vertical bar in template parameter

Is there a way to escape a | in a template? A page title contains a | and I'm trying to use {{cite web|title = Title with | in it|date= ...NE Ent 13:35, 3 October 2015 (UTC)

In {{cite web}}, use %7c inside |url= and | in any other parameter when you want to write a vertical bar. – Jonesey95 (talk) 13:48, 3 October 2015 (UTC)
Alternatively, you can place | within nowiki tags, i.e. <nowiki>|</nowiki>. 2607:FB90:21C9:F62A:0:37:4515:9301 (talk) 17:39, 3 October 2015 (UTC)
@NE Ent: Very often, a pipe in the title of a web page is used to separate the true title of a web page from the name of the website. They might be in either order. So, a web page that is titled like "Guaranteed Roadrunner Trap | Acme" or like "Acme | Guaranteed Roadrunner Trap" would be split into |title=Guaranteed Roadrunner Trap |website=Acme. If you provide the URL of the web page, I can look at it to advise on the best action. --Redrose64 (talk) 20:54, 3 October 2015 (UTC)
Also, {{!}} will do the trick. — Mr. Stradivarius ♪ talk ♪ 04:08, 4 October 2015 (UTC)
Thanks. It's "blah blah blah | Science News" -- I don't what ya'll consider the true title ... I just use what is between the <title> </title> tags on the page. NE Ent 12:31, 10 October 2015 (UTC)

Shutting off pings from specific users

I really do like the pings in general, and I can apparently shut them off overall, but sometimes I'm dealing with a user who has gotten "ping-happy" and stays that way even after I've asked them to stop doing that. I don't want to go as far as reporting the user, but I sure would like to have their pings to me shut off. Is there any way this can happen? Stevie is the man! TalkWork 16:09, 7 October 2015 (UTC)

Probably better place to ask is WT:Echo, but it sounds like a good idea. --Edgars2007 (talk/contribs) 17:30, 8 October 2015 (UTC)
Thanks. I am continuing to get ping-abused from two particular editors. So I would love a solution. Stevie is the man! TalkWork 18:43, 10 October 2015 (UTC)

SQL question

I've been trying to find out the total admin actions per month in order to answer a question at WT:RFA, but I've been having problems running my SQL query on Quarry. My query to find the total admin actions for September ran fine, but when I tried to expand that to all months from January 2001, my query took over 20 minutes to execute and was killed. Is there anything I can do to improve the efficiency of this, or am I better off just making separate database queries for each month from an external script? — Mr. Stradivarius ♪ talk ♪ 11:49, 10 October 2015 (UTC)

Backend is mysql? If log_timestamp isn't indexed, it's likely to be slow regardless of your query. You could try using the IN operator [1] instead of the multiple OR statements. NE Ent 12:29, 10 October 2015 (UTC)
Thanks for the tips. :) Yes, it's mysql. I've tried using the IN operator and also reducing the number of dates from 1000 to 300, so we'll see if that helps. It's waiting in the queue now, but there are no other queries running - I think I might have lost my right to have it executed straight away after the previous two attempts... — Mr. Stradivarius ♪ talk ♪ 13:40, 10 October 2015 (UTC)
Actually, I just saw this example query, which is an approach that should be much more efficient. I'll give that one a try. — Mr. Stradivarius ♪ talk ♪ 14:19, 10 October 2015 (UTC)
You can also sign up for a tool labs account and run the query manually instead of using quarry :) Legoktm (talk) 15:08, 10 October 2015 (UTC)
I have one, actually, but I wanted to do it on Quarry so that other people could easily see the results and maybe modify the query. In any case it will have to wait, as I'm getting "SELECT command denied" errors at the moment. — Mr. Stradivarius ♪ talk ♪ 15:43, 10 October 2015 (UTC)
It's because you are trying to run the query on the private "enwiki" database, not "enwiki_p" which is public. I just ran it on enwiki_p here. --Glaisher (talk) 15:55, 10 October 2015 (UTC)
BTW, you don't have to include USE enwiki_p;. For enwiki it's there by default. Edgars2007 (talk/contribs) 16:56, 10 October 2015 (UTC)
Ah, that makes sense, thanks. — Mr. Stradivarius ♪ talk ♪ 16:58, 10 October 2015 (UTC)

comparing revisions across two or more history pages without missing some in between

When searching old revisions by comparing an older one with a newer one that necessarily are listed on different pages of the page's history, a searcher ordinarily conducts two or more searches, thereby missing the effect of one intermediate revision listed at the bottom of a history page (or two intermediate revisions at the bottom of two history pages, etc.). This is because the older revision selected on the history page is not itself reflected in the top of the resulting diff, but only implied in the midst of everything still older, thus easily missed or misunderstood. It is possible to solve this by manually constructing a URL, but that possibility is easily missed by users.

Example (frequent editing of the article will quickly make this outdated but nonetheless the model applies):

Perhaps this can be solved with a technical solution.

Nick Levinson (talk) 02:00, 4 October 2015 (UTC)

You can list up to 5000 revisions at once in a page history. Careful setting of the "From year (and earlier)"/"From month (and earlier)" options will bring the appropriate revision close to the top; or you can set &offset= This will allow you to get both on the same screen. --Redrose64 (talk) 10:43, 4 October 2015 (UTC)
That's complicated for most users and not explained where most users would likely see it, and I know how to get up to 500 revisions in a page history, not 5,000. Can this be explained with a link from history pages or the procedure technically simplified? Nick Levinson (talk) 00:41, 11 October 2015 (UTC)
If you click a number of revisions then you get a url where the number can manually be changed to anything up to 5000. I suspect the interface links stop at 500 for performance reasons. I don't know whether it's OK to advertise the 5000 method on page histories. PrimeHunter (talk) 08:40, 11 October 2015 (UTC)
If we were at WP:VPR, I'd oppose adding even a little to the page history page with so little need for it. Thankfully, we're not. At least you have some way of getting what you need. Maybe an addition to Help:Page history about this would be warranted, and the page history page already has a link to that page. ―Mandruss  09:55, 11 October 2015 (UTC)
For comparison, user contributions link to Help:User contributions which already says: "The blue numbers list the number of edits displayed on a page: 20, 50, 100, 250 or 500. The number you select replaces n in the links to the previous or next pages e.g. (newer 100 / older 100). Views of up to 5000 edits per page are possible by modifying the URL." PrimeHunter (talk) 10:28, 11 October 2015 (UTC)

Page size dumps

Does anyone know if there is a dump file of en.wikipedia which contains some measure of the size of each page, without having to waste the bandwidth downloading all the page contents. Thanks. HYanWong (talk) 14:04, 11 October 2015 (UTC)

To answer my own question, in enwiki-XXXXX-page.sql.gz ('only' a 1.2Gb download), the penultimate column in the 'page' table is a field named 'page_len', which gives this statistic.HYanWong (talk) 22:09, 11 October 2015 (UTC)
@HYanWong: What do you need the data for? Instead of downloading a dump, it might be better to run an SQL query on Quarry. — Mr. Stradivarius ♪ talk ♪ 01:50, 12 October 2015 (UTC)
@Mr. Stradivarius: I want to find out the page size of about 1.5 million pages (all living organisms), which I'm going to combine with page views from http://dumps.wikimedia.org/other/pagecounts-ez/merged/ to assess some measure of popularity of each species or higher level group. Thanks for the tip about Quarry: is it considered reasonable behaviour to ask for such large datasets via SQL? HYanWong (talk) 07:51, 12 October 2015 (UTC)
I think in this case downloading enwiki-XXXXX-page.sql.gz would be better solution. I once run a query on Quarry, that returned 30,000 pages, it was fine, but you most probably won't get 1.5 million pages returned. But you can try :) --Edgars2007 (talk/contribs) 09:37, 12 October 2015 (UTC)
Thanks. An additional problem in this case is that the query itself contains 1.5 million page titles (grabbed from the wikidata dump), so even sending the query in the first place will be a pain! HYanWong (talk) 10:03, 12 October 2015 (UTC)
Hint (but probably not for this case). Put links to pages in some user-space page, then use pagelinks table joined with page table in Quarry. --Edgars2007 (talk/contribs) 10:29, 12 October 2015 (UTC)
P.s. I would like to document somewhere that the page lengths are present in XXX-page.sql.gz, but not sure where best to add documentation. Suggestions? (edit: maybe best to change the summary in the dump listing to "Base per-page data (id, title, old restrictions, page size etc).", but that's not user editable)HYanWong (talk) 10:05, 12 October 2015 (UTC)

Quotation marks breaking URL display

I've noticed just recently that there seems to have been a change to how external links are parsed that seems problematic. In particular, note how this:

[http://example.com?foo="bar"&bar="baz" Foobar]

renders as

"bar"&bar="baz" Foobar

i.e. that the "bar"&bar="baz" is included in the displayed URL, which is obviously unintended. This can be manually solved, i.e. by changing " to %22:

Foobar

However, since there are many, many such links that would have to be fixed (in no small part Google Books URLs), this is probably something that needs to be dealt with on the software end of things. {{Nihiltres |talk |edits}} 17:09, 11 October 2015 (UTC)

This isn't a new thing - I remember running into this years ago when I was making a template (can't remember which one offhand). — Mr. Stradivarius ♪ talk ♪ 17:21, 11 October 2015 (UTC)
Yes, it's listed at Help:URL#Fixing links with unsupported characters. The example here was also broken when The Internet Archive first archived the page in [2] 14 August 2014. PrimeHunter (talk) 17:36, 11 October 2015 (UTC)
I'm seeing the issue, in particular, in some citations. I don't recall seeing that one before. Possibly something's changed in one of the involved templates? Possibly something in this change? {{Nihiltres |talk |edits}} 18:44, 11 October 2015 (UTC)
Could you give us an example of a place where you're seeing problems? — Mr. Stradivarius ♪ talk ♪ 01:13, 12 October 2015 (UTC)
Nihiltres, the CS1 citation templates were recently (Sep 25 or 26) updated to feature more robust checking of URL parameters in the templates. I recommend reposting your question at Help talk:Citation Style 1 with some examples from articles. It is possible that the enhanced error detection has resulted in a change in the displayed link. It is also possible that the link wasn't working to begin with, and the new display is a side effect that will go away when the link is fixed.
Since the change, Category:Pages with URL errors has increased its population from roughly zero articles, where it had been for over a year, to 13,000+ articles. We have some citation template URLs to fix. All of you VPT watchers are welcome to come over and fix a hundred or so. – Jonesey95 (talk) 19:35, 12 October 2015 (UTC)

Edit summary maxlength

WP:ES#The 250 character limit claims an edit summary is limited to 255 bytes (not characters). I recall seeing some language to the effect that the limit was around 230 so a few multibyte UTF-8 characters wouldn't overflow. However, there is now a hard limit of maxlength="200", although possibly that can be overridden with JavaScript? Has maxlength been reduced in the last few months? Johnuniq (talk) 03:21, 12 October 2015 (UTC)

I must be misremembering because looking at some old source for EditPage.php makes it appear nothing has changed for several years, and "maxlength is overridden in JS to 255 and to make it use UTF-8 bytes, not characters" (what is a UTF-8 byte?). Johnuniq (talk) 06:58, 12 October 2015 (UTC)
Where do you see this maxlength="200"?
The maximum typeable length and the maximum storable length are not the same thing. The maximum storable length has always been 255 bytes. Edit summaries are stored as UTF-8, and a character like the en-dash (–) occupies three bytes when stored in this manner. The maximum typeable length was 200 characters for several years, there was a gadget that could increase this to (I think) 250; but that became redundant when the limit imposed by the MediaWiki software was increased two or three years back. If those 250 or so characters were more than 255 bytes when encoded as UTF-8, the excess would be truncated. The maximum typeable length is presently 255 bytes; if you type an edit summary using pure ASCII characters, you can enter 255 before it stops letting you type; if you type an edit summary using only characters that are three bytes in UTF-8 (e.g. Japanese script), you'll find that it won't let you enter more than 85. So unless you use characters outside the Basic Multilingual Plane that encode as four or more bytes (Unicode code points U+10000 or higher), the practical typing limit is between 85 and 255 characters. --Redrose64 (talk) 08:40, 12 October 2015 (UTC)
Thanks but I think "maximum typeable length is presently 255 bytes" is only correct if JavaScript is enabled—I can only type 200 characters with scripting off. The maxlength="200" is in the html source of a page with an edit summary and is what tells my browser (when JavaScript is disabled) to not accept more than 200 characters in the edit summary box. If anyone wants to experiment, the box following has one physical line starting with "A" and finishing with "Z". The line is 497 bytes = 200 characters. With scripting off, I can paste the line into an edit summary, but previewing shows the summary truncated. For some reason, when JavaScript is enabled, pasting the line does not work (it is truncated).
A—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×—×Z
Johnuniq (talk) 10:29, 12 October 2015 (UTC)
Yes, with JavaScript you can enter UTF-8 characters which use up to 255 bytes, which is was the database field can hold. With JavaScript off you can enter 200 UTF-8 characters regardless how many bytes they use, but on preview or save they are truncated to use at most 255 bytes. The situation is confusing for users without JavaScript but I don't know whether it's possible without JavaScript to make an input field which counts bytes and not characters. The 200 limit without JavaScript is a compromise: Large enough for most edit summaries, and limits the number of times an edit summary must be truncated because it breaks the 255-byte limit by having too many multi-byte UTF-8 characters. Languages with non-Latin scripts have more problems with this. PrimeHunter (talk) 10:52, 12 October 2015 (UTC)
Yes users with Javascript will get a better experience, because this check is not possibly with plain HTML. —TheDJ (talkcontribs) 11:12, 12 October 2015 (UTC)
How do I turn off JavaScript in Firefox 41.0.1? --Redrose64 (talk) 20:51, 12 October 2015 (UTC)
Tools/Add-ons - Click on Disable next to NoScript. Then you have to close your browser and restart it for the disable to take affect. — Maile (talk) 20:57, 12 October 2015 (UTC)
It's not there. I only have two entries: Microsoft .NET Framework Assistant (disabled) and RealDownloader (disabled). --Redrose64 (talk) 22:58, 12 October 2015 (UTC)
NoScript is a Firefox extension Maile has installed. Most Firefox users don't have it. It has controls to allow or disallow JavaScript for each website. PrimeHunter (talk) 23:07, 12 October 2015 (UTC)
In about:config there is javascript.enabled. --Edgars2007 (talk/contribs) 21:01, 12 October 2015 (UTC)

"jump up to ^"

Sometimes when I look at articles and see the citations at the bottom, instead of just having the "^" caret, it will say before it, linked in blue) "jump up to ^", next to every citation. It's ugly and quite redundant. Is this a Wikipedia issue or is it on my end? I use Firefox on a Macmini. 108.54.167.196 (talk) 15:14, 12 October 2015 (UTC)

WP:Village pump (technical)/Archive 140#Jump up!. --Edgars2007 (talk/contribs) 15:20, 12 October 2015 (UTC)

Florence Rush

I added an infobox to Florence Rush and the entire article disappeared (diff), I've reverted for now, can someone please take a look? --The Vintage Feminist (talk) 15:52, 12 October 2015 (UTC)

  Done. You had left in the opening part of an html comment before the death date and age template, The Vintage Feminist, but not the matching close, so everything after that, including the template close, was swallowed in the comment, leaving only an unclosed template call. DES (talk) 15:59, 12 October 2015 (UTC)
Oops, okay thanks. --The Vintage Feminist (talk) 16:08, 12 October 2015 (UTC)

16:29, 12 October 2015 (UTC)

Blank image

 

File:GrahamCube.svg has started rendering blank for no apparent reason. It used to be ok in the article. Clicking through to the image it is still there and no new versions have been uploaded. SpinningSpark 21:37, 12 October 2015 (UTC)

I've fixed the file by adding the type="text/css" attribute to the <style> element. Unlike in HTML, that attribute seems to be required for CSS stylesheets to work in SVG. SiBr4 (talk) 21:51, 12 October 2015 (UTC)
Funny that it used to be ok though. Something must have changed somewhere. SpinningSpark 22:04, 12 October 2015 (UTC)
Yes, the documented behaviour is that in the absence of any other indication, the default is "text/css". --Redrose64 (talk) 23:26, 12 October 2015 (UTC)

Article Feedback Activity Log erased

Some entries in Special:Log have only a single word after the user page, talk page, and contribs links. Those entries were in the Article Feedback Activity Log, and had some text that got erased after AFT became obsolete on March 3, 2014. The entries that were in the Article Feedback Activity Log should be completely erased. GeoffreyT2000 (talk) 01:55, 13 October 2015 (UTC)

Because the Article Feedback extension was uninstalled, its log entry formatter is no longer available, so these logs are presented in a pretty useless raw form. The relevant Phabricator task is phab:T64722. — This, that and the other (talk) 06:10, 13 October 2015 (UTC)

Revising Special:WantedPages

There is fairly universal agreement, at Wikipedia talk:Special:WantedPages, that this automated list is, in its current form, almost worthless. That's because the high-count items are being generated by either signatures on article talk and other pages, or by templates on article talk pages.

What is useful is Wikipedia:Most-wanted articles. That's built using red links from article pages, only. But it's not automatically generated - it's not a special page.

So, a purely technical question: Who has the authority/ability to modify the parameters that generate the Special:WantedPages content? I'm aiming, of course, at getting those parameters changed so that only red links on article pages will count; I'd like to know who has to be convinced that the current content isn't worthwhile. -- John Broughton (♫♫) 21:36, 7 October 2015 (UTC)

WantedPages is hard-coded into the MediaWiki software as including pages from all namespaces. So there is no parameter to change at the moment. The relevant task is phab:T37758.
Personally I think the query pages need a lot of help, since most of them (not just WantedPages) are pretty worthless on large wikis like English Wikipedia. Unless and until any improvements are made, Labs is the best place for developers to create improved alternatives to the query pages. — This, that and the other (talk) 01:21, 8 October 2015 (UTC)
It's used by the ContentTranslation feature, so it is very prominently linked at user contribution pages under "New contribution". Gparyani (talk) 18:08, 11 October 2015 (UTC)
The answer to the specific question you asked is that you need developer-level rights to change that, because it's hardcoded in the software and thus needs a software change to change. (It might also need a change to the database schema, and the like; it doesn't strike me as the type of thing that would be easy to do efficiently with the current software.) Gparyani's response is a good guide as to how you might practically go about getting it changed. --ais523 22:03, 14 October 2015 (UTC)

Is there a way to auto-populate categories through Wikidata/other wiki comparison?

I just created Category:Polish skydivers and Category:Cichociemni, both based on Polish Wikipedia categories (linked through Wikidata). Now, pl wiki categories have quite a few entries, and I expect 20-30% of them are on en wiki already. Is there some way (bot, script) that would automate checking which entries categorized under equivalent Polish categories exist here, and auto-tag them? It seems like something that should be possible to automate. --Piotr Konieczny aka Prokonsul Piotrus| reply here 09:15, 9 October 2015 (UTC)

If nobody has better solution, I have one, but it's quite crappy. Take Autolist, fill the parameters and get the list of Wikidata items. The URL with parameters for Cichociemni is here (it will take some few minutes to open). Then you can open each article and add category by yourself. Yes, it doesn't actually ease the work very much :D --Edgars2007 (talk/contribs) 10:51, 9 October 2015 (UTC)
But anyway -   Like the idea. Would really like to use such tool, which at least gets that list of pages, that needs to be categorised. --Edgars2007 (talk/contribs) 14:22, 9 October 2015 (UTC)
@Magnus Manske: Possible? --Izno (talk) 03:09, 10 October 2015 (UTC)
Well, you can use PagePile. Generate a list of pages on plwiki in a category in QuickIntersection, then filter the pagepile here, by "converting" it to Wikidata, then to enwiki; try "Load from PasteBin" with ID d6653a0c to see what I mean. You now have a list of all enwiki articles that correspond to plwiki articles on Polish skydivers. Convert that back into Wikidata, then load it in autolist, and set "pages in enwiki category Polish skydivers" as "NOT". You will then get all Polish skydivers from plwiki that are on enwiki but not in the enwiki category "Polish skydivers". The parts are all there, you just need to connect them :-) --Magnus Manske (talk) 12:13, 12 October 2015 (UTC)
@Magnus Manske, Izno, and Edgars2007: My eyes glazed a bit reading the how to... any chance somebody (pretty please) could hack together a tool which would make it as easy as typing in/linking the two categories (English and Polish)? --Piotr Konieczny aka Prokonsul Piotrus| reply here 02:49, 13 October 2015 (UTC)
Actually, I think I already wrote this but forgot about it ;-) Is this what you want? --Magnus Manske (talk) 18:21, 13 October 2015 (UTC)
@Magnus Manske: No, it actually does the opposite - it finds articles, which are in plwiki, but not in enwiki. BTW, if we have already started to talk about the tool - sometimes it isn't working, if there is more than 100 results. See here (fill in "WDQ": tree[211][150][17,131] OR claim[27:211]) - 101st result and so on isn't right... And if I tick "Source Wikipedia", then at page 2 the tick is on "Wikidata". --Edgars2007 (talk/contribs) 19:31, 13 October 2015 (UTC)

Page Padding

Am I delusional or was the padding on the mw-body class recently changed? It seems most noticeable on the Main Page where there seems to be a larger white gap along the left edge than I recall seeing in the past. Dragons flight (talk) 20:21, 10 October 2015 (UTC)

No recent changes. But in Vector, the padding does change from 1 to 1.5em when the screen width goes above 982px. It has been doing that form some years now. Have you gotten a bigger screen perhaps? -- [[User:Edokter]] {{talk}} 10:57, 13 October 2015 (UTC)

WP:FLAGCRUFT malicious?

I have recently inserted the WP:FLAGCRUFT js code, and found out it could be malicious. Simple question here: is it really that bad or is there really nothing to worry about here? --JB82 (talk) 14:49, 13 October 2015 (UTC)

What do you mean by "found out it could be malicious"? If you mean you see MediaWiki:Jswarning on your css and js pages then it's a standard warning always shown on those pages regardless of content. PrimeHunter (talk) 15:41, 13 October 2015 (UTC)
Adding JavaScript code to your user JS pages can do pretty much anything to your account (e.g. cause it to make edits, etc.). As such, it's best to ensure that any scripts you place on the page are written by someone trustworthy (or alternatively by an administrator; there's no loss of security there because administrators can edit other people's user JS pages anyway). The message there is just a scary warning to try to prevent people from installing dubious scripts they found outside Wikipedia. --ais523 21:55, 14 October 2015 (UTC)

Why can't I see the deleted version of David Godman

Wikipedia:Articles for deletion/David Godman shows this article was deleted by AfD in July 2008. It was recreated in October 2008 but I can't find the deleted versions. Doug Weller (talk) 14:11, 14 October 2015 (UTC)

It appears that the article was moved to User:Iddli/David Godman. Elockid(Boo!) 14:18, 14 October 2015 (UTC)
That means then that it wasn't deleted in July, I guess. Pretty promotional. Doug Weller (talk) 15:03, 14 October 2015 (UTC)
Click "View logs for this page" at top of the page history to see [6] which shows it was deleted in July 2008, and in October 2008 it was restored, moved to User:Iddli/David Godman, and the resulting redirect was deleted. The recreation was in November 2008‎. It was later deleted again and then restored including the October 2008 redirect from the move, which is why the visible page history currently starts there. PrimeHunter (talk) 16:35, 14 October 2015 (UTC)
Thanks PrimeHunter - I don't think that's a very satisfactory way of handling an article, especially as it lost all its attribution history before October 2008. Doug Weller (talk) 17:38, 14 October 2015 (UTC)
I've done the hist merge, so all the history is at David Godman now. —SpacemanSpiff 17:46, 14 October 2015 (UTC)

Table won't sort

On List of Xbox Live Arcade games, there is a table of 707 rows. (At least, the article says there are 707 games. I haven't counted the rows.) When clicking the arrow next to the text of the "Release Date" column header, there is a delay of a few seconds, and then the arrow will change, as expected; but the table does not sort.

I don't see this behavior when sorting by the other columns; the table sorts normally after clicking one of those arrows. Tarcil (talk) 00:03, 10 October 2015 (UTC)

@Tarcil: I've fixed this by adding data-sort-value for each of the rows, as described at Help:Sorting#Specifying a sort key for a cell. — Mr. Stradivarius ♪ talk ♪ 01:20, 10 October 2015 (UTC)
@Mr. Stradivarius: Thank you! Is there an item on the developers' backlog to automate this for the thousands (I assume) of tables on Wikipedia that don't include this markup? Tarcil (talk) 17:42, 15 October 2015 (UTC)
@Tarcil: No - I think this is more of an editor task than a developer task. Having said that, it would be good to find a way of tracking which tables have columns that can't sort so that they can be dealt with by AWB or by bot. — Mr. Stradivarius ♪ talk ♪ 22:06, 15 October 2015 (UTC)
Works fine for me. Tvx1 01:20, 10 October 2015 (UTC)
Is it really necessary to link all those years? That's the reason the sorting breaks (it would not need a sort key otherwise), and it seems excessive to me. nyuszika7h (talk) 16:06, 11 October 2015 (UTC)

Cascading protection glitch

Are users here who are template editors but not admins able to edit Template:Center? I ask because I appear to have caused a glitch which makes it cascade-protected from a deleted page. Here's how it happened:

  1. I notice that when I edited Template:Center, there was a notice above the edit box that said it was cascade-protected from File:Onion Powder, Penzeys Spices, Arlington Heights MA.jpg. I decide that this protection isn't serving any purpose, as the file is no longer visible on the main page.
  2. I delete File:Onion Powder, Penzeys Spices, Arlington Heights MA.jpg without unprotecting it. I thought this would remove the protection and make pages fetch the file from Commons, all in one edit.
  3. I reduce Template:Center to template-protection.
  4. Template:Center is now displayed as template-protected, but when editing it the notice that says that it is cascade-protected is still there. The difference is that instead of File:Onion Powder, Penzeys Spices, Arlington Heights MA.jpg being listed as a source, there are no sources listed.

I'm guessing that this will have to go to Phabricator, but if there's a way we can resolve it here, that would be nice. — Mr. Stradivarius ♪ talk ♪ 11:09, 13 October 2015 (UTC)

To me it shows the "template protected" message now. Not a template editor or administrator.Jo-Jo Eumerus (talk, contributions) 11:11, 13 October 2015 (UTC)
I think that the glitch may have defeated the protection level detection code that makes the "template protected" message, as it relies on the CASCADINGSOURCES magic word producing some text, but if there are no sources then the output will just be blank. I am also guessing a similar thing happened with the Lua code that checks for cascading protection to see what colour the padlock icons should be. — Mr. Stradivarius ♪ talk ♪ 11:19, 13 October 2015 (UTC)
Also, I should add that after I deleted File:Onion Powder, Penzeys Spices, Arlington Heights MA.jpg I restored it, then tried cascade-protecting it and then unprotecting it again, but it didn't seem to make any difference. — Mr. Stradivarius ♪ talk ♪ 11:12, 13 October 2015 (UTC)
I placed the "Template editor" right on an alternate account of mine and loaded the edit page. I did get the page (I didn't actually try to change anything there), along with the warning that only admins can edit it due to cascade protection. עוד מישהו Od Mishehu 11:37, 13 October 2015 (UTC)
For me (TE) it displays the "cascade-protected" warning alongside the template-protection one, though it seems I am able to edit the template normally: the edit box is pink and unlocked, and there are "save"/"preview"/"changes" buttons, unlike when I try to edit a fully protected page. SiBr4 (talk) 13:39, 13 October 2015 (UTC)
Phabricator would be the place for this, yes. Checking the database, what's going on is basically what you guessed: the deletion of the cascade-protected page should have deleted the protection entry (and other links table entries), but it did not. Unfortunately undeleting the page would have no effect, since the links table entries are tied using the page_id value and undeleting assigns a new page_id rather than reusing the old page_id (T60986 seems to be commenting on that situation). I've fixed this specific issue by deleting the problematic database row, and I don't see any others on enwiki. See T115586 for tracking the underlying cause of the issue. BJorsch (WMF) (talk) 13:08, 15 October 2015 (UTC)

Apostroph problem

Not sure if this is the correct place to report this. I noticed a redlink in the second section of the HMS Chevron article. I copied the word (Ma’apilim) into the searchbox, and from there created a new article and made it a redirect to where other versions of this word, such as Maapilim go. It's still red. Apparently that specific apostrophe (’) gets substituted to another kind (') from search but not from links in articles. No More Mr Nice Guy (talk) 17:01, 14 October 2015 (UTC)

The best way to create a new article from a red link is to click on it. No need to use search. Ruslik_Zero 18:13, 14 October 2015 (UTC)
Generally, we should use straight apostrophes instead of curly ones. You should correct the text in the article. But in my opinion, it would make sense if the software treated straight and curly marks as the same for linking. I think it's a similar problem for capitals. Both cases are treated the same in the search, but in links they're not unless they're at the start of the word (or at least, that's how it used to be). McLerristarr | Mclay1 18:14, 14 October 2015 (UTC)
@No More Mr Nice Guy: I cannot reproduce this in Firefox. If I enter Ma’apilim in the search box and choose "containing..." (which is now needed to avoid the redirect you created at Ma'apilim) then I get a search results page [7] saying: You may create the page "Ma’apilim". Clicking the red link leads me to a page creation window with the curly apostrophe and not to the redirect you created. What is your browser and what do you see at [8]? PrimeHunter (talk) 22:27, 14 October 2015 (UTC)
I was using Chrome but I get the same behavior in Firefox. If I hit "containing..." I get the same results as you describe. If I hit enter it takes me to the redirect with the substituted apostrophe. I could create another redirect that would solve the redlink problem, but this behavior is not what a user would expect, I think. No More Mr Nice Guy (talk) 22:54, 14 October 2015 (UTC)
I still don't understand how you created a redirect with a straight apostrophe after entering a title with a curly apostrophe in the search box. Let's try an example where neither version matches a current page name, like your original case before creating the redirect. If I enter Ma’ap (curly apostrophe) in the search box then I get: You may create the page "Ma’ap". If I click the red link then I go to https://en.wiki.x.io/w/index.php?title=Ma%E2%80%99ap&action=edit&redlink=1 with a curly apostrophe %E2%80%99 in the url. If I enter Ma'ap (straight apostrophe) in the search box then I get: You may create the page "Ma'ap". If I click the red link then I go to https://en.wiki.x.io/w/index.php?title=Ma%27ap&action=edit&redlink=1 with a straight apostrophe %27 in the url. Do you get the same result as me in both cases, or do you get the version with the straight apostrophe regardless whether you entered a curly or straight apostrophe in the search box? I have tested it in Chrome 46.0.2490.71 on Windows Vista and I get the same result as in Firefox. PrimeHunter (talk) 23:14, 14 October 2015 (UTC)
I get the same result. I don't remember what I did, to be honest. It's possible I didn't notice the curliness at first, I guess, so I typed it in rather than copied it? I'm fairly but not 100% sure I copied it. No More Mr Nice Guy (talk) 23:56, 14 October 2015 (UTC)

The Article Traffic Statistics tool need to be updated badly

Hi my name is SeminoleNation and I have been wondering if Wikipedia plans on implementing a stable way of looking at article traffic statistics. The current http://stats.grok.se/ link is very unstable and frequently crashes and will lose information of multiple days at a time. Thank you--SeminoleNation (talk) 21:51, 14 October 2015 (UTC)

Hmm, Analytics folks might be the best to ask. You could ask on the Analytics mailing list for example. --AKlapper (WMF) (talk) 10:21, 15 October 2015 (UTC)

Pages without JS or CSS via Google

Some of you might be interested in this: Especially for slow connections, Google sometimes sends a stripped-down version of a webpage to users, with limited Javascript and CSS. You can see what it looks like here with the article about Oxygen: https://icl.googleusercontent.com/?lite_url=http://en.m.wiki.x.io/wiki/Oxygen

I believe this only happens on the mobile site, but I'm not certain of that. Whatamidoing (WMF) (talk) 00:08, 15 October 2015 (UTC)

Searching for a template following a string

I tried asking a question at AWB tasks, but that may not be the best place (or maybe it was TLDR).

I want to use AWB to improve some references. I can find each of the references because they are all named references, but the field to be replaced is not necessarily unique, mainly because the access date might vary. My goal is to search for a particular named reference, then locate the immediately following cite web template, and replace the contents of that template with a value.

I'm not quite sure how to do the specification in AWB (or if there is some better way to do it).

I've started the creation of a table to identify the replacement strings here (The table currently has only five entries but there will be dozens when I complete it).

Can someone tell me if it is possible to search for a template following a string, or if there is a better way to approach this?--S Philbrick(Talk) 13:50, 14 October 2015 (UTC)

Is the opening reference tag always in the form <ref name="name">? Are you replacing the {{cite web}} template in toto? then:
find:
(\<ref\s+name\s*=\s*"name"\s*\>)\{\{\s*cite\s+web[^\}]+\}\}
where "name" is the reference name
replace:
$1{{cite web |...}}
where ... is the full content of replacement {{cite web}} template
No doubt, this could be optimize but should should work as a starting point. If the {{cite web}} template contains another template, this regex will fail.
Trappist the monk (talk) 14:06, 14 October 2015 (UTC)
It's better to use [^\{\}]+ instead of [^\}]+ in cases like this. That way, if the template you're replacing does contain a nested template, the regex doesn't do anything, instead of replacing the wrong things and breaking stuff. SiBr4 (talk) 14:19, 14 October 2015 (UTC)
Thanks for the very quick responses. Yes, the opening reference tag is always in that form. Yes, I will be replacing the template in toto. I do not believe I have any nested templates but I will either use the second suggestion for code or watch carefully.--S Philbrick(Talk) 14:44, 14 October 2015 (UTC)
I tried it and it worked (at least in the first case). Thanks.--S Philbrick(Talk) 15:00, 14 October 2015 (UTC)

Good news, bad news

@Trappist the monk:

The good news is when I set up AWB and ran against the list of World University Game references I got over 50 hits which is about what I expected.

The bad news is that when I tried to repeat for Pan Am games I got only two hits, while expecting over 50. If I had zero hits I would assume I set up something fundamentally wrong but I get two which suggests the format is correct.

An example of my search for string is:

(\<ref\s+name\s*=\s*"2003 Pan Am"\s*\>)\{\{\s*cite\s+web[^\}]+\}\}

The replacement string is:

$1{{cite web|title=Fourteenth Pan American Games -- 2003|date=Feb 20, 2014|url=http://www.usab.com/history/pan-am-womens/fourteenth-pan-american-games-2003.aspx|publisher=USA Basketball| archive-url =https://web.archive.org/web/20150907195503/http://www.usab.com/history/pan-am-womens/fourteenth-pan-american-games-2003.aspx| archive-date =7 September 2015|dead-url=no|accessdate=15 Oct 2015}}

An example of the page in the list is Rebekkah Brunson; the relevant reference is:

<ref name="2003 Pan Am">{{cite web|last=|first=|date=|title=FOURTEENTH PAN AMERICAN GAMES -- 2003|url=http://www.usab.com/womens/panamerican/wpag_2003.html|work=|publisher=USA Basketball|accessdate=15 Oct 2013}}</ref>

I am a REGEX newbie, but I'm not seeing why that search should not find this string.--S Philbrick(Talk) 17:38, 15 October 2015 (UTC)

When I put your search and replacement strings into AWB and have it search Rebekkah Brunson, it finds and replaces the middle reference in §References (I did not save). If it isn't broken, I can't fix it. Have you inadvertently set something that you should not have set on the Skip tab? What about the case sensitivity checkbox in Options -> Normal settings?
Trappist the monk (talk) 19:45, 15 October 2015 (UTC)
@Trappist the monk:I'm surely not the first person to ask you to fix something that wasn't broken :) seriously, it is probably user error. I'm encouraged that it worked for you, but I didn't think I did anything unusual in terms of settings.
I took a screenshot of my skip settings File:Temp_AWB_Pan_Am_issues_-_skip_page.JPG and my normal S&R settings File:Temp AWB Pan Am issues - normal settings page.JPG. Does that suggest anything?--S Philbrick(Talk) 20:18, 15 October 2015 (UTC)
Sorry, but I think that I'm at a loss to explain why it works for me and not for you. The rule, as you have typed it here, works for me. Was that a direct copy/paste from AWB? Have you tried putting the rule and what it is supposed to find in the AWB Regex tester (Tools->Regex tester)? If that works then the problem is somewhere in your settings. You might save a copy of your settings (File->Save settings [as...]) to a sandbox page so that others can try to duplicate your problem using your settings.
Trappist the monk (talk) 21:49, 15 October 2015 (UTC)
Thanks for your responsiveness. I don't recall whether I copied the rules from AWB or from my sandbox. I'll try using the tester you mentioned. I'll also try starting clean and only using one rule.--S Philbrick(Talk) 21:56, 15 October 2015 (UTC)
  Resolved
@Trappist the monk:I figured out the problem. Some of my "search for" strings had a space character at the end of the string. I guess that means it looks for the citation template followed by a space, which happened only twice. The fact that you could get it to work was a helpful clue. Thanks for looking into this. I now have a few hundred edits to do. Yaay.--S Philbrick(Talk) 16:06, 16 October 2015 (UTC)

Why do all pages look blank?

