[go: nahoru, domu]
Jump to content

Wikipedia:Village pump (technical)

From Wikipedia, the free encyclopedia

This is an old revision of this page, as edited by 88.89.219.147 (talk) at 06:21, 24 April 2016. The present address (URL) is a permanent link to this revision, which may differ significantly from the current revision.

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

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


Unresolved issue is archived - How to "UN-Archive"?

Greetings, A problem awaiting resolution is now archived here and tracked at phabricator:T126553.

If it's not possible to pull back from archive, does it need to be re-posted again?

Regards,  JoeHebda (talk)  19:04, 27 February 2016 (UTC)[reply]

@JoeHebda: Just post the new question, with a link back to the archived thread. --Redrose64 (talk) 21:39, 27 February 2016 (UTC)[reply]
If I re-post again here, will it be archived again before the issue is solved? Just wondering... JoeHebda (talk)  02:19, 29 February 2016 (UTC)[reply]
Possibly; archiving bots do not take any notice of whether a thread has been replied to or not. Threads on this page are archived if they've not been posted to for five days. So if nobody responds here before the next bot run that occurs after 10:30, 5 March 2016 (UTC), it will be archived. --Redrose64 (talk) 10:30, 29 February 2016 (UTC)[reply]
Redrose64 – Thanks for the update. Yesterday I did repost the new VPT archive link to Wikipedia talk:User scripts#Fix needed: Gadget only works with Vector skin, script error which was orginally posted on Feb. 15th. Wondering if there might be a better place to post where the issue can be fixed? Without waiting for months for a resolution? If I knew anything about scripts (which I do not) I would attempt to fix myself. Regards,  JoeHebda (talk)  13:39, 29 February 2016 (UTC)[reply]

1) ω Awaiting a solution – also see: MediaWiki:Gadget-mobile-sidebar and Wikipedia:Gadget.  JoeHebda (talk)  13:45, 29 February 2016 (UTC)[reply]

@JoeHebda: You can use {{DNAU}} to prevent archiving. nyuszika7h (talk) 21:26, 29 February 2016 (UTC)[reply]

2) ω Awaiting a solution — JoeHebda • (talk) 12:08, 2 March 2016 (UTC)[reply]

3) ω Awaiting a solution — JoeHebda • (talk) 18:12, 3 March 2016 (UTC)[reply]

FWIW, i copied User:Brion VIBBER's script locally, and verified it works with "monobook" and "modern" after minor changes. i had to modify about 5 lines to use mw.util.$content instead of $('#content'), and add a line to incease z-index specifically for monobook: for some reason the left edge of the mobile image was overshadowed by the right edge of the content in monobook (i'm sure there's more . you can try User:קיפודנחש/mobile-sidebarcopy.js to verify it indeed works with other skins (you'll probably want to disable the gadget, or you may get 2 of them...). peace - קיפודנחש (aka kipod) (talk) 20:55, 3 March 2016 (UTC)[reply]
קיפודנחש – Thanks for helping. I know about testing computer programs in other languages but nothing about js scripts. If you or anther editor could explain the steps to test, I would be willing to try testing. So far, I did un-check existing Gadgets option to disable Mobile sidebar preview option. Regards, — JoeHebda • (talk) 21:36, 3 March 2016 (UTC)[reply]
to test the modified script, add to your personal script file the following line (if you copy from edit screen, *do not* include the "code" tags):
importScript('User:קיפודנחש/mobile-sidebarcopy.js');
(if the file does not exist yet, create it, add the line, and save. more detailed instructions and explanations about personal scripts can be found in WP:JS). peace - קיפודנחש (aka kipod) (talk) 22:15, 3 March 2016 (UTC)[reply]

קיפודנחש – Followed the test instructions & same results = Mobile sidebar preview works only with Vector skin & not the other skins. I did the Purge and logout for each; not seeing the little toolbar icon to activate preview. I know it is running the test version because I have the Gadgets one unchecked with Vector & icon shows & works correctly. — JoeHebda • (talk) 22:37, 3 March 2016 (UTC)[reply]

you are correct. it turned out to be more elaborate than i thought - the code Brian wrote displayed the phone nicely, but placing the icon for turning it on/off on the menu turned out to be a bitch... anywhoo, i at least made it so that you can use it now, though it's not really pretty when you use a skin other than vector. peace - קיפודנחש (aka kipod) (talk) 00:10, 4 March 2016 (UTC)[reply]
@JoeHebda: update: could not get the mobile on/off icon to look acceptable on monobook, modern and cologne blue, so on these skins, the on/off switch is text only (top row for mo*, left-menu under "Edit" for cologne). peace - קיפודנחש (aka kipod) (talk) 16:34, 4 March 2016 (UTC)[reply]
@קיפודנחש: – So far, it looks good and okay to me. For Cologn Blue, I am confused because I can not find any "Edit" anywhere on the page; not on top or left sidebar. I do see a "Mobile view" but that is just for switching the entire page between Mobile/Desktop views. Wondering where to look? — JoeHebda • (talk) 16:48, 4 March 2016 (UTC)[reply]
@JoeHebda: i practically never use cologne, and not aware of all its idiosyncrasies. for me, using this skin, when i am on a page i am allowed to edit, i see in the left-hand menu column "Edit". i verified that this is so even as anon (you can test any skin by appending to the address line "?useskin=<skinname>". this works for anons too. names are [ vector monobook modern cologneblue ]). i also verified that the script causes "Mobile" to appear there even for anons. there might be something in your prefs that causes it not to show, or maybe it *does* show and for some reason you missed it. HTH, peace - קיפודנחש (aka kipod) (talk) 17:58, 4 March 2016 (UTC)[reply]

