in other templates. --  Gadget850 talk 20:28, 12 February 2015 (UTC)[reply]
Then why do we have {{nbsp}}? I thought it was replacing the bad version (from HTML?). Bubba73 You talkin' to me? 23:49, 12 February 2015 (UTC)[reply]
Presumably by "from HTML" you mean the &nbsp; entity. Why do you describe this as the "bad version"? There is nothing wrong with it; it is recognised by every browser (except possibly some from 20+ years ago); and indeed, that entity is at the core of {{nbsp}}, so if &nbsp; was in some way "bad", then disguising it in a template must be just as bad, if not worse.
Anyway, {{nbsp}} exists because it was created and hasn't (yet) been taken to TFD. The original documentation said "template to minimize code used for multiple consecutive non-breaking spaces", which sounds like a primary reason that it was created. The template still does that - given a parameter (and used in carefully supervised conditions), it can reduce the amount of wiki markup - if you want four non-breaking spaces, you can use {{nbsp|4}} instead of &nbsp;&nbsp;&nbsp;&nbsp;. But used without a parameter, it is pointless, much less efficient than the entity, and breaks certain bots, scripts and templates (such as the citation templates). --Redrose64 (talk) 09:16, 13 February 2015 (UTC)[reply]
Yes, the HTML version is what I call the bad version. I can remember to open and close with a pair of curly braces, but I can't remember the format of the HTML version. It is easy for the editor to get wrong and if there is no space between it and the text to the sides, it makes reading the text difficult for the editor. And it is not pointless without a parameter - it makes it easier for the editors to edit and not make mistakes.

Bubba73 You talkin' to me? 00:36, 14 February 2015 (UTC)[reply]

Unpaired font-size tag on Template:Copyviocore?

The font size of the header of {{Copyviocore}} has suddenly got much bigger (which is fine by me), but a weird </font-size> tag has sprung up at the same time. There's nothing in the history of the template to explain the change; indeed, it shows up in historic versions too. Can anyone explain? Fix? Many thanks, Justlettersandnumbers (talk) 23:45, 12 February 2015 (UTC)[reply]

Oddly did not show until I purged. Changed invalid </font size> to </span>. --  Gadget850 talk 00:05, 13 February 2015 (UTC)[reply]
It seems to have been added by Moonriddengirl during edits attempting fix some problem in 2011! Mark Hurd (talk) 01:30, 13 February 2015 (UTC)[reply]
It sounds like Sanitizer.php fixed the invalid html before but has stopped doing it. PrimeHunter (talk) 02:03, 13 February 2015 (UTC)[reply]

Is this more than coincidence?

Yesterday amongst other corrections to pages with ref errors, I made one to Browser hijacking. Today my notebook is unusable, as things keep popping up in MSIE, Firefox and Opera. Is it just coincidence, or is there some secret stuff hidden there? It could be coincidence, because yesterday I also ran the weekly Windows update. (W7). Also, that notebook is falling to pieces, literally, so anything could be going on. But worth checking just the same. -- Unbuttered parsnip (talk) mytime= Fri 08:53, wikitime= 00:53, 13 February 2015 (UTC)[reply]

Looking through the source editing window, I don't see anything out of the ordinary. I'm not aware of any way to hide stuff within a Wikipedia page, but I'm no expert, so take that as you will. Supernerd11 Firemind ^_^ Pokedex 02:30, 13 February 2015 (UTC)[reply]
I know a few weeks ago I was correcting a software page, and it was just XML, nothing else. But when I saved, there was a 'nothing-changed' report and it didn't show up on my contributions list, so I can't even tell you what it was.
And if your machine doesn't work tomorrow @Supernerd11: you'll know why! -- Unbuttered parsnip (talk) mytime= Fri 13:52, wikitime= 05:52, 13 February 2015 (UTC)[reply]
A long shot but could your browser be trying to look ahead and cache pages that are e.g. linked from ones you visit? I'm not going to look for myself but Browser hijacking might have links that lead to, well, browser hijacking; deliberately (unlikely) or to sites which link to them, or even to sites about browser hijacking which have been tempting targets for hackers (especially given how ELs can suffer serious link rot). My first guess though was coincidence, possibly something picked up somewhere else, possibly Windows needing a rebuild or reinstall.--JohnBlackburnewordsdeeds 06:28, 13 February 2015 (UTC)[reply]

I don't know whether this is the right spot for this but.

Some pages refuse to let me post. Right now I am trying to edit [[25]] and here. I press save edit and nothing happens. Please help. I am using a phone and chrome browser Wikipedia. Not an app. I have no idea what is triggering this. It might have something to do with references but I've posted at least 5 references before without an issue. Thanks. Don't forget to ping me. -DangerousJXD (talk) 01:11, 13 February 2015 (UTC)[reply]

OK I think I know what caused it. I got it to work now. Finally! And added a bucket load of references.-DangerousJXD (talk) 02:01, 13 February 2015 (UTC)[reply]

The Amazing Spider-Man (2012 film) vandalism

Hi, looks like this page was vandalized yesterday but not spotted, and other changes have been made since, can someone look at ? Thanks GrahamHardy (talk) 13:07, 13 February 2015 (UTC)[reply]

Could have been a mistake. Anyway, I reverted to last known-good version. Any subsequent edits will have to be redone. -- [[User:Edokter]] {{talk}} 13:31, 13 February 2015 (UTC)[reply]

Knowledge of personal pronoun preference data base

There is an interest in tracking progress toward gender equality by dynamically updating statistical measure of self-identified "gender pronoun usage preference" on user pages to confirm trend of progress to achieve gender parity. This is likely admin only access to technical user data. Does anyone know how easy/how difficult it would be to gather and then print out these statistics? LawrencePrincipe (talk) 21:16, 13 February 2015 (UTC)[reply]

@LawrencePrincipe: Actually, you can access it using the "gender" magic word like so: {{gender:username|he|she|unset}}
So if you enter {{gender:Philosopher|he|she|unset}}, you get he, indicating hat I have my preferences set to "he". I do the same thing to you and see that yours is "unset". – Philosopher Let us reason together. 00:14, 14 February 2015 (UTC)[reply]
@Philosopher: On a one-by-one basis that works, though the stats are needed system wide on a weekly and/or monthly basis. That is, out of the total number of new users per month, what ratio self-identified as female and what ratio self-identified as male system-wide. Can it be done system-wide for being dynamically updated? LawrencePrincipe (talk) 00:35, 14 February 2015 (UTC)[reply]
Listed at Wikipedia:Database reports/User preferences but out of date. --  Gadget850 talk 00:53, 14 February 2015 (UTC)[reply]
@Gadget850: Those stats are quite fine, but quite dated: "User preferences statistics; data as of 23:12, 19 June 2014 (UTC)." Is there a way to update them, or get a month-by-month graph? LawrencePrincipe (talk) 02:01, 14 February 2015 (UTC)[reply]
Looks like this was run on the old Toolserver. I don't know of a replacement. --  Gadget850 talk 02:23, 14 February 2015 (UTC)[reply]

signature

Why I can not sign correctly? does anybody know? M.Sakhaie 11:29, 14 February 2015 (UTC)

Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(technical)&oldid=647086488"