Whenever I go to any page, it looks like there is no content in the page at all (i.e. it looks like a blank page). However, when I click "Edit", I see the wikitext. What happened? Gparyani (talk) 18:06, 14 October 2015 (UTC)

Something went wrong with the weekly software update. It was immediately reverted and stuff should get back to normal now. —TheDJ (talkcontribs) 18:08, 14 October 2015 (UTC)
I saw this for a moment: File:Missing_text.PNG. I also saw "Due to high database lag, changes newer than 170 seconds may not appear in this list" on Special:MyContributions. Gparyani (talk) 18:10, 14 October 2015 (UTC)
Seems to be effecting User pages in particular. Blethering Scot 18:11, 14 October 2015 (UTC)
A few minutes ago, I started having the same problem in Google Chrome with the idea lab, which I didn't have an hour earlier. I actually posted this answer from Mozilla Firefox. Blackbombchu (talk) 18:13, 14 October 2015 (UTC)
That problem has already ended for me. Blackbombchu (talk) 18:17, 14 October 2015 (UTC)
A significant number of pages have been cached as blank during the outage. A WP:PURGE will get it back to normal. I've had to do a revert of a vandalism bot once for this, too. Mamyles (talk) 18:15, 14 October 2015 (UTC)
(edit conflict) David Godman seems to be affected...was about to use CSD A3 but stopped. Gparyani (talk) 18:16, 14 October 2015 (UTC)
I've just purged that page. Should be back up. Mamyles (talk) 18:16, 14 October 2015 (UTC)
Thanks for reporting this. This is currently being handled and the corresponding bug report is phab:T115505. As more information on the reasons becomes available that bug report will receive updates. Sorry for the inconvenience! --AKlapper (WMF) (talk) 18:17, 14 October 2015 (UTC)
It appears to be affecting every page that has been updated since the problem started. Belarusian presidential election, 2015, EMC Corporation, and Roses Are Red (My Love) are all affected, just to name a few. Portal:Current events also appears completely blank for me.--Tdl1060 (talk) 18:23, 14 October 2015 (UTC)

Follow-up: there is now an incident report. the wub "?!" 11:23, 17 October 2015 (UTC)

Phabricator request, please

 

Would someone please file a Phabricator request and mark it as being of the lowest priority? I've never wandered over there. As you can see in the image, it's possible for the letter at the top of a section to be at the bottom of a column when the rest of the section's on a different column; I'm basically asking for a top-section letter always to be in the same column as the first article link within that section. If this were implemented, the "L" would always be immediately above Lawler's Tavern (not below and to the left), and the "P" would always be with Peninsula Village Historic District. Nyttend (talk) 14:38, 16 October 2015 (UTC)

PS, I'm in the middle of moving a lot of the articles in the pictured category to a subcategory, so things will look different from in the screenshot, although perhaps you can create the same appearance by de-maximizing your browser window. Nyttend (talk) 14:41, 16 October 2015 (UTC)

As it happens, we just published a blogpost on the Wikimedia Blog documenting the process. Perhaps that would be useful? JSutherland (WMF) (talk) 14:57, 16 October 2015 (UTC)
First section here. --Edgars2007 (talk/contribs) 16:37, 16 October 2015 (UTC)
This is a known issue, that cannot be fixed right now, because browsers don't allow us to control this behavior yet... See phab:T104541TheDJ (talkcontribs) 18:39, 16 October 2015 (UTC)
@Nyttend: Please feel encouraged to wander over there - everybody is welcome to report a bug and it's not complicated. :) --AKlapper (WMF) (talk) 15:23, 17 October 2015 (UTC)
Thanks, but no point if this is a RESOLVED CANTFIX or whatever :-) Nyttend (talk) 21:48, 17 October 2015 (UTC)

Blanked articles

Lately, I've been coming across articles which have been appearing blanked (see Gorseddau Tramway), even though they haven't been blanked. Any idea, what's up with this? GoodDay (talk) 14:22, 17 October 2015 (UTC)

#Why do all pages look blank?. --Edgars2007 (talk/contribs) 14:27, 17 October 2015 (UTC)
Here's one, I had to fix up. Also, had to fix this. -- GoodDay (talk) 14:31, 17 October 2015 (UTC)
As said at #Why do all pages look blank?, you only have to purge affected pages. Any edit will also do. PrimeHunter (talk) 14:48, 17 October 2015 (UTC)
@PrimeHunter: The Incident Report says all affected pages have been auto-purged. Browser cache issue? - NQ-Alt (talk) 14:51, 17 October 2015 (UTC)
It says "Ori deploys hook to purge blank pages". I don't know whether that was supposed to purge all blank pages and be done by now. The three pages reported here had not been edited for months. I haven't seen blank pages myself for a while. PrimeHunter (talk) 15:07, 17 October 2015 (UTC)
Yes, it should be done by now. If you run into a blank page, can you give me a link and NOT PURGE it? Legoktm (talk) 17:43, 17 October 2015 (UTC)

Ramanaw mallam

Trying to rename this to Ramanaw Mallam but not allowed, suspect someone's circumvented something here...GrahamHardy (talk) 15:04, 17 October 2015 (UTC)

@GrahamHardy: Salted by Yunshui. You can request removal at WP:RFPP - NQ-Alt (talk) 15:09, 17 October 2015 (UTC)
I have now moved it to Ramanaw Mallam without redirect. De728631 (talk) 15:10, 17 October 2015 (UTC)

Disable Visual editor globally

Wikipedia:Help desk#3 questions

mw.user.options.set('visualeditor-enable',0); in global.js does not seem to disable it anymore. Any ideas? - NQ-Alt (talk) 17:27, 17 October 2015 (UTC)

Template help needed (I think)

Please see User_talk:NeilN#Infobox_mystery. --NeilN talk to me 18:09, 17 October 2015 (UTC)

Resolved by SiBr4 and NQ. Thanks! --NeilN talk to me 19:00, 17 October 2015 (UTC)

Automated archiving fails

Can anyone see why, for example, Talk:Bengalis, Talk:Crimean status referendum, 2014 and Talk:Charlotte's web (cannabis) are not being automatically archived? they looks to me like they are set up correctly. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:40, 17 October 2015 (UTC)

@Pigsonthewing: Archive locations changed as a result of subsequent page moves. Hopefully fixed now. [9] [10] [11]- NQ-Alt (talk) 19:15, 17 October 2015 (UTC)
@NQ-Alt: Good catch, thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:23, 17 October 2015 (UTC)

Long pages data needs updating

In the absence of User:Snaevar, how can we get Wikipedia:Database reports/Long pages updated? the data there is now around two months old, and very inaccurate. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:19, 15 October 2015 (UTC)

Here you are. Should be the same criteria. If (when) you need an update, just say. --Edgars2007 (talk/contribs) 15:01, 15 October 2015 (UTC)
@Edgars2007: Just the job; thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:20, 17 October 2015 (UTC)
@Edgars2007: I've set up automated arching on most of those pages, which has now run. Could you regenerate the stats, please? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:12, 18 October 2015 (UTC)
  Done. --Edgars2007 (talk/contribs) 11:26, 18 October 2015 (UTC)

Twinkle broken?

In the last view days, I've noticed that Twinkle is mostly broken. I've looked the the archives and found a discussion, but the "solution", turning off Content Translation, did not work. In fact, Content Translation was already turned off in my preferences. I've already cleared my JavaScript file and use [Wikipedia:Twinkle/Preferences], which I haven't done until today. I've also reset my preferences and turned Twinkle back on, but nothing works. TheFarix (talk) 12:46, 18 October 2015 (UTC)

Seems to be working now. Not sure what was going on. —Farix (t | c) 12:52, 18 October 2015 (UTC)

Block lengths

An administrator issued a block today, for 59 days. Today being the 19th of a 31 - day month the 19th of November would be 31 days hence, and 28 days further on, making up the 59 days, would bring us to 17 December, since November has 30 days. However, the block notice displayed on - screen when the editor attempts to edit says the block will expire on 18 December. Why the 24 - hour difference? The time at which the block expires is the same time as when the administrator made it.

Also, some administrators do funny blocks such as 2 years, 3 months, 4 hours, 5 minutes, 6 seconds, an orange and a lemon. How does the software handle those? 62.140.210.135 (talk) 15:43, 18 October 2015 (UTC)

Sorry, it's the 18th. Explained. But I'm still interested in how the odd blocks mentioned are handled. 62.140.210.135 (talk) 15:45, 18 October 2015 (UTC)

How would you know what the block notice says when the blocked editor tries to edit? - NQ-Alt (talk) 15:59, 18 October 2015 (UTC)
Based on what the OP has said it is clear they are User:Vote (X) for Change. I have blocked them for block evasion. HighInBC 16:09, 18 October 2015 (UTC)

Where can I go for coding help?

I'm having trouble coding a template for a wiki that uses wikimedia-based markup. All the people I usually ask for help either cannot solve my problem, or are too busy to answer. Where is the appropriate place to go to ask for help with my specific problem?

Thanks, Iustinus (talk) 19:51, 18 October 2015 (UTC)

@Iustinus: Do you mean it is a Wikimedia wiki, i.e. a wiki run by the Wikimedia Foundation which runs Wikipedia? Or is it just a wiki using the same MediaWiki software as wikis of the Wikimedia Foundation? There is a general help desk for MediaWiki at mw:Project:Support desk. At Wikipedia:Help desk you may sometimes find help but can also be told it's only for Wikipedia. We do have Wikipedia:Reference desk/Computing which is for anything computer related. PrimeHunter (talk) 20:14, 18 October 2015 (UTC)

16:02, 19 October 2015 (UTC)

Bot borked

Cross posted from here:

Hey, this edit failed to add the nomination link to the article FAC to the talk page, for some reason. I'm not sure how to fix it either, can you take a look? I need it to work so that the Signpost's weekly compilation script works. ResMar 16:27, 19 October 2015 (UTC)

Blocked!

Hi all... I just spent about an hour trying to figure out why I had been blocked from editing. It was apparently an IP address block. I couldn't even edit my own talk page. So eventually, I had a brainstorm and rebooted my computer, DSL modem and router, and now I'm back. I'd like to know how to keep this from happening. Is there a FAQ about it? Tfr000 (talk) 22:55, 17 October 2015 (UTC)

Wikipedia:IPEXEMPTCONDITIONS NE Ent 23:23, 17 October 2015 (UTC)
Ok, I guess I'm looking for something more than that. The hour was spent trying to append some tag to my talk page (blocked from editing) and filling out some arbitration form (it insisted that I was not blocked, and it would not submit itself). I looked for something else, an email address maybe, and found nothing. What I'm saying is, I followed all the procedures and was defeated. "Unblock Ticket Request Systems" which refuse to help aren't very useful. Tfr000 (talk) 12:22, 18 October 2015 (UTC)
Maybe Help:I have been blocked (this is linked at the bottom of Mediawiki:Blockedtext, the message you see when you try to edit while blocked). In your situation, the IRC channels, pointed to on that page, would probably have been the fastest way to get help. Tip: when you have a problem, copy any messages you get, or take screenshots. It's hard for people to figure out what happened after the fact if you can't provide details. I'm not familiar with the workings of UTRS, but I presume it wouldn't let you submit a request because it didn't see any block against your account (because your account wasn't blocked; the IP address you happened to be using was). This seems to me like a pretty glaring shortfall in UTRS for people accidentally affected by blocks. I don't know if there are technical reasons behind it, but I would suggest someone look into changing that, given that UTRS is one of the places we point people to if they're blocked. --71.119.131.184 (talk) 07:45, 19 October 2015 (UTC)
Thanks, however, I'm either not finding or not noticing any IRC channels on either Help:I have been blocked or Mediawiki:Blockedtext. Tfr000 (talk) 22:14, 20 October 2015 (UTC)
Oh, sorry, it's in the last section, which is collapsed by default. If you click/tap/whatever "show" it'll appear, or you can just go to Wikipedia:IRC for a bunch more information on the IRC channels. Anyone else think the sections on Help:I have been blocked don't need to be collapsible? --71.119.131.184 (talk) 20:55, 21 October 2015 (UTC)

Pageview Stats down again

Pageview stats at http://stats.grok.se are down again. Stats have not compiled since October 11. Thus, the following is a summary the datefiles that are currently not compiled.

See update above.--TonyTheTiger (T / C / WP:FOUR / WP:CHICAGO / WP:WAWARD) 04:11, 15 October 2015 (UTC)

Page view statistics

Page view statistics only shows up to October 11, 2015. --Jax 0677 (talk) 16:30, 17 October 2015 (UTC)

It is still not working. Please repair it, it is now a week behind. Johnsmith2116 (talk) 23:51, 19 October 2015 (UTC)

Obviously this has become a critical issue for more than just a few users. Is there any work on addressing the issue? — Preceding unsigned comment added by Dstone1029 (talkcontribs) 14:27, 21 October 2015 (UTC)

Page view stats

It has been pointed out multiple times here that the page view stats function is down, and has not reported article views since October 11. User:TonyTheTiger says "updated" or "see thread above". What does that mean? The tool is still not working. Is someone trying to fix this? --MelanieN (talk) 15:08, 20 October 2015 (UTC)

When TonyTheTiger says "Updated" in #Pageview Stats down again, it just means he has updated his own list in that section of dates with no page view stats. Comments like "see thread above" just means there is a related thread earlier on the page. It doesn't imply that thread has a solution. The page view stats at http://stats.grok.se/ is an external tool made by a single volunteer editor User:Henrik who rarely replies to questions or makes edits to Wikipedia. I don't think anyone else knows when he is working on the tool. Note that the tool is not a feature in the MediaWiki software and is not run by the Wikimedia Foundation. The editors of the English Wikipedia have just agreed to add some links to it, notably in MediaWiki:Histlegend which is displayed at top of page histories. The tool uses publicly available data and others could make an alternative tool but I don't know any that work. There was once work on a tool at toollabs:wikiviewstats but I don't know whether there is current work to try to make it operational. PrimeHunter (talk) 15:42, 20 October 2015 (UTC)
Thanks for the explanation. This is such a valuable tool, it would be good if he shared its management with someone else - since nobody can be available all the time. --MelanieN (talk) 16:08, 20 October 2015 (UTC)
I think I read something about pageview rebuilding (which could be done by WMF) in some of Magnus Manske tools. But maybe I'm wrong. --Edgars2007 (talk/contribs) 16:12, 20 October 2015 (UTC)

Is not being able to view the stats to see how many views a DYK gets during its allotted slot on the main page a deliberate decision? Mine never get that many but I still liked to see how many views it got and how many others got too.  — Calvin999 17:29, 21 October 2015 (UTC)

So, when is this expected to be solved? Its a really important tool.--Makeandtoss (talk) 19:35, 21 October 2015 (UTC)

Page view statistics

This tool is not showing any data after October 12, why is that? --Makeandtoss (talk) 13:59, 21 October 2015 (UTC)

@Makeandtoss: #Pageview Stats down again - NQ-Alt (talk) 14:18, 21 October 2015 (UTC)
See the previous thread "Pageview Stats down again" on this page. --AKlapper (WMF) (talk) 14:40, 21 October 2015 (UTC)

spurious nbsp with infobox cyclist

There seems to be a bug in template:Infobox cyclist wherein if there is a template:Convert nested within, a search for nbsp gives a false positive. For example, a search for nbsp has hits for Annie Ewart, Alison Tetrick, Andrea Dvorak, Ariane Horbach, Richard Lamb, Lauren Komanski, Laura Van Gilder, Claudia Lichtenberg, Allie Dragoo, Sara Mustonen (cyclist), Joël Zangerlé, Chris Froome, and many many more articles with this issue. It is a minor bug, but if a Wikipedian searches for articles with spurious "nbsp" strings, this bug generates a lot of false positives. —Anomalocaris (talk) 16:44, 19 October 2015 (UTC)

nowiki doesn't stop the html character entity for nbsp so this post should be viewed in the wikitext and not the rendered page to see what I talk about. If {{Infobox cyclist|weight = 58 kg}} is entered at Special:ExpandTemplates then it produces code for the infobox followed by [[Category:Pages using infobox cyclist with atypical values for height or weight|W]]<span style="display:none;visibility:hidden;color:transparent;">58&nbsp;kg</span>. In this ending code,   is converted to &nbsp; by Module:Infobox cyclist tracking. I don't know Lua so this is as far as I got. PrimeHunter (talk) 17:58, 19 October 2015 (UTC)
That's mad code. I work on cycling articles a lot and I have never used those values. I can't work out why I would want to. Two things should be done. First, the code should recognise {{convert}}. Second, it should stop outputting invisible span elements for no apparent reason. (The code could also usefully be documented, of course...) Relentlessly (talk) 18:14, 19 October 2015 (UTC)
Module:Infobox cyclist tracking is made by User:Frietjes. I don't know the background. The issue happens for {{convert}} because it places a non-breaking space &nbsp; between number and unit in the output. This is intentional and in accordance with WP:UNIT. My simplified example shows it is this &nbsp; which is converted by Module:Infobox cyclist tracking to cause the reported issue. PrimeHunter (talk) 19:19, 19 October 2015 (UTC)
PrimeHunter: I don't understand where to find your simplified example, but, whatever the requirements of WP:UNIT, this problem is not showing up for other infoboxes such as Template:Infobox professional wrestler, Template:Infobox football biography, and Template:Infobox sportsperson. Interestingly, Template:Infobox golfer, Template:Infobox NFL player, and Template:Infobox ice hockey player all perform Metric/Imperial conversions without being asked. Why aren't these conversions exported to all human infoboxes having height/weight parameters? —Anomalocaris (talk) 21:08, 19 October 2015 (UTC)
My simplified example is {{Infobox cyclist|weight = 58 kg}} in my first post (see the wikitext to see it uses nbsp). I don't know other templates than {{Infobox cyclist}} (via code in Module:Infobox cyclist tracking) which adds non-displayed tracking code like this, but I haven't looked for it. Maybe Frietjes can say why it does it. PrimeHunter (talk) 21:21, 19 October 2015 (UTC)
PrimeHunter: I see what you mean. I created User:Anomalocaris/sandbox with your snip of code, and when I display the page source, in addition to everything one would expect, it generates the garbage line
<p><span style="display:none;visibility:hidden;color:transparent;">58&amp;nbsp;kg</span></p>
So if a &nbsp; character is passed to {{Infobox cyclist}}, it generates the non-displaying nbsp garbage that is found in a general search. But {{Infobox professional wrestler}} doesn't generate any such garbage line. —Anomalocaris (talk) 23:07, 19 October 2015 (UTC)
It's not &nbsp; specifically: any format that doesn't match what was expected by the module author will output broken HTML. That includes any use of {{convert}}. Relentlessly (talk) 23:14, 19 October 2015 (UTC)
Yes, and I don't know whether there is any parameter value that will avoid the hidden code at the end. The only thing special about &nbsp; is that the module converts it to &amp;nbsp;. That's normally a way to display the code for a html character entity in a html document instead of displaying the actual character. PrimeHunter (talk) 23:33, 19 October 2015 (UTC)
I removed the hidden code output, it was useful when we were cleaning up these articles, but since no one is cleaning them up right now, there is no need for the code right now. Frietjes (talk) 14:24, 20 October 2015 (UTC)

Read in another language button on mobile Wikipedia showing in Urdu

On my android phone, on Google chrome, when I'm logged in, if I scroll to the bottom of the page on the mobile site, the button that usually says Read in another language says ٻي ٻوليءَ ۾ پڙهو instead. I looked on Google translate and this is apparently Urdu, although it couldn't translate what it said. I haven't changed my language settings and it doesn't say this when I'm logged out, so I was just wandering why this is happening. Thanks.  Seagull123  Φ  19:37, 19 October 2015 (UTC)

It's in Sindhi language which Google Translate doesn't know. MediaWiki:Mobile-frontend-language-article-heading/sd displays it so it must mean "Read in another language" like MediaWiki:Mobile-frontend-language-article-heading. I don't know why you see it. PrimeHunter (talk) 20:09, 19 October 2015 (UTC)
@PrimeHunter: Thank you. Google translate isn't the most accurate translator out there is it? But it isn't really bothering me, I was just curious why randomly, it would use Sindhi language for just one part of the page. Thanks again.  Seagull123  Φ  22:55, 19 October 2015 (UTC)
I also use Google Translate. When it failed to translate I suspected that Urdu was a wrong guess so I did a Google search on the string. This revealed it was probably a language with code sd, and I found the string at MediaWiki:Mobile-frontend-language-article-heading/sd – which is a MediaWiki default and therefore not searchable in the MediaWiki namespace. I knew to look there because http://en.m.wiki.x.io/wiki/Example?uselang=qqx displays "(mobile-frontend-language-article-heading)" in the location where normal mobile pages say "Read in another language", at least for me. PrimeHunter (talk) 23:41, 19 October 2015 (UTC)

Strange breakage when editing a category

 

I went to Category:Immediate children/Tritylodontidae with the intention of adding {{db-catempty}}, but on clicking "edit this page", I got the screen as shown here. Ignore the pale blue backgrounds to two tabs: that's custom CSS because of my eyesight. But why is the puzzleball inside the border, why are the top menu and tabs pushed down, where are the sidebar links - in fact, why is it such a mess? And where does that text "The page that you are currently viewing contains information ..." (and the table below it) come from? --Redrose64 (talk) 21:08, 19 October 2015 (UTC)

I think, some editnorice contains {{Taxonomy key}}. --Edgars2007 (talk/contribs) 21:28, 19 October 2015 (UTC)
That's the actual Indeed. It comes from {{Immediate children category}} due to {{Editnotices/Namespace/Category}}. Template talk:Automatic taxobox might be a place to discuss its layout. DMacks (talk) 22:07, 19 October 2015 (UTC)
I was previewing the fix when Legoktm beat me to the save.[21] PrimeHunter (talk) 22:37, 19 October 2015 (UTC)

Favicon

In Internet Explorer 11 in Windows 10, visited Wikipedia pages show a piece of paper with a folded corner rather than the Wikipedia favicon in the browsing history. GeoffreyT2000 (talk) 22:22, 19 October 2015 (UTC)

Is this happening for anyone else in IE11 on Windows 10? Whatamidoing (WMF) (talk) 16:51, 22 October 2015 (UTC)
It's working fine for me. Rchard2scout (talk) 10:57, 31 October 2015 (UTC)

Category:Pages with incorrect ref formatting

Category:Pages with incorrect ref formatting normal contents has fewer than 300 entries, of which mostly are drafts which I ignore. Yesterday it suddenly started lots of entries, and they are articles which haven't had edits for weeks. I think it's still going up – current 8,700+ entries. What happens? Unbuttered parsnip (talk) mytime= Tue 17:08, wikitime= 09:08, 20 October 2015 (UTC)

Is this due to phab:T114898, perhaps? Relentlessly (talk) 09:18, 20 October 2015 (UTC)
It happened since yesterday (Oct 19). Since working – current now 9200–ish Unbuttered parsnip (talk) mytime= Tue 18:18, wikitime= 10:18, 20 October 2015 (UTC)
The category may have started being filled since yesterday but around the 9 October I noticed articles that hadn't previously been displaying ref errors were suddenly adorned with red warnings? I know there are other articles that are not (yet) appearing in that category that still have errors showing. SagaciousPhil - Chat 10:30, 20 October 2015 (UTC)
Most probably it's the new cite error. See also here. Edgars2007 (talk/contribs) 10:41, 20 October 2015 (UTC)
I'll wait. Instead I'll use wikidata. Oh no, that's broken too! I'll go for a shower then lunch! Unbuttered parsnip (talk) mytime= Tue 18:54, wikitime= 10:54, 20 October 2015 (UTC)
I don't know what you usually do with the category but you could consider working on the new cite error. The error is supposed to say:
The named reference "$1" was defined multiple times with different content (see the help page).
However, the code for "Display 'cite_error_references_duplicate_key' next to the affected ref" at phab:T114898 moved the error message to a place where the help link is not displayed for some reason. Can somebody help with that? To see the difference, compare cache:https://en.wiki.x.io/wiki/A._P._J._Abdul_Kalam#References (currently cached 19. Oct 2015 17:06:55 GMT, error message displayed after references and has help link) to the current rendering of that revision [22] (error message displayed at reference 12 without help link). The message is made by MediaWiki:Cite error references duplicate key which calls {{Broken ref}}. The help link is added in Template:Broken ref/lang. It works for other cite error messages, for example MediaWiki:Cite error references no text in [23]. Maybe it should be reported at phab. PrimeHunter (talk) 13:04, 20 October 2015 (UTC)
I try to mend Category:Pages with incorrect ref formatting entries. If there a couple of dozen a day, then I try to mend then. If 10,800+ then I wouldn't if try.
Error message is Cite error: Invalid <ref> tag; name "$1" defined multiple times with different content (see the ). I.e. "$1" works, but end is (see the ). without any parameter. Unbuttered parsnip (talk) mytime= Tue 21:37, wikitime= 13:37, 20 October 2015 (UTC)
The problem is not about omission of a parameter but about stripping of wikilinks. For testing I have added a simple unpiped link directly in the message [24]. The added code says (see [[Help:Cite errors/Cite error references duplicate key]]), but the rendering of a page built after the change [25] only says "(see )" in that location. There is no sign of the link or link text in the html of the rendered page which just says (see ). PrimeHunter (talk) 14:27, 20 October 2015 (UTC)
The error message is accidentally being parsed twice (or perhaps once-and-a-half), which causes this behavior. Sorry about it. phab:T116149. Matma Rex talk 12:02, 21 October 2015 (UTC)
This should be resolved now. Pages might continue to display the broken message until they are purged. Matma Rex talk 15:27, 21 October 2015 (UTC)
Thanks for the fast fix. PrimeHunter (talk) 18:31, 21 October 2015 (UTC)
Does the 32,000+ entries in Category:Pages with incorrect ref formatting going to get them remove? If it takes me say 30 a day, it will take me more until my 100th birthday to clear, if I last that long. Unbuttered parsnip (talk) mytime= Fri 04:55, wikitime= 20:55, 22 October 2015 (UTC)
Now that there is a way to track the errors, I hope somebody will make a tool to help fix them. In most cases I have examined manually, it is the same references with small variations in the reference text. Those references should be merged. It may be hard for a bot to decide but I would like a semi-automated tool where you see two reference texts with the same name and can decide whether to merge using the first text, merge using the second text, or assign different ref names to them. Sometimes there are more than two reference texts with the same name and it may get more cimplicated. PrimeHunter (talk) 22:00, 22 October 2015 (UTC)
If the ref name is invoked and two definitions are given different names then there is also the problem of which version to use in invocations. PrimeHunter (talk) 22:09, 22 October 2015 (UTC)
What I've seen is that, frequently, the cause of this in articles using sfn is an sfn template missing p= for one page reference. For example, a single {{sfn|Solomon|2011|2}} when other references are properly formatted like {{sfn|Solomon|2011|p=3}}. Squeamish Ossifrage (talk) 17:43, 28 October 2015 (UTC)
If, as with your examples, one were {{sfn|Solomon|2011|2}} and the other was {{sfn|Solomon|2011|p=3}} there would be no "Cite error" message,[1][2]

References

  1. ^ Solomon, 2011 & 2.
  2. ^ Solomon 2011, p. 3.
since the emitted ref names are different. If, however, the second one was {{sfn|Solomon|2011|p=2}}, they would then generate the same ref name, and so you would see an error message. --Redrose64 (talk) 08:38, 29 October 2015 (UTC)

I want to reuse a photo in another Wikipedia article

The photo appears in Flat Rock Historic District but when I copy from the infobox exactly to St. John in the Wilderness (Flat Rock, North Carolina), or at least when I believe I do, I get "invalid designation".— Vchimpanzee • talk • contributions • 14:07, 20 October 2015 (UTC)

  Fixed The left-hand vertical bar on the template was missing, possible during the copy. It's fine now. — Maile (talk) 14:14, 20 October 2015 (UTC)
Thanks.— Vchimpanzee • talk • contributions • 14:15, 20 October 2015 (UTC)

Patrolling my new creations

About a year ago, I created Forty-fifth State and several dozen similar redirects, and ONUnicorn has helpfully patrolled most or all of them in the last few minutes. Why is patrolling possible for these pages? I'm an administrator, and I was an administrator when these pages were created; aren't my creations autopatrolled? Nyttend (talk) 16:14, 20 October 2015 (UTC)

Someone has nominated all those redirects for deletion. For some odd reason, whenever a redirect is nominated for deletion, it shows up in the new pages feed as an unpatrolled new article. I occasionally go through the new pages feed and mark them all as patrolled just to get them off the feed so that actual new articles are easier to find. You may want to comment on the RFD discussion. ~ ONUnicorn(Talk|Contribs)problem solving 16:17, 20 October 2015 (UTC)
You already did comment I see. So never mind that part. But yeah; changing a redirect into an article, DAB page, or nominating it for deletion adds it to the new pages feed. Annoyingly, it adds it as of the date it was originally created, so those of us who work from the back of the feed forward find ourselves dealing with former redirects first instead of the oldest actual new articles. It's quick and easy to go through and check patrolled on all the redirects that have been nominated for deletion and makes it easier to see what's actually on the new pages feed that needs attention, and that's what I was doing. ~ ONUnicorn(Talk|Contribs)problem solving 16:20, 20 October 2015 (UTC)
See phab:T92621: "Implement addition of un-redirected pages to Special:NewPages and Special:NewPagesFeed". When we nominate a redirect for deletion, we break the redirect. The software doesn't care and would have a hard time telling exactly what happened to a redirect. It just sees it's no longer a redirect. PrimeHunter (talk) 16:36, 20 October 2015 (UTC)

This isn't really a bug. It's to stop vandals/COI pushers getting around new page patrol by creating redirects and then editing them into articles (or worse, editing other people's redirects into articles). --ais523 20:59, 20 October 2015 (UTC)

Blacklisted…

Website [now hidden in the code screen...] is blacklisted, at least since this edit in 2013. I’d like to find out, why. But the website is not mentioned on MediaWiki:Spam-blacklist. I’ve tried searching for it on Wikimedia’s Spam blacklist but that list seems not organized in any way so it is quite impossible to search in it—especially if you also want to find out exactly when (and why) a specific website was added to that list. --Corriebertus (talk) 09:11, 21 October 2015 (UTC)

@Corriebertus: MediaWiki talk:Spam-blacklist#www.kavkazcenter.com - NQ-Alt (talk) 09:20, 21 October 2015 (UTC)
Thanks. --Corriebertus (talk) 10:32, 21 October 2015 (UTC)

Give out Deletion to Quality Awards and log at Hall of Fame

Please see Wikipedia:Bot_requests#Give_out_Deletion_to_Quality_Awards.

A one-time-run would be totally acceptable here.

Is there any way either a bot or someone with a user script or automated or semi-automated skills, can help out here ?

Thank you,

Cirt (talk) 09:22, 21 October 2015 (UTC)

Can anyone help me out with above, please? — Cirt (talk) 07:59, 29 October 2015 (UTC)

Urgent request: unblock IP address for training session

Hello! I am doing a training session at the Africa School of Excellence in Tsakane in Gauteng, South Africa. We are about to get a few children to register from this IP address. Please can someone unlock the IP restriction for the next 5 hours. You can block it again thereafter. I have done this before, but I am not sure of the correct procedure. Thanks!! Isla Haddow (talk) 06:14, 22 October 2015 (UTC)

You will have to tell us the IP address, or the block message that comes up. The correct procedure is probably to request at WP:AN. Graeme Bartlett (talk) 07:07, 22 October 2015 (UTC)
Yes, it's not a VPT matter. --Redrose64 (talk) 07:44, 22 October 2015 (UTC)

Thanks! I have asked at the admin page. Here is the IP address just in case: 41.147.55.76 Isla Haddow (talk) 07:54, 22 October 2015 (UTC)

@Islahaddow: I am not sure which "IP restriction" this refers to. Account creation throttle? Or is editing blocked from that IP in general? If this is about account creation, see meta:Mass account creation#Requesting temporary lift of IP cap. If the event is about to start, you could go to the IRC channel #wikimedia-tech on Freenode IRC and ask for help (general information on using IRC). --AKlapper (WMF) (talk) 12:19, 22 October 2015 (UTC)
It does seen unnecessarily complex- and leaves the trainer with a room full of excited individuals, and can't start because they can't save. The start of a training session usually starts by sorting out incompatible browsers so the trainer already has a credibility problem. Can we apply our minds to making the system more trainer friendly while keeping it secure. I am thinking of two scenarios- the prepared and publicised session, and the impromptu session where an average user just says to guys in the office- '..Look Wikipedia is really easy, we'ĺl pop over to Wetherspoons and I'll show you...'. -- Clem Rutter (talk) 12:40, 22 October 2015 (UTC)
@ClemRutter: "Sorting out incompatible browsers" sounds interesting, would you have an example (browser, browser version, a functionality/action that did not work as expected) to share? As Wikimedia sites run the MediaWiki software in its latest version, mw:Compatibility#Browser support matrix lists browser support levels. --AKlapper (WMF) (talk) 11:47, 23 October 2015 (UTC)
Always IE and probably 7- the point is on a training course the host will bring out a collection of 'retired' laptops from the store- these will have an have been set up by their tech support with a corporate software image years ago. With austerity- nothing is thrown out, and IT support is a thing of the past. A rule of thumb is: If the page title of an assessed article appears in Black- then there is no Javascript- and the editor will not work fully, specifically I can't get at the Cite templates. So we quickly look for Opera, Firefox, Chrome-- and assess what level of functionality is available there. Then we make do. Does that help? -- Clem Rutter (talk) 12:18, 23 October 2015 (UTC)

Thanks for your thoughts and help ... I agree Clem Rutter, the process does seem unnecessarily complex and opaque, and ultimately frustrating. I would be very keen for there to be a clear process that leads from all the "training" resources, so that the scrabbling doesn't have to happen when you are actually at the place, and can identify the IP that needs unblocking. AKlapper (WMF), the IP restriction was for mass account creation - and I couldn't find that meta page, not matter how hard I tried. But I have it now, and will use it in future. Thanks!! Isla Haddow (talk) 07:32, 23 October 2015 (UTC)

Citation error

Back in February, this edit created a citation error in Assassination of John F. Kennedy: "Cite error: Invalid <ref> tag; name "FOOTNOTEStokes19792" defined multiple times with different content (see the help page)." The current version still contains the error. It appears to me that various citation were moved around but not changed, so I cannot understand what created the error. I don't even see "FOOTNOTEStokes19792" in the article as a reference name. Thoughts? Thanks! - Location (talk) 13:41, 22 October 2015 (UTC)

Those errors arose because there were two version of the {{sfn}} template pointing differently to the same page in the source:
{{sfn|Stokes|1979|p=2}}
{{sfn|Stokes|1979|pp=2}}
which get translated to this:
<ref name="FOOTNOTEStokes19792">[[#CITEREFStokes1979|Stokes 1979]], p. 2.</ref>
<ref name="FOOTNOTEStokes19792">[[#CITEREFStokes1979|Stokes 1979]], pp. 2.</ref>
Same name, different content.
Trappist the monk (talk) 13:53, 22 October 2015 (UTC)
I have added this case to the help page.[26] PrimeHunter (talk) 14:18, 22 October 2015 (UTC)
Thanks again! - Location (talk) 15:27, 22 October 2015 (UTC)
This bug persisted at LNWR Lady of the Lake Class from 29 May 2014 until just now, through 10 edits by 6 unique registered editors, and none of them solved it until now. Editors who apparently didn't know how to fix it include User:Redrose64, an Administrator who now ranks 201 in Wikipedia:List of Wikipedians by number of edits, and User:Peter Horn, an Autopatrolled Reviewer who now ranks 629 in the same list. I studied the problem for a long time and was completely stumped, and was going to report it on this page when by coincidence User:Location happened to report the same bug. User:PrimeHunter, thank you for adding to the help page, but that is not an adequate solution, because from the article that brought me here we know this bug stumps major wikipedians. The real solution is perhaps that |p= and |pp= should be name-mangled differently in constructing ref names. —Anomalocaris (talk) 06:55, 23 October 2015 (UTC)
@Anomalocaris: It is not true that "this bug persisted at LNWR Lady of the Lake Class from 29 May 2014 until just now, through 10 edits by 6 unique registered editors", nor is it true that I "apparently didn't know how to fix it". This is because the MediaWiki software did not have means for detecting such problems until very recently, no earlier than 8 October 2015, and certainly no later than 04:41, 13 October 2015 - I think that PrimeHunter (talk · contribs) knows when it actually went live. People didn't start talking about it at Help talk:Cite errors until until 19:52, 18 October 2015 (UTC) Only two edits have been made to that article since then - both by the same person. Similarly with Assassination of John F. Kennedy, where the first edit that was not vandalism (or reversion of that) since the error message went live was on 15 October 2015. --Redrose64 (talk) 08:47, 23 October 2015 (UTC)
There are more details at phab:T114898. The patch was merged on 18 October, according to the Phabricator task. — Mr. Stradivarius ♪ talk ♪ 09:13, 23 October 2015 (UTC)
That Phab task concerns pages that lack a <references /> tag. This thread is about pages with a <references /> tag, where the message has been displayed since no later than 04:41, 13 October 2015. --Redrose64 (talk) 09:17, 23 October 2015 (UTC)
The relevant task is phab:T85386 - NQ-Alt (talk) 09:18, 23 October 2015 (UTC)
Ah yes, you're right. — Mr. Stradivarius ♪ talk ♪ 09:25, 23 October 2015 (UTC)
Judging by what Matma Rex (talk · contribs) wrote on Wed, Oct 7, 2:54 PM, this went live with MediaWiki 1.27.0-wmf.2 which was rolled out to English Wikipedia on Thursday, 8 October 2015 - just over two weeks ago. --Redrose64 (talk) 09:48, 23 October 2015 (UTC)
@Redrose64: Thank you for the clarification on when the bug was introduced. I still believe it should be a high priority to fix the bug and not just warn people about it. —Anomalocaris (talk) 15:44, 23 October 2015 (UTC)
Anomalocaris, if I understand correctly, what happened was not the "introduction of a bug" but rather the introduction of detection logic for existing errors or at best dangerous practices. (If a two refs define cites use the same name but neither is ever referred to by name (no <ref name=ABC /> is present) then no actual problem occurs, but it is an error waiting to happen, and so an error is now reported for this case along with other cases where the same ref name is defined twice. This wiki-coding mistake was common but rarely detected as there was no easy way to notice it, only a through review of all named refs in the wiki-code would have found it. Now it produces a large red error message and the accumulated errors must be fixed one-by-one. I have already fixed a few. I don't see how any system-level code change could cure these long-existing wiki-coding errors. DES (talk) 16:11, 23 October 2015 (UTC)
User:DESiegel: I fixed LNWR Lady of the Lake Class by changing an occurrence of {{sfn|Nock|1952|p=47–48}} to {{sfn|Nock|1952|pp=47–48}} so that both forms would not exist together in the article. It is improper formatting to use |p= with multiple pages or |pp= with single pages, but that's beside the point. The bug is that if an article happens to have two {{sfn}} templates identical except that one uses |p= and one uses |pp=, a ludicrous, difficult-to-diagnose error message is generated. I fail to see how LNWR Lady of the Lake Class as it existed prior to my fix, and prior to the introduction of the bug that made the error message show up in the first place, is "an error waiting to happen". —Anomalocaris (talk) 16:55, 23 October 2015 (UTC)
The new error message correctly warns if a ref name is defined twice with different content. The software has always ignored and still ignores the second definition, and displays the first definition for both references. In some cases the two references are completely different and it's a big error to ignore the second definition so a warning is good there. In other cases the two definitons are about the same source but use different wording. In such cases it matters less. The error message cannot be expected to determine whether the error is important or not, and it cannot detect whether the error comes from a template with the way MediaWiki works. If there are two identical definitions of a ref name then the error message is not produced. {{sfn|Nock|1952|p=47–48}} produces <ref name="FOOTNOTENock195247–48">[[#CITEREFNock1952|Nock 1952]], p. 47–48.</ref>, while {{sfn|Nock|1952|pp=47–48}} produces <ref name="FOOTNOTENock195247–48">[[#CITEREFNock1952|Nock 1952]], pp. 47–48.</ref>. The two definitions of "FOOTNOTENock195247–48" say respectively "p." and "pp." where the second definition with "pp." will be ignored, so the error message is correctly generated. The difference is minor but an automated error message cannot, and probably should not, ignore errors just because they seem minor to a human. If you want {{sfn}} to produce different ref names in the two cases, or to check whether p= is inappropriately used with an interval or pp= with a single page, then you can post a suggestion to Template talk:Sfn. A correct error message should not be removed everywhere just because some occurrences of it are caused by a template which could be improved. However, the error message should maybe be made less visually dominant. PrimeHunter (talk) 18:37, 23 October 2015 (UTC)

I want to draw attention to this edit at the help desk. This seems to be a case in which {{sfn}} is supposed to produce two identical ref names with slightly different content (one supporting a postscript). I'm not sure if there is any way for the error msg generation logic to detect this case, or what can be done to fix the calls to sfn to avoid the error message. Advice would be welcome. DES (talk) 20:36, 23 October 2015 (UTC)