קיפודנחש – Yes, I did find it now. I had to set my custom CSS font-size to 155 percent for Cologne Blue skin to help overcome a line-height issue making that skin very hard to read. And the Mobile preview sidebar works okay there as well. Just a FYI, I found a report of User skin preferences at Wikipedia:Database reports/User preferences#Skin. Thanks so much for getting Mobile to work correctly with these skins! IMO not seeing the little toolbar icon is a lesser (non-structural) issue.

Now in order to go live with this script, what needs to be done? Once it is out there I will check On in Gadgets. Then, could the Mobile sidebar preview line of Gadgets be moved out of the Testing and development section? Perhaps move into Appearance section? Cheers! — JoeHebda • (talk) 19:13, 4 March 2016 (UTC)[reply]

@JoeHebda: i saw this more as an experiment, to gauge the amount of change needed to make mobile-sidebar work with other skins. ideally, User:Brion VIBBER would patch the script on meta to work with all skins (TBH, Brion needs me to tell him how to do that exactly as much as i need my grandson to teach me to suck eggs). if, for any reason, Brion doesn't do it, enwiki can decide that limiting this gadget to vector is acceptable. the very last resort would be to change the gadget on enwiki to consume Brion's css file from meta, and the modified js file from my userspace (or some copy of it). this last option is both undesirable and unlikely. if none of those options is executed, people who do not use vector can consume my modified script privately, the same way i listed above and you tried. peace - קיפודנחש (aka kipod) (talk) 20:31, 4 March 2016 (UTC)[reply]
@קיפודנחש: – Thanks for the answer. It sounds complicated, but I think I understand. For now, I will leave the test js in place and wait until resolved. Cheers! — JoeHebda • (talk) 13:51, 5 March 2016 (UTC)[reply]

Template styles

Hey all.

So, since the Jerusalem Hackathon I've been working on a very very longstanding issue we have had with styles. The current WIP solution: template styles.

That new feature allows attaching proper CSS to templates, finally getting rid of limited and troublesome inline styles in elements of the page. The upsides:

  • No redundancy: If you use a style on more than an element, simply place them all in the same class and share styles;
  • Even less redundancy: if you have a family of templates sharing styles, you can put all the styles in the one "library" template and include that;
  • Proper style sheets means support for @media blocks, finally (thus making it possible for styles to work right on mobile, amongst other things);
  • Smaller and easier to maintain templates;
  • Style sheets becomes available to anyone being able to edit the template, rather than just admins; and
  • No more need to edit common.css and have to deal with the crap of caching, or the risk of breaking the site.

The downsides... well, I can't think of any yet but I'm sure we'll find some eventually.  :-)

There is a test wiki in Labs where the current WIP extension is deployed. Feel free to register (not with your real Wikimedia credentials!) and play with it. Beware that this is my testing ground and thus may randomly be unstable. The main page there is a transcluded template with styles if you want a quick and easy example to look at. There is no documentation (yet) but the jist of it is simple: add a <templatestyles>...</templatestyles> element to a template containing a style sheet, and that style will be prepended to any page that transcludes that template (just once, even if the template is transcluded multiple times, including recursively). Not unlike TemplateData, you can actually put the templatestyles element on a subpage - but I expect that would be a bad idea in this case since it affects transclusions and thus should track the template's own protection level.

This pet project of mine has a lot of buy-in and support from the WMF (tracked in phabricator at T483) so once we're happy about how it works I've no doubt it will end up being deployed in production fairly quickly. — Coren (talk) 15:05, 14 April 2016 (UTC)[reply]

Sweeeeet! Hope this gets deployed soon, so we can properly style the Main Page. -- [[User:Edokter]] {{talk}} 15:52, 14 April 2016 (UTC)[reply]
Extreme enthusiasm \m/ =) \m/  fredgandt 16:46, 14 April 2016 (UTC)[reply]
Best thing EVAH ! Will take us 5 years to get rid of the existing cruft I'm sure, but this will be awesome. —TheDJ (talk • contribs) 17:56, 14 April 2016 (UTC)[reply]
Yeah, this looks very good. --Edgars2007 (talk/contribs) 17:57, 14 April 2016 (UTC)[reply]
FYI, I experimented with this using the Wide image template. {{Wide image}}. —TheDJ (talk • contribs) 16:58, 18 April 2016 (UTC)[reply]
Wonderful! Sounds like an awesome change. Kudos to you for implementing/coming up with this. APerson (talk!) 02:35, 20 April 2016 (UTC)[reply]

I have only one thing to say: Lua Lua Lua.