DES, that's a great example showing the need to fix {{sfn}}. User:PrimeHunter: I agree that if a ref name is defined twice with different content, there should be an error message. I simply say that it is incorrect for {{sfn}} to take two different parameter sets and mangle them to the same ref name. As an additional example of what could go wrong, suppose we have two books, both written by authors named Nock, one in 198 and one in 1989. {{sfn|Nock|198|p=97}} produces <ref name="FOOTNOTENock19897">[[#CITEREFNock198|Nock 198]], p. 97.</ref>, while {{sfn|Nock|1989|p=7}} produces <ref name="FOOTNOTENock19897">[[#CITEREFNock1989|Nock 1989]], p. 7.</ref>. {{sfn}} needs to be rewritten so that calls to this template with different parameters don't mangle to the same ref name. —Anomalocaris (talk) 20:45, 23 October 2015 (UTC)
Of course one call to {{sfn}} can't know what other calls there might be on the same page. The case Anomalocaris mentions above could be handled if sfn output a ref name such as FOOTNOTENock1989_7 for one case and FOOTNOTENock198_97 for the other. Alternatively, this fairly rare case (we don't have many articles citing two sources 1000 years apart with identical author names) could be handled just as we handle Smith, John and Smith, Jane both with works in 2002. One is dated as 2002a. If the older source is cited as {{sfn|Nock|198a|p=97}}, the issue does not occur. None of this would fix the ps case, however. DES (talk) 21:13, 23 October 2015 (UTC)
This is not something that {{sfn}} can fix. The template cannot see outside itself. I does not know that any other {{sfn}} templates exist. I suppose that it is possible to add a parameter that would instruct MediaWiki to ignore this one {{sfn}} template but , as it stands, {{sfn}} (and the {{harv}} templates) is not broken.
Trappist the monk (talk) 20:54, 23 October 2015 (UTC)
From what I read here, the only solution for the particular case seems to replace the subsequent occurences of {{sfn}} with <ref name="FOOTNOTEAuthorYearPage"/>, if I want them to point to the same note without retyping the long quote in ps= parameter. Is it correct? However, this solution will only work, as long as the format of ref names generated by sfn does not change. As Anomalocaris argues for changing this format, I guess it will not stay like this for long. Is any permanent solution possible? --Off-shell (talk) 21:01, 23 October 2015 (UTC)
The whole point about {{sfn}} is that you shouldn't need to use the <ref name="..." /> syntax (in fact you shouldn't need to use any form of <ref> tag) - and you shouldn't need to worry about the ref name either. Any change to the way that {{sfn}} works should be discussed at Template talk:sfn, not here. We spent some time in getting it to work satisfactorily and predictably; it has been stable for two and a half years now. --Redrose64 (talk) 22:54, 23 October 2015 (UTC)
Unfortunately it does not work anymore the way it worked before. If it should help, I will repost my question at Template talk:sfn. --Off-shell (talk) 23:08, 23 October 2015 (UTC)
(edit conflict) Quite true, Redrose64 an editor using {{sfn}} in accordance with its documentation shouldn't need to worry about how it works internally. And considered on its own, it works just fine. But now its interaction with the new cite error detection logic reveals that in some cases sfn produces ref names that worked, but technically broke the documented interface, or at least violated the best practices, unless I have misunderstood things. Makign two sfn calls to the same source, one using the ps= parameter, and one not (as recommended in the sfn docs) produces two ref names with non-identical content -- as it must for sfn to work as it is documented to do in regard to ps=. In this case the "error" of producing two identical ref names is harmless, but there is no obvious way to tell the error detecing logic that. Perhaps if that logic simply ignored all ref names that start with "FOOTNOTE"? Its not a likely human-generated name for a ref, and sfn (and other tempaltes that emit or react to these refs) could be edited to use a less likely string instead -- all behind the scenes. That might also stop the p vs pp error reports, which are a bit hard to track down. I suppose sfn could be made to detect if pp is used with a single page number and issue a more helpful error message. No doubt before this it seemed harmless or at least not worth the effort of detecting.
But in acy case we do need to address the interaction of sfn and the new logic, to avoid problems such as those reported above, with sfn using ps. DES (talk) 23:10, 23 October 2015 (UTC)
Here is a proposal for the next software version: Define an additional optional parameter for the ref tag, e.g. check=no, which will explicitely exclude the tag from the content check. Then add an equivalent parameter to the harv/sfn(p) templates, e.g. {{sfn|Author|Year|check=no}}.--Off-shell (talk) 23:27, 23 October 2015 (UTC)
(edit conflict)A good idea, but as I understand it, by the time the error check code is operating, any such detail has already been stripped from the ref name. To change the processing to pass such a parameter through to where it is needed would be a much more basic change, would require re-working of the entire ref-handling sub-system and a matching level of testing, in short a much larger change than telling the error detecting code to simply ignore ref names that match a specific pattern. DES (talk) 23:38, 23 October 2015 (UTC)
Template:Sfn#Additional comments or quotes: .7Cps= encourages users to enter parameters where the template will produce the same ref name with different definitions. The feature deliberately relies on the second and later definitions being ignored. See xkcd Workflow. {{sfn}} or its documentation should be changed to get rid of this feature. If there are many uses then maybe a bot could be coded to clean them up. sfn could add a parameter to indicate it isn't the first use for that reference. In such cases it could produce <ref name="FOOTNOTEAuthorYearPage"/> to invoke another definition (unfortunately this will fail if there isn't actually another definition). Currently it produces something like <ref name="FOOTNOTEAuthorYearPage">definition2</ref>, where definition2 will be different from the other definition if they don't have the same ps=. In theory, MediaWiki:Cite error references duplicate key could be coded to detect "typical sfn output" in the ref name and omit the error message in that case. One of the problems with this hack is that users with other interface languages in preferences would still get the error message, unless we also make a similar MediaWiki:Cite error references duplicate key/xx for their language. Another problem is that other Wikimedia wikis often copy our templates, but they may not have a self-destructing error message. PrimeHunter (talk) 23:32, 23 October 2015 (UTC)
I think that there is value in detecting and showing differences in the content of named references. Simply ignoring any <ref name="FOOTNOTE...">I don't know that it's possible for MediaWiki to know that {{sfn|Author|year|ps=text}} should link to a citation that is also linked to by a {{sfn|Author|year}}. Following up on what I wrote before, we could add an |mwignore=true parameter to {{sfn}} which then would render it like this:
{{sfn/sandbox|Author|2006|p=34|mwignore=true}}
<ref mwignore="true" name="FOOTNOTEAuthor200634">[[#CITEREFAuthor2006|Author 2006]], p. 34.</ref>
I hacked Module:Footnotes/sandbox to set mwignore="true" just to see what happens. MediaWiki emits a too many names error so that would need to be fixed at MediaWiki also.
But that isn't a good solution either. The defined problem is two {{sfn}} templates pointing to the same citation and the same page or location but one has |ps=some text and the other does not. MediaWiki builds a link between the article text and the {{reflist}} where the {{sfn}} result is rendered. The link between the subscript and the {{sfn}} is by the <ref name="FOOTNOTE..."> tag. If we create a new parameter for {{sfn}} that appends an identifier to the end of the name="..." value then the name is unique and MediaWiki doesn't choke on it and the {{sfn}} links are the same pointing to the same citation. The identifier could be any text. So, then:
{{sfn/sandbox|Author|2006|p=34|id=a}}
<ref name="FOOTNOTEAuthor200634a">[[#CITEREFAuthor2006|Author 2006]], p. 34.</ref>
For each unique quote on page 34 of Author 2006, use a different identifier.
Trappist the monk (talk) 23:53, 23 October 2015 (UTC)

Here is a sandboxed version of this possible solution

  1. Some nonsense.[1] {{sfn/sandbox|Stokes|1979|p=2|ps=. A long and tedious quote of little import}}
  2. Same as #4.[1] {{sfn/sandbox|Stokes|1979|p=2|id=a}}
  3. Different nonsense from the nonsense in #1.[1] {{sfn/sandbox|Stokes|1979|p=2|id=b|ps=. And another bit of drivel}}
  4. Same as #4.[1] {{sfn/sandbox|Stokes|1979|p=2|id=a}}
  5. Link to #1 reference.[1] {{sfn/sandbox|Stokes|1979|p=2|name-only=yes}}
  6. Some more nonsense.[2] <ref name="quote in short ref">{{harvnb|Stokes|1979|p=2}}. A restatement of the long and tedious quote of little import.</ref>
  7. Same cite as at #6 without retyping the quote.[2] <ref name="quote in short ref" />

References

  1. ^ a b c d e Stokes 1979, p. 2. A long and tedious quote of little import Cite error: The named reference "FOOTNOTEStokes19792" was defined multiple times with different content (see the help page).
  2. ^ a b Stokes 1979, p. 2. A restatement of the long and tedious quote of little import.
  • Stokes (1979). Nothing worth reading here. {{cite book}}: Invalid |ref=harv (help)

Trappist the monk (talk) 00:09, 24 October 2015 (UTC)

So if I understand it correctly, mwignore="true" is what I meant by check=no above. Your 2nd solution with id parameter means that another footnote is produced, and if I really want to refer to the same footnote, I have to copy the whole text in ps parameter. What about this alternative: provide an option for {{sfn}} telling it to generate a bare reference <ref name="..."/>. Compared to writing <ref ...> directly, the advantage is that I don't need to know how the ref name is constructed by sfn and the name will automatically be adjusted, if the sfn code changes. But this will be the responsibility of the editor to ensure that the initial reference exists. A deletion of the initial reference will break these ones. A bot may then look at such cases, search for the initial reference in the history of the page, and fill it at the first occurence of such a reference in the current version of the article. --Off-shell (talk) 00:11, 24 October 2015 (UTC)
That is possible (item 5 in the list above) though the code produces <ref name="FOOTNOTEStokes19792_"></ref> and not <ref name="FOOTNOTEStokes19792_"/>. A shorter but still meaningful parameter name would be better. The documentation being what it is, does anyone know if it is possible to produce a self-closing tag with frame:extensionTag{}?
Trappist the monk (talk) 03:06, 24 October 2015 (UTC)
frame:extensionTag uses the same code internally as the #tag parser function. That is, its arguments are passed through to the Cite extension without ever being constructed into an HTML tag string. Also, the Cite extension checks to make sure that <ref name="foo" /> generates the same output as {{#tag:ref||name=foo}}. If you were substituting, there would be a difference in the wikitext you would get back from frame:extensionTag('ref', '', {name = 'foo'}) and frame:preprocess('<ref name="foo" />'), but otherwise using frame:extensionTag should work fine. — Mr. Stradivarius ♪ talk ♪ 07:17, 24 October 2015 (UTC)
The solution with name-only would be fine with me. It does not matter for the user how the empty tag is formed by sfn "behind the scenes", as long as the checker treats it in the same way (hopefully it must, according to XML standards). In addition, the id option is indeed also needed for citing different excerpts from the same source page. So I think both options should be implemented. --Off-shell (talk) 07:25, 24 October 2015 (UTC)

Usage of ps= in {{sfn}}

The intent of {{sfn}} was that you should need no more than six parameters: up to four authors, a year, and any one of |p=|pp=|loc=, and for a long time, those were all that it offered. It seems that in a drive for commonality, its documentation is now shared with {{harvnb}} etc. which always had a broader range of parameters. It's not necessary to make {{sfn}} even more complicated with more parameters. In the (rare) cases that a page number is insufficient, the |loc= parameter provides all you need:

Statement 1.{{sfn|Smith|2015|loc=p. 12, para 1}} Statement 2.{{sfn|Smith|2015|loc=p. 12, para 4}}

If this still isn't accurate enough, and you need to quote something, use <ref>{{harvnb}}</ref> and put it after the closing double brace. --Redrose64 (talk) 07:40, 24 October 2015 (UTC)

Redrose64, do you suggest to remove the parameter ps= completely? I don't see any magic in the number 6. If it were "10 parameters", I would have no problem with that. But if my problem can be solved with the current tools, it would be fine. Could you provide a solution for the particular example from my original request? I make a footnote with a long quote inside it. Later I want to reuse the footnote without copying the quote. --Off-shell (talk) 08:19, 24 October 2015 (UTC)
See items 6 and 7 in my list. It works because an editor can choose a unique name and use the self-closing <ref name="unique name" /> tag.
Trappist the monk (talk) 10:25, 24 October 2015 (UTC)
OK, thank you Trappist the monk. In a previous posting above, Redrose64 specifically noted that "the whole point about {{sfn}} is that you shouldn't need to use the <ref name="..." /> syntax (in fact you shouldn't need to use any form of <ref> tag) - and you shouldn't need to worry about the ref name either." The new proposal means in fact the opposite - going back to ref tags. --Off-shell (talk) 21:22, 24 October 2015 (UTC)
Don't try to stretch {{sfn}} to do something that it wasn't designed for. The (up to) six parameters that it's normally used with are all combined to make a ref name that is specific to that particular combination, none of the six are ignored. As soon as you start using params that don't contribute towards the ref name, you potentially get exactly the sort of issues that started this off. Ask yourself: "do I really need to give more than author(s), year, and page number?" If so, it's time to look at another way. --Redrose64 (talk) 22:53, 24 October 2015 (UTC)
Redrose64, by that argument the |ps= parameter shouldn't be supported, or should accept only "." and " " (to switch between CS1 and CS2 behavior). But ps= not only permits longer additions, but the documentation explicitly recommends using them for quotes. Should we remove that recommendation, and change all uses of ps= for such purposes to some other citation method? If not, we are luring editors into creating reasonable markup that now displays a nasty erro message which is not easy to track down for those not technically sophisticated and aware of that {{sfn}} does behind the scenes. Personally I don't see why you are so strict about the limiting data passed to sfn to year, author, and page. Many cites, particularly to off-line sources, benefit with a quote -- I might even say most off-line cites should have one. I don't use harvard or sfn style cites much, so perhaps I am missing a nuance. DES (talk) 00:45, 25 October 2015 (UTC)
There are good reasons why the intent of {{sfn}} was that all passed-in information should be included in the ref name. Let's assume that your article has
The design was altered in 2015 at the insistence of the government in order to reduce costs,{{sfn|Smith|2015|p=12|ps="The minister stated that no more money was available"}} originally having been designed to have every luxury imaginable.{{sfn|Smith|2015|p=12}}
which displays as
The design was altered in 2015 at the insistence of the government in order to reduce costs,[1] originally having been designed to have every luxury imaginable.[1]

References

  1. ^ a b Smith 2015, p. 12"The minister stated that no more money was available" Cite error: The named reference "FOOTNOTESmith201512" was defined multiple times with different content (see the help page).
This is out of chronological order so you might rewrite it, interchanging two phrases:
The building was originally designed to have every luxury imaginable,{{sfn|Smith|2015|p=12}} but the design was altered in 2015 at the insistence of the government in order to reduce costs.{{sfn|Smith|2015|p=12|ps="The minister stated that no more money was available"}}
which displays as
The building was originally designed to have every luxury imaginable,[1] but the design was altered in 2015 at the insistence of the government in order to reduce costs.[1]

References

  1. ^ a b Smith 2015, p. 12. Cite error: The named reference "FOOTNOTESmith201512" was defined multiple times with different content (see the help page).
Both reflists have one ref that has two backlinks. Now ignoring the red error messages, we also notice that the quote is absent from the second ref. Unexpected losses due to a simple reordering of text has been a problem with using |ps= ever since it was added to {{sfn}} - something that was not discussed at Template talk:Sfn but was apparently done so that it would be more like {{harvnb}}. The {{harvnb}} template does not have the problem, because it does not construct <ref>...</ref> tags, and so does not generate a named ref. This is how I would do it:
The building was originally designed to have every luxury imaginable,{{sfn|Smith|2015|p=12}} but the design was altered in 2015 at the insistence of the government in order to reduce costs.<ref>{{harvnb|Smith|2015|p=12}}: "The minister stated that no more money was available"</ref>
which displays as
The building was originally designed to have every luxury imaginable,[1] but the design was altered in 2015 at the insistence of the government in order to reduce costs.[2]

References

  1. ^ Smith 2015, p. 12.
  2. ^ Smith 2015, p. 12: "The minister stated that no more money was available"
Two refs, one backlink each, one with the quote. --Redrose64 (talk) 08:25, 25 October 2015 (UTC)
Very well, Redrose64, that makes some sense. But in that case we need to fix existing uses of |ps= in accord with your example above, and we need to change the documentation on {{sfn}} to stop encouraging this use of the ps= parameter. Ideally we need to change sfn to emit an error message of its own when ps is so used (anything more than a single space or period, perhaps). And for all that we really should have consensus. DES (talk) 16:59, 25 October 2015 (UTC) @Redrose64: DES (talk) 17:19, 25 October 2015 (UTC)

I have been using AWB to find and clean up some of the uses of {{sfn}} with a non-empty ps= parameter. (see my recent contribs.) However, I found a case of the use of {{sfnm}} with multiple quotes in Argentina. I am not sure how this sort of thing should be fixed. Any suggestions? DES (talk) 00:22, 26 October 2015 (UTC)

Spaces replaced with underscores in error msg

This was mentioned above. Here is an example (since fixed) of a cite error message showing the offending ref name with underscores, while the actual name uses spaces. This makes searching for the issue in the wiki-text a bit harder, although far from impossible once one realizes the issue. Can the message be modified to convert underscores to spaces for display? Failing that, or perhaps better, cna the help page mention this issue? DES (talk) 15:55, 24 October 2015 (UTC)

The error message should show the reference name as it is in the wiki text. The error detector should not convert underscores to spaces because underscores may legitimately be used in a reference name.
Trappist the monk (talk) 16:04, 24 October 2015 (UTC)
Trappist the monk, from the comments above, and from those in the linked help desk thread, I understood that when this error message is generated, the ref name has already been formatted for use in a URL, which includes converting spaces to underscores, and numeric encoding of special characters (which luckily are rarely used in ref names). While we could convert underscores back to spaces for the message, the previous transformation is lossy, and we can't know what the actual name was at that point. Is that correct, or have I misunderstood? DES (talk) 19:12, 24 October 2015 (UTC)
You would need to ask someone who is familiar with the code that makes the error message. From a user's point of view, as you yourself saw, the existing error message can be less than helpful to the non-technical among us. I suspect that it isn't that difficult to keep the original around until after the error checking is done.
Trappist the monk (talk) 19:42, 24 October 2015 (UTC)
There is a suggestion at Help talk:Cite errors#ThisCite error: Invalid .3Cref.3E tag.3B name defined multiple times with different content to change underscores to spaces in the error message here. Spaces are currently converted to underscores by MediaWiki before the error message is called. This can be tested by previewing <ref name="a b">foo</ref><ref name="a b">bar</ref> in mainspace. The code for MediaWiki:Cite error references duplicate key currently makes no conversion but just says what was passed to it as $1. PrimeHunter (talk) 23:11, 24 October 2015 (UTC)

Watchlist

Selecting the namespace and then clicking go in the watchlist does not change the URL. GeoffreyT2000 (talk) 13:47, 22 October 2015 (UTC)

So what? The watchlist is restricted to the chosen namespace, at least for me in Firefox. I have never seen reason to post a watchlist link others could click to only see a given namespace. PrimeHunter (talk) 13:54, 22 October 2015 (UTC)
I presume, were I the sort, that one might want to put together a bookmark (or several) arranged by namespace. That can't be done if the URL does not change depending on namespace. Which, I'm pretty sure there is a way to manipulate the URL such that you can set up such a bookmark, but it's not user-friendly if you can't access it by doing so in the context of the current watchlist. --Izno (talk) 13:58, 22 October 2015 (UTC)
Alternatively, as the task provided notes, a bot (AWB if not a full bot) could manipulate the URL to get a list of articles... etc. --Izno (talk) 14:00, 22 October 2015 (UTC)

Not getting notices

Hi to all, earlier today I left a message on a new user's talk page explaining some policies. When I logged back on just now, I went to his page abd saw that he had replied. The problem is that he had linked my name in his reply, and I didn't receive a notice that he had, like I usually do. I use mobile view, so to me any notification appears as a number in a red square at the little bell symbol at the top of the screen. I got a notification yesterday for practically the same thing, but not today. Is anybody else having this? Thanks. White Arabian mare (Neigh) 18:22, 22 October 2015 (UTC)White Arabian mare

Hi White Arabian mare - Are you referring to this edit? - NQ-Alt (talk) 18:24, 22 October 2015 (UTC)
Yes, that's it. Weirdly, I got your ping just then, but not that one earlier. 😛 White Arabian mare (Neigh) 18:29, 22 October 2015 (UTC)White Arabian mare
@White Arabian mare: According to Wikipedia:Notifications, for a ping to work properly, the post containing a link to a user page must be signed; if the edit does not add a new signature to the page, no notification will be sent. See the related discussion at Wikipedia talk:Notifications#Why didn't it work? - NQ-Alt (talk) 18:32, 22 October 2015 (UTC)
Thank you, I have never heard that before. White Arabian mare (Neigh) 18:34, 22 October 2015 (UTC)White Arabian mare
More at Wikipedia:Village pump (technical)/Archive 133#Missed pings and Wikipedia:Village pump (technical)/Archive 132#Ping notifications working? - NQ-Alt (talk) 18:36, 22 October 2015 (UTC)
And dozens of other threads. Omitted signature is probably the most common reason for failed notification, and the one we get most queries about. --Redrose64 (talk) 18:41, 22 October 2015 (UTC)
@Redrose64: Might want to add this to this page's editnotice then? Gparyani (talk) 19:20, 23 October 2015 (UTC)

I thought IPs were unable to create articles

I thought IPs were unable to create articles. But it's happening, for example Lola LC91. Comments? Help? -- Diannaa (talk) 22:52, 22 October 2015 (UTC)

@Diannaa: they can create talk pages and an eligible user could move them to mainspace. –xenotalk 23:00, 22 October 2015 (UTC)
(edit conflict) They can't. They can, however, create pages in Talk: space, which confirmed users may then move to mainspace. --Redrose64 (talk) 23:01, 22 October 2015 (UTC)
Thanks, I missed that, -- Diannaa (talk) 23:03, 22 October 2015 (UTC)
For the record, they can also create pages in one non-talk namespace: The Draft namespace. This is more common. PrimeHunter (talk) 23:22, 22 October 2015 (UTC)
Yes, I've seen lots of those from my work trying to clear out Category:AfC submissions declined as copyright violations -- Diannaa (talk) 01:35, 23 October 2015 (UTC)

Reporting - Category:WikiProject assessments not updating

Greetings: Yesterday I added Category:WikiProject assessments to Wikipedia:WikiProject Catholicism/Assessment. It is still not updated even though I did the purge on both pages. I do not know if something is broken so I all I can do is report the problem here and ask for help. Thanks. JoeHebda (talk) 14:42, 23 October 2015 (UTC)

@JoeHebda: It is listed under "W" [27] - NQ-Alt (talk) 14:47, 23 October 2015 (UTC)
Woops! sorry I forgot about that pipe to Cath. Now fixed &   Done. Thank you! JoeHebda (talk) 14:53, 23 October 2015 (UTC)

"Due to high database server lag, changes newer than 44 seconds may not appear in this list."

What causes this message to show up on my contributions list? Gparyani (talk) 19:18, 23 October 2015 (UTC)

If a lot of database writes are done in a short period of time, it can cause lag (some details at database replication). In this case it was phab:T116425. Legoktm (talk) 03:24, 24 October 2015 (UTC)

Case sensitivity

The performer field in Special:Log is case sensitive. When you start typing, it will only give suggestions that start with what you type and have matching capitalization. GeoffreyT2000 (talk) 04:02, 24 October 2015 (UTC)

It's case-insensitive on first letter, which means that it matches the case-sensitivity of user names. --Redrose64 (talk) 07:18, 24 October 2015 (UTC)

Sort and prune my watchlist?

Is there a way to sort my watchlist so that I can get it down to a more reasonable size? Firstly to remove all the redlinks and secondly to sort by when last I edited the pages, so I can remove those I have not touched since <date x>. It's rapidly heading towards 20,000 pages, most of which I no longer care about. Roger (Dodger67) (talk) 10:39, 24 October 2015 (UTC)

I use the raw watchlist for this purpose.Jo-Jo Eumerus (talk, contributions) 11:32, 24 October 2015 (UTC)
That's unfortunately useless, it gives no meaningful information about the items in the list, not even identifying redlinks. It seems the "primitiveness" of watchlists is a perennial issue, I've found quite a few posts like mine dating back years. How difficult is it really to take a watchlist entry, search a user contributions log for the most recent occurrence of the page title and return the diff or even just the timestamp - rinse, repeat, all the way down the list - then sort the output list by the timestamp? It seems on the face of it to be a fairly simple database query operation. Roger (Dodger67) (talk) 13:51, 24 October 2015 (UTC)
This is some Javascript I was handed a while ago by someone on Wikidata.
$('#editwatchlist-ns0 .mw-htmlform-flatlist-item').each(function (x, y) { if ($('a:first', y).is('.new')) { y.firstChild.checked = true; } })
The above, if executed from the browser console line (ctrl + shift + k in Firefox), checkmarks every page on Special:EditWatchlist which is redlinked and in the main namespace. (You could change ns0 right in the beginning for other kinds of pages based on their namespace ID e.g. ns1 is talk space.) Requires jquery. Not guaranteed to work because I've had it for a while... --Izno (talk) 14:17, 24 October 2015 (UTC)
Thanks Izno, unfortunately "executed from the browser console line" is gobbledygook to me. I use Chrome, if that helps. Roger (Dodger67) (talk) 17:19, 24 October 2015 (UTC)
@Dodger67: In Chrome, when on page https://en.wiki.x.io/wiki/Special:EditWatchlist, use ctrl + shift + i to open the developer tools. Click Console (the right-most tab across the top of the developer box). Paste the above code after the input cursor ">" at the bottom and press enter. Presto, all article-space red links on your watchlist have been selected. Worldbruce (talk) 04:36, 26 October 2015 (UTC)

Please add this to Special:ActiveUsers article

Please add this to the first sentence of the [[Special:ActiveUsers]] article:

The total number of active editors on the English Wikipedia is [[Special:ActiveUsers|{{NUMBEROFACTIVEUSERS}}]].
Here is how it renders: The total number of active editors on the English Wikipedia is 121,692.

The number is dynamic on a page refresh, and the number is a clickable wikilink.

The page is locked, and there is no Talk page. This addition would be a great asset to that article. Thank you. Cheers! {{u|Checkingfax}} {Talk} 13:28, 24 October 2015 (UTC)

Checkingfax special pages are not articles and have no talk pages. They are part of the Mediawiki software. I'm not sure how hard such a feature would be to add, but it would take a developer to alter the software. What would the link link to, as you visualize it? There is not usually a good reason for a page to link to itself (except for section links). DES (talk) 14:24, 24 October 2015 (UTC)
Actually, we can do it locally: the text at the top of Special:ActiveUsers is defined by MediaWiki:Activeusers-summary and MediaWiki:Activeusers-intro. — Mr. Stradivarius ♪ talk ♪ 16:20, 24 October 2015 (UTC)
What's the difference between MediaWiki:Activeusers-summary and MediaWiki:Activeusers-summary/en-gb? My WP interface language preference is set to en-English not en-GB-British English. I just made an (unrelated) edit the apparently language-independent file and no change in Special:ActiveUsers result (even with various cache purges), then edited /en-gb and it took effect. DMacks (talk) 17:02, 24 October 2015 (UTC)
MediaWiki:Activeusers-summary affects users with en as language at Special:Preferences, and MediaWiki:Activeusers-summary/en-gb affects en-gb. Your edits were only a minute apart. It sounds like a timing coincidence if your edit to MediaWiki:Activeusers-summary only took effect after the edit to MediaWiki:Activeusers-summary/en-gb. PrimeHunter (talk) 20:12, 24 October 2015 (UTC)

My bad, DES. Here is my revised suggestion:

Please add this to the first sentence of the [[Special:ActiveUsers]] article:

The total number of active editors on the English Wikipedia is [[Special:ActiveUsers|{{NUMBEROFACTIVEUSERS}}]].
Here is how it renders: The total number of active editors on the English Wikipedia is 121,692.

The number is dynamic on a page refresh or purge, and it's a wikilink to the Special:ActiveUsers page.

The Special:ActiveUsers page is locked, there is no Talk page, and I have floundered to find the proper suggestion box for "Special Pages". This addition would be a great asset to that article. Thank you. Cheers! {{u|Checkingfax}} {Talk} 18:47, 24 October 2015 (UTC)

If you add ?uselang=qqx or &uselang=qqx to a url then it shows names of used interface pages in the MediaWiki namespace. For Special:ActiveUsers it diplays "(activeusers-summary)" and "(activeusers-intro: 30)" at the top, meaning MediaWiki:activeusers-summary, and MediaWiki:activeusers-intro with the parameter $1=30. MediaWiki pages have talk pages but often they haven't been created or aren't active. {{Edit fully-protected}} can be used to get admin attention. PrimeHunter (talk) 20:01, 24 October 2015 (UTC)
Okay now it's using the non-language-specific item as the header. Maybe some wacky caching or...something or another. DMacks (talk) 20:10, 24 October 2015 (UTC)

Test rendering of SVG files

Is it possible to get a rendering of SVG files without actually uploading them? I frequently get a problem with Inkscape that "invisible" objects are generated in the drawing. They are not rendered either in Inkscape itself or in my browser, but they are rendered by the Wikiengine in the thumbnail as black rectangles. It would be good if I could find them before uploading multiple versions of the same image. SpinningSpark 22:33, 24 October 2015 (UTC)

There is (or was) a tool, possibly on commons, that was provided by Jarry1250 (talk · contribs). I don't recall what it is called, nor if there is a version on Wikipedia. --Redrose64 (talk) 22:49, 24 October 2015 (UTC)
Got it, it's at http://tools.wmflabs.org/svgcheck/ SpinningSpark 23:13, 24 October 2015 (UTC)
Another option is to install librsvg on your own computer and use that to check, as that's the SVG converter currently used on Wikimedia wikis. Anomie 12:52, 26 October 2015 (UTC)

Hello,

I am trying to debug an issue I am seeing with interlanguage links on de:Steinkohle. If I visit in incognito, I correctly see a list of 30 languages to choose from. However, if I am logged in and have Language settings > Display > Display language set to "Deutsch" (the default setting for this "de" article), I see no languages at all. Changing the setting to `English` causes all languages to appear again.

Notes: The list of language should be coming from wikidata:Q732607. I do not see any of the old interlanguage links in the content of the article. One suspicious item is this: If I view wikidata:Q732607 in incognito window, I see a different set of "In more languages" than I see if logged in.

This seems like a bug to me. Can anyone confirm / or explain whats going on here?

Thanks, -- JonathanCross (talk) 13:36, 25 October 2015 (UTC)

It works for me. What is your browser? Do you have "Kompakte Sprachlinks" ("Compact language links") enabled at de:Special:Preferences#mw-prefsection-betafeatures? Do you see the language links when you are not logged in and not in incognito? Which language links at wikidata:Q732607 are different for incognito? Is the issue only at de:Steinkohle or also other German pages? PrimeHunter (talk) 14:21, 25 October 2015 (UTC)
Interesting. If I look at it when logged in as NSH002, I see no language links, but I see the full list as NSH001, or if not logged in at all. In both accounts no beta features enabled, and normal Chrome browser (not incognito). One curious feature is that I get the green "Stelle deine Frage" alert at the top as NSH002 but not as NSH001. I made a lot of tweaks to my settings on de:NSH001, but it was a long time ago and can't recall the details. As far as I can remember, the only setting I've changed on de:NSH002 is to enable Navigation popups. --NSH002 (talk) 14:47, 25 October 2015 (UTC)
More info: if I change the language setting to English on de:NSH002, I see the full list. Going incognito also shows the full list. --NSH002 (talk) 15:03, 25 October 2015 (UTC)
+ problem only occurs on de:Steinkohle, not on any other de:wp article I've tried on de:NSH002. --NSH002 (talk) 15:13, 25 October 2015 (UTC)

List of top-class stubs

According to Wikipedia:Version 1.0 Editorial Team/Assessment#Statistics, there are 4,214 top-importance stubs. I would like a list of these because they would be ideal candidates for the WP:TAFI project. Any chance you could help me out?--Coin945 (talk) 03:08, 26 October 2015 (UTC)

@Coin945: I've made a list for you on Quarry. — Mr. Stradivarius on tour ♪ talk ♪ 08:22, 26 October 2015 (UTC)
I can't thank you enough. :D--Coin945 (talk) 08:25, 26 October 2015 (UTC)
@Coin945: No worries. I've now also made a wiki-friendly version at User:Mr. Stradivarius/Top-importance stubs. — Mr. Stradivarius ♪ talk ♪ 09:30, 26 October 2015 (UTC)
There are some interesting finds in that list. I'm surprised that Choreography is as short as it is, for example. — Mr. Stradivarius ♪ talk ♪ 09:45, 26 October 2015 (UTC)
BTW, I already posted something similar. Those are Stub-Class vital articles, that have top or high importance in some of the wikiprojects. --Edgars2007 (talk/contribs) 11:35, 26 October 2015 (UTC)

@Coin945: I've made a couple of similar searches for you as well: top-importance articles less than 3000 bytes, and articles less than 3000 bytes in top-level categories. — Mr. Stradivarius ♪ talk ♪ 02:24, 30 October 2015 (UTC)

  Like--Coin945 (talk) 15:50, 30 October 2015 (UTC)

500 Internal Server Error

Recently, there was a 500 Internal Server Error on Wikipedia. GeoffreyT2000 (talk) 15:24, 26 October 2015 (UTC)

See phab:T116593 for the corresponding bug report. --AKlapper (WMF) (talk) 16:03, 26 October 2015 (UTC)

Wikipedia down

Just a few minutes earlier, Wikipedia was down the with the message "Exception caught inside exception handler". It just came up a minute or two ago. Did anyone else see this? --Biblioworm 15:24, 26 October 2015 (UTC)

Yep (though I don't remember if that was exactly what the error message said). Even the Main Page was down for this one. --IJBall (contribstalk) 15:25, 26 October 2015 (UTC)
Based upon this and this, it is safe to say you were not alone in seeing the error message. --Allen3 talk 15:27, 26 October 2015 (UTC)
This didn't only affect English Wikipedia. I got the same error message on both Wikipedia and on Commons. --Stefan2 (talk) 15:36, 26 October 2015 (UTC)
I tried the Simple English Wikipedia, Metawiki and Wiktionary and they were all down. --Rubbish computer (HALP!: I dropped the bass?) 18:10, 26 October 2015 (UTC)
Curious what Wikipedia's electric bill is for their Server farm. Raquel Baranow (talk) 18:21, 26 October 2015 (UTC)

Images and responsive design

It's really time Wikipedia image embedding sytnax is overhauled to allow for responsive design. Rather than displaying images to universally fixed pixel widths, there should be scope to allow images to scale gracefully on smaller screen sizes such as tablets and mobile phones. This would be especially useful for wider images. I see some work has gone on to improve the Gallery feature and I've found this to be quite effective in displaying images across mobile devices. However, the Gallery tag seems to be highly contentious and almost always gets deleted by the more fussy editors. It would be much more desirable if the standard image embedding just dealt with this properly. I would suggest that

  • images are styled with CSS rather than the old HTML width parameter
  • below a given screen size breakpoint (somewhere around the 400px mark), the image width switches to a max-width of 100%.

Is this the best place to post this suggestion, or should I be posting elsewhere? Cnbrb (talk) 17:38, 26 October 2015 (UTC)

This is already tracked in T90914 and also a bit in T69695. Short story, it's freaking simple in theory, but terribly hard in en.wp practice :) —TheDJ (talkcontribs) 18:29, 26 October 2015 (UTC)
Thanks DJ - oh dear! I'll try to keep an eye on that. Cnbrb (talk) 19:09, 26 October 2015 (UTC)

18:04, 26 October 2015 (UTC)

Image viewer fails to load image

Hey, all, it looks like the Image Viewer is failing to load images for me. When I click on an image, it begins to load (blurry and low-res), but then switches to a gray screen that says "Sorry, the file cannot be displayed, there seems to be a technical issue. You can retry or report the issue if it exists. Error: could not load image from <url of image>". This has happened on every image I've tried to click on so far, in multiple articles (currently speed of light, and others). I haven't filed a Phabricator bug yet; I did a quick search and couldn't find any that seemed relevant, but figured it'd be better to post here before possibly filing a duplicate. Thanks, 73.133.12.111 (talk) 18:05, 26 October 2015 (UTC)

Hm, nevermind, seems to be working now. 73.133.12.111 (talk) 18:07, 26 October 2015 (UTC)
Sounds like phab:T115563 and the reasons for the problem are still being investigated. If you are tech-savvy and know your web browser's tools, your help is welcome (see that task at the bottom). --AKlapper (WMF) (talk) 18:20, 26 October 2015 (UTC)
AKlapper, it happened again, with error messages like:
Cross-Origin Request Blocked: The Same Origin Policy disallows reading the remote resource at https://up.wiki.x.io/wikipedia/commons/2/25/Coyoteinacanoe.png. (Reason: CORS request failed). <unknown>
Cross-Origin Request Blocked: The Same Origin Policy disallows reading the remote resource at https://up.wiki.x.io/wikipedia/commons/2/26/Coyote.jpg. (Reason: CORS request failed). <unknown>
Cross-Origin Request Blocked: The Same Origin Policy disallows reading the remote resource at https://up.wiki.x.io/wikipedia/commons/9/9b/Hermes_Logios_Altemps_33.jpg. (Reason: CORS request failed).
Stopped once I CTRL-F5ed to reload. 73.133.12.111 (talk) 19:15, 28 October 2015 (UTC)
Interesting. That sounds like "Access-Control-Allow-Origin" related stuff. Could you add a comment in Phabricator task phab:T115563 please? --AKlapper (WMF) (talk) 19:25, 28 October 2015 (UTC)

Hi. Is it possible to use mw.wikibase.getEntity(frame.args[3] or "") in Module:External links 2 to get information from any wikidata item?--ԱշոտՏՆՂ (talk) 21:21, 26 October 2015 (UTC)

Or maybe there is a better way--ԱշոտՏՆՂ (talk) 21:38, 26 October 2015 (UTC)
@ԱշոտՏՆՂ: Why do you want to get data from any Wikidata item? It will be easier for us to help you if we know what you're trying to do. — Mr. Stradivarius ♪ talk ♪ 23:22, 26 October 2015 (UTC)
Dear Mr. Stradivarius: There are two standardized literary forms of Modern Aremnian. But we use the same wiki (hy.wiki.x.io). It is not possible to link two articles to wikidata item, but I want to use this template in both articles. For example like this {{external links|Q=Q937}}--ԱշոտՏՆՂ (talk) 23:32, 26 October 2015 (UTC)

Problem with rendering of tex vline

Referring to User talk:Eric Kvaalen#vline. Usage of the tex element \vline, as in

 

using MathML, results in

Failed to parse (Conversion error. Server ("http://mathoid.svc.eqiad.wmnet:10042") reported: "Error:["TeX parse error: Undefined control sequence \\vline"]"): \int_e^{\infty}\frac 1{t(\ln t)^2}dt={\frac{-1}{\ln t}\,\vline\,}_e^\infty

Using PNG it renders fine. Any idea anyone? - DVdm (talk) 12:46, 27 October 2015 (UTC)

vline is not a mathcommand. It's a table building command and not part of the math rendering tex subset. It works historically in PNG mode (there is a ticket about such markings somewhere...), but it's not correct. You'll want \vert, \rvert, \lvert \left \right \mid or \divides, depending on the semantic meaning that you require...
For instance using \Bigg\vert:  
TheDJ (talkcontribs) 15:50, 27 October 2015 (UTC)
Thanks a bunch. User Eric Kvaalen, it looks like the usage of \vert is already covered in Help:Displaying a formula but if you think we can use this example, do go ahead. Cheers - DVdm (talk) 16:50, 27 October 2015 (UTC)

Have people always get notified when they receive a response to their post.

It looks like it's not very hard for the people who have the capability of changing notifying algorithms to figure out how to change the algorithm of getting notified when you get pinged to an algorithm that will notify you every time a post in that section with more indentations gets created that's under a post signed by you but above the next post that has as many or fewer indentations. It might be even easier but not as useful to figure out how to create an algorithm that notifies you every time a section whose top post is signed by you gets edited by somebody other than you. Blackbombchu (talk) 16:59, 27 October 2015 (UTC)

There is a project to do this. It is called WP:Flow. You can see it in action at most talk pages on mw:. Hmm, it looks like I have notifications about posts on 15 different pages over there. Perhaps I'll even find time to read them all this week.  ;-) Whatamidoing (WMF) (talk) 17:31, 27 October 2015 (UTC)

Find me a new teammate?

Hi everyone,

This is a quick heads-up that my team is hiring someone for the mw:Discovery team. The job description is here, but the ideal candidate is not only an experienced editor, but also someone who speaks at least one language other than English and who is nicer than I am.  ;-) If you are interested, then please apply soon; if you know someone who would be good, then please pass the information along. Whatamidoing (WMF) (talk) 17:11, 27 October 2015 (UTC)

Characters replaced with #

In the past few days, I've noticed a couple of cases in which an edit replaced various characters in an article (parentheses, brackets, periods, etc.) with # characters. I seem to recall that the Visual Editor was doing something like this at one time, but these don't seem to be VE edits—at least they're not tagged as such. (I don't know whether it's significant that both editors were merely trying to disambiguate a wikilink.) Anyone have a clue to what's going on here? Deor (talk) 18:05, 27 October 2015 (UTC)

Looks to me like a browser extension is interfering with something. —TheDJ (talkcontribs) 18:23, 27 October 2015 (UTC)
Agree with TheDJ, it was happening for some people before VE went live. --Redrose64 (talk) 23:36, 27 October 2015 (UTC)
It looks like both examples used Dab solver although the first didn't say so in the edit summary. PrimeHunter (talk) 00:33, 28 October 2015 (UTC)
It is the cross-site scripting XSS in Internet Explorer. The browser will say "Internet Explorer has modified this page to help prevent cross-site scripting." at the bottom of the screen in IE9, 10, and 11 and the top in IE8 and earlier. GeoffreyT2000 (talk) 01:35, 28 October 2015 (UTC)
I can confirm both edits were done by IE11 users.
  • Bug T34013 (2011) for WMF to fix it (they said they wont accept patches).
  • XSS Filter Edit filter (2012) about adding another warning or preventing saving.
  • OAuth (2015) is alpha quality software that WMF has allocated no budget, confusing to users with weird permissions, no support for IP editors, and my requests keep expiring. Only good for identification.
Good Luck getting anywhere with WMF leadership. Wikiconference 2015 was an eye opening experience for me where I kept saying "I did this X years ago" and hearing organizations I was rejected by were now in partnerships with the WMF. — Dispenser 02:09, 28 October 2015 (UTC)
I'd also add that many users will not understand why their edits were reverted, because it only occurs after saving the page, and show preview will show everything correctly. Gparyani (talk) 17:38, 28 October 2015 (UTC)
Well, in the two cases I cited, I reverted the messed-up edits and then performed the edits that the editors were trying to make. (One of them wrote a thank-you note on the article's talk page.) Yes, I can see that someone might be surprised, but one would think that even a cursory glance at the article after the edit was saved would clue an editor in that something had gone wrong. Deor (talk) 20:36, 28 October 2015 (UTC)

wikitable IPA no longer working?

Got logged out this morning, and noticed that setting a table to class "wikitable IPA" no longer forces the contents to display in a font that supports the IPA. Is it broken, or has something changed? Thanks — kwami (talk) 20:34, 28 October 2015 (UTC)

Nothing has changed. But the IPA class is added through javascript in Common.js. -- [[User:Edokter]] {{talk}} 10:24, 1 November 2015 (UTC)

Wikidata descriptions

Hello, apologies for the delayed notice, but this is headsup that, on the mobile version, Wikidata search results will be showing underneath the article name while searching. This has already been active on beta for a while, how it works is that for a person on their mobile device, when they start typing in the search box, the description will show underneath each item like the example in the image.

 
Wikidata showing in search results on English Wikipedia while searching for the word "Alexandria"

. Thanks--Melamrawy (WMF) (talk) 11:29, 29 October 2015 (UTC)

Contributions from IP ranges

When placing a range block it would be handy to be able to see contributions from a range of IP addresses. For example it would be great if Special:Contributions/151.20.0.0/17 could show all contributions from IP addresses in that range. Is there a technical reason why this is not possible? Or some privacy issue perhaps? — Martin (MSGJ · talk) 12:06, 29 October 2015 (UTC)

Special:Preferences#mw-prefsection-gadgets has: Allow /16, /24 and /27 – /32 CIDR ranges on Special:Contributions forms, as well as wildcard prefix searches (e.g., "Splark*") (report issues) PrimeHunter (talk) 12:24, 29 October 2015 (UTC)
Many thanks, looks useful — Martin (MSGJ · talk) 12:45, 29 October 2015 (UTC)
(edit conflict) I have wondered about that too. Searching 151.20.0.0/16 and 151.20.0.0/24 work perfectly, but 151.20.0.0/17 fails. The ones that work correspond to 151.20.*.* and 151.20.0.*, so it looks like the internals are working with the asterixes, as /17 can't be text wildcarded. Just a guess. - DVdm (talk) 12:34, 29 October 2015 (UTC)
 
Screen-shot showing issue with banner obscuring page title.

All the best: Rich Farmbrough, 15:29, 29 October 2015 (UTC).

Dismiss the banner? — Martin (MSGJ · talk) 15:38, 29 October 2015 (UTC)
For all our readers? I never get it, when I point out a problem with the interface, I get solutions that hide the symptom for me. I feel like an ageing climate scientist trying to warn of an impending ice age, and getting a rug tucked around my knees, and a cup of hot coco.
Thanks for the coco! But that's not really the point.
All the best: Rich Farmbrough, 16:23, 29 October 2015 (UTC).
Looks like a meta:CentralNotice. —TheDJ (talkcontribs) 16:59, 29 October 2015 (UTC)
I'll cut and paste this there. All the best: Rich Farmbrough, 22:45, 29 October 2015 (UTC).

SUL issue

I am regularly finding that I am not signed in to sister projects. Is this happening to other people? All the best: Rich Farmbrough, 15:32, 29 October 2015 (UTC).

Have you switched off third-party cookies? --Stefan2 (talk) 15:50, 29 October 2015 (UTC)
All second level domains have user, token and session cookies which have the same values as the other domains.
All the best: Rich Farmbrough, 16:19, 29 October 2015 (UTC).

Watchlist and your opinion

Hello technical people. Please watchlist Wikipedia:Requests for adminship/Optional RfA candidate poll and drop by to give your views. Thank you kindly. Anna Frodesiak (talk) 19:51, 29 October 2015 (UTC)

Automatically detecting edit types -- New project. Help wanted.

 
Type-augmented article history (mockup)

Hey folks. The Revscoring Team and I have been working on vandalism detection and article quality prediction for Wikipedia/Wikidata. We have some high quality classifiers online (see m:ORES) and their use is growing (see m:R:Revscoring#Tools that use ORES). The next classifier that we think will prove valuable for wikiwork is one that, given a revision ID, can tell you what type of change was made in that edit. Just to give you a sense for what such an edit type classifier could do, I've mocked up a type-augmented article history page. See the thumbnail on the right.

So, in order to do this well, I need your help. Before we can detect edit types, we need to figure out what types are important and build a taxonomy. We've been working through past research (e.g. Daxenberger & Gurevych (2012)), documentation around standardized edit summaries (e.g. Wikipedia:Edit summary legend), and tools that produce edit summaries (e.g. MediaWiki:Gadget-defaultsummaries.js). We've started to flesh out a Taxonomy that includes all potential edit types. We need your help to make sure we didn't miss anything important.

Eventually, we'll also need help labeling a random sample of edits by the type of change that was made. We're planning to use Wikipedia:Labels to make that work easy. If you'd be interested in helping us with that, please add your name to the list of volunteers on the campaign description page: Wikipedia:Labels/Edit types --EpochFail (talkcontribs) 21:15, 29 October 2015 (UTC)

Blank Afd page

Why is Wikipedia:Articles for deletion/512k day appearing blank? Oiyarbepsy (talk) 04:33, 30 October 2015 (UTC)

It's appearing fine for me. Whatever it was, it must have been temporary. — Mr. Stradivarius ♪ talk ♪ 05:06, 30 October 2015 (UTC)

Reducing the load of WP:TAFI unofficial-manager Northamerica1000

Hello VP Technical Wikipedians! TAFI is a Wikiproject that aims to encourage collaborative improvement of shoddy articles on important topics. Our fantastic colleague @Northamerica1000: has been doing various manual tasks for us for a very long time in order to keep the project afloat. North wants to decrease her/his load, and I want to help in any way I can. I come here to ask for your brains and creativity in coming up with ways to help automate the tasks through bots etc. If you can think of any ways to help us, or require more information on the innerworkings of how things currently work at TAFI, please comment below. :D--Coin945 (talk) 09:58, 30 October 2015 (UTC)

Northamerica1000 TAFI's current manual tasks:

  • Adding templates to approved/unapproved nominations
  • Updating the articles for improvement page, including the addition of new approved entries and removing entries that are scheduled
  • Scheduling, including creating/populating the necessary subpages, writing captions, adding images
  • Notify Wikiprojects when articles within their scope are a weekly selection – {{TAFI project notice}}
  • Update article talk pages when collaborations have finished – {{Former TAFI}}. This has not been performed lately.
  • Removing {{TAFI}} from atop articles when collaborations have concluded. The template is added by Theo's Little Bot, but it appears that it must be manually removed.
  • Sending weekly notifications to users on the notifications list. This is typically done using mass messaging.
  • Updating project accomplishments
  • Periodic updates to the automation and templates page as needed
  • Periodically checking the members page to make sure things are properly formatted
  • Fixing template errors that occur from time-to-time.
  • Updating project pages periodically when project processes change, such as instructions and templates
  • Promoting the project by sending invitations – {{Today's article for improvement invitation}}

Discussion

Note that the tasks above are also located at the TAFI project's Project management page. North America1000 10:48, 30 October 2015 (UTC)

I'd recommend cross-posting at WP:BOTREQ. Jenks24 (talk) 12:55, 30 October 2015 (UTC)
Also see the project's Automation & templates page for an overview of present and former bot tasks. North America1000 13:04, 30 October 2015 (UTC)

red box with deletion-log no longer visible (to some wikipedians) , unless deletion was "within the last 24 hours"

Apologies if this has been brought up before; searching for red box deletion log 2015 gives a lot of unrelated hits.  :-)

The usual this-page-was-deleted redbox seems to have been significantly modified. At least for people who are viewing pages whilst logged out. Ping User:JAaron95, with whom I believe I discussed this briefly on #wikipedia-en-help connect livechat a few weeks ago, with respect to ruWiki policies at the time.

example#1 , Pauscal programming language

example#2 , Genesis Mining cloud computing company

The usual redbox is visible:

  • Sorry, this page was recently deleted (within the last 24 hours). The deletion and move log for the page are provided below for reference.
    • 17:50, 29 October 2015 User:DGG, deleted page Genesis Mining (Recreation of article by bitcoin sock.)
    • 19:03, 1 September 2015 DGG, deleted page Genesis Mining (G11: Unambiguous advertising or promotion: A7 - startup company without credible claim of notability, created by Orangemoody sock, non-reliable sources for central claims, already speedy-deleted once)
    • 23:03, 27 April 2015 RHaworth, deleted page Genesis Mining (A7: Article about an eligible subject, which does not credibly indicate the importance or significance of the subject)

However, I can find no enWiki discussion[34] on the change in behavior for eliminating the display of the redbox after that 24-hour window has passed. Can someone point me to where this happened? metaWiki? phabricator? Bueller? Bueller? Bueller? p.s. I believe, from the livechat conversation, that logged-in editors can still see the example#1 redbox deletion log info at Pauscal immediately. 75.108.94.227 (talk) 15:28, 30 October 2015 (UTC)

I checked it as an unlogged-in editor on another computer. It was visible to me. DGG ( talk ) 15:36, 30 October 2015 (UTC)
Thanks DGG... and just to be sure, by "it" you mean you clicked https://en.wiki.x.io/wiki/Pauscal , and you (without clicking anything else) saw a red box containing the words "WP:PROD: Nominated for seven days with no objection"? Because I don't see the redbox anymore. I used to, on other articles.
e#1. what I see at Pauscal:
  • <div id="mw-content-text" lang="en" dir="ltr" class="mw-content-ltr">
  • <div class="noarticletext mw-content-ltr" dir="ltr" lang="en">
  • <table id="noarticletext" class="plainlinks fmbox fmbox-system" role="presentation"><tr>...
e#1_prime. what I would expect to see at Pauscal :
  • <div id="mw-content-text" lang="en" dir="ltr" class="mw-content-ltr">
  • <div class="mw-warning-with-logexcerpt mw-content-ltr" dir="ltr" lang="en">
  • <p>Page was deleted... deletion log message...
e#2. what I see at Genesis Mining, similar to what I would expect but with the new 'last 24 hours' sentence which I presume is recently added to mediawiki:
  • <div id="mw-content-text" lang="en" dir="ltr" class="mw-content-ltr">
  • <div class="mw-warning-with-logexcerpt mw-content-ltr" dir="ltr" lang="en">
  • <p>Sorry, this page was recently deleted (within the last 24 hours).
Maybe this is skin-related, or even browser-specific somehow? Firefox 38.2 and whatever the default skin, is what I'm using at present. 75.108.94.227 (talk) 16:01, 30 October 2015 (UTC)
So, the page Pauscal should look different when not logged in. It does for me. Anyway, [35] - it's summarised as "Reduced the DOS potential of 404 page floods" -- zzuuzz (talk) 16:03, 30 October 2015 (UTC)
Ah, thanks zzuuzz. For those who want the shorthand summary, the change was made because
  1. there are a lot of search-enginess and web-crawlers and such
  2. most of them are IP-only (they don't login when crawling wikipedia pages)
  3. many of the off-wiki crawlers misbehave, trying to crawl deleted pages or more simply, pages-which-never-existed
  4. displaying the deletion-log (and maybe even checking whether to display) is a relatively expensive database hit , in terms of number of pushups the server-kitties must do , compared to simply displaying a link to the deletion-log
  5. theoretically ... or maybe even actually ... some adversary might try to exploit that weakness of the server-kitties , by making a crawler that endlessly reloaded millions of deleted pages
Therefore, as of mediawiki 1.26.something, only pages that were deleted within the past 24 hours show the deletion-log, and all other non-existent pages (whether deleted or just never created) only show a hyperlink to the deletion-log (which might or might not be empty). This is easier on the server-kitties, because the list of recently-deleted-articles is quite small at any given time, and the boilerplate no-such-page-found text is cleanly cacheable for the most part. 75.108.94.227 (talk) 16:24, 30 October 2015 (UTC)
 
That's it: the server kitties. They all died long ago, thanks to Brion VIBBER, because people overused {{Infobox}}, {{Asbox}}, etc.  :-) Nyttend (talk) 16:29, 30 October 2015 (UTC)

The deletion and move log still appear when creating a new talk or draft page, and also when viewing the (empty) page history. GeoffreyT2000 (talk) 19:12, 31 October 2015 (UTC)

Pageview stats case insensitive

I noticed today that the stats tool ([stats.grok.se]) is not differentiating between different capitalization cases for page view stats. I have not noticed before whether or not the stats are case-sensitive, but Wikipedia is, so the case-insensitivity is a significant flaw. If this can be fixed, it should be. For example, the stats for apple ([36]) are identical to the stats for APPLE (([37]). This is unhelpful because APPLE is currently up at RfD and the pageview tool is returning stats for an entirely different page, if not both pages combined. Can this be repaired? Ivanvector 🍁 (talk) 18:38, 30 October 2015 (UTC)

Not very likely, because the service is not being maintained and just kept in keep alive mode by it's developer. The analytics team at WMF is currently testing the new pageview API. That's where future development likely will occur. —TheDJ (talkcontribs) 18:59, 30 October 2015 (UTC)
The same question was asked about 3 weeks ago: here. In response to an unrelated question about the tool about a week ago, User:PrimeHunter said, "The page view stats at http://stats.grok.se/ is an external tool made by a single volunteer editor User:Henrik who rarely replies to questions or makes edits to Wikipedia. I don't think anyone else knows when he is working on the tool. Note that the tool is not a feature in the MediaWiki software and is not run by the Wikimedia Foundation. The editors of the English Wikipedia have just agreed to add some links to it, notably in MediaWiki:Histlegend which is displayed at top of page histories. The tool uses publicly available data and others could make an alternative tool but I don't know any that work." Akld guy (talk) 19:26, 30 October 2015 (UTC)
It's been case-insensitive for as long as I can remember. Nothing's wrong, and nothing's different. The number of hits for apple is smaller than what you're seeing, because it does indeed mix the hits for apple and APPLE. For example, the results for http://stats.grok.se/en/latest/Christopher_Lloyd should be slightly higher than the 509,328 hits recorded at WP:5000, since it will include results for christopher lloyd as well as results for Christopher Lloyd. Of course, if we could add such a functionality, it would be nice; I'm not trying to dismiss your idea. Nyttend (talk) 19:58, 30 October 2015 (UTC)
Alright, well we'll have to keep those things in mind when we talk about hits in deletion discussions. Thanks all. Ivanvector 🍁 (talk) 16:24, 31 October 2015 (UTC)

Any tool for counting transclusions of many templates?

I know about the transclusion count tool, but it only works one template at a time. When trying to judge if any upmerged stubs are over the 60-article line for being split out into having their own categories, it's a pretty tedious task to have to go through every single one you want to check in turn -- especially when there are, in some cases, literal hundreds at a time. Is there any way to check for a whole bunch at once? Or am I out of luck? Buttons to Push Buttons (talk | contribs) 17:06, 31 October 2015 (UTC)

@Buttons to Push Buttons: do you need it regularly or in quite rare cases? Could probably help. You could give that bunch of templates in interest for first try. Feel free to put them in my talk page, so this page doesn't get spammed :) --Edgars2007 (talk/contribs) 10:12, 2 November 2015 (UTC)

HOVER LINK: HIDDEN ERROR: Usage of "_VALUE_" is not recognized

When hovering over the link to Michael Laucke, the image in the infobox is picked up fine, but the introductory page text one is accustomed to view, does not render, and reads "HIDDEN ERROR: Usage of "_VALUE_" is not recognized". The link works however. When one gets to the Michael Laucke article, the same problem occurs when hovering over Elton John. Many thanks in advance for your kind help; it is always much appreciated. --Natalie.Desautels (talk) 03:27, 1 November 2015 (UTC)

It appears to be outputted by the infobox in an invisible (style="display:none;") span element, using Module:Check for unknown parameters. I'm looking into it further, but I lack proficiency in Lua, so someone else may be able to help better. {{Nihiltres |talk |edits}} 03:46, 1 November 2015 (UTC)
Made a quick stopgap edit that might help. The problem will still need to be fixed elsewhere, but this should reduce its visibility slightly, hopefully. {{Nihiltres |talk |edits}} 04:09, 1 November 2015 (UTC)
@Checkingfax: Thanks very kindly Nihiltres. Hovering over links to the article Michael Laucke is now working properly. The hover link to Elton John has the same problem; I understand it has to be fixed elsewhere. Thanks you once again; very much appreciated. I look forward to following this issue and it's broadly implemented resolution. --Natalie.Desautels (talk) 06:55, 1 November 2015 (UTC)'
@Natalie.Desautels and Nihiltres: Is this error seen when using Popups, or is it some other setup? I can't reproduce it myself, so it's hard to know what to fix. — Mr. Stradivarius ♪ talk ♪ 13:54, 1 November 2015 (UTC)
When "Navigation popups" is enabled at Special:Preferences#mw-prefsection-gadgets, I see "Word1 word2 word3." when hovering over User:PrimeHunter/display none. The source says Word1 <span style="display:none;">word2</span> word3., so on the page it renders as "Word1 word2 word3." without "word2" displayed – at least for me in a normally working browser. The issue for Michael Laucke was fixed by Nihiltres' stopgap edit to a template used there. A broad fix would be for Popups to recognize style="display:none;" and omit display. PrimeHunter (talk) 16:41, 1 November 2015 (UTC)

I noticed the issue using Hovercards. While my stopgap edit fixed the immediate issue, there's still the underlying problem that Module:Check for unknown parameters is outputting a generic error message (rather than one specific to the particular unknown parameter[s]), and given that it might be producing such improper error messages elsewhere; it seems advisable to take defensive measures like forcing non-category error messages from the module to be wrapped in something with class="error". {{Nihiltres |talk |edits}} 20:06, 1 November 2015 (UTC)

@Checkingfax: Thank you again, Nihiltres, for fixing the issue on the Michael Laucke page. I have seen this error on a few pages, such as Louis Armstrong, although the "hidden error" preventing good rendering on this page is different, and it seems that Elton John's infobox output is now fixed. best wishes, --Natalie.Desautels (talk) 07:47, 2 November 2015 (UTC)
I've fixed Module:Check for unknown parameters. The bug was a regression that occurred after a fix for templates that use the blank parameter (| = foo). — Mr. Stradivarius ♪ talk ♪ 11:35, 2 November 2015 (UTC)

Recently, the GaLLERY tag display seems broken to me, instead of having it roll horizontally across the page like a FLATLIST, it is appearing as a bulletted vertical list. This affects both Wikipedia, and on Wikipedia category pages that contain images, and Wikimedia Commons

Did someone modify whatever CSS this uses (and break it)?

-- 70.51.44.60 (talk) 05:27, 1 November 2015 (UTC)

link please, which browser, which os, which versions of those, and I'm assuming logged out ? —TheDJ (talkcontribs) 08:37, 1 November 2015 (UTC)So
I know gallery tags doesn't have anything to do with Special:Prefixindex and Special:Allpages, but maybe related to phab:T32965? And I have also encountered with this problem, but everything is fine after some 0.5 sec. --Edgars2007 (talk/contribs) 14:23, 1 November 2015 (UTC)
Ah, this is because of phab:T117328. The IP probably had Javascript disabled.. A fix is on its way. —TheDJ (talkcontribs) 15:45, 1 November 2015 (UTC)
Yes, that's exactly it. Why the change? (It's also why the searchbox is borked, since the searchbox hasn't worked in years, after then change was implemented requiring JavaScript, or else the searchpage would result) -- 70.51.44.60 (talk) 08:54, 2 November 2015 (UTC)

5,000,000 articles logo needs to be put in place

There was a logo election, and a logo was selected to temporarily replace the wikiglobe on the sidebar once the article count exceeded 5,000,000. Now that the 5 million mark has been reached, a tech-savy admin is needed to put the logo in place. See Wikipedia:Requests for comment/5 millionth article logo.

Thank you. The Transhumanist 12:53, 1 November 2015 (UTC)

This can't be done locally by enwiki admins (see the instructions on mediawiki.org). The WMF tech staff are on the ball, though, and the logo is already up. — Mr. Stradivarius ♪ talk ♪ 13:45, 1 November 2015 (UTC)
Actually it was done locally, with this edit to Common.css. This is an easier way to do it for a short-term change. the wub "?!" 13:55, 1 November 2015 (UTC)
I stand corrected. That's a lot of my original post that has to be struck! — Mr. Stradivarius ♪ talk ♪ 14:02, 1 November 2015 (UTC)
http://www.wiki.x.io/ still shows that we have 4,999,000+ articles. When will this be updated? Gparyani (talk) 18:39, 1 November 2015 (UTC)
A meta admin can update it at meta:Www.wiki.x.io template. I don't know whewther changes go live right away. PrimeHunter (talk) 19:01, 1 November 2015 (UTC)
@Gparyani: The portal is updated semi-manually based on m:List of Wikipedias/Table, which is updated automatically by a bot each midnight UTC. The 5 millionth article missed the cutoff by a bit, but I made an exception just this once, since the milestone is bound to get some attention. – Minh Nguyễn 💬 19:54, 1 November 2015 (UTC)
(edit conflict) I noticed that the page still showed 4,999,000+ articles even after the recent edit. I purged it just now and it says 5,000,000+, but the front page still has the old value cached. Gparyani (talk) 19:56, 1 November 2015 (UTC)
The front page has now been updated. Gparyani (talk) 20:21, 1 November 2015 (UTC)

Counting growth in article types: Did we just have a 37% spike in Women architects?

The Wikipedia:Meetup/Women in Red/3 / Guggenheim / WikiD edit-a-thon campaign, by my rough hand-count, appear to have increased membership in Category:Women architects, including national subcategories, by 37%! (451=pre-event; 619= post-event; % Change = 37%) Is anyone able to technically confirm or correct this with tools / bots / coding magic?--Pharos (talk) 19:18, 1 November 2015 (UTC)

Well, [THIS LINK] shows new names which have been added to List of women architects. There are quite a few new ones. —Anne Delong (talk) 20:16, 1 November 2015 (UTC)
@Pharos: If you know the right URL to input, you can view categories sorted by the date that articles were added to them. Here's the result for Category:Women architects. There were 34 articles added to it on 22 October. — Mr. Stradivarius ♪ talk ♪ 09:27, 2 November 2015 (UTC)

Log out

If you are logged in on two computers and you log out on one of the computers, you are also logged out on the other computer. GeoffreyT2000 (talk) 22:55, 1 November 2015 (UTC)

Yes. Logging out sends a message to the Wikimedia servers that causes all previously-issued cookies for your login to be invalidated. That has been the case for as long as I can remember. I suspect that it's intentional: if you log in on a public computer, but forget to log out, you merely need to log in anywhere and immediately log out again. --Redrose64 (talk) 23:55, 1 November 2015 (UTC)
@GeoffreyT2000: To amplify that, see comment by csteipp posted Jul 11 2013, 9:18 PM. --Redrose64 (talk) 18:56, 3 November 2015 (UTC)

Help someone fix a small bug in your code project: It's Google Code-In time!

Help someone fix a small bug in your code project: It's Google Code-In time!

  • Are you a developer and have small, self-contained, "easy" bugs in your Wikimedia code that you would love to get fixed?
  • Would you enjoy helping someone port your template to Lua?
  • Does your gadget use some deprecated API calls?
  • Does the documentation of your code need some improvements?
  • Do you enjoy mentoring to a new contributor fixing small tasks?

Google Code-In (GCI) will take place again in December and January: a contest for 13-17 year old students to provide small contributions to free software projects. Wikimedia will apply again to take part and would like to offer a wide range of tasks. Just one example: Multimedia saw some impressive achievements in last year's contest!

Tasks should take an experienced contributed about two-three hours ("beginner tasks" also welcome which are smaller) and can be of the categories Code, Documentation/Training, Outreach/Research, Quality Assurance, and User Interface/Design. For more information, check the wiki page and if something is unclear, please ask on the talk page!

Thank you, --AKlapper (WMF) (talk) 23:59, 1 November 2015 (UTC)

WP shortcuts appear to be broken

I noticed just now that WP:YESPOV does not take me to the correct sub-chapter of WP:NPOV. And when I checked the main "see also" for that same chapter, WP:ASSERT, it didn't take me to the right place either. -- Kendrick7talk 09:18, 2 November 2015 (UTC)

Sounds like Wikipedia:Village pump (technical)/Archive 140#Links to sections on other pages not going to beginning of section and threads linked back from that. --Redrose64 (talk) 10:00, 2 November 2015 (UTC)
When I clicked on that link, it didn't go to the beginning of the section, but scrolled nearly halfway further down the page. I've noticed this problem on other pages too but it's intermittent, and doesn't seem to be connected to a particular browser or OS. Ivanvector 🍁 (talk) 14:52, 2 November 2015 (UTC)
Kendrick7, are you using Safari?
Ivanvector, are there collapsed/hatted sections on that page? Whatamidoing (WMF) (talk) 19:17, 2 November 2015 (UTC)
Nope, using Chrome on Windows 10. -- Kendrick7talk 19:24, 2 November 2015 (UTC)
Using Firefox on Windows 7 and Chrome on Windows 10 here. No, there do not seem to be collapsed sections on the page. Ivanvector 🍁 (talk) 22:30, 2 November 2015 (UTC)
There's a known bug in (at least some versions of) Safari that causes redirects to not point always to the correct place. (The bug's in Safari, not in MediaWiki.) Chrome and Safari are both WebKit browsers, but I don't know if they both have the same bug. I'm inclined to guess that the problem is something else. Whatamidoing (WMF) (talk) 23:43, 2 November 2015 (UTC)

Is Wikipedia development slowing down?

As can be seen in the chart below, between the time WP reached 1 million and 2 million it was about a year and a half. Between 2 and 3 million it took almost 2 years, between 3 and 4 million almost 3 years went by, and to reach 5 million over 3 more years.

Aritlce count milestones Date Data size
1 million 1 March 2006 ?
2 million 9 September 2007 ?
3 million 17 August 2009 ?
4 million 13 July 2012 ?
5 million 1 Nov 2015 ?

So, article creation is slowing down.

But what about development as a whole? By how much has WP been growing in words or bytes?

Can someone track down the size figures for those dates, to finish up the chart?

I look forward to your answers and comments. The Transhumanist 12:26, 2 November 2015 (UTC)

 
Total article text in English Wikipedia, measured in gigabytes (compressed)
Wikipedia:Size of Wikipedia#Size of the English Wikipedia database. The chart you're looking for (total amount of article text) already exists—the trend has been virtually constant for a decade. ‑ iridescent 12:41, 2 November 2015 (UTC)
I believe that chart is based on the size of the database rather than readable article content. This means that it is affected by change in formatting trends (e.g., using bulkier citation templates instead of manual formatting, or removing interwiki links in favor of hosting that information at Wikidata). WhatamIdoing (talk) 19:20, 2 November 2015 (UTC)
Thank you. The Transhumanist 20:19, 2 November 2015 (UTC)

Template FlagATHCH

See 2010 European Athletics Championships – Women's 800 metres - in the 'Results' section the country of the participants is repeated: Mariya Savinova - Russia (Russia). Is there something wrong with the FlagATHCH template? Ssu (talk) 13:17, 2 November 2015 (UTC)

The {{FlagATHCH}} documentation says the first parameter should be a three-letter code, not a country name, so I have edited the article. -- John of Reading (talk) 13:32, 2 November 2015 (UTC)

Wrongly initialised chevrons in mobile version of articles

Another user reported this on the Help Desk, and I confirmed it using Chrome on a Google Nexus (android). When a page first loads, all the chevrons point downwards but the content is collapsed. When a section is then opened, the chevron still points downward. Only when the section is then collapsed again does the chevron point upwards. Maproom (talk) 14:01, 2 November 2015 (UTC)

I see the same, and I have created a ticket in Phabricator for this problem. —TheDJ (talkcontribs) 18:39, 3 November 2015 (UTC)

Redirects

Some redirects, such as Talk:Robert J Hirsch -> Talk:Robert J. Hirsch and Talk:List of If Loving You Is Wrong (TV series) episodes -> Talk:List of If Loving You Is Wrong episodes are not listed in Special:WhatLinksHere for the target page. GeoffreyT2000 (talk) 14:56, 2 November 2015 (UTC)

The redirects are from moves today. I guess the WhatLinksHere entries are delayed in the job queue per Anomie's post in #Article categories not always saving completely. PrimeHunter (talk) 15:15, 2 November 2015 (UTC)
I found that Talk:Robert J Hirsch wasn't listed at Special:WhatLinksHere/Talk:Robert J. Hirsch, so I did a WP:NULLEDIT on Talk:Robert J Hirsch and now it's listed. Null edits don't fix the category problem. But it's definitely a problem if WhatLinksHere is not accurate unless those in the know take appropriate action - an article might get marked as an {{orphan}} but more seriously, an image or template might be considered orphaned and so eligible for WP:CSD. --Redrose64 (talk) 19:37, 2 November 2015 (UTC)

16:43, 2 November 2015 (UTC)

When an article is blanked by vandals, the transclusion backlinks disappear from the backlinks list. The article is immediately restored by a bot, but the transclusion backlinks are still missing in 'What Links Here' (or the API). A current example is David Livingstone which was recently blanked -- there are no transclusion backlinks in the WLH. There should be because the article has templates (for example {{Internet Archive author}}). Eventually the transclusions will re-appear but it can take up to 7 days. Is this expected behavior? -- GreenC 21:20, 2 November 2015 (UTC)

After reading this page it seems the problem is delay in the queue and the solution is a WP:NULLEDIT. I've made a feature request with ClueBot to add a NE after a page blank->restore. -- GreenC 21:43, 2 November 2015 (UTC)

IPv6 incompatibility: information thingamabob on user talk pages for anonymous users

Hi everyone! When creating a user talk page for an IP address, there's always a box telling us that it's the discussion page for an IP user, yadda yadda, and a bunch of links to WHOIS, WMFLabs, and the RIRs. Great. No problem, very useful. You can see it here, I don't know where exactly it comes from, probably somewhere in the MediaWiki namespace. However, it doesn't show this box when creating a user talk page for an IPv6 address, such as here. Any chance this can be added? Thanks! Rchard2scout (talk) 21:52, 2 November 2015 (UTC)

The MediaWiki page is MediaWiki:Anontalkpagetext. Relentlessly (talk) 22:02, 2 November 2015 (UTC)
It starts with MediaWiki:Newarticletext which is diplayed by the software on page creations. It contains this:
<!--The below two ifeq-cases detects if it is an IP-user page 
        and then shows the MediaWiki:Anontalkpagetext.
-->{{#ifeq: {{#expr:{{PAGENAME}}}} | {{PAGENAME}}
  | 
  | {{#ifeq: {{#expr: {{PAGENAME}}}} | {{#expr: {{PAGENAME}}+deliberatesyntaxerror }}
    | 
    | {{MediaWiki:Anontalkpagetext|caller=MediaWiki:Newarticletext}}
    }}
  }}
It was designed for IPv4 and uses that {{#expr:{{PAGENAME}}}} doesn't produces an error message for IPv4 addresses which only contain digits and periods. See Template talk:IP-user other. There are also certain unusual usernames which give false positives by not producing an error. However, IPv6 addresses do produce an error message and are therefore falsely identified as usernames. Is there a reliable way to detect IP addresses without a complicated bunch of nested string functions? Maybe a module function written for the purpose? It could also be used in {{IP-user other}} and {{IP-talk}}. PrimeHunter (talk) 23:54, 2 November 2015 (UTC)
I'm not a Lua coder but I have made {{IPv6 other}} to test for IPv6 addresses by using Module:String#match with a long Lua pattern. If others can do something better then go ahead. I haven't used the template anywhere yet. PrimeHunter (talk) 04:36, 3 November 2015 (UTC)
If somebody wants just to copy and adapt the code to new module, then here is one (last comment). --Edgars2007 (talk/contribs) 12:48, 3 November 2015 (UTC)
Oh, Swedish Wikipedia has it already :) --Edgars2007 (talk/contribs) 12:50, 3 November 2015 (UTC)

Women's college basketball team roster formats with "Lady" in team name

Some NCAA basketball teams have gendered nicknames, e.g. the University of Massachusetts Minutewomen, but one of the templates commonly used for NCAA team articles requires that "men's" or "women's" be added, even though it's unnecessary because we can assume that the Lady Volunteers are a women's team. Someone tried to fix this situation previously, but the effort was dropped after one of the editors was blocked, and it's not yet done. Could someone try to help? See Wikipedia:Village_pump_(technical)/Archive_128#Template_question for the first effort.

I discovered this issue at User talk:Sphilbrick while leaving a note on an unrelated matter. I quote the relevant chunks of that talk page:

Hey. Has there ever been any template for those few women's teams that use the "Lady" in their names to override the "women's" bit? I noticed those created in the past say "Lady <team> women's basketball team", which wrong, long and a mouthful. Seems like I remember the same problem with infobox formats. — Wyliepedia 03:37, 2 November 2015 (UTC)

@CAWylie: Yes, and no. Sorry for the useless answer, but my monitor just dies so I have to get a replacement. I'll provide a more useful, though still complicated answer in a couple hours.--S Philbrick(Talk) 13:09, 2 November 2015 (UTC)
@CAWylie:There is sort of the ability to suppress "women" if the team name includes "Lady", but it's a bit of a kludge and it doesn't always work.
If you look at the source of {{Infobox NCAA team season/name}} you will see that there is a list of teams for which a conversion takes place. It is a terrible way to accomplish this, and worse, even though it seems to work in the example for UMass, if you check out 2014–15_UMass_Minutewomen_basketball_team, you'll see that it doesn't correctly work there.
It does work in some cases, for example 2013–14_Central_Arkansas_Sugar_Bears_basketball_team.
If you check out Wikipedia:Village_pump_(technical)/Archive_128#Template_question You'll see I started trying to solving the problem. Unfortunately, the editor with the technical experts expertise got himself blocked.--S Philbrick(Talk) 15:45, 2 November 2015 (UTC)

As far as I can tell, the issue is that the infobox is supposed to begin with a link to the program's article, e.g. 2014–15 Connecticut Huskies women's basketball team begins with a link to Connecticut Huskies women's basketball, but with at least some of these gendered team names, we get silly links like UMass Minutewomen women's basketball instead of the correct UMass Minutewomen basketball. Because we get different results between 2014–15 Central Arkansas Sugar Bears basketball team and 2014–15 UMass Minutewomen basketball team, even though the coding is essentially identical, something's wrong that ought to be fixed. Nyttend (talk) 23:27, 2 November 2015 (UTC)

Page curation

Redirects that were edited into articles remain in the page curation queue only for a short amount of time. GeoffreyT2000 (talk) 01:16, 4 November 2015 (UTC)

Patrol log

Why is the patrol log shown by default for logged out users, but hidden by default for logged in users? GeoffreyT2000 (talk) 01:00, 4 November 2015 (UTC)

More precisely, it's hidden by default for autoconfirmed users who have a "Show patrol log" option at Special:Log. IP's and new users don't have an option at all. They always see the patrol log. I'm not sure of the reason but it's probably connected to autoconfirmed users having the patrol user right to mark others' edits as patrolled. PrimeHunter (talk) 02:34, 4 November 2015 (UTC)

Issue when redirects on Wikipedia and redirect on Commons have the same title

While working on some issues with file name having the same names as files on Wikimedia Commons, I seem to have ran across a rather annoying technical bug; it seems that if there is a file namespace redirect on both Wikipedia (at least here on the English one) and Wikimedia Commons, when looking up the title on Wikipedia if the redirect on Wikipedia is subdued (by a XFD or speedy deletion notification of some sort), the redirect page on Wikipedia will be completely bypassed, redirecting the viewer to the Wikipedia local page for the Commons file with no way to view any part of the Wikipedia redirect. Here's the example I ran across: File:Cesare Mori.jpg; on Wikipedia, it is a redirect to File:Cesare Mori3.jpg that currently has a CSD tag placed on it, but on Commons, it (Commons:File:Cesare Mori.jpg) is an active redirect to Commons:File:Cesare Mori2.jpg. When looking up File:Cesare Mori.jpg on Wikipedia, the redirect in Commons is called instead of allowing the viewer to see the redirect page on Wikipedia, bringing the reader to Commons:File:Cesare Mori2.jpg. If effect, I attempted to go to view the redirect page for the Commons redirect directly (Commons:File:Cesare Mori.jpg) to see if there was a way to go to the page on the English Wikipedia from there, but to no avail. I think the issue may be that if there is a soft redirect on Wikipedia when there is an active redirect on Commons with the same name, this issue occurs. Steel1943 (talk) 04:16, 4 November 2015 (UTC)

The url stays the same at File:Cesare Mori.jpg and you are not really redirected but just see content and tabs corresponding to the redirect target at Commons. The English Wikipedia page can be accessed by manually adding ?redirect=no to the url: https://en.wiki.x.io/wiki/File:Cesare_Mori.jpg?redirect=no. I haven't tested it but I would guess the same happens for any file page which exists at the English Wikipedia without a local file or file redirect. PrimeHunter (talk) 04:35, 4 November 2015 (UTC)

Category membership issues

Categories not placing items in the File namespace correctly.

Category:Candidates for speedy deletion as hoaxes

This category has been stuck a 2 for days, even though there are no pages tagged as a hox. What's wrong?--Bbb23 (talk) 04:45, 18 October 2015 (UTC)

  • @Bbb23: That is odd. On the main category page (Category:Speedy deletion) it lists the current pages under hoaxes as 1P, 2F. The 1 page is accounted for. I am assuming the 2F stands for 2 files so it must be something in the File namespace that is triggering it. I just don't know why it is not displaying those files in the category listings. The same thing is happening with Category:Candidates for speedy deletion for unspecified reason. It is showing 1F on the main category page but no listings. --Stabila711 (talk) 05:38, 18 October 2015 (UTC)
  • This seems like a much larger issue and is probably worth a bug report. I would file one myself but I don't know how to do that. The File namespace is not being allocated to categories correctly. This can be seen multiple times on the speedy deletion page. Even a category marked for deletion as empty (Category:Wooldridge Monuments images) says it contains 10 files and even warns that it doesn't appear empty even though it is. I am going to copy and paste this thread over to VPT and request a bug report be filed. --Stabila711 (talk) 06:02, 18 October 2015 (UTC)

This thread was copied from the Help Desk. Would someone with more technical knowledge take a look at this and post a bug report if necessary? Thank you. --Stabila711 (talk) 06:02, 18 October 2015 (UTC)

Thanks, Stabila711.--Bbb23 (talk) 11:26, 18 October 2015 (UTC)
Some cashing issue. WP:NULLEDIT also didn't help. Wasn't there some API call to force update? --Edgars2007 (talk/contribs) 06:53, 18 October 2015 (UTC)
User:Edgars2007: You might be thinking of titles=name of page with namespace&action=purge&forecelinkupdate. This doesn't help when used on the category, though. --Stefan2 (talk) 23:12, 19 October 2015 (UTC)
Maybe :) Thanks. --Edgars2007 (talk/contribs) 05:41, 20 October 2015 (UTC)
It appears the file count is not updated when a file is deleted, at least in the discussed cases. Category:Candidates for speedy deletion as hoaxes was empty but still said 2 files at Page information and in the parent category a second before I added a file as a test.[44] The count correctly changed to 1 right away. I removed the category again (without deleting the file) and the count correctly changed to 0. Google's cache shows Category:Wooldridge Monuments images did contain 10 files previously, for example File:Wooldridge Monuments 2.JPG. This and all the others are currently at Commons. Page information still claims 10. I haven't made tests to change this. PrimeHunter (talk) 11:44, 18 October 2015 (UTC)

Thanks PrimeHunter. The addition and subtraction of a file in the Wooldridge Monuments category fixed that one as well. So this definitely seems like a caching issue in the File namespace. Now that we know exactly how to fix the problem, does that deserve a bug report or is that something that is just going to have to be lived with and fixed when it comes up? --Stabila711 (talk) 20:14, 18 October 2015 (UTC)

My suspicion is that this is another effect of phab:T115586 preventing links tables from being updated correctly. Anomie 13:10, 19 October 2015 (UTC)

Article categories not always saving completely

I started noticing this a few days ago, but every now and then when I make a category addition or change to an article, the category doesn't get updated with showing the article in its list. This is even considering lag. I've had to re-save articles to get them showing in categories. Is anyone else seeing this? Stevie is the man! TalkWork 17:11, 25 October 2015 (UTC)

Please give an example (always give an example when possible no matter what you report), so we can see how the category was added. You appear to use HotCat a lot. Maybe the relevant link table isn't always updated when categories are added via the API. If this is the case then any manual edit, including a null edit but not a purge, should update the link table. PrimeHunter (talk) 17:43, 25 October 2015 (UTC)
Pardon me for not providing an example, as I was initially just interested if others were seeing the issue, not making an official report. Jeffersontown, Kentucky was the last one, adding Category:1797 establishments in Kentucky. I already did the null edit to fix it, though. Stevie is the man! TalkWork 18:02, 25 October 2015 (UTC)
Thanks, that was with HotCat.[45] I haven't been active with categories lately and rarely use HotCat. PrimeHunter (talk) 19:24, 25 October 2015 (UTC)
Yeap, it's Hotcat. Strangely enough, from out of the blue, somebody just asked me about this problem on my talk page, and I fixed it the same way. I'll report this to Hotcat if it's not reported already. Stevie is the man! TalkWork 21:05, 25 October 2015 (UTC)
I note that MediaWiki was recently changed to put the updating of links and categories after an edit onto the job queue, rather than performing them immediately. If the job queue is being slow, that could be a cause of this as well (e.g. see T116001, particularly this comment). Anomie 13:00, 26 October 2015 (UTC)
I have noticed this lag recently, just in the past couple of weeks, as I regularly clean out Category:Pages with citations using unsupported parameters and similar tracking categories. About one out of every twenty articles will remain in the category after I make a change that should remove it from the category. When I visit the article, the tracking category is not listed at the bottom, but until I do a null edit, the article stays on the category page. This is something new that has never happened to me in tens of thousands of similar edits over the past couple of years. The bug report linked immediately above has a technical but pretty clear explanation by Bawolff. – Jonesey95 (talk) 00:03, 29 October 2015 (UTC)
This is a pain in the a**. I just reverted this edit, which had put Category:Canadian table tennis players inside itself, and no matter what I do, it's still showing as a subcategory. --Redrose64 (talk) 09:15, 2 November 2015 (UTC)

Misreporting category size

It seems that Mediawiki has started misreporting category size in the past week or so. Mediawiki states that Category:All Wikipedia files with the same name on Wikimedia Commons is at 608, while it quite obviously is at 324. Is this a known issue? Magog the Ogre (tc) 00:21, 26 October 2015 (UTC)

See #Categories not placing items in the File namespace correctly. above. PrimeHunter (talk) 01:08, 26 October 2015 (UTC)

Wikipedia:Sockpuppet investigations/Turnbull35 is in the category "Open SPI cases". The SPI is in fact archived and not in open status. According to Amalthea, according to this discussion, when this occurs (a relatively new phemenon), one should do a null edit to the category page. I did that, but nothing changed. The next step recommended by Amalthea was to notify someone here. Thanks for looking at this.--Bbb23 (talk) 15:29, 3 November 2015 (UTC)

@Bbb23: This sounds like another instance of #Category membership issues above. — Mr. Stradivarius ♪ talk ♪ 15:48, 3 November 2015 (UTC)
@Mr. Stradivarius: I'd completely forgotten I'd filed that report here, mainly because at least with respect to CSD, although I don't patrol CSD as often as I use the WP:SPI table, I never saw the problem again. Assuming the issue here is connected to the other, is there a resolution, an outstanding bug? I'm not good at following these things having to do with the Wikimedia software.--Bbb23 (talk) 15:53, 3 November 2015 (UTC)
  • @Stefan2: My recollection is I used to do that and sometimes it worked and sometimes it didn't, which is principally why I took it to Amalthea. (My recollection also is that when it did work, it sometimes came back not too longer after, don't remember the exact time frame.) Did I misunderstand Amalthea's instructions?--Bbb23 (talk) 18:07, 3 November 2015 (UTC)
@Bbb23: Yes, you need to purge the SPI page, not the category page. It's always worked that way, as far as I can remember. (So it seems that this isn't actually an instance of the category members bug that others have brought up - sorry, I should have read your original post more closely.) — Mr. Stradivarius ♪ talk ♪ 23:25, 3 November 2015 (UTC)
I do think there's /some/ new problem: The recategorization is caused by an edit of the page, not through a parser function or transcluded template. A null edit or maintainance task shouldn't be necessary then. Amalthea 08:59, 4 November 2015 (UTC)
Getting annoyed with this, so I filed phab:T117679. --Redrose64 (talk) 11:41, 4 November 2015 (UTC)

Delay in displaying and filling new categories

Why do some newly created categories (eg by year) still show up as red-links, and also a new category is still empty after I have added an article to it. Eg Bocaue Pagoda Tragedy which I altered from Category:1993 in the Philippines to Category:1993 disasters in the Philippines and created the latter as it was a (non-natural!) disaster. But despite the Pagoda article now having the new category and it showing as blue-linked, clicking on the link at the bottom of the article gets the new category but it is empty! I could understand if there was a short delay until slave servers are updated with new or updated articles and categories, but this was done yesterday! NB: I found a somilar problem after creating Category:1503 in Russia, but it seemed to require a space at the end of the standard country by year template before the final brackets. Hugo999 (talk) 03:41, 5 November 2015 (UTC)

If a page has a red link to another page that exists then a purge of the first page will fix it. See #Category membership issues for discussion of missing entries on category pages. PrimeHunter (talk) 03:58, 5 November 2015 (UTC)

JavaScript gadget problems

The problem I am about to describe may have something to do with Wikipedia:Village pump (technical)/Archive 141#Twinkle broken? or Wikipedia:Village pump (technical)/Archive 141#Pages without JS or CSS via Google, but editors there state the issues there have been resolved. Yet for me, the revert vandal, rollback, and revert good faith edits buttons that should be above each current edit 100% of the time now only appear intermittently. The RefToolbar and CharInsert also have this problem, even though all three gadgets are turned on in my preferences. Vycl1994 (talk) 06:09, 20 October 2015 (UTC)

If it's intermittent, that suggests that the problem isn't at your end but something elsewhere is being slow - either the bits server itself or your connection to it. --Redrose64 (talk) 08:54, 20 October 2015 (UTC)
Its been a smooth couple weeks in regards to RefToolbar, CharInsert, and Twinkle. Whatever was wrong seems to have been resolved. Thanks, Vycl1994 (talk) 02:35, 5 November 2015 (UTC)

I was perusing MOS:HEAD, and stumbled across the following:

"Before changing a section heading, consider whether you might be breaking existing links to that section."

It occurred to me that I know how to determine what pages link to a particular page (via the "What links here" link at the left of every page), but I have no clue how to determine what existing links there may be to a particular section. How do I see what links to a section, so that I may follow the advice given, and not break things if I fix a heading? Thanks! 1980fast (talk) 02:01, 26 October 2015 (UTC)

There is no easy way in MediaWiki to find links to a specific section of a page, so the only way to be certain is to check all the links in "What links here" individually. Personally, I wouldn't bother doing that all of that, as it would be a lot of tedious work for not much probable benefit. Instead, I would check any articles in "What links here" that you think are likely to contain links to that section, and I would also click "Hide transclusions" and "Hide links" so that only redirects to the page show up, and then check all of those redirects to see if they point to that section. User:WildBot (owned by Dispenser) used to check for these, but it hasn't edited in a couple of years. Does anyone know if there are any bots or tools out there at the moment that check for broken section links? — Mr. Stradivarius ♪ talk ♪ 05:57, 26 October 2015 (UTC)
Wildbot is actually Josh Parris's. What I recommend is making redirects for sections you want to link to. This way they'll be easy to fix (made rdcheck years ago). I also have some code that'll replacement section links with their equivalent redirect if they're close enough (The Glossary#G issue). — Dispenser 15:13, 27 October 2015 (UTC)
1980fast (talk · contribs), MediaWiki has regular expression searches. It would now be easy to write a template to do this.
  1. Go to a page that actually has a link to the desired pagename section (or make one, and go there).
  2. Experiment with {{regex}} there until your pattern matches the wikitext that is the hunted section-link.
  3. Release it into article space, and get your list. — CpiralCpiral 00:16, 5 November 2015 (UTC)
It may miss some links and give some false positives but you can make a blank search to get the main search form and then enter the expected link text in quotation marks like "Wikipedia:Manual of Style#Section headings". Click "Everything" to search all namespaces like [46]. PrimeHunter (talk) 11:54, 26 October 2015 (UTC)
Haven't tested myself, but there is such tool. --Edgars2007 (talk/contribs) 11:59, 26 October 2015 (UTC)
this is not 100% foolproof, but in most cases, using the "insource:" (don't forget the colon) magic serachword should do it. include the article name and the section name in quotes: i.e., serch for insource:"Article name#Section name". e.g., to find all the links to WP:CENSOR, use [47]. i chose this example simply because it was the first section-link i found... as i said, this is not perfect. specifically, if there are any redirects to the article (*not* to specific sections), it's possible that someone linked to the section using [[redirect name#section name]], and this search will not find it. there are probably other links you will miss. peace - קיפודנחש (aka kipod) (talk) 17:32, 5 November 2015 (UTC)

Watchlist subpages

Is there a way to watchlist "all subpages of a given page" (for namespaces that support subpages) rather than having to explicitly and separately watchlist each subpage? DMacks (talk) 06:07, 3 November 2015 (UTC)

Not that I'm aware of. This would be a good idea for a gadget, though. The question is, what would be the best interface for it? A dropdown menu next to the star icon, perhaps? — Mr. Stradivarius ♪ talk ♪ 13:17, 3 November 2015 (UTC)
Actually, just putting it in the "More" dropdown menu (in the Vector skin) might be best. — Mr. Stradivarius ♪ talk ♪ 13:20, 3 November 2015 (UTC)
This works well for, say, five subpages, but consider how many subpages there are of Wikipedia:Articles for deletion, or somesuch. Any tool would need to have a limit on how many pages could be watched in one go. Relentlessly (talk) 13:36, 3 November 2015 (UTC)
"All subpages of a given page" ought to include subpages that haven't been created yet. That's obviously much more difficult and would need changes to the core software. -- John of Reading (talk) 07:41, 4 November 2015 (UTC)

Javascript puts logo with banner back

MediaWiki:Common.css was edited today to remove the rules that used to replace the Wikipedia globe logo in the corner of every page with a celebratory logo with a red banner. Thus, the logo with the banner should no longer appear. However, I'm still seeing the logo with the red banner, if I enable Javascript in my browser. The rules are no longer in common.css, but some javascript on the page adds CSS rules to the DOM that put the celebratory logo there. This shouldn't happen.

Why does this happen, and how do we fix this? I don't have enough webdev-fu to debug this. Flushing the browser caches didn't help. ?action=purge didn't help. It was suggested that changes to common.css may take up to five minutes to take effect, but it's been over 50 minutes since the edit. It is possible that what you see depends on your user preferences or skin. If you could help, I'd appreciate it.

I have cross-posted this problem from the #wikimedia-tech channel on the Freenode IRC network. – b_jonas 18:49, 3 November 2015 (UTC)

It's gone for me.. are you logged in, out, what browser, did you try restarting it ? —TheDJ (talkcontribs) 18:52, 3 November 2015 (UTC)
Reproducible as an anonymous user. The pages loads this url, which is not updated with the new content. Seems caching gremlins have returned to Common.js and Common.css. —TheDJ (talkcontribs) 19:01, 3 November 2015 (UTC)
The stylesheet cache rolls over every 5 minutes and was updated swiftly. The module update got skipped due to a rare race condition between the hundreds of web servers. It seems the old revision of the page was effectively re-cached under the new revision ID. I've done another update to mitigate the issue for now. –Krinkle 19:23, 3 November 2015 (UTC)
Krinkle has explained that the "module" he's referred to is the "site" Resourceloader module, which indeed loads a copy of the styles from javascript for arcane and historical reasons.
The banner logo is gone now, thank you. I hope it's fixed for everyone else as well, if not, please mention it here. – b_jonas 10:20, 4 November 2015 (UTC)

Adding Template:Main Page banner to the Mobile site front page

 You are invited to join the discussion at Talk:Main Page#Adding Template:Main Page banner to the Mobile site front page. Thanks. Mz7 (talk) 21:07, 4 November 2015 (UTC)

Move log

The "revert" link does not appear next to entries in the move log that are moves over redirects. GeoffreyT2000 (talk) 02:03, 5 November 2015 (UTC)

Please include examples in reports so others don't have to search for them. Move over redirect, Move not over redirect. I don't know why a move over redirect has no revert link. PrimeHunter (talk) 03:31, 5 November 2015 (UTC)
After some epic archaeology in the MediaWiki codebase, it looks like this is simply due to an oversight by Aaron Schulz in 2008 (the important commit that removed "revert" links from move log entries was r33234). So you could file a Phabricator task if you would like it changed. All the same, one has to remember that the revert link doesn't perform any undeletions of the redirect that was there before. — This, that and the other (talk) 08:38, 5 November 2015 (UTC)

Recent contributions from a group of editors

Is there a good tool to track recent contributions from a group of usernames listed on, say, a meetup page, without going through wikimetrics or an education program extension?--Pharos (talk) 14:52, 5 November 2015 (UTC)

distance around .OGG files?

A minor issue, that has irritated me for some time: see the anthem in the infobox in Germany. On my display (vector skin, Firefox 41.0.2) the grey bar icon to play the .OGG-file overlaps the preceding text "(English: "Song of Germany")" - most of it is visible, but the lower pixels of "g" and "y" are hidden by the .OGG bar. I tried adding "border" to the file tag, as that is supposed to add a tiny white border around File links, but that didn't have a visual effect. What's the best way to re-arrange those 2 elements a bit to get some distance between them? Note: I found Wikipedia:Village_pump_(technical)/Archive_64#Rendering_of_ogg_player which may be related, but the bug request was declined for a technical reason. GermanJoe (talk) 14:23, 25 October 2015 (UTC)

It's not just you. I have Modern skin, same version of Firefox 41.0.2. I also have IE 11. Same thing on both browsers. I also toyed around with the Zoom, because sometimes that can clear oddities. Zoom has no effect on this. No answers, but it's not just Firefox. — Maile (talk) 14:34, 25 October 2015 (UTC)
Village pump (technical)/Archive 141
Anthem: Song of Germany
With JavaScript disabled I see another dark grey media player box which doesn't overlap. When JavaScript is enabled and I reload the page, I briefly see the dark grey media player box and it doesn't overlap. Then the box disappears and the content below it moves up to fill the hole. Then a light grey media player box appears instead and the content below it moves down, but the box overlaps the text above it a little. With no infobox but just '''Anthem:''' Song of Germany<br />[[File:German national anthem performed by the US Navy Band.ogg]] I see something similar:
Anthem: Song of Germany
However, this case starts out with more space between the text and the dark grey box, so when the light grey box appears instead and moves up, it doesn't move enough to overlap the text. PrimeHunter (talk) 15:07, 25 October 2015 (UTC)
It's the same behaviour for me: First a darker bar during loading, which has some distance and looks OK. And then another light-grey version of the bar is displayed with the mentioned overlap problem. Java has its own set of display elements of course, but I am really not that experienced with the technical details. GermanJoe (talk) 15:24, 25 October 2015 (UTC)
Java and JavaScript have absolutely nothing to do with each other. --71.119.131.184 (talk) 16:41, 25 October 2015 (UTC)
I added a line break <br /> to move the player down a line. I then found that the United States article already uses exactly that same fix. I randomly checked Mexico and Brazil, both had the same problem and I applied the same fix. It looks like this is a widespread issue. Do we need to open this as a bug on Phabricator? It's really strange that the light-grey music player renders a few pixels too high when it's inside an infobox. In fact, I think it should probably be a pixel or two lower even when it's not in an infobox. I don't like the way it still touches the bottom of g and y. Alsee (talk) 08:18, 28 October 2015 (UTC)
Yes, please file a bug in phabricator -- a few of us are working on improvements to the media playback interface, and this would be nice to fix up properly. Sounds like a CSS styling issue, probably combining funky styles from the infobox with funky styles from the player... --brion (talk) 20:10, 6 November 2015 (UTC)

List of Wikipedians by article count

Hello, WP:List of Wikipedians by article count has stopped updating since past few weeks. Although the bot is getting activated everyday, it is not updating anything. Can someone please give this a look? I also placed a request here about the same issue. Many thanks. Arun Kumar SINGH (Talk) 07:50, 30 October 2015 (UTC)

Thanks for raising this, Arun. I did some checks yesterday, after creating about 15 or so new articles, and the totals have not changed. The page is updated around 1am each morning. Lugnuts Dick Laurent is dead 13:01, 30 October 2015 (UTC)
Yes, please also note this recent discussion at WP:AN about this topic that AKS referenced, especially the comments by ErrantX, as ErrantX's comment seem to be directly relevant to the technical side of this issue. --IJBall (contribstalk) 18:27, 30 October 2015 (UTC)
Bump. Lugnuts Dick Laurent is dead 14:23, 2 November 2015 (UTC)

Hi. Right, so I'm the only person who maintains this report. This means that the report is pretty much entirely independent of the Wikimedia Foundation. It also means that if you don't ping me about the report not updating, no amount of bumps or noticeboard threads will help. :-)

I will take a look in the next day or two to see what's going on. The script has occasionally been throwing uncaught exceptions, but I'm still not totally sure why. I imagine this is related, though. Thank you for the concrete examples of incorrect stats. Those should help in diagnosing the issue and verifying a complete fix. --MZMcBride (talk) 14:02, 6 November 2015 (UTC)

Import limits

See WP:AN#Edward Sims Van Zile, or once that gets archived, see the section edited in this diff. The article had been copied from simple: without proper attribution, and someone suggested importing the other page to resolve the problem, but we had to pick a different route because it's not possible to import from simple:. Currently, the only normal Wikipedias whence we can import pages are German, Spanish, French, Italian, and Polish, plus the Nostalgia English Wikipedia, Meta, and a couple of technical WMF projects. (1) What technical barriers prevent us from importing from other projects, i.e. what would have to be changed to enable us to import from elsewhere? (2) Why was it decided to have such a small list of possible source projects? Nyttend (talk) 14:10, 30 October 2015 (UTC)

I think it was mostly limited for concerns about the quality of the content being imported (sourcing, copyright violations etc). If you can find consensus to expand the list, you can file a phabricator ticket to define additional wgImportSources. —TheDJ (talkcontribs) 15:09, 30 October 2015 (UTC)
@Nyttend and TheDJ: I'm working on some code to allow any wiki to import from any other (see phab:T17583). Fingers crossed, it should be done in the next month or two. — This, that and the other (talk) 09:02, 6 November 2015 (UTC)

Main Page oldid pages lack revision info – they look like live pages

The old version (oldid) pages of the Main Page are lacking the revision info and links usually found at the top of such pages, making them look like current pages. Example: 13 January 2003 Main Page version. Is this normal?

This is also the case with the oldid pages of for example the sv: and simple: front pages, but not with for example da: and no:. --Pipetricker 15:10, 4 November 2015 (UTC)

We change some things for the display of the main page. The info is hidden in the Vector and Monobook skins by display: none !important; for #contentSub in MediaWiki:Vector.css and MediaWiki:Monobook.css. Compare to https://en.wiki.x.io/w/index.php?oldid=585725&useskin=modern. I don't know whether it was different previously but the live version of Main Page in Vector currently has an empty <div id="contentSub"></div> so I don't know why we hide it. PrimeHunter (talk) 15:59, 4 November 2015 (UTC)
The Vector code was added by TheDJ 5 July 2009.[48] It may have been based on Monobook code from 9 August 2008 with edit summary "Fix padding weirdness on top of Main Page".[49] MonoBook also currently has an empty <div id="contentSub"></div> on Main Page. PrimeHunter (talk) 16:11, 4 November 2015 (UTC)
Archived versions at [50] and [51] also have an empty contentSub on those dates. PrimeHunter (talk) 16:18, 4 November 2015 (UTC)
Just to confirm, yes, the mainpage Vector directives were either directly copied ( like most of the first few days content of Vector ) from the monobook adaptations. —TheDJ (talkcontribs) 22:19, 5 November 2015 (UTC)

Safari not displaying images in articles

As of this morning, I noticed for the first time that Safari is not displaying images in the articles. I know Safari has WikiMedia issues, but I've had this new Mac for weeks, doing quite a bit of work on Wikipedia, and I don't recall seeing this issue before. Anyone else notice a recent problem? Works fine when I download Firefox, of course. Shawn in Montreal (talk) 14:44, 5 November 2015 (UTC)

Works fine for me on for example Sean Combs, Normandy landings. I have Safari as my backup and normally use Chrome. I am running Mavericks -- Diannaa 🍁 (talk) 20:13, 5 November 2015 (UTC)
Okay thanks. I'm on Yosemite. Shawn in Montreal (talk) 20:15, 5 November 2015 (UTC)
It's been fine for me on El Capitan. Imzadi 1979  20:26, 6 November 2015 (UTC)

Has Wiki Code change Prevented saving as "Web Page Complete?"

I periodically save some Wikipedia pages to my hard drive, to have a record of how before and after edits appeared, which helps me a lot to understand template coding, as well as other forms of structure in Wikipedia.

For the last few months, however, when I click to save as "web page complete," the result does not include most of the photo images that are in the article too. Instead, where the photos are located in the article, there is complex code, but no actual photo.

This has been going on only for two or three months now. Since I don't have the same problem when I save pages from other websites, I am assuming it is the result of some kind of Wikipedia code revision. I use the latest version of Firefox for my browser and have Windows 10 OS or Windows 7, on the computers I am using to access Wikipedia. I get the same "save as" result with both of those OSs.

Wondering if this will be a permanent situation in Wikipedia, or if some pending revision of your software will enable me to save Wiki pages as I did before -- as a web page complete?EditorASC (talk) 00:11, 6 November 2015 (UTC)

"complex code"... can you try describing that again... :) —TheDJ (talkcontribs) 07:51, 6 November 2015 (UTC)
It works for me, except the logo is missing. Firefox 42.0 on Windows Vista 32-bit. Try saving Couch in a folder. Which images are missing? Do you get a subfolder called something like "Couch - Wikipedia, the free encyclopedia-files"? If so, are the images in that folder? I for example get "300px-2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartme.jpg", a copy of https://up.wiki.x.io/wikipedia/commons/thumb/c/c2/2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg/300px-2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg. Can you view source (Ctrl+U) of the saved page and copy-paste the code where an image is missing? For example something like:
<div class="thumb tright">
<div class="thumbinner" style="width:302px;"><a href="https://en.wiki.x.io/wiki/File:2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg" class="image"><img alt="" src="Couch%20-%20Wikipedia,%20the%20free%20encyclopedia-files/300px-2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartme.jpg" class="thumbimage" srcset="//up.wiki.x.io/wikipedia/commons/thumb/c/c2/2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg/450px-2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg 1.5x, //up.wiki.x.io/wikipedia/commons/thumb/c/c2/2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg/600px-2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg 2x" data-file-width="3888" data-file-height="2592" height="200" width="300"></a>
<div class="thumbcaption">
<div class="magnify"><a href="https://en.wiki.x.io/wiki/File:2009-05-16_Main_office_lobby_at_Hampton_Forest_Apartments.jpg" class="internal" title="Enlarge"></a></div>
A three-cushion couch in an office lobby</div>
</div>
</div>
PrimeHunter (talk) 13:40, 6 November 2015 (UTC)

User profile

Special:UserProfile, which is displayed when you click on the username in a mobile diff, displays just plain text with blue (unvisited) or purple (visited) links instead of putting "Edited the page" and "Thanked by" in two boxes and making the links black. GeoffreyT2000 (talk) 02:26, 6 November 2015 (UTC)

Please include in example in reports. Here is one: http://en.m.wiki.x.io/wiki/Special:UserProfile/GeoffreyT2000. It sounds similar to Wikipedia:Village pump (technical)/Archive 136#Some sort of presentation bug in April where Jdlrobson posted when it was fixed. Today I don't see boxes at any tested wikis, except a "Talk to" box. PrimeHunter (talk) 13:15, 6 November 2015 (UTC)

So how come I have the "underline links" option set to "always" in my Preferences, yet the links in the article ToCs are no longer underlined? (The rest of the links are underlined as they should be). Can this be fixed, please, or do I need to add yet another hack to my css sheet? I'm using Monobook skin, if this matters. Thanks.—Ëzhiki (Igels Hérissonovich Ïzhakoff-Amursky) • (yo?); November 6, 2015; 15:48 (UTC)

Me too - Vector skin. Also affects Commons. Optimist on the run (talk) 16:43, 6 November 2015 (UTC)
It's caused by Mediawiki's CSS treating the TOC as a table. It seems to be discussed in phab:T92481, but I'm not sure if that's the correct ticket or not. Relentlessly (talk) 17:11, 6 November 2015 (UTC)
I can confirm, the issue is related to the change that was the result of T92481.. Once more we have found a difference in the browser implementations :) —TheDJ (talkcontribs) 19:51, 6 November 2015 (UTC)

View Test in the skin:

When I select "Underline links: Skin or browser default" at Special:Preferences#mw-prefsection-rendering, I see underlined TOC links in Cologne Blue and no other skins. With "Underline links: Always" I also see it in Modern, so the setting is able to affect the TOC in some circumstances. PrimeHunter (talk) 17:45, 6 November 2015 (UTC)

VisualEditor News #5—2015

Read this in another languageSubscription list for this multilingual newsletter

 
Did you know?
You can use the visual editor on smartphones and tablets.

 

Click the pencil icon to open the editor for a page. Inside that, use the gear menu in the upper right corner to "Switch to visual editing".

The editing button will remember which editing environment you used last time, and give you the same one next time. The desktop site will be switching to a system similar to this one in the coming months.

You can read and help translate the user guide, which has more information about how to use the visual editor.

Since the last newsletter, the VisualEditor Team has fixed many bugs, added new features, and made some small design changes. They post weekly status reports on mediawiki.org. Their workboard is available in Phabricator. Their current priorities are improving support for languages like Japanese and Arabic, making it easier to edit on mobile devices, and providing rich-media tools for formulæ, charts, galleries and uploading.

Recent improvements

Educational features: The first time you use the visual editor, it now draws your attention to the Link and ⧼visualeditor-toolbar-cite-label⧽ tools. When you click on the tools, it explains why you should use them. (T108620) Alongside this, the welcome message for new users has been simplified to make editing more welcoming. (T112354) More in-software educational features are planned.

Links: It is now easier to understand when you are adding text to a link and when you are typing plain text next to it. (T74108, T91285) The editor now fully supports ISBN, PMID or RFC numbers. (T109498, T110347, T63558) These "magic links" use a custom link editing tool.

Uploads: Registered editors can now upload images and other media to Commons while editing. Click the new tab in the "Insert Images and media" tool. You will be guided through the process without having to leave your edit. At the end, the image will be inserted. This tool is limited to one file at a time, owned by the user, and licensed under Commons's standard license. For more complex situations, the tool links to more advanced upload tools. You can also drag the image into the editor. This will be available in the wikitext editor later.

Mobile: Previously, the visual editor was available on the mobile Wikipedia site only on tablets. Now, editors can use the visual editor on any size of device. (T85630) Edit conflicts were previously broken on the mobile website. Edit conflicts can now be resolved in both wikitext and visual editors. (T111894) Sometimes templates and similar items could not be deleted on the mobile website. Selecting them caused the on-screen keyboard to hide with some browsers. Now there is a new "Delete" button, so that these things can be removed if the keyboard hides. (T62110) You can also edit table cells in mobile now.

Rich editing tools: You can now add and edit sheet music in the visual editor. (T112925) There are separate tabs for advanced options, such as MIDI and Ogg audio files. (T114227 and T113354) When editing formulæ and other blocks, errors are shown as you edit. It is also possible to edit some types of graphs; adding new ones, and support for new types, will be coming.

On the English Wikipedia, the visual editor is now automatically available to anyone who creates an account. The preference switch was moved to the normal location, under Special:Preferences.

Future changes

You will soon be able to switch from the wikitext to the visual editor after you start editing. (T49779) Previously, you could only switch from the visual editor to the wikitext editor. Bi-directional switching will make possible a single edit tab. (T102398) This project will combine the "Edit" and "Edit source" tabs into a single "Edit" tab, similar to the system already used on the mobile website. The "Edit" tab will open whichever editing environment you used last time.

Let's work together

If you can't read this in your favorite language, then please help us with translations! Subscribe to the Translators mailing list or contact us directly, so that we can notify you when the next issue is ready. Thank you!

Whatamidoing (WMF) 04:16, 30 October 2015 (UTC)

Visual editor putting in HTML markup?

Take a look at this edit.[52]. This was the first edit by a new user, using Visual Editor, and added many bogus <div> and <span> tags. No idea what that editor did, but all those HTML tags probably were not inserted by hand. Possible Visual Editor bug, or there's some way a new editor can get it very confused. John Nagle (talk) 20:32, 30 October 2015 (UTC)

This happens if you do "Select All" and from the Style menu, select "language" and just pick a language. —TheDJ (talkcontribs) 23:17, 30 October 2015 (UTC)
Maybe they were trying to translate it? — Mr. Stradivarius ♪ talk ♪ 00:37, 31 October 2015 (UTC)
To Sinhala? --Redrose64 (talk) 09:25, 31 October 2015 (UTC)
Mr. Stradivarius, that sounds like a very plausible guess. I've requested that the actual purpose (which is to label text that isn't in the wiki's usual content language) be clarified in phab:T117460. Whatamidoing (WMF) (talk) 19:12, 2 November 2015 (UTC)

@Whatamidoing (WMF), Nagle, and Mr. Stradivarius: Already filed bug report on this... T116887, but was marked as closed as not a VE error, but user error. Already filed and closed as user error were empty span tags and div tags. T110995 for unnecessary font tags is still open. Content Transcrapulator (CX) does the same thing. One has to file separate bug reports for VE and CX as they use different versions of parsoid. T110173 is the ticket for unnecessary span tags in CX. I continue to see these errors every day by different editors. Alot of errors are picked up by CheckWiki or I notice while fixing other CheckWiki errors. A group of us file bug reports that are mostly closed as user error or just closed for no reason. The rest we get emails when they put it further back on the queue. Bgwhite (talk) 01:16, 4 November 2015 (UTC)

I believe this diff does actually represent a user error. That's why the subject of phab:T117460 is finding a way to tell users that they are (probably) about to make a mistake.
I haven't followed anything with CX; the only thing I hear about it is that it's not using VisualEditor. I should try it out again and see whether it's compatible with NoScript these days. Whatamidoing (WMF) (talk) 01:57, 4 November 2015 (UTC)
This looks like a "developer in denial" situation. Clearly, VE is inserting tags in article space it has no business inserting. It's violating WP:HTML: "Most HTML can be included by using equivalent wiki markup or templates; these are preferred within articles, as they are simpler for most editors, and less intrusive in the editing window.". HTML is supposed to be used only in templates, not where editors have to explicitly deal with it. Also, <div> and <span> aren't on the allowed list of HTML tags in WP:HTML. John Nagle (talk) 05:17, 4 November 2015 (UTC)

WP:HTML#div and WP:HTML#span both work for me.

Most HTML can be included by using equivalent wiki markup or templates; these are preferred within articles VE has no way to know whether there is an acceptable substitute for a particular piece of HTML it is dealing with.

HTML is supposed to be used only in templates I have dealt with it in some small amount in a handful of articles. --Izno (talk) 12:37, 4 November 2015 (UTC)

John, I realize that it's frustrating to see a big mess like that. I've cleaned up my share of such problems, as you would expect of anyone who's spent so much time at Wikipedia. But the real problem isn't the HTML.
Izno's analysis is correct, but try this as a thought exercise: Let's pretend that WP:HTML were a policy rather than an essay, and that it was magically possible for software to comply with the directly conflicting requirements of every wiki,[1] and that it was possible for the software to know which standard HTML elements map to which local templates (and which HTML elements don't have local templates), so that if a community preferred templates, then those templates would automagically be substituted. Let's also say that this system is working perfectly, and that this wiki had chosen to always use {{lang|si}} instead of the equivalent HTML tags.
Let's say that in this template-based system, the same user went to the same page and did exactly the same thing, and that you checked the diff. Then you would have seen every element on the page wrapped in the {{lang|si}} template, rather than every element on the page wrapped in the HTML tag. It would still be a mess, and it would still need to be fixed.
That's because the real problem is that the user didn't understand the tool. The real problem isn't the exact method used to document the fact that the user was confused; the real problem is that the software was unclear to the user. Therefore, the most relevant and practical solution is: Make the tool be less confusing. Making the tool's purpose more obvious might result in what we actually want, which is neither HTML tags that incorrectly label the whole page as Sinhalese nor local templates that incorrectly label the whole page as Sinhalese, but instead a page that does not have all the English text marked as being in Sinhalese through any method at all. Whatamidoing (WMF) (talk) 21:16, 4 November 2015 (UTC)
  1. ^ For example, WP:HTML was forked some years ago from m:Help:HTML in wikitext, which takes the opposite approach. It opens with a lengthy list of explicitly permitted HTML.
@Nagle and Whatamidoing (WMF): I used to see 1-2 <h2> tags a month, now with VE I see upwards 10 a day (9 for today). When I fix them later, I'll expect to see the <h2> tags surrounded by <p> and <li> tags, with <br> tags inside the <h2> tags every few days. Why are those tags surrounded or inside, who knows. Yes, a bug report has been filed and has been ignored. Sorry Izno, but VE does know there is an acceptable substitute for a section heading.
Whatamidoing's answer goes along with what they say at developer's conferences... 1) wikicode is going away and only HTML/VE. 2) WMF does not have to abide by community MOS.
WP:HTML is an essay, but WP:ACCESSIBILITY says use wikicode when possible and WP:MOS says use HTML sparingly. But, Whatamidoing uses a Meta help page which says what HTML codes are possible, not what should be used. Meta help page trumps community MOS. Both VE and CX are littering pages with HTML, dates wikilinked, templates (including cite templates) left in foreign languages and other messes. This goes against what Wikicode was originally created for, to make editing a Wiki page relatively simple. Whatamidoing is only here for the users of VE. The peons who cleanup VE's messes don't matter to the WMF, and thus doesn't matter to Whatamidoing either. Bgwhite (talk) 06:36, 5 November 2015 (UTC)
Technically, VisualEditor doesn't know the wikitext for ==Section headings==. It "thinks" in HTML. VisualEditor sends <h2> headers to Parsoid, whose sole purpose is to turn HTML into wikitext (and vice versa). If that's not happening, then it's either a bug and it needs to be fixed, or it's the return of phab:T93754's problem with a RESTbase race condition—and it still needs to be fixed. I don't see any bug reports or recent comments from you on Phab about the h2 header problem, but I'll make sure it gets on the list. Please feel free to drop by WP:VEF or my talk page whenever you want more information about a problem, or even to have someone complain loudly remind the team about one that you think they're ignoring. Whatamidoing (WMF) (talk) 07:41, 5 November 2015 (UTC)
Whatamidoing (WMF) T113173 is the ticket filed for <h2> problem. It was closed as a duplicate of T116460. I'm already a subscriber on the ticket. Today's list of articles with header problems are Ahoy (greeting), Cross merchandising, Gentamicin, Gill Kalan, Knowledge-based decision making, Tigecycline, Transcendent (television series), William Spiggot, Groth Law Firm. Tigecycline and Chloroquine also contain <li> tags in an already defined wikicode list, thus it breaks accessibility guidelines by creating a list made up of multiple lists. Bgwhite (talk) 09:48, 5 November 2015 (UTC)
Thanks for the link. I've got a little list of bugs for a meeting tomorrow. Whatamidoing (WMF) (talk) 20:38, 5 November 2015 (UTC)

Pages with DOIs inactive since 2014 category

  Resolved

Could somebody please tell me what is causing the deleted category Category:Pages with DOIs inactive since 2014 to appear on Real ear measurement. Thanks, JMHamo (talk) 00:02, 7 November 2015 (UTC)

Caused by ref 3 which has |doi_brokendate= set. Keith D (talk) 00:47, 7 November 2015 (UTC)
More specifically, that {{cite journal}} has |doi_brokendate=2014-03-28 --Redrose64 (talk) 00:56, 7 November 2015 (UTC)
@Keith D and Redrose64: Thank you. I've removed |doi_brokendate=2014-03-28 from the citation and it no longer appears. JMHamo (talk) 01:21, 7 November 2015 (UTC)
Deleting an error message does not resolve the error. I have recreated the tracking category with some explanatory text. It looks like that article was in a draft state until the last few days, so the category was not applied until the article moved to mainspace. I reported the broken DOI to crossref.org, which may be able to re-establish a link to it. – Jonesey95 (talk) 16:58, 7 November 2015 (UTC)

Bullet randomly appearing

Hi, yesterday I created American Morgan Horse Association. I used the infobox organization at the top of the article and filled in the most-needed parameters (I plan to fill in more later). The association is headquartered in Shelburne, Vermont and every time I look at the article, a bullet pops up in the infobox in front of the words "Shelburne, Vermont". I've checked it several times and can't find any error, and the bullet disappears after a couple of seconds. This is very weird and I've never had it happen before. White Arabian mare (Neigh) 00:39, 7 November 2015 (UTC)

The location parameter produces different code because {{Infobox organization}} has this:
| label19 = Location
| class19 = label
|  data19 = {{Unbulleted list
             | 1 = {{comma separated entries
                    | 1 = {{#if:{{{location_city|}}}    |<span class="locality">{{{location_city}}}</span>}}
                    | 2 = {{#if:{{{location_country|}}} |<span class="country-name">{{{location_country}}}</span>}}
                   }}
             | 2 = {{{location|}}}
             | 3 = {{comma separated entries
                    | 1 = {{#if:{{{location_city2|}}}    |<span class="locality">{{{location_city2}}}</span>}}
                    | 2 = {{#if:{{{location_country2|}}} |<span class="country-name">{{{location_country2}}}</span>}}
                   }}
             | 4 = {{{location2|}}}
            }}
I don't see a bullet in Firefox 42.0 on Windows Vista but other browsers may react differently. The infobox produces this wikitext for the cell:
<td class="label">
<div class="plainlist"><ul><li>Shelburne, Vermont</li></ul></div></td>
The rendered page gets identical html apart from line breaks. MediaWiki:Common.css contains:
/* Unbulleted lists */
.plainlist ol,
.plainlist ul {
    line-height: inherit;
    list-style: none none;
    margin: 0;
}
"none" prevents <ul>...</ul> from producing a bullet but your browser may render the original page long enough to see the bullet before the css is applied.
Here is <div class="plainlist"><ul><li>Shelburne, Vermont</li></ul></div>:
  • Shelburne, Vermont
Here is <div><ul><li>Shelburne, Vermont</li></ul></div>:
  • Shelburne, Vermont
I never see a bullet in the first but permanently (as expected) see a bullet in the second where I removed the plainlist class. PrimeHunter (talk) 01:22, 7 November 2015 (UTC)

Ok, thank you. White Arabian mare (Neigh) 02:58, 7 November 2015 (UTC)

Redirect templates in mobile

In the mobile Wikipedia, "DISPLAY CATEGORY OVERVIEW →" and the redirect templates below it are not shown when clicking on "Page issues" in a redirect page such as Angular units. GeoffreyT2000 (talk) 22:07, 7 November 2015 (UTC)

Your post is apparently about the difference between the desktop https://en.wiki.x.io/w/index.php?title=Angular_units&redirect=no and the mobile http://en.m.wiki.x.io/w/index.php?title=Angular_units&redirect=no. The redirect page contains {{redr|from move|from plural}} which is displayed on the desktop redirect page but not the mobile. {{redr}} uses {{mbox}} which in mainspace adds the classes metadata and ambox, both of which are not displayed in mobile as far as I can tell from https://github.com/wikimedia/mediawiki-extensions-MobileFrontend/blob/master/resources/skins.minerva.content.styles/hacks.less. I find it more odd that the mobile redirect page displays (at least for me in Firefox on Windows Vista) a cut off version of File:Angle measure.svg, the first image on the redirect target Angular unit. PrimeHunter (talk) 23:15, 7 November 2015 (UTC)

Mickopedia

When viewing a page such as Mickopedia on Mickopedia (see e.g. [53]) in Internet Explorer, it says that it is HTTP 403 Forbidden because it requires you to log in. GeoffreyT2000 (talk) 23:58, 7 November 2015 (UTC)

Wikipedia has nothing whatever to do with Mickopedia, and it is very unlikely that anybody here can answer any technical questions about it. --ColinFine (talk) 00:14, 8 November 2015 (UTC)
I'm also unable to access any tested pages at the site except http://mickopedia.stroma.org. There are lots of Wikipedia:Mirrors and forks with no association to the Wikimedia Foundation which runs Wikipedia. The sites may break, close or restrict access at any time with nothing we can do about it. PrimeHunter (talk) 01:07, 8 November 2015 (UTC)

Superprotect is gone

Superprotect was introduced by the Wikimedia Foundation to resolve a product development disagreement. We have not used it for resolving a dispute since. Consequently, today we are removing Superprotect from Wikimedia servers.

Without Superprotect, a symbolic point of tension is resolved. However, we still have the underlying problem of disagreement and consequent delays at the product deployment phase. We need to become better software partners, work together towards better products, and ship better features faster. The collaboration between the WMF and the communities depends on mutual trust and constructive criticism. We need to improve Wikimedia mechanisms to build consensus, include more voices, and resolve disputes.

There is a first draft of an updated Product Development Process that will guide the work of the WMF Engineering and Product teams. It stresses the need for community feedback throughout the process, but particularly in the early phases of development. More feedback earlier on will allow us to incorporate community-driven improvements and address potential controversy while plans and software are most flexible.

We welcome the feedback of technical and non-technical contributors. Check the Q&A for details.

Quim Gil, Engineering Community Manager @ Wikimedia Foundation -- 17:33, 5 November 2015 (UTC)

Thankyou! Alsee (talk) 06:48, 8 November 2015 (UTC)

Do we normally delete old MediaWiki interface pages that are no longer in use?

See the CSD request at MediaWiki talk:Antispoof-name-conflict. No idea what we normally do in cases like this, thought it better to come here and ask first rather than just delete it. Jenks24 (talk) 03:35, 24 October 2015 (UTC)

Looks like we conflicted, sorry :S I think it's fine to delete under G6 given nothing uses it anymore. Legoktm (talk) 03:38, 24 October 2015 (UTC)
No worries. I just wasn't sure if we marked them historical or something, but I'm sure you'd know better than me. I'll delete the talk page. Jenks24 (talk) 03:44, 24 October 2015 (UTC)
I'd have liked to have seen this kept as historical. All the best: Rich Farmbrough, 21:37, 8 November 2015 (UTC).

Search in Google

Hi. Please tell me how to remove it?. -- Дагиров Умар (talk) 21:22, 7 November 2015 (UTC)

For others searching for this: the article is at ce:Соьлжа-ГӀала, you can see the problem in this Google search, and the "Коьрта агӀо" part seems to mean "Main page" in Chechen. @Дагиров Умар: I don't see anything strange happening at the ce:Соьлжа-ГӀала page, so I can only assume that this is a mistake on Google's part. We don't control how Google displays their search results, so I'm afraid that you will have to ask them about it. Best — Mr. Stradivarius ♪ talk ♪ 10:08, 8 November 2015 (UTC)
Google's title choice can be hard to predict or affect. Many sites have a poor title in <title>...</title> so it often makes sense to use something else, but it's done completely automatically with varying results. See https://support.google.com/webmasters/answer/35624?hl=en and https://yoast.com/google-page-title/. I have no idea how to get Google to remove the unwanted text. PrimeHunter (talk) 13:49, 8 November 2015 (UTC)
I see that you recently created MediaWiki:Pagetitle. Maybe wait for some time. But anyway it is strange - I see, that previously you have used the translatewiki version, which is the same for Wikipedia. --Edgars2007 (talk/contribs) 14:13, 8 November 2015 (UTC)

Lowercase title

Template:R from Java package name displays a lowercase "r", even though it has {{lowercase title}} within includeonly tags. See Template talk:Lowercase title#.3Cincludeonly.3E.7B.7Blowercase_title.7D.7D.3C.2Fincludeonly.3E_doesn.27t_work. GeoffreyT2000 (talk) 15:46, 8 November 2015 (UTC)

As I asked there, why is it a problem? --Redrose64 (talk) 16:02, 8 November 2015 (UTC)
It happens because it transcludes itself. It could be avoided by testing wether {{PAGENAME}} is R from Java package name but I fail to see a problem needing a fix. PrimeHunter (talk) 17:08, 8 November 2015 (UTC)
There was more potential for confusion at Wikipedia:Template messages/Redirect pages which displayed "template" in the title so I have used {{main other}} to only apply lowercase in mainspace.[54] PrimeHunter (talk) 17:34, 8 November 2015 (UTC)

Whole content of Wikipedia as PDF files

Hi.

Instead of using web crawlers to fetch pdf versions of all the articles in Wikipedia, is there any other option available? Possible reasonable fee would also be ok in this case.

And if using web crawlers is the only option, what would be the guidelines for using that option to e.g. avoid unnecessary load on Wikipedia servers?

Kind Regards,

Jukka Pesonen — Preceding unsigned comment added by Jukkapesonen (talkcontribs) 10:59, 27 October 2015 (UTC)

@Cscott: One that you might be able to answer? I guess the question is, Jukka, why do you want to do this? — This, that and the other (talk) 11:25, 27 October 2015 (UTC)

Wikipedia content would be used for developing and testing full text search technologies with pdf preview capability locally in our company. Respecting the great service provided by Wikipedia community, we would like to get the content with no interference to the service. Assuming naturally that this kind of utilization of the content would be ok in the first place?

If we can make this happen, I believe it would not be too hard to convince our management to consider a small donation to Wikipedia.

Kind Regards,

Jukka Pesonen — Preceding unsigned comment added by Jukkapesonen (talkcontribs) 08:56, 28 October 2015 (UTC)

Jukkapesonen, Wikipedia&more is available for free download at dumps.wikimedia.org.... the most recent HTML version of English Wikipedia is at dumps.wikimedia.org/other/static_html_dumps/current/en/. We even have a page of Torrent links for English and a few other languages. We have a page on how to get and use the Wikipedia:Database_download. You should probably also check our page on Wikipedia:Copyrights, especially the section on Reusers' rights and obligations. It's all free to download, but there are conditions on reuse. If you redistribute any of it, basically you need to credit the original authors, you need to label it as licensed under Creative Commons Attribution-Share-Alike License 3.0 or later, if you make modifications you have to indicate it's been modified, and you must license your changes under the same terms. Alsee (talk) 13:19, 28 October 2015 (UTC)
I think the html dump would be the easiest to use as there are many tools available for coverting html to pdf. Roger (Dodger67) (talk) 06:48, 29 October 2015 (UTC)

Which tool is Wikipedia using for pdf rendering and is it free to use by others too? If it is not available, what would be the second best option?

Kind Regards,

Jukka Pesonen — Preceding unsigned comment added by 87.100.243.238 (talk) 18:58, 29 October 2015 (UTC)

Jukka, the last I heard, there are bugs in the rendering of WP's xml to pdf, which the html rendering does not suffer from. --Ancheta Wis   (talk | contribs) 22:03, 7 November 2015 (UTC)
We use mw:Offline content generator to generate PDFs from Parsoid-format DOM stored in RESTBase. You can install it on your own servers and run it against wmf content to reduce the load on our servers. Our request rate to the PDF endpoints ranges between 1 to 4 requests per second. If you keep your third-party use to a small fraction of that, we shouldn't be too affected by the additional load. C. Scott Ananian (talk) 21:28, 9 November 2015 (UTC)

Prompt for edit summary when only using default section summary

I have selected the preference for the wiki software to warn me if I try to save an edit with a blank edit summary, and I think it has helped me. Until today (or at least quite recently) It treated a summery that contained the defualt section summary on a section edit (/* Section name */) as if it were blank, and still prompted me for a fuller edit summary. Today it doesn't, but it does still propmpt if the summary is blank. I prefer the previous behavior. Is this an intentional change? I use FireFox on MS Windows 7 (64 bit). DES (talk) 00:20, 3 November 2015 (UTC)

It prompted me when I tried it just now. Do you happen to have some user script or gadget that changes the summary somehow? Anomie 13:37, 3 November 2015 (UTC)
subtle point: as far as i know, the logic is not necessarily "default section name". the logic compares the "pre-filled" summary with the content of "summary" field, prompts you if they are equal. the difference may seem meaningless, but under some conditions it isn't: for instance, if you choose the gadget to add "edit" link to the lead section, you'll see /* top */ as the summary when editing the lead section. however, this summary was added by the gadget, and is not "pre-filled", as far as mw software is concerned, so you won't get prompted if you leave it as is (you *will get prompted if you delete it - "nothing" is the pre-filled value from mw POV). there are probably other such cases. this works the other way too, i believe: e.g., when you hit "undo", there's elaborate pre-filled summary, which is not "section name". mw will still prompt you to fill one if you leave it unchanged. peace - קיפודנחש (aka kipod) (talk) 15:16, 4 November 2015 (UTC)
This seems to be the opposite of phab:T19416. We should add a way to override the change that it made. Jackmcbarn (talk) 02:29, 9 November 2015 (UTC)
For gadgets it's simple enough: if the hidden wpAutoSummary form field matches the md5 hash of the summary, it will prompt for a summary. If not, it won't. So have your gadget update the field appropriately and you're good. That probably doesn't help much for the "edit lead section" gadget, though, or other things using the summary URL parameter instead of JS on the edit form. Anomie 16:47, 9 November 2015 (UTC)

TOC

The TOC in Wikipedia:Village pump (proposals) has a longer gap between consecutive sections than most other pages. GeoffreyT2000 (talk) 01:43, 7 November 2015 (UTC)

This is likely related to the thread "Underlined links" above. Just scroll up to see where the situation stands at this time. You might also want to cut and paste this to that thread to keep everything together. MarnetteD|Talk 01:52, 7 November 2015 (UTC)
It's because __TOC__ is placed in a table there. The table has other content but would render the same as now with only this:
{| 
|-
| __TOC__
|}
PrimeHunter (talk) 02:13, 7 November 2015 (UTC)
The same effect is seen at User talk:Bgwhite, which doesn't have an explicit __TOC__ but does have an unclosed table (and two unclosed <div> which are not the culprits). I've been stalking that talk page for over three years, and the TOC was normally-spaced until recently. I'm sure that it's a recent styling change in MediaWiki. --Redrose64 (talk) 22:21, 7 November 2015 (UTC)
That sounds right. The 3 October version at https://web.archive.org/web/*/https://en.wiki.x.io/wiki/Wikipedia:Village_pump_%28proposals%29 has normal TOC gaps while 7 November has large gaps. Our own 3 October revision [55] with current rendering has large gaps. PrimeHunter (talk) 23:31, 7 November 2015 (UTC)
The problem (as in the other thread) is the addition of display:table-cell to Mediawiki's styling of the table of contents. It's clearly a not-totally-thought-through change. Relentlessly (talk) 23:41, 7 November 2015 (UTC)
Is there a hack I can add to my css/js scripts to fix this? Optimist on the run (talk) 07:44, 8 November 2015 (UTC)
It also affects any page that uses {{TOC hidden}}, since that is a __TOC__ wrapped in a collapsible table. --Redrose64 (talk) 00:14, 9 November 2015 (UTC)
Unless I'm much mistaken, Redrose64, it affects any page with a table of contents, not merely those wrapped in a table. Relentlessly (talk) 16:26, 9 November 2015 (UTC)
No - it doesn't affect (for example): Wikipedia:Village pump (technical); User talk:Redrose64; Wikipedia:WikiProject Women writers or indeed User talk:Relentlessly. These all have normal spacing, not the broader spacing seen at Wikipedia:Village pump (proposals); User talk:Bgwhite; or Wikipedia:WikiProject Women. --Redrose64 (talk) 20:32, 9 November 2015 (UTC)
Oh, I see. So all TOCs have more spacing than previously, but those ones have even more. It's caused by the border-spacing property on the table, for what it's worth: elements with display:table cell will get that border spacing. The use of that display property feels like an ill-thought through kludge to me. Relentlessly (talk) 22:11, 9 November 2015 (UTC)

New maps service for Wikipedia

Dear community, I would like to get your feedback on the future of Wikipedia Maps. As you might have heard, our maps server is now operational, and is used on Wikivoyage and the Android app. We do not yet have enough maps servers to enable it for all Wikipedia articles, in part because we need community to expressly say this is a high priority, in part because community needs to establish what must happen for maps to be adapted, and lastly because we should have a <map> wiki markup to support direct maps usage. Be forewarned that most of our time was spent developing the tiling server, whereas the map itself (style) can use more work - we even have this proposal from a volunteer to help us with it (please support). So, to move ahead, we need:

  • Blockers: Decide what changes must happen for maps to be used anywhere in Wikipedia.
Hopefully this list is short, otherwise it will be tough to allocate substantial resource without a clear community buy-in.
  • Labs: Decide on the migration of the map popups at the top of the articles to the new underlying tiles.
  • Articles: Decide which capabilities should the <map> tag support.
  • Prioritization: Plan which of these features should we work on next. What use cases will be the most valuable from the start?

Please send your general maps feedback, comments, and questions here. Thanks! --Yurik (WMF) (talk) 08:15, 7 November 2015 (UTC)

Come one people, show your enthusiasm about maps !!!! —TheDJ (talkcontribs) 23:08, 9 November 2015 (UTC)

Article patrolled twice

Hi, I created the article Roan Allen a while back. It was originally patrolled by Wgolf. A few minutes ago, I got a notice that it was patrolled by Eeekster. I have no idea how this happened. I had always thought, one new article, one patrol. How did it show up on new pages feed or wherever? Are all my articles going to have to be patrolled again (which seems like a waste of time)? Or was it the random page patrol? White Arabian mare (Neigh) 23:55, 8 November 2015 (UTC)

I'm not exactly sure what you mean by "patrolled", because User Wgolf does not show up in the edit history. Did he click that notice that says "mark this as patrolled"? It looks to me like User Eeekster tagged it for bare URLS. New articles appear on all kind of of links. On the left-hand side of the article click on "What links here", and you'll see what I mean. People look at new articles for all kinds of reasons, and they edit an article, tag it in this case, wherever they feel the need arises. Hope this helps. — Maile (talk) 00:07, 9 November 2015 (UTC)
He clicked the "mark this page as patrolled" tab, I guess. I got a notification that said "User:Wgolf patrolled Roan Allen". Anyway, somebody else removed the tag and filled in the bare URLs with Refill. White Arabian mare (Neigh) 00:22, 9 November 2015 (UTC)
@White Arabian mare: There's only one entry in the patrol log. @Maile66: to find that, go to the page history, at the top click View logs for this page, in that select "Patrol log" in the first dropdown and click Go. --Redrose64 (talk) 09:09, 9 November 2015 (UTC)

Looks like it was some kind of computer or system error then. I know I got two patrolled notices. Thanks, White Arabian mare (Neigh) 15:46, 9 November 2015 (UTC)

Helping page creators get title capitalization correct

I have never ceased to be amazed at the number of biographical articles created about people with names like "First middle last", even though, in the lead sentence, they magically become "First Middle Last". :-) I suppose that's because editors start by searching for the names, entering them in lower case as I suppose people do when searching, to see whether an article already exists. When told that an article doesn't exist and offered the chance to create one, they go ahead without noticing that the article that they're about to create is cased incorrectly. I wonder if we can warn them prominently at the time when they're about to create the article. Maybe the page creation view can include a field that displays the name as currently entered, and gives them a chance to change it right there in place? Regarding either of those options, however, note the section I began immediately above this one, where I note that page creators don't see the instructions anyway, and asking if that can be fixed. —Largo Plazo (talk) 01:57, 9 November 2015 (UTC)

I'm unsure about the techno stuff, but I'll give you the way I create new pages; by writing their correct titles on my userpage and then clicking the redlinks! 😉 I think that what you are saying about the search fields does happen all the time though, or else some people may get in a hurry or get excited. It depends on the user. White Arabian mare (Neigh) 02:03, 9 November 2015 (UTC)
In any case it is trivial for a new page patroller or other experienced editor to move the page to the correct title, so I don't see this as a major issue. DES (talk) 03:02, 9 November 2015 (UTC)
It is a problem for users who create an article very soon after registering, as even though they can create pages, they can't move them until they are autoconfirmed. In my opinion, the best fix for this would be a new MediaWiki extension that automatically checks articles as they are being created to root out problems like a lack of references, promotional wording, and yes, problems with the title. This would take some time to develop, even if the the resources were allocated for it, which they aren't at the moment as far as I'm aware. But as a stopgap solution we could add title detection to MediaWiki:Newarticletext and output a big warning message for titles of more than one word where all the words apart from the first start with lowercase letters. (The problem being that with this approach it is very hard to work out which titles are names of people and which aren't.) — Mr. Stradivarius ♪ talk ♪ 07:37, 9 November 2015 (UTC)
Is this really such a big problem ? I mean we just move the article at some point and done.. As opposed to potentially scaring away people with yet more big warning messages.. —TheDJ (talkcontribs) 07:41, 9 November 2015 (UTC)
Well, yes and no. The biggest problem that I've noticed is that instead of waiting for someone else to come along and move the article, new users will often start a new article with the same content at the correct title, resulting in a cut and paste move. On the other hand, there are bigger problems on WP, and having a big bold warning might well be overkill. — Mr. Stradivarius ♪ talk ♪ 10:03, 9 November 2015 (UTC)
Trivial but also just a nuisance for something that could possibly be avoided. As for scaring new users off with warnings, it might be interesting to learn how many people have been disturbed to realize that the title of their new article is written incorrectly and worried that it would be irreversible. Then we could ask which is the better way to jar users. —Largo Plazo (talk) 10:46, 9 November 2015 (UTC)

As helpdesk and OTRS agents can attest, there are quite a few messages from new editors who've managed to figure out some of the basics of editing and assume, incorrectly but understandably, that they should be able to edit the title as well. Many spend some time trying to figure it out and then reach out to the help desk or an email to OTRS in frustration. It would be nice if there were a better solution. However, the suggestion to "one them prominently" raises a red flag. We have lots of prominent warnings and I'm not excited about creating yet another one.

I have another thought although I can identify a problem with it; maybe somebody can resolve. If we change the suggestion to:

Click "First Middle Last" or "First middle last" to create an article with that name

they will almost always choose the first one.

The obvious problem is that if we offer them "Guinea Pig" and "Guinea pig", they are likely to choose the first, which would be wrong. If we knew a simple rule for identifying names of humans, we might be able to come up with a workable role for suggesting alternatives.--S Philbrick(Talk) 14:46, 9 November 2015 (UTC)

We can't restrict the set of choices because then they wouldn't have any way to create Guy de Maupassant, Honey in the Rock, De Polignac's conjecture, etc. —Largo Plazo (talk) 14:58, 9 November 2015 (UTC)
The suggestion was only where all the searched words apart from the first start with lowercase letters. PrimeHunter (talk) 16:06, 9 November 2015 (UTC)

Contents of big Categories

The webpage https://en.wiki.x.io/wiki/Category:Human_name_disambiguation_pages?from=E shows Aa, Ab, Ac, Ad, Ae etc in its Contents. Should not it show Ea, Eb, Ec, Ed, Ee etc?SoSivr (talk) 13:03, 9 November 2015 (UTC)

The large TOC is not a MediaWiki feature but made with {{Large category TOC}}. That limits what is possible but I have made a test version at User:PrimeHunter/sandbox5 with anchors. I added it to Category:Human name disambiguation pages below the normal version. In my Firefox, the anchor link scrolls the long TOC line so the right end is placed at the anchor. I therefore placed each anchor letter at the end of the sublist for that letter. It seems unavoidable that browsers will scroll vertically to place the whole sublist at top of the page. A large TOC could be placed higher on the category page if we want more of the leading category text to be visible after this scrolling. If we want the short TOC line to remain visible then it must be moved or copied below the scrolling TOC line. If we copied it then we could keep the original functionality (going to top of the page with no scrolling in either direction) in one version and the new in the other. This test version certainly has some disadvantages but I don't know how to do better with the available features. PrimeHunter (talk) 15:48, 9 November 2015 (UTC)
Note the alternative {{Large category TOC 2}} which avoids horizontal scrolling by allowing line wrapping. The long TOC line is kept shorter for this reason, with five subentries for each letter instead of 26. Five may be OK on Category:English footballers with 17000 pages. 26 would probably be better for Category:WikiProject California articles with 52000 pages. PrimeHunter (talk) 16:00, 9 November 2015 (UTC)
There is a good reason for not changing the behaviour of templates like {{LargeCategoryTOC}} - regardless of how far through the category you have got, you always have a quick link to an earlier stage, so it would be wrong to suppress the Aa-Dz links if you were currently viewing E. --Redrose64 (talk) 20:24, 9 November 2015 (UTC)
@PrimeHunter:The test version ("Contents of big Categories") does not seem to work on Android Firefox. But I can go to e.g. "Ev" manually by typing https://en.wiki.x.io/wiki/Category:Human_name_disambiguation_pages?from=Ev (to overcome this problem which is more intense for bigger categories) @Redrose64: ("Contents of big Categories") If one views the letter "E" one need not suppress the Aa-Dz links but the latter could somehow go at the end: Ea, Eb,Ec ...,Fa,Fb, ...,Za,Zb, ..., Zz, Aa, Ab, ..., Da,Db, ..., Dz.SoSivr (talk) 14:18, 10 November 2015 (UTC)
The "E" in the second TOC goes to https://en.wiki.x.io/wiki/Category:Human_name_disambiguation_pages?from=E#E. What do you see there? I see a scrolling horizontal box at top of the page where the visible part starts out from Db to Ez in my browser. I can scroll it from "*" at the left end to "Zz" at the right. PrimeHunter (talk) 14:29, 10 November 2015 (UTC)
@PrimeHunter:Sorry, before you even wrote it I found out myself that IT WORKS! Earlier I had looked at the address https://en.wiki.x.io/wiki/Category:Human_name_disambiguation_pages?from=E .SoSivr (talk) 14:42, 10 November 2015 (UTC)

17:19, 9 November 2015 (UTC)

Community Wishlist Survey

Hi everyone!

The Community Tech team at the Wikimedia Foundation is focused on building improved curation and moderation tools for experienced Wikimedia contributors. We're now starting a Community Wishlist Survey to find the most useful projects that we can work on.

For phase 1 of the survey, we're inviting all active contributors to submit brief proposals, explaining the project that you'd like us to work on, and why it's important. Phase 1 will last for 2 weeks. In phase 2, we'll ask you to vote on the proposals. Afterwards, we'll analyze the top 10 proposals and create a prioritized wishlist.

While most of this process will be conducted in English, we're inviting people from any Wikimedia wiki to submit proposals. We'll also invite volunteer translators to help translate proposals into English.

Your proposal should include: the problem that you want to solve, who would benefit, and a proposed solution, if you have one. You can submit your proposal on the Community Wishlist Survey page, using the entry field and the big blue button. We will be accepting proposals for 2 weeks, ending on November 23.

We're looking forward to hearing your ideas!

MediaWiki message delivery (talk) 21:30, 9 November 2015 (UTC)

watchlist no longer exists

I finally found this area and hereby report that my watchlist that I counted upon for years no longer exists. I tried to fix it and only destroyed what remained I then had. Maury (talk) 06:08, 10 November 2015 (UTC)

Is Special:EditWatchlist/raw blank? If you still see something there then make an offline copy before trying to change anything. PrimeHunter (talk) 12:53, 10 November 2015 (UTC)

Spam filter triggering without entering anything

Today on IRC UY Scuti and I tried to help Littlemartin66 with the strangest problem. It seems like whenever Littlemartin66 tries to edit a page the spam blacklist filter warning is triggered without them doing anything else. This can be seen in this screenshot [63]. In addition to not being able to edit articles, the same warning comes up when they try to edit their own talk page. This can be seen here: [64]. Since Littlemartin66 is not autoconfirmed I am making this post on their behalf since the page is semi-protected.

They have tried to edit in both Safari and in Chrome. They are using a Macbook Pro with OS 10.7.5 and the Ad Block Plus extension. They are not blocked and they haven't triggered any edit filters (at least not any public ones). They are not using the visual editor. Any insight anyone can provide would be greatly appreciated it. --Stabila711 (talk) 09:50, 10 November 2015 (UTC)

A user at Wikipedia:Teahouse/Questions/Archive 405#Blacklist/blocked-link error in error? said at Wikipedia:Teahouse/Questions/Archive 407#Follow up to "Blacklist/blocked-link error in error?" question yesterday that their problem went away when they removed all Chrome extensions. PrimeHunter (talk) 13:06, 10 November 2015 (UTC)
Pinging Littlemartin66
Thanks PrimeHunter. We never thought the extensions would have caused that. Good to know for the future. If they come back and still have the problem I will post again. --Stabila711 (talk) 13:48, 10 November 2015 (UTC)

Wikimania 2016 scholarships ambassadors needed

Hello! Wikimania 2016 scholarships will soon be open; by the end of the week we'll form the committee and we need your help, see Scholarship committee for details.

If you want to carefully review nearly a thousand applications in January, you might be a perfect committee member. Otherwise, you can volunteer as "ambassador": you will observe all the committee activities, ensure that people from your language or project manage to apply for a scholarship, translate scholarship applications written in your language to English and so on. Ambassadors are allowed to ask for a scholarship, unlike committee members.

Wikimania 2016 scholarships subteam 10:47, 10 November 2015 (UTC)

Cleaning up gadgets

I propose we delist any gadget older than 2-3 years with fewer than 5 users... They can still be loaded by users from userscripts, but they won't clutter up the Gadget list. —TheDJ (talkcontribs) 22:20, 3 November 2015 (UTC)

The only way to clean up that list is to purge the database of user preference flags still pointing to old, deleted gadgets. -- [[User:Edokter]] {{talk}} 23:26, 3 November 2015 (UTC)
I was just going to ask why we had a gadget called "-contribsrange" in addition to the usual "contribsrange", along with several other similar cases. This looks like it needs to be made a Phabricator task. — Mr. Stradivarius ♪ talk ♪ 23:32, 3 November 2015 (UTC)
Looking at phab:T117120, I see that this has already been fixed. I'm not sure whether the patch made it out in time for Thursday's update, though. — Mr. Stradivarius ♪ talk ♪ 23:40, 3 November 2015 (UTC)
It didn't. Legoktm (talk) 23:46, 3 November 2015 (UTC)
Is this patch going to remove the de-listed items from the database of user preference flags? Or only hide them in the list? WhatamIdoing (talk) 02:07, 4 November 2015 (UTC)
It just hides them in the list. — Mr. Stradivarius ♪ talk ♪ 02:27, 4 November 2015 (UTC)
Yuck polution.. :D Well let's see what the list looks like after the next train.. —TheDJ (talkcontribs) 08:18, 4 November 2015 (UTC)
Counts should be accurate now. Legoktm (talk) 08:19, 9 November 2015 (UTC)
There are still some in the list with fewer than five users. Whatamidoing (WMF) (talk) 18:12, 11 November 2015 (UTC)

A list

Hi! I am hungarian, I speak english only a little bit. I woluld like a list of en:Category:County seats of the United States with all articles name. AWB is not supported for me... Thank you! --B.Zsolt (talk) 21:32, 5 November 2015 (UTC)

You can ask some AWB user here. Or you can try catscan for yourself. --Edgars2007 (talk/contribs) 22:39, 5 November 2015 (UTC)
Answered over at User_talk:B.Zsolt#List. You can actually do this without permission to use AWB. The Quixotic Potato (talk) 01:25, 11 November 2015 (UTC)

Error page

Some of us have been trying to help a user and could use some help ourselves. Details are found on the user's talk page. In a nutshell, when in PreferencesUser profileInternationalisation, the user clicks on "More language settings", and instead of the language pop-up the user gets an error page. User appears to browse with IE-8 in Windows NT. Pop-up Blocker seems to have no effect. Anybody know what's happenin'?  Paine  04:53, 10 November 2015 (UTC)

ahum, windows NT ????? Anyway, it seems that once more, a bunch of ES5 keywords have snuck into the main modules, killing the IE8 JS parser. I think we are VERY close to dumping JS on IE 8 altogether... This simply happens too often and the install base is tiny. I'll see if i'm able to find the issues, or recommend dumping JS functionality for the platform. @Edokter:, since he often dealt with such issues. —TheDJ (talkcontribs) 07:26, 10 November 2015 (UTC)
Not much I can do if the code is part of core modules. If IE8 is the only ES5-incompatible browser, then by all means, disable javascript support for IE8. -- [[User:Edokter]] {{talk}} 11:53, 10 November 2015 (UTC)
Thank you both, again, for your help, and I hope you won't mind putting this into simpler terms for the newuser and me:
  1. If js support is disabled for IE8, then will the newuser still see the "More language settings" link? and, if so,
  2. Where will that link take them when clicked?
  3. Also, when would this take effect?
 Paine  23:10, 10 November 2015 (UTC)

Strange problem with {{reflist}}

I was trying to change Michael Laucke. The references look found, but when I add |30em it gives many errors. What is going on? – Unbuttered parsnip (talk) mytime= Wed 06:07, wikitime= 22:07, 10 November 2015 (UTC)

When you remove that param, all the refs are bulleted instead of numbered. Not good. --Redrose64 (talk) 22:14, 10 November 2015 (UTC)
This edit fixed the bullet problem, and will have fixed the other one too; but I would advise against re-adding |30em since all the refs are long and constraining them to narrow columns would be detrimental. --Redrose64 (talk) 22:32, 10 November 2015 (UTC)

Creating an extra parameter by using the 'extra' field

Hey. Could someone help me with this, please? It is regarding infobox parameters. Thanks in advance. Rehman 23:25, 10 November 2015 (UTC)

can't get into my account

I got off WP a couple of hours ago without logging out. I use the remember me feature, and normally the system recognizes me. When I got onto Wikipedia a couple of minutes ago, I wasn't logged in. I am White Arabian mare. I can't get into my account, although I've done nothing wrong and haven't been blocked. I've never even been warned. I just can't get into my account. I'm using the correct password and I keep getting an incorrect password notice. Is anyone else having this? I'm pretty upset.2600:1005:B112:DFA1:38A4:C51B:4664:42E3 (talk) 02:17, 11 November 2015 (UTC)

You might have been logged out because you had been logged in for more than 30 days - the cookie that stores your login info expires after that. Alternatively, if you (or someone else) logged into your account from another IP address, that would also log you out. Or it may be that you deleted the cookie on your computer somehow. If you changed your browser settings before you tried to edit Wikipedia again, that might be it. As for not being able to log back in, did you try the obvious things? Check that you don't have caps lock on, and make doubly sure that you type the password in correctly. Otherwise, if you didn't specify an email address there's not a lot else you can do, I'm afraid. — Mr. Stradivarius ♪ talk ♪ 06:40, 11 November 2015 (UTC)
Also, your account isn't blocked or locked, so it seems to be just the password that is preventing you from logging in. — Mr. Stradivarius ♪ talk ♪ 06:43, 11 November 2015 (UTC)
I'm pretty sure that logging into your account from another IP address will not of itself log you out from other sessions where you were logged in, but I do know that logging out from any IP address will log you out on all machines and devices where you are currently logged in, whether they are associated with that IP or not. --Redrose64 (talk) 10:41, 11 November 2015 (UTC)
Correct. Logging into an account via one IP address does not log you out of other sessions (many of us I am sure are logged in on more than one device at the same time, I have three active sessions via different IPs right now). Your account is not locked, or blocked and the IP address is not blocked (locally or globally). Try Special:PasswordReset to get a new password, if you have email set. QuiteUnusual (talk) 11:19, 11 November 2015 (UTC)
User:White Arabian mare does not have email set and doesn't even have email according to the userpage. Some browsers may store an incorrectly entered password and replace what the user writes later. Try another wiki like wiktionary:. Passwords never expire. If you don't have a working password then you have to create a new account. Comparison of webmail providers has many free services that only require access to a browser (and remembering a password!) when you want to use mail. PrimeHunter (talk) 12:20, 11 November 2015 (UTC)

We should probably move this thread back to the talk page - the IP claiming to be White Arabian mare won't be able to edit this page right now as it's semi-protected. — Mr. Stradivarius ♪ talk ♪ 16:58, 11 November 2015 (UTC)

Watchlist

Items in the watchlist, such as this recent edit to User talk:GoingBatty, have an invisible Unicode character between the link and the semicolon. This character is encoded as "%E2%80%8E" in URLs. GeoffreyT2000 (talk) 03:29, 11 November 2015 (UTC)

Do you consider this a problem or do you have some question? It would be great if you could elaborate. Which link? Plus I don't see a semicolon? --Malyacko (talk) 11:16, 11 November 2015 (UTC)
It's when pagenames are seen on watchlists. It's a U200E left-to-right mark (lrm). Some characters like Arabic can cause right-to-left writing until a letter in a left-to-right writing system is encountered, or something like that. Pagenames on watchlists are followed by semicolon, space, timestamp. None of these will return to left-to-right, so I guess lrm is always placed just in case. Compare these lines:
العَرَبِية‎; 12:00
العَرَبِية; 12:00
They are identical apart from a lrm before the semicolon in the first. Many browsers will change direction in the second and display 12:00, space, semicolon, Arabic text. PrimeHunter (talk) 11:53, 11 November 2015 (UTC)

Stop hiding the instructions when a user creates a page

How many users ever see the advice and instructions given at the top of the page when they go to create a page? In the browser, as soon as the page displays, it scrolls up so that the content entry area is at the top. Should this behavior be eliminated? —Largo Plazo (talk) 01:49, 9 November 2015 (UTC)

It doesn't scroll for me. What is your browser, how do you create articles, and what is the url at the time? When I create articles by clicking a red link like Title of article, the top says "Creating Title of article" in big bold, and the url is https://en.wiki.x.io/w/index.php?title=Example_page&action=edit&redlink=1 with no anchor and no scrolling in my Firefox. PrimeHunter (talk) 02:03, 9 November 2015 (UTC)
Suppose I search for "Blah de Blah". The search results tell me "You may create the page "Blah de Blah", but consider checking the search results below to see whether the topic is already covered." When I click the red link, I'm at https://en.wiki.x.io/w/index.php?title=Blah_de_Blah&action=edit&redlink=1. The editing toolbar is at the top of the viewport. I get the same result in both Chrome and Firefox on Windows 7, and I got the same result when I clicked your Example_page link. I'm using the Vector skin. —Largo Plazo (talk) 02:10, 9 November 2015 (UTC)
Have you enabled wikEd at Special:Preferences#mw-prefsection-gadgets and is wikEd active with a yellow pencil icon in the top right corner? PrimeHunter (talk) 02:21, 9 November 2015 (UTC)
Yes and yes, except that the pencil's black and white, not yellow. —Largo Plazo (talk) 02:33, 9 November 2015 (UTC)
OK. wikEd scrolls for me as you describe but only when it's active with a yellow pencil. Try to also disable it at Gadgets. Do you scroll if you log out and click Draft:Blah de Blah? Unregistered users cannot create mainspace pages so Blah de Blah cannot be tested. How about when you are logged in and click https://simple.wiki.x.io/w/index.php?title=Blah_de_Blah&action=edit. I guess you haven't changed account settings there. PrimeHunter (talk) 03:54, 9 November 2015 (UTC)
If you're seeing a black-and-white pencil, then you might be in the regular wikitext editor. That pencil icon was added so that you can switch to the visual editor if you want to (e.g., to add and delete table columns, which is much easier there). Whatamidoing (WMF) (talk) 17:36, 11 November 2015 (UTC)
I'm not sure which icon Largo Plazo was seeing. There are a confusing number of pencil icons at the top right of something. I was referring to the wikEd icon   or   at the top right of the whole window (to the right of "Log out") for users with wikEd enabled at Special:Preferences#mw-prefsection-gadgets. I have VisualEditor disabled at Special:Preferences#mw-prefsection-editing but now see that when it's enabled and the source editor is used in a namespace where VisualEditor works, there is a black-and-white pencil icon above the top right of the edit text area. "Advanced" in the toolbar File:RefToolbar 2.0a.png makes a search-and-replace icon with a yellow pencil and a magnifying glass right below it. And the mobile interface has a black-and-white pencil icon to edit at the top right of the page text and section headings. PrimeHunter (talk) 23:32, 11 November 2015 (UTC)

Using Cyrillic with circumflexes

Hello. To represent Максимовичівка — historical Ukrainian language orthography by Mykhaylo Maksymovych I need Cyrillic characters combined with Circumflex. Unfortunately I found following issue: all this characters displaying with shifted circumflex (see below; reproduced at least by FF and IE on Windows). There is trick that used by Russian Wikipedia: placing that characters into span tag with style text-transform:uppercase or text-transform:lowercase removing shift. Using template to insert these symbols is inconvenient but also I found yet another problem: this trick reduces shift for dotted circle but keep it for Cyrillic letter. (P.S.: It looks there is no shift for dotted circle in enwiki but it there is in ukrwiki. It's very strange. I'm confused).

So, question: Is there any way using Cyrillic with circumflexes in Wikipedia?

Tests / Examples

  • Latin symbols with embedded (not combined) circumflexes, often using as Максимовичівка replacement because bugs: â , ê , û , î , ô , ŷ .
  • Circumflex just, circumflex with placement, circumflex with Cyrillic letters: ̂  ; ◌̂  ; а̂ , е̂ , и̂ , о̂ , у̂ .
  • Circumflex with placement, upper and lower Cyrillic letter: ◌̂  ; А̂  ; а̂ .
  • Trick: circumflex combined with placement, upper and lower Cyrillic letter in <span style="text-transform:{upper,lower}case"></span> tags:
    • ◌̂ ; ◌̂
    • А̂ ; А̂
    • а̂ ; а̂

Artem.komisarenko (talk) 21:39, 10 November 2015 (UTC)

You did not need an edit protected template for this discussion. No other comments. --Izno (talk) 21:48, 10 November 2015 (UTC)
Your examples look good in Linux (FF and Chromium). We had the same problem with the accent marks on Cyrillic letters in Windows a few years ago, but when I checked recently the problem in Windows remains only when an accented letter is in a heading (the problem was that the accent mark was shown over the next letter). So support for these combining characters is improving. --V111P (talk) 22:34, 12 November 2015 (UTC)

template problem

Hi, I'm trying to amend Template:Johor_Darul_Takzim_F.C._squad, but when I click on the little E to edit, it tries to create it, any ideas ? Thanks GrahamHardy (talk) 14:38, 11 November 2015 (UTC)

Fixed. The name= parameter has to be exactly the same as the actual template name otherwise it won't work. Jenks24 (talk) 14:42, 11 November 2015 (UTC)
But when I click on the template in for example Muhammad Al-Hafiz I get the same problem, is that because I need to wait a while ? Thanks GrahamHardy (talk) 14:59, 11 November 2015 (UTC)
I think so. Is it fine now? It's working for me form Al-Hafiz's article. Jenks24 (talk) 15:32, 11 November 2015 (UTC)
Yep, looks good now, thanks for your help... GrahamHardy (talk) 18:41, 11 November 2015 (UTC)
Wikipedia:Database reports/Invalid Navbar links by Bamyers99 lists errors like this for many templates but currently not those using {{Football squad}}. I have requested it at User talk:Bamyers99/Archive 2#Football squad templates with invalid Navbar links. All the new errors are usually fixed quickly by Redrose64 when the report is updated each month. Let's see if we can make more work for December. PrimeHunter (talk) 00:36, 12 November 2015 (UTC)
Yay! --Redrose64 (talk) 11:31, 12 November 2015 (UTC)

GoogleAnalyticsObject

I found this in a page:

<script> (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','//www.google-analytics.com/analytics.js','ga'); ga('create', 'UA-61654406-1', 'auto'); ga('send', 'pageview'); </script>

What is it, does anyone know? Justlettersandnumbers (talk) 23:10, 11 November 2015 (UTC)

Looks like something squirted in by a browser extension. --Redrose64 (talk) 23:20, 11 November 2015 (UTC)
The code is for Google Analytics. Websites can insert it in the <head>...</head> tag of their pages to get traffic data for their site from Google. I'm not aware of the Wikimedia Foundation using it anywhere. Editors don't have access to add the code inside <head>...</head>. You failed to say where you saw it but I see you removed a version from the wikitext of an article.[65] It has no effect there. It was added in [66] by the article creator Pfc2016 with no explanation. Maybe they hoped to get traffic data for the article. A search also found copies at User:Jefrymaster, User:Thanhcaraven, User:Yogeshindian/sandbox. It was added by those users and seems harmless but maybe we should remove it so visitors don't think they are being tracked by Google. PrimeHunter (talk) 00:01, 12 November 2015 (UTC)
Thank you both. It was the "it has no effect" that I wanted to be sure of. Justlettersandnumbers (talk) 08:46, 12 November 2015 (UTC)

Firefox and anchors

Is the reoccurring issue with Firefox not hitting the exact anchor (header) spot when clicking on a wikilink ever going to get fixed? --NeilN talk to me 04:40, 12 November 2015 (UTC)

Annoying indeed, but confirming the url bar a second time gets you there, as you probably noticed. A timewaster for sure. Samsara 04:57, 12 November 2015 (UTC)
It's a difficult problem to fix, but it's in the next sprint of the reading team (it was bumped from the previous). There are deeper (architectural) problems tracked in phab:T53736 and firefox specific issues in phab:T67468. —TheDJ (talkcontribs) 08:49, 12 November 2015 (UTC)

Quicksurveys

Hello Folks, as mentioned earlier in the reading team updates here and here. Quick surveys is a new tool that helps with gathering data and understanding users. Extension documentation is available on mediawiki, and more information about the research is found on the research documentation page. The first survey will be deployed tomorrow, it will run on 0.005%, meaning it will load 5 times out of each 100K page requests. As usual results and stats will be shared. This will run until 5000 survey results are collected. This is the first time the tool is used, let's keep an eye on the experiment and the results; who doesn't love data :-) --Melamrawy (WMF) (talk) 19:31, 11 November 2015 (UTC)

Hi everyone. This is Leila from Research. One clarifying point to what Melamrawy (WMF) said above (Thanks for that, btw!). We are not 100% sure that the survey will go out tomorrow at this time. If everything goes smoothly today in terms of the deployment to testwiki, it will, otherwise, it will be postponed to a later date. Nevertheless, it was important to communicate that the survey is coming up, whether it's tomorrow or in few days. --LZia (WMF) (talk) 19:50, 11 November 2015 (UTC)

Drafts still showing up in Google searches

Back in July there was a consensus to remove drafts from Google searches. Wikipedia:Village pump (proposals)/Archive 126#Userpage drafts shown in search engines. It's not clear if this was intended to cover userspace only, or draft space as well. Regardless, material in both areas (draft space and user space) is presently being displayed on Google searches unless it's specifically marked as noindexed. Can this be fixed? Thanks, --Diannaa 🍁 (talk) 20:24, 11 November 2015 (UTC)

It seems no one merged the patch yet. I prodded the folks subscribed to the ticket. P.S. I still think this is an incredibly stupid thing to do.. More evidence of our visionless processes, rules and regulations akin to slowly boiling the Wikipedian frog, while the real world passes us by. Just wanted to have said that for posterity. —TheDJ (talkcontribs) 20:43, 11 November 2015 (UTC)
Thank you, I hope someone gets it done. -- Diannaa 🍁 (talk) 02:51, 13 November 2015 (UTC)

Section headings missing on Finding Dory

On Finding Dory the section headings are missing. The "edit" tabs are there. See:
http://tinypic.com/r/1zb8lqp/9

I tried purge and refresh. Using latest Google Chrome, and Internet Explorer. Ping me back. Cheers! {{u|Checkingfax}} {Talk} 14:00, 12 November 2015 (UTC)

Do you have this problem on other pages? Whatamidoing (WMF) (talk) 23:46, 13 November 2015 (UTC)
Whatamidoing (WMF), I did not have any problems on any other pages but Finding Dory is now back to normal. Cheers! {{u|Checkingfax}} {Talk} 02:32, 14 November 2015 (UTC)

Blank spaces left when second-level headings follow floating boxes

TOC problem

When Javascript is fully loaded at Cliff Robertson or Dina Merrill (I suppose, there are more pages), there is blank space between TOC and first section (till the end of infobox), as there would be {{clear}} after the lead. Using Windows XP, (one of) latest Firefox both logged-in (Monobook), and logged-out. --Edgars2007 (talk/contribs) 16:46, 12 November 2015 (UTC)

I checked, and I am only seeing this on Firefox, and not on IE on Windows. Frietjes (talk) 16:47, 12 November 2015 (UTC)
I'm seeing it in both Chrome and IE, for example Bill Medley. Oddly enough it appears to only affect certain pages on IE. Hzh (talk) 16:51, 12 November 2015 (UTC)
ETA: it seems to be affecting all pages on IE now. Hzh (talk) —Preceding undated comment added 16:57, 12 November 2015 (UTC)
It looks like the floating of headers near boxes and similar elements doesn't work. The error also occurs for subheaders near images further down in articles - it's not only the TOC and infobox at top, that are affected. GermanJoe (talk) 17:06, 12 November 2015 (UTC)
I am seeing this as well, in Safari. This change must have been made recently, some time in the last hour or so. Maury Markowitz (talk) 17:06, 12 November 2015 (UTC)
Seems to be a new issue and affecting a wide range of pages. It has just done it to me (Firefox) after a trivial edit to Henschel Hs 293 . I happened to have both 'before' and 'after' versions still open on different tabs. Even going back to view old versions, I now see the large gap. Andy Dingley (talk) 17:07, 12 November 2015 (UTC)
It's this CSS rule:
.infobox,
.last-modified-bar,
h2 {
  clear: both;
}
that is, all second-level headings are now preceded by an automatic {{clear}}. It should be apparent not just with infoboxes, but also where any right-floated box (such as an image) is taller than the section in which it is placed, such as the image stacks at Picturesque Palestine, Sinai, and Egypt and Nepal Sambat or the portal boxes at Storm Abigail#See also. It also makes {{TOC right}} look ridiculous, see for example List of closed railway stations in Britain: S. --Redrose64 (talk) 17:21, 12 November 2015 (UTC)
Hey Andy! That article is particularly good example. Maury Markowitz (talk) 17:12, 12 November 2015 (UTC)

(ec)*It's not just infobox/TOC related, although that's going to be the problem in most articles. It also shows up in articles with vertical image galleries to the right of the text. Until today, the text "flowed" along next to the images; now all the images display above the next section break. See Edmond Hamilton (images listed immediately after infobox); Algis Budrys (images listed within article section). The Big Bad Wolfowitz (aka Hullaballoo) (talk) 17:16, 12 November 2015 (UTC)

Got it filed a bug and am pinging deployers. —TheDJ (talkcontribs) 17:16, 12 November 2015 (UTC)
I've just seen this TOC problem. Current changes have made the TOC wrong. I then looked from 6 months ago. It looked an old version (of infobox), but it changes while I was looking. – Unbuttered parsnip (talk) mytime= Fri 01:18, wikitime= 17:18, 12 November 2015 (UTC)
I know it's bad form to say "me too" but I just went round in circles for 10 minutes on Oxford Street wondering what on earth I'd done to screw up the layout - turns out it was this. As Hullabaloo says, it affects any right aligned image that previously overhung into the following section - on Oxford Street, the image of the 100 Club now puts a big gap in the middle of the article. Ritchie333 (talk) (cont) 17:29, 12 November 2015 (UTC)
until this is fixed, this edit to my personal common.css appears to override the problem. Frietjes (talk) 17:31, 12 November 2015 (UTC)
Should we make the same edit to MediaWiki:Common.css until the external issue is fixed? PrimeHunter (talk) 17:54, 12 November 2015 (UTC)
Definitely. epic genius (talk) 20:29, 12 November 2015 (UTC)
I have made the edit as a temporary fix until phab:T118475 is fixed.[67] PrimeHunter (talk) 20:51, 12 November 2015 (UTC)
PrimeHunter, your edit was reverted, probably because this turns off the default 'clear:right' in infoboxes, which is only noticeable when the infobox isn't the first floating item on the page. it seems as though there is no perfect solution outside of fixing the bug upstream. Frietjes (talk) 21:31, 12 November 2015 (UTC)

Template alignment and V T E nav

Hi, I'm having trouble fixing this template which breaks the layout of pages likes this as the content does not flow on the left alongside the template. How can this be fixed? Also, how can the V*T*E nav at the bottom of most templates be added to this one? (This template should probably be a sidebar, but I'm unsure if that can handle a table as content.) Thanks.--Cpt.a.haddock (talk) (please ping when replying) 17:20, 12 November 2015 (UTC)

Your problem is related to the section above this one. —TheDJ (talkcontribs) 17:24, 12 November 2015 (UTC)
Yes, the flow issue is the same as above. Just ignore it for now. See {{Navbar}} for ways to make V T E links. PrimeHunter (talk) 17:29, 12 November 2015 (UTC) PrimeHunter (talk) 17:29, 12 November 2015 (UTC)
@PrimeHunter: @TheDJ: Thank you both. Frietjes has sorted out the table issue by using an {{aligned table}} in a sidebar and the VTE issue appears to have also been resolved as a result. Cheers.--Cpt.a.haddock (talk) (please ping when replying) 18:06, 12 November 2015 (UTC)

Infoboxes

Has something recently been changed in the way that infoboxes are displayed? I see a lot of ugly white space, as apparently the first section after the lead only starts below the infobox (if present). Far as I remember, that was not the case up till recently. --Randykitty (talk) 17:33, 12 November 2015 (UTC)

@Randykitty: See parent thread. --Redrose64 (talk) 17:35, 12 November 2015 (UTC)
Oops, didn't see that, thanks! --Randykitty (talk) 18:39, 12 November 2015 (UTC)

Article displays with excess white space, but only when I'm logged in to Wikipedia

I noticed today that an article I recently edited was showing up with a huge amount of excess white space - but only when I was logged in. When I logged back out it went back to looking normal. Maybe there's a simple reason for this that is eluding me, but if there is something I can to make it go away (like change a setting) I'd like to know. The article in question is Susan Slade which I edited a couple days ago and it looked normal at that time. I am using the same browser and computer I used to do the edits (Chrome 46.0.2490.86 m and Win 7). Today when I looked at the article while logged into Wikipedia, there was a large amount of white space now showing between the bottom of the Contents box and the start of the next section (Plot), which didn't start until below the infobox: film . There weren't any other edits since my last one that might have introduced the space to the page. I tried clearing out my cache, restarting the browser, and logging out and back into Wiki. I did notice that when I was logged out and viewed the page, it returned to normal (white space gone). But when I logged back in, the white space immediately returned. Can someone explain why this is happening and if there's something I can do to make the article look normal to me when I am logged in? So far I have only noticed this on the one article. Thank you, TheBlinkster (talk) 17:40, 12 November 2015 (UTC)

As above —TheDJ (talkcontribs) 17:41, 12 November 2015 (UTC)
Thank you, I noticed that after I posted and tried to take out the "new section" but hit an edit conflict with your reply. I will refer to the discussion above. Just adding my case to the pile. TheBlinkster (talk) 17:43, 12 November 2015 (UTC)
Also as above --RaphaelQS (talk) 19:04, 12 November 2015 (UTC)
Also experiencing the same issues, I actually reported it below and then realized it was here!  . –Davey2010Talk 19:25, 12 November 2015 (UTC)
Initially my experience was the same, but now the white space is there whether I'm logged in or not. --Andreas Philopater (talk) 20:49, 12 November 2015 (UTC)

It's happening after images, too

See here ... I just put that one in just now. It happens regardless of whether I'm logged in, in both Firefox and Chrome. Daniel Case (talk) 20:18, 12 November 2015 (UTC)

Me too. I just edited Westchester Square – East Tremont Avenue (IRT Pelham Line) and there is a huge white space below the infobox, because it's pretty long. epic genius (talk) 20:25, 12 November 2015 (UTC)
Yes; but it's not a space below the infobox or after images: as I noted above, it's effectively a {{clear}} before all second-level headings regardless of what they are preceded by. It's most noticeable with infoboxes and image stacks, because they tend to be taller than a section. --Redrose64 (talk) 21:55, 12 November 2015 (UTC)

Common.css

I need to see some cases (screenshots) of the above problems. The solution that was implemented in Common.css introduced other problems as infoboxes and H2 headings were no longer allowed to clear. Really people... TEST this kind of code first. -- [[User:Edokter]] {{talk}} 21:16, 12 November 2015 (UTC)

Based on some Random article clicks to Chi Darki, Cheverny, Socket (video game) and so on, millions of articles now have a large gap before the first section heading. Do you not have the gap in your browser? I didn't expect my temporary common.css fix [68] to be perfect but if it fixes millions of articles then problems in a far smaller number of other articles may be acceptable until the underlying problem outside editor control is fixed. PrimeHunter (talk) 21:58, 12 November 2015 (UTC)

Hurray, the bug has been fixed! Rfassbind – talk 21:34, 12 November 2015 (UTC)

I'm still seeing it on e.g. Henschel Hs 293. — Mr. Stradivarius ♪ talk ♪ 21:46, 12 November 2015 (UTC)
I am still having the same problem as before - I checked on both Susan Slade and Dina Merrill. Tried clearing my history, logging out, etc. Pages display fine when I am logged out but same white space issue I described above when I log in. TheBlinkster (talk) 21:50, 12 November 2015 (UTC)
Yes, it's still there. I verified that it's not there when I log out. Also, eliminating the TOC with "notoc" doesn't do anything. BMK (talk) 21:54, 12 November 2015 (UTC)
It wouldn't, because it's not a TOC issue. The problem is with second-level headings. --Redrose64 (talk) 21:56, 12 November 2015 (UTC)
User:Edokter, User:Epicgenius, User:Redrose64, User:Mr. Stradivarius, ... this hack is working for me, but almost certainly depends on the order in which these scripts are executed. someone could probably clean this up. the real solution is to not add this to the style sheet in the first place. Frietjes (talk) 22:00, 12 November 2015 (UTC)
It doesn't appear to be fixed at this moment in time in IE 11, even when I do an Alt-F5. It appears to work in Chrome 47. Both tested on Windows 7 64-bit. • SbmeirowTalk22:01, 12 November 2015 (UTC)
@Frietjes: The MediaWiki:Common.css fix seems to work for me, but thanks anyway. epic genius (talk) 01:57, 13 November 2015 (UTC)

The bug has returned, unfortunately.Rfassbind – talk 22:03, 12 November 2015 (UTC)

I've put in another solution; one that doesn't force the clear to none. -- [[User:Edokter]] {{talk}} 22:05, 12 November 2015 (UTC)
That seems to have solved it. For now at least. --Andreas Philopater (talk) 22:08, 12 November 2015 (UTC)
Seems good now. BMK (talk) 22:09, 12 November 2015 (UTC)
Looks good now, thanks! The Big Bad Wolfowitz (aka Hullaballoo) (talk) 22:46, 12 November 2015 (UTC)

Articles that have back-to-back infoboxes in the lead e.g. 2015 UCLA Bruins football team are now stacked left-to-right instead top-to-bottom.—Bagumba (talk) 23:20, 12 November 2015 (UTC)

yes, the hack work-around in MediaWiki:Common.css effectively adds 'clear:none' to the infoboxes. I was working on a javascript fix, but I was certain that the backend javascript would be fixed sooner. Frietjes (talk) 01:07, 13 November 2015 (UTC)

Although most of the pages look good now (thanks!), I am seeing a chunk of white space midway down the page on You Light Up My Life (song)#Debby Boone's cover where there are two columns of table to the left of an infobox to the right. Reducing the width of the table (like setting width=75%) did not help so I left it alone. Not sure if this relates to the same issue or a different one but just thought I'd mention it since I came across it. Thanks, TheBlinkster (talk) 19:35, 13 November 2015 (UTC)

Seems to be fixed now, thanks! TheBlinkster (talk) 22:40, 13 November 2015 (UTC)

Year articles

In year articles such as 2015, the "in other calendars" box appears to the left of rather than below the "by topic" box. GeoffreyT2000 (talk) 23:06, 12 November 2015 (UTC)

Looks to be the same problem I reported above at #Common.css.—Bagumba (talk) 23:27, 12 November 2015 (UTC)

Template errors on articles dedicated to Presidents of the United States

I notice that "series" templates are suddenly crowding the lead sections of multiple articles such as Franklin D. Roosevelt, Barack Obama, and George W. Bush. If anyone can find a solution, please speak. Dustin (talk) 01:00, 13 November 2015 (UTC)

This is the same issue as above. The only currently known fix we can make at the English Wikipedia is to revert a temporary fix to an external problem, but the temporary fix improves far more articles than it damages. The external problem will hopefully soon be fixed so we can avoid all damage at the same time. PrimeHunter (talk) 01:15, 13 November 2015 (UTC)
Strike that. phab:T118475 says the external problem has been fixed and already deployed. I have reverted our temporary fix in MediaWiki:Common.css and everything seems to work now! Franklin D. Roosevelt doesn't have boxes stacked horizontally instead of vertically, and Chi Darki doesn't have a big space before the first section heading. PrimeHunter (talk) 01:27, 13 November 2015 (UTC)
Still seeing horizontally stacked boxes in Chrome, but not IE. Hzh (talk) 01:41, 13 November 2015 (UTC)
It works for me in Chrome. Try Wikipedia:Bypass your cache#Google Chrome to reload everything including css. PrimeHunter (talk) 01:46, 13 November 2015 (UTC)
Reloaded according to instruction, but still the same with the horizontal stacking in Chrome, and same after rebooting. No problem in IE or Firefox. Everything else I've checked appears to be normal, except the horizontal stocking issues on some pages. Hzh (talk) 03:59, 13 November 2015 (UTC)
I tried Chrome and had the same problem you did, even after trying the "bypass cache" instructions (using Mac). However, if I go to History->Show Full History->Clear browsing data and cleared everything (it might be "Cached images and files" that is really all that is needed though), the problem went away.—Bagumba (talk) 04:46, 13 November 2015 (UTC)
Excellent, works now. Thanks.Hzh (talk) 11:13, 13 November 2015 (UTC)

4,895,000 articles

If anyone is seeing an outdated article count for the English Wikipedia on www.wiki.x.io, please see m:Wikimedia Forum#www.wiki.x.io portal, m:Wikipedia.org Portal Improvements, and finally phab:T118544. – Minh Nguyễn 💬 03:12, 13 November 2015 (UTC)

Malformatted signature after Template:uw-wrongsummary

So this is kind of a petty bug but it's driving me crazy. After placing a subst'd template warning on a user's talk page, I'm in the habit of adding my signature after the template by typing a space and four tildes, like so: {{subst:uw-wrongsummary|Article}} ~~~~

But with Template:Uw-wrongsummary, this renders one's signature on a separate line below the warning, in monospace font, inside the indented box one would normally use to display code samples, etc. Like so:

  Constructive contributions to Wikipedia are appreciated, but a recent edit of yours to the page Foo has an edit summary that appears to be inaccurate or inappropriate. Please use edit summaries that accurately tell other editors what you did, and feel free to use the sandbox for any tests you may want to do. Thank you. AtticusX (talk) 04:46, 13 November 2015 (UTC)

See what I mean? This is the only user warning template I know of that's behaving this way, but I'd be out of my comfort zone trying to fix it. A workaround that was recommended to me was to put my signature right up against the template, without a space between: ...}}~~~~ ...But old habits die hard, because for years I've been putting a space before my four tildes and it's never caused me any problems till now. AtticusX (talk) 04:46, 13 November 2015 (UTC)

try it again. The end of the template code had the opening noinclude on its own line instead of right after the end of the code. -- George Orwell III (talk) 05:02, 13 November 2015 (UTC)
You are a gentleman and a scholar, and I thank you from the bottom of my OCD. AtticusX (talk) 05:07, 13 November 2015 (UTC)
You're welcome. Please note that back on Nov. 4, 2015, the change that I just made was, in effect, undone so there is a chance my change will be reverted as well at some point -- George Orwell III (talk) 05:16, 13 November 2015 (UTC).
About the undoing, I've started a thread at Wikipedia talk:Template messages/User talk namespace#Uw-wrongsummary template loop. — Mr. Stradivarius ♪ talk ♪ 05:39, 13 November 2015 (UTC)

Template sfn

Can an editor who know what they're doing review the use of {{sfn}} on Paris? It seems to be inducing error messages (e.g. see current reference 249 in the reference section at the bottom) NE Ent 02:07, 14 November 2015 (UTC)

All the parameters for each instance used for the same short footnote need to match. In this case some had p=300-301 while others had pp=300-301. I changed them all to the latter to reflect a range of pages (pp=) instead of a single page (p=). -- George Orwell III (talk) 02:49, 14 November 2015 (UTC)

Help in fixing a "cite error"

I found a big red cite error message for reference #1 here American Water Spaniel#References. My skills in fixing these is limited and I couldn't track down the problem so any help from those of you who know what to do will be appreciated. MarnetteD|Talk 23:56, 4 November 2015 (UTC)

The error message correctly said: '"spiotta30" defined multiple times with different content'. One of the definitions was for page 38 so I changed it to "spiotta38". PrimeHunter (talk) 00:33, 5 November 2015 (UTC)
Many thanks PrimeHunter. MarnetteD|Talk 00:58, 5 November 2015 (UTC)

The change in sensitivity of templates including {{harvnb}} to Cite error: Invalid <ref> tag; name "..." defined multiple times with different content (see the help page) is worse than reported in the recently archived discussion on this page. In the revision of 03:58, 6 September 2015 of Mandore (instrument), errors of this type are generated for the silliest of reasons. {{harvnb}} is used with parameter sets identical except that one set uses |p= and the other uses |page=, generating the errors for "mcdonald08p11" and "mcdonald08p14". {{harvnb}} is used with parameter sets identical except for trailing blanks in the |url= and |title= parameters plus a different |accessdate=, generating the error for "vam.ac.uk", but in a test edit, making the accessdates the same did not get rid of the error; one has to also remove the trailing blanks, which should be ignored! —Anomalocaris (talk) 09:07, 10 November 2015 (UTC)

Just landed here after trying to find out more about these errors. I consult the Cleanup Listing for Wikiproject: Musicans to check what needs fixing every week and was very surprised to find over 500 new entries today. Normally this stands around a 100. On closer inspection, I find that the vast majority of these are classed as 'reference errors' - the ones described above. My question is, do I ignore this as this is a technical error or do I tackle them? Karst (talk) 10:17, 10 November 2015 (UTC)
@Anomalocaris: the "sensitivity of templates" has not changed in any way. Unlike {{sfn}}, the {{harvnb}} template does not generate ref names, those are manually created. It is the MediaWiki parser that has changed, and it looks at what is enclosed by the <ref>...</ref> tags before any templates are expanded. This particular article has <ref name=mcdonald08p11>{{harvnb|McDonald|2008|page=11}}</ref> and <ref name=mcdonald08p11>{{harvnb|McDonald|2008|p=11}}</ref> and in these cases, one uses |page= the other |p= (although both expand to <ref name=mcdonald08p11>[[#CITEREFMcDonald2008|McDonald 2008]], p. 11</ref>). Similarly for <ref name=mcdonald08p14>{{harvnb|McDonald|2008|page=14}}</ref> and <ref name=mcdonald08p14>{{harvnb|McDonald|2008|p=14}}</ref>
@Karst: In a case like this, where the only difference is whether |page= or |p= is used, harmonise them. It doesn't matter which you use, just make sure that they are the same. --Redrose64 (talk) 16:42, 10 November 2015 (UTC)
@Redrose64:Thanks for the tip! Came across it today here. Karst (talk) 15:47, 12 November 2015 (UTC)
@Karst: Wrong way around - |p= is for single pages; this is a page range, so |pp= is the param to use. --Redrose64 (talk) 16:02, 12 November 2015 (UTC)
@Redrose64: I stand by my comment 100%. Because of changes to the MediaWiki parser, the {{harvnb}} template has become more sensitive to the Cite Error Bug. It is a bug. It says at Template:Harvnb, "The parameters |page= |pages= and |Ref= exist as aliases for |p= |pp= and |ref= respectively." Here, Wikipedia is testifying to all editors that it will treat |page= and |p= identically.
I believe that it is confusing to other editors and sloppy to define the same ref name more than once, even with identical definitions. (For that matter, it is also confusing to other editors and sloppy to have ref names defined that are not used elsewhere in the article.) However, since Wikipedia does allow two or more identical definitions of the same ref name, Wikipedia is obligated to be consistent about this, and avoid the silliness of treating {{harvnb}} parameters |page= and |p= differently. To preserve Wikipedia's integrity with users and editors and stop barraging them with these nonsensical error messages, there is a simple workaround. As template {{harvnb}} is processed, a new first step should be to convert all aliased parameters to their unaliased names. —Anomalocaris (talk) 18:55, 10 November 2015 (UTC)
The internal structure of {{harvnb}} has not changed. The way that it is processed and expanded is also no different. {{harvnb|McDonald|2008|p=11}} expands to [[#CITEREFMcDonald2008|McDonald 2008]], p. 11, and did exactly the same prior to 8 October. --Redrose64 (talk) 19:34, 10 November 2015 (UTC)
Thanks for the clarification, now for a user experience, perhaps. I have started to go through the Cleanup listing I mentioned above, and I am finding that all the errors are in relation to 'ref name=' errors. One page for Alabama (band) had a large number of book references (with page numbers) but none of them gave me an error. Instead, someone had numbered various Billboard articles and forgot to add the next number. So there were two different articles, both with the tag 'ref name=billboard6'. Most of the pages have a variety of other issues too. Now, it would be very useful if a bot like Refill could address these issues, but my guess it that they will all need to be fixed manually. Karst (talk) 22:12, 10 November 2015 (UTC)
@Redrose64: When one component of a system is "improved", this often necessitates other components of a system being improved to compensate for side-effects of the improvement of the first component. That is the case here. The way that {{harvnb}} is processed and expanded should be improved to address side-effects of other changes, to preserve Wikipedia's integrity with users and editors and stop barraging them with these nonsensical error messages. —Anomalocaris (talk) 23:00, 10 November 2015 (UTC)
One instance of {{harvnb}} cannot know what any other instance of {{harvnb}} has been given as its parameters. There is nothing that can be done to that template that might remove these messages. --Redrose64 (talk) 23:04, 10 November 2015 (UTC)
Right, and there is nothing the template could do even it it knew about the other call. This will for example also produce the error message because one of the definitions of "T" has a space after "cite book": <ref name="T">{{cite book|title=Example}}</ref><ref name="T">{{cite book |title=Example}}</ref>. There must be identical strings inside the named ref before it's evaluated. But the idea of a named reference is to only define it once and invoke it the other times with <ref name="T"/>. mw:Extension:Cite#Multiple uses of the same footnote says "You can either copy the whole footnote, or you can use a terminated empty ref tag". This text goes back to 2006.[69] MediaWiki hasn't checked for it before and some documentation pages at the English Wikipedia have encouraged bad practices or not warned about it. PrimeHunter (talk) 00:40, 11 November 2015 (UTC)
@Redrose64: You said before, "It is the MediaWiki parser that has changed, and it looks at what is enclosed by the <ref>...</ref> tags before any templates are expanded." I apologize, I replied without fully processing this. If this is true, then you are right, this is completely beyond the reach of {{harvnb}}. The MediaWiki processor should at least ignore leading blanks after <ref name=...> and trailing blanks before </ref>. This won't help with blanks anywhere inside the {{harvnb}} template, or with {{harvnb}} invoked with identical entries except that one uses an alias for a parameter name. But it would be a step in the right direction. And now I have a new related bug to report! —Anomalocaris (talk) 00:55, 11 November 2015 (UTC)

This Cite Error Bug does not occur in the User space. See User:Anomalocaris/sandbox, which is identical to revision of 03:58, 6 September 2015 of Mandore (instrument), but lacks the error messages of that version. The meta-bug that the Cite Error Bug does not occur in the user space should be fixed so that the basic bug can be better analyzed by users in their sandboxes. —Anomalocaris (talk) 00:55, 11 November 2015 (UTC)

The English Wikipedia has deliberately disabled the error message in many namespaces. MediaWiki displays MediaWiki:Cite error references duplicate key when the error is encountered. Like many other cite error messages it calls {{Broken ref}} which uses namespace detection to only render error messages in these namespaces: main, category, template, file, help, portal, wikipedia, draft. The messages are still produced but with a class brokenref which has display: none in MediaWiki:Common.css. This means you can display it in all namespaces with this in your CSS:

.brokenref {display: inline !important;}

PrimeHunter (talk) 01:22, 11 November 2015 (UTC)
H:SHOWCITEERROR has a variation of the above code. PrimeHunter (talk) 01:24, 11 November 2015 (UTC)

I do not understand the cite error bug at the current version of Saumarez Parish, New Brunswick. <ref name="cp2011"> is defined once and used once. It is linked a third time in the output of template {{Canada census}}, but it's above my paygrade to figure out how that is even possible, let alone how it generates the cite error. —Anomalocaris (talk) 21:26, 13 November 2015 (UTC)

Just as you said: that cite name is also transcluded within Template:Canada census in a ref with different content. Changing the ref name in the article will fix the problem (try "cp2011-1" or something). Ivanvector 🍁 (talk) 21:42, 13 November 2015 (UTC)
User:Ivanvector: Thank you, that fixed it. I believe this would be a difficult error for most Wikipedia editors to diagnose. It is also a burdensome to prevent. Template:Canada census does not have any <ref>s of its own. How is anyone expected to track down all the <ref name>s all the way down the template call chain? In Fortran, we never had to worry about this, because each subroutine's variables were local to itself, and variables that were intended to be shared had to be explicitly passed in subroutine calls or stored in COMMON. Wikipedia is a wonderful and amazing system, but recent changes to MediaWiki has created a bunch of Heffalump traps. —Anomalocaris (talk)
Anomalocaris: someone else will have to help diagnose if this is a bug or not, I'm not all that up to speed on our template language. It looks to me like the template pulls references from a subpage, Template:Canada census/reference. The way I traced it was by looking at the reflist in the article in the revision you posted, which indicates three inclusions (a,b,c), and clicking on c highlighted the reference within the template, and so I deduced that the template must be using the same ref name. It's easier to fix the article than the template, unless this is a widespread problem. However I think this is coincidence, and not a result of any recent software changes. Ivanvector 🍁 (talk) 22:41, 13 November 2015 (UTC)
Based on a quick experiment trying to fix with a regex, it is a widespread problem, and sometimes articles use a named reference provided by the template to reference facts in the article body, and sometimes they provide a (clashing) separate definition in the article. For example in the article Port Hope, Ontario, cp2011 is redefined, but cp2006 is used from the template. Ugh! William Avery (talk) 22:56, 13 November 2015 (UTC)
It is indeed a widespread problem. Wikipedia search for cp2011 has 571 results, of which the first 100 are all articles about places in Canada generating the error message
Cite error: Invalid <ref> tag; name "cp2011" defined multiple times with different content (see the help page).
This is not a problem that should be addressed one at a time. There are 571 articles just with cp2011, and undoubtedly thousands of others with the same problem and different variables. It was a design error to have templates automatically transclude <ref> names into articles. It was an even worse design error to have this happen without such <ref> names explicitly named in the template documentation, warning users that those names were reserved and not to be used in any article using the template. This mess was here all along; it is only with the modifications to the MediaWiki software that the mess is obvious. But now that we know about it, it needs to be cleaned up in a systematic way. —Anomalocaris (talk) 04:59, 15 November 2015 (UTC)

Treatment of edited redirects at New Pages feed

As mentioned here: alterations to redirect pages place an entry in NewPagesFeed, whether or not the redirect itself is removed. A particular problem at present, with dozens of RfD discussions relating to redirects created by Neelix, which are piling up at the older-date end of the new pages queue.

How it should work for edits to redirect pages:

  • When the redirect is not removed, the page shouldn't appear in New pages feed at all
  • When the redirect is removed - treat as equivalent to a new article being created:
    • It should appear in new pages feed sorted by the date the redirect was removed, not the date the page was originally created
    • the entry should show the username of the editor ("B") who removed the redirect, not the one ("A") who first created the page
    • when marked as patrolled, the "patrolled" notification & any talk page message should go to B not A.

It doesn't seem good enough not to attempt to fix this major bug. Could someone with the access please file a report as appropriate?: Noyster (talk), 12:44, 13 November 2015 (UTC)

A redirect is a redirecting page. This is done by placing #REDIRECT [[Target]] as the first text on the page. Our process for nominating redirects for discussion places something else above the redirect code to deliberately break the redirect and display a text box instead (example). If #REDIRECT [[Target]] occurs later on the page then it's just normal wikitext where # makes a numbered list with one entry:
  1. REDIRECT Target
Suppose MediaWiki didn't list this at NewPages. Mischievous users could avoid NewPages scrutiny of new articles by turning a redirect into an article with #REDIRECT [[Target]] somewhere on it, or good faith users might just forget to remove the existing redirect code. Should we accept such NewPages bypassing? Is there a good solution against it which would work across all languages and wikis? Or should we request a MediaWiki feature which identifies a non-redirecting page with very constrained content used in redirect discussions at the English Wikipedia? A general system like a category or magic word saying "This non-redirect should be treated like a redirect" could be abused on articles. Maybe there could be a page in the MediaWiki namespace (which can only be edited by administrators) to specify allowed content of non-redirects that shouldn't appear on NewPages. The only allowed content could for example be a specific template with parameters. PrimeHunter (talk) 13:59, 13 November 2015 (UTC)
I see the point about RfDs, but please note that if it is considered necessary to keep RfDs in the new page patrolling queue, then the rest of my plea only becomes more of an issue - the date and creator's name assigned to the queue entry need to be corrected for the RfDs as well as for the genuine new articles that replace redirects: Noyster (talk), 10:16, 14 November 2015 (UTC)

Changes to HTML Tidy

User:Edokter, User:TheDJ, User:PrimeHunter, User:Redrose64, ... was there a change to HTML tidy or something? it used to be that if I wrote

  • one
  • two
  • three

or

  • one
  • two
  • three

the result was the same. i.e., that the empty dots would be stripped, and the output would be the same as without the empty items. but this is no longer the case? this is causing problems for many templates which previously relied on the this feature (e.g., Template:A-Z multipage list). Frietjes (talk) 19:59, 13 November 2015 (UTC)

It is a recent MediaWiki Tidy change. See the 2 November Tech News, which links to this mailing list message, in turn linking to phab:T49673 and gerrit:246148. SiBr4 (talk) 20:14, 13 November 2015 (UTC)
It was never a feature (but abused none the less). But please do not change hlist, because it is not the root of the problem. Hlist should behave like regular wiki/html lists, which also show empty list items now. Fixes should be implemented at template level, or using CSS associated with the template. -- [[User:Edokter]] {{talk}} 21:52, 13 November 2015 (UTC)
Personally I never used empty list items, and if I found them, I removed them. --Redrose64 (talk) 22:10, 13 November 2015 (UTC)
if that's the way it is, then I suppose we just have to use more parser functions. Frietjes (talk) 23:26, 13 November 2015 (UTC)
Wait a second... unless I'm mistaken, the end behavior is rendered exactly the same as before except for those instances where the default display: list-item; was being "changed" (as is the case with hLists).

Before the recent change, empty LI tags were stripped and therefore never made it to the final rendering.

Now, empty LI tags are given a class attribute set to mw-empty-li and that class name is now defined with a display: set to none - resulting in the same non-display of empty LI tags as before the change although they are no longer being stripped.

This 'same behavior in the final rendering through a new means' is evidenced by an input of...

* one
* two
* three
* <!--four-->
* five
* six

... resulting in ...

  • one
  • two
  • three
  • five
  • six

... note the nullification of four thanks to the new .css addition.

So if the "default" is still not to display empty LI tags, shouldn't hList (and any other similar alterations to the core display:list-item; definition) mirror the current default behavior? Applying hlist to the previous example...

<div class="hlist">
* one
* two
* three
* <!--four-->
* five
* six
</div>

... results in ...

  • one
  • two
  • three
  • five
  • six

... which indicates the disparity between the previous & current final rendering behavior of typical LI element usage versus the previous to current final rendering behavior under hList's usage. -- George Orwell III (talk) 01:56, 14 November 2015 (UTC)

George Orwell III, just checked and adding a rule to my common.css removes the extra dots. Edokter, is there a reason why we can't include something like this in MediaWiki:Common.css, perhaps restricted to hlist? Frietjes (talk) 16:04, 14 November 2015 (UTC)
That CSS is new to me; it was not there initially. That changes matters. I'll try and resolve the clashing definitions in an elegant manner. -- [[User:Edokter]] {{talk}} 19:14, 14 November 2015 (UTC)
Right; I missed it at first as well.

And to be clear Edokter, I'm in total agreement with you about revisiting Modules and Templates taking advantage of the previous stripping (now, the hiding) of empty LI tags regardless. If you re-read the discussion under the Phabricator Task, it seems the powers that be plan on removing that .css addition once the majority of "problematic" local Templates/Modules are identified and addressed accordingly as well.

Anyway, all things seem "as before" again at the moment -- Thanks for your attention on this. -- George Orwell III (talk) 23:45, 14 November 2015 (UTC)

Text from quote template not wrapping around audio clip

The text from a {{quote}} is not wrapping around the adjacent embedded audio clip here: Michael Laucke#Style and influences. I have checked for stray code but cannot find anything that would prevent the quoted text from wrapping all the way to the right hand margin around the audio file. Or, do quotes have a different right hand margin? Please advise and ping me back. Thank you. Thanks. Cheers! {{u|Checkingfax}} {Talk} 02:37, 14 November 2015 (UTC)

UPDATE: I RTFM at {{quote}} and it is a known "bug" and solution #1 did not apply, so I tried solution #2 which worked for a moment but then the text jumped back to the left. Here is what I inserted: |style=overflow:inherit;=
So, I still need help fixing this. Cheers! {{u|Checkingfax}} {Talk} 02:49, 14 November 2015 (UTC)

You inserted the overflow in the wrong place; it should be placed in the quote template. -- [[User:Edokter]] {{talk}} 12:56, 14 November 2015 (UTC)

E-mail unsubscription?

(previously posled at WP:Village pump (miscellaneous). I am getting already for a week or longer every day, sometimes several times per day, Wikipedia alerts (those on top of every page) that my e-mail has been unsubscribed due to multiple delivery failures. It says exactly "Your registered e-mail address <...> has been unsubscribed due to multiple message delivery failures. You can verify your e-mail address again". Verification works every time (basically, I perform it after every alert). My e-mail is corporate. I did not have complaints about it recently; in particular, I am subscribed to several Wikipedia mailing lists, and the delivery works just fine. Does anyone else get it, or is this a problem at my end? Thanks.--Ymblanter (talk) 09:19, 14 November 2015 (UTC)

Are you saying you see the message when viewing a page at enwiki? In one of the standard web browsers? Is it a message visible (to you) at the top of the page and looking like like a banner asking for donations? Or, is it something to do with the line at the very top of each page showing your user name, alerts, messages, talk, etc. Is this on a computer or a mobile device? Does the message identify the mailing list? Johnuniq (talk) 09:46, 14 November 2015 (UTC)
This is an alert of the notification system (the left of the two small squares between my user name link and the talk page link, the same one which also tells me that someone mentioned my name on wiki). I am using a laptop, Win-7 and FF last version browser. I do not see any reason to think that it is mailing list related whatsoever, in my understanding the message says my e-mail is not linked to my Wikimedia account anymore. For the record, my account was first registered on Meta in the pre-SUL time. Meta also has the notification system (I believe all WMF projects now have) but does not say anything strange.--Ymblanter (talk) 09:50, 14 November 2015 (UTC)
The message is MediaWiki:Notification-bouncehandler-flyout from mw:Extension:BounceHandler. This is the first time I hear about it. PrimeHunter (talk) 11:15, 14 November 2015 (UTC)
Indeed, this is the message I get.--Ymblanter (talk) 11:21, 14 November 2015 (UTC)
Just happened again 15 minutes ago.--Ymblanter (talk) 15:36, 14 November 2015 (UTC)
Ping Legoktm and 01tonythomas.... — This, that and the other (talk) 12:38, 14 November 2015 (UTC)

Hi, I'm a bit busy today, but I filed phab:T118648 to track this for now, and will look at the logs when I have some free time. Your email is bouncing something, which is why it's getting unsubscribed... Legoktm (talk) 18:36, 14 November 2015 (UTC)

Thanks. Normally, it does not bounce things, and I have never got this complaint in the past eight years, though I use the same e-mail from the beginning.--Ymblanter (talk) 18:43, 14 November 2015 (UTC)
I've just deployed some debugging... Can you reconfirm your email address for me please Ymblanter? And then when it gets unsubscribed again, we should have some actual useful information to find out why... Thanks! Reedy (talk) 20:41, 14 November 2015 (UTC)
Great, thanks. Now it says "your e-mail was confirmed on 14 November 2015 at 15:37", and I can not reconfirm it (other than by removing it by hand first). Or do you need smth else? If I get an unsubscription notice, I will post here.--Ymblanter (talk) 20:52, 14 November 2015 (UTC)
Ah. I didn't actually check. Let me send you some test emails to try and force an unsubscribe action from MW. Reedy (talk) 20:57, 14 November 2015 (UTC)
As of now, I got two of your e-mails.--Ymblanter (talk) 21:00, 14 November 2015 (UTC)

Infobox needs missing file?

Jalan_Lapangan_Terbang_Baru_Bintulu is in the category "Articles with missing files". I think its because of something in the infobox but I can't figure it out. Do you guys know what to do? Thanks in advance, The Quixotic Potato (talk) 11:58, 14 November 2015 (UTC)

I have added the undocumented marker_image = none. I will add it to Template:Infobox road/doc#Route information. PrimeHunter (talk) 12:17, 14 November 2015 (UTC)
Thank you! <3 The Quixotic Potato (talk) 13:09, 14 November 2015 (UTC)
Please don't. We periodically look through a category generated by {{infobox road}} looking for missing marker graphics and create them. When you suppress them that way, they never get created through that system, PrimeHunter, The Quixotic Potato. Imzadi 1979  08:56, 15 November 2015 (UTC)
@Imzadi1979: Every day I learn something new. Would you be so kind to add that information to the template documentation and look at this diff? The Quixotic Potato (talk) 09:36, 15 November 2015 (UTC)
@Imzadi1979: Deliberately generating red file links displaying 100px at top of infoboxes is a strange system, and having no documentation for it is certain to confuse editors. A better method would be to use mw:Help:Extension:ParserFunctions##ifexist to test whether the required image exists (use Media: and not File: to also detect Commons images). If it doesn't exist then don't add it but add a hidden category for the specific purpose instead of polluting Category:Articles with missing files. If editors want a red file link to upload a file with the right name then there could be a template parameter to force a file link, only intended for previewing. PrimeHunter (talk) 12:50, 15 November 2015 (UTC)

Scrub-out two edits with IP number? & broader comment

I mistakenly made two edits after Chrome failed on my mobile and I was logged out without my knowledge. I'd like the two edits scrubbed completely out of the history -- the edit here and the edit immediately prior -- because the IP number is easily associated by context with my user name.

I'm not sure I've actually posted before while not logged on but I do know the mobile-edit process is more 'fluky', in ways like the above circumstance, than is the desktop one. Specifically, the warning that one is logged out before one posts an edit is almost invisibly small. I'm using the 'Desktop' view to look at and hence edit pages because otherwise sections are all closed up, amongst other drawbacks. I don't even know if there's an Edit key tho that's a slightly different issue I know. If there's a better place to engage on the general mobile-edit situation, suggestions are welcome.

Thanks. Swliv (talk) 16:47, 14 November 2015 (UTC)

Done, though next time email User:Oversight - with regards to the mobile interface you can try leaving comments on mw:MobileFrontend/Feedback. -- Mentifisto 17:33, 14 November 2015 (UTC)
Neat as a pin, the old folks'd say. THANKS!! Swliv (talk) 18:12, 14 November 2015 (UTC)

Disable Page Curation toolbar

Is it possible for a user to deactivate the Page Curation toolbar? This thread from 2012 recommends "clicking on the X", but I don't see an X on the toolbar as it appears today. Tevildo (talk) 00:32, 15 November 2015 (UTC)

I think they meant to minimize the toolbar by ticking the top icon/button (looks like →|) first and then click on the X to completely close it (To re-open the Curation toolbar once its "closed", you should have a link somewhere in your sidebar's Tools menu letting you do that). -- George Orwell III (talk) 00:56, 15 November 2015 (UTC)
That's fixed it, thanks. Tevildo (talk) 00:59, 15 November 2015 (UTC)

Copy and paste table using VE

A few days ago I was able to add a table by copying from an Excel file and pasting in an edit box (Visual Editor)

Example

Yesterday and today I have been unable to.

I don't believe this functionality is identified in the manual. That manual states you can drag a CSV file, but I was able to copy from the screen and paste into the edit box.

I made sure to reboot my computer in case I had something hung up. Any thoughts on what might've changed?--S Philbrick(Talk) 17:14, 31 October 2015 (UTC)

Hi S. It still works for me, using Google Docs in Safari 9 on Mac OS 10.10. (I don't have Microsoft Excel.) How similar is the content that you're trying to paste this time? Any odd formatting, charts, images, etc.? What's your browser or OS? Whatamidoing (WMF) (talk) 21:24, 4 November 2015 (UTC)

@Whatamidoing (WMF):My operating system is Windows 7; my browser is Chrome; no changes to either of those recently (other than the obvious occasional automatic updates)

I tried three examples in a sandbox which should provide some useful information.

My normal procedure is to drop in a header, a source with reference and a legend template first using edit source, then follow up with the visual editor edit, drop in the table, then follow up with an edit source to add in the class="wikitable"

I tried that (except for the last step) three different ways:

  1. I tried it in Mozilla copying from Excel, and that drops the data in but not in a well-formed way.
  2. Next I tried the same thing in chrome. In this case there is nothing below the legend because when I copy and paste it's like I didn't do anything.
  3. The third time, I used chrome but copied from Google docs. This time it worked as it should. I show the table in the first part of that section as it appears after simply dropping it in. I added it again and then added the class="wikitable" table code to make it look right.

I hope this will provide insight into what the problem is. The bad news is it seems to have something to do with Excel, which is a bit of a pain because you don't have it. For what it's worth when I said I copied from Google Docs, I first grabbed the data from Excel dropped in it to Google Docs and then copied it from Google Docs so it's very much the same data but it used to work when I did it from Excel.

On the one hand your advice might be to use Google Docs. If I have to, I have to, but I get the data from an external source (an NCAA database), drop it into Excel and do a fair amount of processing. I have separate routines for ordinary four year players, players with five years and one redshirt year, players with five years and one partial year, players with four years but different schools etc. I might be able to do all that Google Docs but I sure would be a lot happier if I could continue to use Excel.--S Philbrick(Talk) 22:21, 4 November 2015 (UTC)

The next question in the de-bugging process is: What happens if you try this with an older file (one that worked a couple of weeks ago)? The point they want to sort out is whether the file format (or something) has changed, e.g., due to a new version in Excel. Whatamidoing (WMF) (talk) 17:15, 11 November 2015 (UTC)

I think the short answer is that nothing has changed regarding the version of Excel. I can't quite do exactly what you suggest, so I will elaborate in the hopes that this will provide what you need.

  • Some time ago, I purchased a copy of Office 2013 for Windows
  • I am running Windows 7
  • The Excel version appears to be Excel 2013
  • I haven't personally, affirmatively, made any change to the version of Excel. I am vaguely aware that Microsoft updates their OS on occasion. I don't know whether any of those updates affected Excel. I don't recall seeing any such update, but I'm not in a position to definitively state that Microsoft hasn't tinkered with my software. I think it's unlikely.

My process is as follows:

  1. I open a spreadsheet which has a tab called "player stats"
  2. I then go to the NCAA site http://web1.ncaa.org/stats/StatsSrv/careersearch and search for a player
  3. Unfortunately, the resulting page doesn't generate a unique url. As an example, search for "Janning" which brings up three players, and select "Janning, Marissa"
  4. That should display her stats
  5. I then copy the stats, and paste into the Excel file
  6. The results are not formatted the way I want, so I have formulas and references within Excel to produce the desired output, plus generate a section heading and a citation.
  7. I edit the player article to create the section, add a template and a source using "Edit Source" example edit
  8. I then copy my formatted stats from the Excel page, and paste using a VE edit example edit
  9. That produces an ugly table, so I follow that with an "Edit source" to add {|class="wikitable" as well as a link (because I don't think I can add a table using VE which contains links) example edit
  10. The point (at long last) is that I don't save each players stats as a separate file, I keep reusing the same sheet over and over again. I had success for roughly 100 edits, then it stopped working.)

More to come, but I have to run to a meeting.--S Philbrick(Talk) 18:21, 11 November 2015 (UTC)

This sounds suspiciously similar to phab:T117859 (which uses MS Word rather than Excel). Does it sound similar to you, too? Whatamidoing (WMF) (talk) 22:54, 13 November 2015 (UTC)
My initial reaction is that it sounds similar, but I'll check closer in the morning. I have Word, so I can try to reproduce tables in Word.--S Philbrick(Talk) 03:32, 14 November 2015 (UTC)
My morning has slipped away with Neelix issues; and I have to be away for much of the day. I am very appreciative that you are looking into this, and I will get to it later this evening.--S Philbrick(Talk) 15:01, 14 November 2015 (UTC)
No rush. Please {{ping}} me or leave a note on my user talk page whenever you have had time to test it. I appreciate your help in sorting out the details here. Whatamidoing (WMF) (talk) 01:46, 16 November 2015 (UTC)

Wrong image template, how to fix?

I uploaded this using the wrong template (or more specifically, selected the wrong file). I tried to edit it after saving... yeah, right. What to do? Maury Markowitz (talk) 14:11, 10 November 2015 (UTC)

@Maury Markowitz: The file description page has three templates. Of those, {{Non-free reduce}} was added by another user after you created the page, the other two are {{Non-free use rationale 2}} and {{Non-free game screenshot}}. Which of these is wrong, and in what way is it wrong? --Redrose64 (talk) 17:18, 10 November 2015 (UTC)
The one I put on was screenshot, because I thought I was actually uploading this image. But I selected the wrong one of the two files in the Select file, and didn't notice. For this image it should have been the "official cover art" template.
This is an excellent example of why these templates are bad, IMHO. Once saved they turn into gogglygook, a fine example of a write-only language. DYK's template is similarly bad in this regard.Maury Markowitz (talk) 17:22, 10 November 2015 (UTC)
So all you should need to do is replace {{Non-free game screenshot}} with {{Non-free video game cover}}. --Redrose64 (talk) 17:57, 10 November 2015 (UTC)

That only changes the license tag, there are a number of tags within the description area that are different too. I'll have to copy and paste them from another page. Maury Markowitz (talk) 12:00, 16 November 2015 (UTC)

Looks like someone did it for me, thanks whoever that was! Maury Markowitz (talk) 12:10, 16 November 2015 (UTC)

Try this: type "help:wikitable" in the searchbox. I get two suggestions: Help:Wikitable (a redirect) and Help:Wikitable sortable (a redlink with no history). I could not find anything regarding this issue in the archives, which brings me to believe that the page may be a ghost. jonkerztalk 01:12, 12 November 2015 (UTC)

Wikitable sortable redirects to Help:Sorting so it looks like a bug for cross namespace redirects. Watched article redirects to Help:Watchlist, and Help:Watched article also shows up in search suggestions. PrimeHunter (talk) 01:25, 12 November 2015 (UTC)
I'm impressed you figured that out in like a minute. I've reported the bug on Phabricator. Thanks! jonkerztalk 02:04, 12 November 2015 (UTC)

Weird diff

I was just going back to Sir and Star to make an edit and saw this diff. I was not editing in that section when I made that save. I was only editing in the External links section. None of that junk renders or shows up in the edit window now. Was this a transient bug? Please ping me back. Thank you. Cheers! {{u|Checkingfax}} {Talk} 02:04, 14 November 2015 (UTC)

More weirdness: You won't be able to see all the clutter and blanking now as the diffs are showing my actual edits. Cheers! {{u|Checkingfax}} {Talk} 02:31, 14 November 2015 (UTC)
Was your intended purpose to add <font color="blue"> to an encyclopedia article? Whatamidoing (WMF) (talk) 01:55, 16 November 2015 (UTC)

Interesting category problem

I just created this category. I did so by adding it to the bottom of an existing article, RelayNet then clicking on the resulting redlink and creating the category page. I then added two other pages to the same category using HotCat.

I have used this exact mechanism countless times in the past. However, this time the original page was never added to the category. If I go to the RelayNet page I see the category and its link is now blue, but if you click through, no RelayNet. I tried it from another machine with the same result.

I have never see this before. Anyone know what's going on?

Maury Markowitz (talk) 16:23, 14 November 2015 (UTC)

It loos fine for me. Likely, you have a caching issue.--Ymblanter (talk) 16:37, 14 November 2015 (UTC)
It just updated for me also. Very odd. Maury Markowitz (talk) 17:08, 14 November 2015 (UTC)
If it happens again then try a null edit of the article. You can wikilink a category by placing a colon in front. When creating categories, please add at least one parent cateogry, e.g. Category:Bulletin board systems and Category:Computer networks. PrimeHunter (talk) 22:17, 14 November 2015 (UTC)
The same problem occurred when I recategorised Bocaue Pagoda Tragedy from Category:1993 in the Philippines to Category:1993 disasters in the Philippines on 4 November as it was a (non-natural!) disaster; after creating the latter category. But it still appears in Category:1993 in the Philippines although the article shows it to be in the blue-linked category Category:1993 disasters in the Philippines. Hugo999 (talk) 04:02, 16 November 2015 (UTC)
@Maury Markowitz, Ymblanter, PrimeHunter, and Hugo999: This is Wikipedia:Village pump (technical)/Archive 141#Category membership issues. --Redrose64 (talk) 09:58, 16 November 2015 (UTC)
Also Wikipedia_talk:HotCat#Question, I think. Ottawahitech (talk) 11:53, 16 November 2015 (UTC)

Timeline template modification

I would like to make a timeline for the History of Christianity in the First Century using the timeline template in the Eagles (band) article:

Eagles_(band)#Band_members

Everything is working out fine except that minus 10 to 100 is outside the allowed timeline range for this template. I am using 1890 as a stand in for minus 10 and 2000 as a stand in for 100. Is there a template I can use that allows the real year numbers?

HowardMorland (talk) 20:19, 15 November 2015 (UTC)

Help:EasyTimeline syntax#DateFormat says formats with day and month are not allowed before 1800, but year-only can be -9999 to 9999. If this doesn't help then please post the attempted code. PrimeHunter (talk) 22:34, 15 November 2015 (UTC)

Is there an editor interaction analyser that works?

This Scottywong tool isn't working on my iPad in Safari. --Anthonyhcole (talk · contribs · email) 00:54, 16 November 2015 (UTC)

Two boolean shorthand templates

I got tired of trying to remember or re-study the exact behavior of {{Yesno}}'s various fiddly parameters. So, I created two shorthand templates for the most common uses that override the default behavior:

  • {{yesno-yes}} – always returns "yes" (or the specified replacement result in |yes=) unless an explicit negative value is given; i.e., it evaluates to "yes" even when the value is empty or missing.
  • {{yesno-no}} – always returns "no" (or the specified replacement result in |no=) unless an explicit positive value is given; i.e., it evaluates to "no" even when the value is present, as long as it does not contain anything that resolves to "yes".

This should save any regular template editor a lot of time.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  06:22, 16 November 2015 (UTC)

@SMcCandlish: Thanks! I just put them to use this morning; I didn't realize they were so new. Hopefully I used them correctly! — This, that and the other (talk) 06:59, 16 November 2015 (UTC)
I think I made them a week or two ago; I forgot to announce until now, but hopefully they'll be useful for everyone. I was finding that it took over 5 minutes on average for me to mentally re-parse the underlying template's table of parameter interactions to be certain I was getting the intended result each time (often also having to set up test-case sandboxes), for either of these two uses cases (by far the two most common intents), so just providing wrappers for them seemed sensible. I figure it will save me alone at least a solid person-hour per month.  — SMcCandlish ¢ ≽ʌⱷ҅ʌ≼  07:19, 16 November 2015 (UTC)

A little template problem

I tried to fix Template:Welcome-unconstructive so that it doesn't say "Hello 192.168.1.1 and welcome to Wikipedia" when posted to anonymous users. But now it doesn't seem to be greeting anybody by name, IP or registered! What have I done wrong? — This, that and the other (talk) 11:13, 16 November 2015 (UTC)

It looks like {{IsIPAddress}} is not designed for substitution and should say {{{{{|safesubst:}}}#invoke:IPAddress|...}}. PrimeHunter (talk) 14:07, 16 November 2015 (UTC)
It is now substable. — Mr. Stradivarius ♪ talk ♪ 14:21, 16 November 2015 (UTC)
Ha! Silly me. Thanks to both of you. — This, that and the other (talk) 14:35, 16 November 2015 (UTC)

Citation needed

In mobile, when hitting on an inline citation and then citation needed, the text of the inline citation is shown after "citation needed", e.g. "citation needed '123RF Review', Photobuyer Guide, August 22, 2010" in 123RF. GeoffreyT2000 (talk) 15:28, 16 November 2015 (UTC)

To clarify the problem, if you go to http://en.m.wiki.x.io/wiki/123RF#Licensing and click [citation needed] as your first click on the page then you get a black box at the bottom saying "citation needed". If you click a reference number then the box displays the reference, for example this for [10]: "123RF Review", Photobuyer Guide, August 22, 2010. But if you have first clicked any reference and then click [citation needed] then the box displays both "citation needed" and the last reference you clicked. That certainly seems wrong but I think the bigger question here is: Why does a click on [citation needed] stay on the page and display a box repeating the text you clicked, instead of going to the link target Wikipedia:Citation needed. This apparently happens for all inline cleanup tags with a link in <sup>...</sup>. Try for example those at http://en.m.wiki.x.io/wiki/Wikipedia:Template_messages/Cleanup#Inline_with_article_text_6. Each click just displays the clicked text in a black box at the bottom instead of going to the linked page. Is this a broken feature that assumes any link in <sup>...</sup> is a reference? Or is it a strange intentional bandwidth saving feature to not take mobile users to another page when they click an inline cleanup tag? Only tested in Firefox on Windows Vista. PrimeHunter (talk) 17:23, 16 November 2015 (UTC)

Read more

Hello everyone, just heads up that as mentioned in the latest reading team updates, read more is a feature that will be tested soon on mobile and desktop beta. Based on Extension:RelatedArticles, relevant articles will be suggested to readers to support further engagement. Readers behaviors will be analyzed and shared during the beta testing. Thanks!--Melamrawy (WMF) (talk) 15:44, 16 November 2015 (UTC)

19:39, 16 November 2015 (UTC)

I keep wishing there were a way to "upvote" edits.

A way to add a very quick "thank you" note in the edit log, rather than having to add a note to a talk page somewhere. The big application is when someone corrects a mistake, or otherwise improves an edit I made, and I want to show that I've seen and approve of the correction. Basically, it would

  • Provide warm fuzzies to editors in smaller doses than barnstars. New editors, in particular, might find it very reassuring and motivating.
  • Demonstrate consensus agreement with an edit, as opposed to having to infer it from the absence of additional edits.
  • Show that someone is paying attention to this page.

If I and another editor are going back and forth over something and they create a version that we agree on, I'm happy when I find a lingering small typo just so I can add an edit log entry giving thanks for the previous edit. Everyone can then see our "edit battle", capped with a final edit by one of us which the other approves of.

Even bots could use the feedback for algorithm training. "Yes, that really was vandalism, good catch."

Another application would be in watchlists and contributions pages, an "edits since the last one you upvoted".

(Yes, I know this would be a huge change to MediaWiki.) 71.41.210.146 (talk) 04:17, 8 November 2015 (UTC)

Registered users have an easy way to thank each other for specific edits without making a new edit. See Wikipedia:Notifications/Thanks. An IP address can be shared by many users and one user can have many IP addresses so the system is limited to registered users. PrimeHunter (talk) 04:28, 8 November 2015 (UTC)
Nifty! I never knew about that. I was imagining a more public version, where third parties could see the thanks. Basically A makes an edit *and B concurs*. (You'd presumably only list two or three explicitly before collapsing to a number.)
I confess I have no idea why IP address confusion constitutes a reason to limit use of the Thanks system any more than it limits use of user talk pages. 71.41.210.146 (talk) 05:49, 8 November 2015 (UTC)
I wish there was a way to thank IP's! It seems, to me anyway, that IP's either make the best or the worst edits, and I'd like to encourage the good ones!!! --MurderByDeadcopy"bang!" 17:55, 11 November 2015 (UTC)
It's part of a larger feature Wikipedia:Notifications which is limited to registered users although some of the parts could also be done for IP's with some confusion. One reason for not doing it may be that thanks are less important than user talk discussions so there is less willingness to accept confusion over IP addresses. Also, the current Notifications system can deliberately only be seen by the user. If an unregistered user changes IP address then they would completely lose access to Notifications to the old IP address, while they can still view its user talk page if they know the IP address or can find it in a page history. PrimeHunter (talk) 06:08, 8 November 2015 (UTC)
@PrimeHunter: But is there a reason that IP addresses couldn't thank signed in users? That might be a really low-threshhold participatory activity for our engaged readers, that might encourage more people to go to history pages, which in turns shows how the process of content development works. It also could be the source of warm-fuzzies for our contributors, helping them feel more of their work is acknowledge. Participation would likely be really low at first, because of how hard it is to get to history pages. Sadads (talk) 13:57, 12 November 2015 (UTC)
Yes, but it's WP:BEANS. --Redrose64 (talk) 15:34, 12 November 2015 (UTC)
I have to confess, I have no idea how WP:BEANS is relevant. That refers to unintentional consequences of telling people not to do things, while the thanks system seems to be entirely about providing a way to do something. 71.41.210.146 (talk) 07:43, 16 November 2015 (UTC)
The Thanks are public and can be seen and searched by anyone: Special:Log/thanks. Finnusertop (talk | guestbook | contribs) 07:57, 16 November 2015 (UTC)
Beans is about not telling people how to do things when we don't want people to do those things without understanding the consequences. Consider: every thank adds an item to your notification list. --Redrose64 (talk) 11:48, 17 November 2015 (UTC)
"Thank" does not appear next to bot edits in the revision history either. GeoffreyT2000 (talk) 15:26, 17 November 2015 (UTC)

Large amount of HTTP 504 responses on API

I do a lot of API querying in the context of WP:STiki and WP:5000. In particular, some of the reports in the theme of the latter (like WP:TOPRED) have been unable to generate this week because the API returns HTTP 504 (timeout) for a query. I'll note that for the overall script to fail, a single query must fail 10 consecutive times, and I installed a slightly pause between these retries. The queries are well-formed (I plug the failing URL into a browser and get the expected output post-failure) and it is always a different query that times out. I've never encountered this before in years of report generation. I've been trying to run the script occasionally for over a day now, with no luck, encountering the same issue. Anyone know what is going on? Thanks, West.andrew.g (talk) 18:31, 16 November 2015 (UTC)

Can you dump the response headers ? They include headers which indicate the the cache server the request is hitting etc. That might be helpful in further analyzing. —TheDJ (talkcontribs) 19:03, 16 November 2015 (UTC)
And what queries are you running? Legoktm (talk) 00:29, 17 November 2015 (UTC)

Module:String and references

One of the parameters in {{Infobox NRHP}} is an 8-digit number that uniquely identifies every site on the National Register of Historic Places (NRHP). Using this number, one can generate a link to informative PDFs of text and images verifying that this site is on the NRHP. The |refnum= parameter of the infobox is supposed to be used for this number, but over time its usage has been abused to include things such as references, other text, etc. While ideally it would be best to just go through and move all information besides the 8-digit number to some other place in the infobox, but having many tens of thousands of transclusions, it would be impractical to do this. Instead I have used Module:String to replace the 8 digit number with the external link to the pdf mentioned above. This works in almost all cases, except for one that has been brought up at Template talk:Infobox NRHP#refnum cannot have ref without a number. This case is when the refnum parameter includes only a reference but no 8-digit number. Something in the way Wikipedia handles references makes this case a false positive for a match for an 8-digit number, and it generates an erroneous link, as on Folkert Mound Group.

Articles like this are clearly in error, and we already have a cleanup category (added when refnum is just blank) that I want to place these articles in automatically using the infobox, but I am unable to figure out how to do so using Module:String. I want any input that includes an 8 digit number to be linked to the external link, but I want to catch the case where there is just a reference in the parameter and put those articles in a cleanup category. Does anyone know how I might go about doing this? Thanks!--Dudemanfellabra (talk) 21:20, 16 November 2015 (UTC)

Dudemanfellabra, I added a check to make sure it only replaces digits at the beginning of the string. the other option may be to use module:unstrip which accesses the lua unstrip function. Frietjes (talk) 21:55, 16 November 2015 (UTC)

Problem with Horizontal TOC on Chrome

it was reported to me that {{Horizontal TOC}} is rendering in a vertical orientation in Chrome, for example see List of Presbyterian churches in the United States. I have Firefox and Internet Explorer installed, and see the intended horizontal lists in those browsers. I do not have Chrome installed, so I cannot check it, or diagnose any problems. anyone have any ideas about what is going on? does this have anything to do with recent changes to how lists are rendered? Frietjes (talk) 21:47, 16 November 2015 (UTC)

I can confirm the issue in Chrome 46.0.2490.86 on OS X El Capitan and I have taken a screenshot to demonstrate it. –Fredddie 22:00, 16 November 2015 (UTC)
User:Edokter, User:Redrose64, ... could this be related to the recent changes to the site-wide css? Frietjes (talk) 22:08, 16 November 2015 (UTC)
@Frietjes:, I just added .hlist .toctext { display:inline !important;} to my personal CSS, and it was magically fixed in Chrome. Should we add this to MediaWiki:Common.css? Plastikspork ―Œ(talk) 23:34, 16 November 2015 (UTC)
Could this be related to phab:T92481, which added display: table-cell to .tocnumber and .toctext in mediawiki.legacy/commonPrint.css? Laoris (talk) 00:17, 17 November 2015 (UTC)
Probably. Can't say for sure it is entirely at fault however - it does seem to be an isolated issue for Chrome users for example. Plus, the "base settings" for the toc class name and toc id value are screwy to begin with...
#toc,.toc {
  display: inline-block;
  display: table;
  zoom: 1;
  *display: inline; /* somebody thinks this helps things given the first 2 display hacks
                          to get around IE7 which isn't even supported anymore but it's like talking
                          to brick wall when comes to getting nonsense like this fixed */
  padding: 7px;
}
... but I believe the specific issue(s) in this case comes down to the use of the {{Horizontal TOC}} template & it's reliance on the .nonumtoc css settings found further down in MediaWiki:Common.css file from the hList settings.

I'm pretty sure something is now "off" in that definition set thanks to one recent change or the other and that is where User:Edokter will need to add something like...

.hlist.nonumtoc .toctext (
    display: inline;
}
... in order to "properly" rectify this 'specific' template 'related' issue. -- George Orwell III (talk) 05:51, 17 November 2015 (UTC)
George Orwell III and Plastikspork, thank you. it looks like we just need an admin like PrimeHunter or Edokter to make the change in MediaWiki:Common.css. I am pretty sure it's both the "nonumtoc" and otherwise which are having problems, but the "nonumtoc" is the most widely used with hlist (as far as I can tell). Frietjes (talk) 16:07, 17 November 2015 (UTC)
It is a little more complicated than that... .nonumtoc is not the culprit; it is the use of display:table-cell; which seems to take precedence in Blink (Chrome and Opera). I don't know why, but that should probably be reverted. In the mean time, I'll add a temporary fix to Common.css. -- [[User:Edokter]] {{talk}} 17:03, 17 November 2015 (UTC)
The horizontal TOC at List of Presbyterian churches in the United States is still a bit weird because now the numbers are visible. The temporary fix in Common.css needs to be placed above the .nonumtoc .tocnumber { display: none; } rule so that TOC numbers will still be hidden in horizontal nonumtoc TOCs. —Laoris (talk) 17:23, 17 November 2015 (UTC)
yes, for nonumtoc, the display for the number should be none. Edokter, can you fix this? thank you! Frietjes (talk) 18:08, 17 November 2015 (UTC)
Done. -- [[User:Edokter]] {{talk}} 18:50, 17 November 2015 (UTC)

Recent changes for WikiProjects

Is there a way to get a recent changes page for articles tagged within a WikiProject? {{WikiProject Devon}} is tagged on all the WikiProject's article's talk pages and the closest I can get to a recent changes is this, but it only lists edits to talk pages, not the articles themselves. Ideally I'd like to see changes to both. Many thanks to any help, Jolly Ω Janner 22:57, 16 November 2015 (UTC)

Wikipedia:Database reports/WikiProjects by changes is not quite what you want.
Wavelength (talk) 23:31, 16 November 2015 (UTC)
Indeed, I'm look for something more similar to Special:Watchlist. Jolly Ω Janner 00:05, 17 November 2015 (UTC)
@Jolly Janner: Here is what you want. I created a page with all the articles whose talk pages are tagged with the {{WikiProject Devon}} template, so now you can use the related changes feature to monitor those pages. You will probably want to ask someone who has AWB update the list of articles routinely so that new articles will get added to the watchlist. Hope that helps! --WayfaringWanderer (talk) 03:36, 17 November 2015 (UTC)
Wikipedia:Wikiproject devon/Article list is the list of articles that seeds the custom watchlist, by the way. --WayfaringWanderer (talk) 03:39, 17 November 2015 (UTC)
Many thanks, as this is exactly what I was looking for. This will be a useful window on the WP Devon page for users to patrol. Jolly Ω Janner 03:46, 17 November 2015 (UTC)

I do this for both WP Louisville and WP Kentucky, but I also include the talk pages. With AWB, it's pretty easy to build and save the list:

  1. Open AWB, then Select Source: Category and enter "WikiProject {your project's name} articles"
  2. Click "Make list".
  3. Right click over list, and click "Convert from talk pages"
  4. Click "Make list" again. (now you have all the project's pages and their talk pages)
  5. If you don't have "Keep alphabetized" set, then right click over list, and click "Sort alphabetically".
  6. Go to the menu, click 'List | Save list..." and proceed to save the list as a file.
  7. Open up the file you just saved in a text editor, Ctrl-A and Ctrl-C.
  8. Edit the wiki page you're keeping the list on (and running "Recent changes" against), Ctrl-A and Ctrl-V, and save the page.

Stevie is the man! TalkWork 21:10, 17 November 2015 (UTC)

Possible problem with DRN's "Request dispute resolution" button

Conversation here about how at least two editors have experienced a problem with the auto-form not appearing. Not sure about what is causing this problem, so I am going to ask the people on here if they have a clue about this? Cheers, Drcrazy102 (talk) 01:19, 17 November 2015 (UTC)

You need to enable the Form for filing disputes at the dispute resolution noticeboard gadget in preferences. 106.51.28.175 (talk) 12:06, 17 November 2015 (UTC)
Wikipedia talk:Dispute resolution noticeboard/request#Protected edit request on 17 November 2015 106.51.28.175 (talk) 13:06, 17 November 2015 (UTC)

Template:Geological_range is broken

See Template:Geological_range, Lokotunjailurus, Panthera leo spelaea (hidden for now), Miomachairodus --Ilya (talk) 17:37, 17 November 2015 (UTC)

+1:This is really bad, Sadads (talk) 17:41, 17 November 2015 (UTC)

My guess is that the problem is with the recent edit to Template:Period_start. Template editor or admin should revert. Mamyles (talk) 17:47, 17 November 2015 (UTC)

Sadads, Mamyles, moved the tfd notice inside the noinclude. Frietjes (talk) 18:04, 17 November 2015 (UTC)
Thanks for that, & I have done something similar at Template:Geologic Ages Inline (which needs checking by someone more expert than me). However, someone has tried to fix Template:Geological range which I don't think was wrong but now looks wrong. - David Biddulph (talk) 18:34, 17 November 2015 (UTC)
looks fine now. Frietjes (talk) 21:29, 17 November 2015 (UTC)

New Pageview API

If you're interested in page views, then you may be interested in this recent announcement from the Analytics team: https://lists.wikimedia.org/pipermail/analytics/2015-November/004529.html Whatamidoing (WMF) (talk) 19:30, 17 November 2015 (UTC)

Recent changes page

Hi, I suggest a feature for the RC page. I usually try to catch vandalism and revert it. The RC page is a convenient way to do so. However, I want to be able to see the major edits (like the bold text meaning large byte count change). This eliminates the small edits. Therefore, I suggest a feature be added that only shows bold changes (large changes in red and green) so it will be easier to catch section/[age blanking vandals, and others who post gibberish. Thanks, SireWonton (talk) 01:18, 18 November 2015 (UTC)

Multiple error reports of The Wikipedia Adventure (WP:TWA)

Greetings, From my watchlist, I see reports of TWA errors at Wikipedia talk:The Wikipedia Adventure (bottom of the page) and at Wikipedia:Teahouse/Questions#The_Wikipedia_Adventure_Getting_Stuck_on_Automated_Messages. From one of the messages, it looks like it may be being worked on.

We've logged this bug and are working to fix it: https://phabricator.wikimedia.org/T118823 Ocaasi (WMF) (talk) 22:35, 17 November 2015 (UTC)
I don't know if anyone here at VP needs to get involved, but I thought to at least report it here. Thanks. JoeHebda (talk) 01:04, 18 November 2015 (UTC)

Hi all. Been on a long break, learning the ropes again. Have I missed some change, or is there a reason I can't seem to make an external link in an edit summary anymore? I swear it used to be possible. Or perhaps I'm losing my mind. Thanks! Ignatzmicetalk 04:05, 18 November 2015 (UTC)

Nope, never was possible. Too much opportunity for mischief. TenOfAllTrades(talk) 04:11, 18 November 2015 (UTC)
Well then. Time to rest my brain for a while, I suppose. Ignatzmicetalk 04:16, 18 November 2015 (UTC)
Welcome back.--S Philbrick(Talk) 15:49, 18 November 2015 (UTC)

Chrome's feedback and Watchlist's shortcuts coincide

Has Wikipedia society resolved this? Just in case you don't know a crash report window pops up if you click Alt+Shift+I in Chrome--Dixtosa (talk) 12:27, 18 November 2015 (UTC)

Yes, not just Wikipedia. Try it on any other page, pressing that will make it pop up. — Preceding unsigned comment added by SireWonton (talkcontribs) 01:14, 19 November 2015 (UTC)

Tool for ordering lists

Hello. Some people have the habit of ordering chronological lists top->bottom from the most recent item to the oldest, which is unusual for – say – filmography lists. Has anybody written an alogorithm to turn them upside down? I know it's quite a basic thing in any programming language, but I'm quite rusty with those. :) Thanks for any replies. --Mihai (talk) 13:57, 18 November 2015 (UTC)

What lists? Can you give an example? --Dixtosa (talk) 14:29, 18 November 2015 (UTC)
I have found the need to do this on several occasions. However, I think it is relatively easy to do semi-manually:
  1. copy list to spreadsheet
  2. add an index column
  3. sort largest to smallest
  4. paste results back to Wikipedia
Suppose an automated tool could make this a little easier, but given how easy it is to do now, I prefer that scarce developer tools be prioritized on other issues.--S Philbrick(Talk) 15:47, 18 November 2015 (UTC)
@Dan Mihai Pitea and Sphilbrick: Here is a function that reverses the lines in the selected text in the text box, you can add it to your common.js file. Unfortunately, adding buttons to the WikiEditor's toolbar is not as simple as it should be [75],[76]. You can call the function from the browser's console instead -- after saving the code to your common.js file, select several lines in the textbox, press F12 to open the console, and enter reverseLines() (or if you didn't save it to your common.js yet, enter the middle 3 lines from the code below). --V111P (talk) 04:01, 19 November 2015 (UTC)
function reverseLines() {
    var $ta = $('#wpTextbox1');
    var text = $ta.focus().textSelection('getSelection').split('\n').reverse().join('\n');
    $ta.focus().textSelection('encapsulateSelection', {pre: text, peri: '', post: '', replace: true});
}

Can someone explain to me why File:FLLlogo.jpg is showing an older image, even though I just uploaded a brand new picture (since they just re-branded)? The other pictures that I updated (File:FIRST Robotics Competition (logo).png, File:Jr FLL logo.jpg, & File:FIRST Tech challenge logo.png) all changed. Is this a bug or did I do something wrong? Elisfkc (talk) 20:10, 18 November 2015 (UTC)

Your computer probably has the old one cached. Try a WP:BYPASS. --Redrose64 (talk) 20:41, 18 November 2015 (UTC)
It wasn't even working when I used Chrome incognito, which usually gets around the cached problem. However, it's correct now so everything is fine. Elisfkc (talk) 20:52, 18 November 2015 (UTC)

Category:Pages with incorrect ref formatting – 200 new overnight

There not actual new, some several weeks old. All seem to be connected to media (TV mostly). Null edit make it go away, nor real edit. – Unbuttered parsnip (talk) mytime= Thu 16:53, wikitime= 08:53, 19 November 2015 (UTC)

Someone has tried to add remove a reference to a navbox, {{NBCUniversal}}. -- John of Reading (talk) 08:57, 19 November 2015 (UTC)
You're right, I just saw it, small blue writing on a blue background. Unbuttered parsnip (talk) mytime= Thu 16:59, wikitime= 08:59, 19 November 2015 (UTC)
Fixed it. -- John of Reading (talk) 09:01, 19 November 2015 (UTC)