This would be better if it came with a lua library that would allow adding such styles without a hacky extensiontag: 

frame:extensionTag{ name = 'templatestyles', content = 'some text', args = { name = 'foo', group = 'bar' } }

Potentially something like mw.templatestyles or mw.styles or mw.html.styles. This would make the conversion of existing modules much more efficient. 08:28, 21 April 2016 (UTC) — Preceding unsigned comment added by 197.218.89.56 (talk)

Technical help for expert review?

BMJ, the publishers of the British Medical Journal and other top-tier biomedical journals, have kindly recruited the best minds they can get to review Parkinson's disease.

We began the review by passing the article, in a Word document, from one reviewer to the next by email. Each made proposed changes to the article text and left comments in the document, using Word's "Review" and "Track changes" features.

At that point we needed to start a discussion, and Word isn't ideal for that. So I pasted the relevant paragraphs from the Word document into the left column of a wiki table, and the reviewers' comments into the right column, where the discussion could happen (here). I manually applied background colours to distinguish deletions from additions in the left column, using <span style="background:#xxxxxx">.

That discussion has now begun but one of the many things I've learned during all this is, the top researchers and theorists spend a lot of time in the air (travelling to conferences, lectures, meetings), and it is then, free from the demands of job and family, when they do their reviewing.

So, yesterday, I pasted that table into Word and have made it available to the reviewers (here). Now they can download a copy before they get on a flight, and email it back to me with their comments when they're online again, and I'll transcribe their comments into the wiki table.

This may be as simple as it gets but I just thought I'd put this before you, in case you may have any thoughts on a better technical approach for next time. (BMJ have offered to do more of these.) I'm finding the construction of the wiki table tedious (particularly highlighting the deletions and additions), and transcribing offline comments from the Word document into the wiki table will be a small chore. The wiki table pastes easily into Word with highlighting and formatting intact, but not vice versa. --Anthonyhcole (talk · contribs · email) 04:59, 17 April 2016 (UTC)[reply]

Forgive my being simplistic, but it seems you're unnecessarily convoluting a procedure MediaWiki software was designed to handle. That is, multiple editors working on and discussing an article. By taking the text out of the wiki, working on it and bringing a new version back, the history and attribution will be effectively corrupt, entirely aside from the whole process being clearly taxing. I think it's especially important to maintain a genuine edit history for attribution in this case, since there appears to be a not minor possibility of a conflict of interest concern. fredgandt 06:55, 17 April 2016 (UTC)[reply]
Fred, you make a very good point about conflict of interest. What we're doing here is, under the oversight of one of the world's most prestigious journal publishers and editors, applying a highly stringent peer review model to a Wikipedia article, one element of which is transparency. BMJ's editor-in-chief, Fiona Godlee is an expert and has published widely on peer review. She and the organisation under her are strident evangelists for open science and open access. In our first meeting we agreed the reviewers would have to be named and declare their COIs and potential COIs. I think that's the best way to deal with reviewers' interests. Their peers will review their declarations so, hopefully that scrutiny will keep them honest. It's not perfect but I think transparency and a public declaration of interests is the least ineffective approach in peer review. What appears in the article is up to the Wikipedia editors, and Wikipedia's best biomedical editors are a tough crowd.
The edit history of the Word document is saved by the "Track changes" feature and the remainder of the process happens on-wiki, so the review's edit history is preserved.
I'd love to do it all on-wiki but these people need to be able to work on this stuff offline. --Anthonyhcole (talk · contribs · email) 10:37, 17 April 2016 (UTC)[reply]
Sounds good Anthony; if Word is able to keep track of the changes for attribution, and they're comfortable editing with it, the major hurdle is the work you're doing to rebuild the table, correct? Notepad++ can do some nifty things which might be handy. Are you any good with Regular expressions? Could the use of the new fangled Visual Editor version of Wikipedia help? I might be able to build you some dandy web app to help, but would need to see the practical workflow and sleep on it. fredgandt 10:50, 17 April 2016 (UTC)[reply]
(I don't know where I got the idea you mentioned COI. I'm in another conversation about this; it must have been someone there.)
Let's all sleep on it for a while. Just articulating my problem here has helped me to put it in perspective - it's pretty trivial, really. Considering what the reviewers and editors are contributing.
I'm using the VE for half my edits, I think, and the reviewers are only using the VE. They'll have to use wikitext when the review moves to the article's talk page, but I'm so glad they can begin with Word and Visual Editor. --Anthonyhcole (talk · contribs · email) 11:05, 17 April 2016 (UTC)[reply]
I did mention COI, but only as a possible concern. It's mainly attribution that first struck me as a big problem, but I see now that the final wiki editing will take place after the review. Anyhoo - the technical issue of building and rebuilding the table - not easier with the VE? I must admit to not having even tried it.
Notepad++ has an export feature that outputs HTML describing the syntax highlighting which could save you a lot of work. I'm not certain (experimentation required) but you should be able to paste the replies you get into NP++, then save as a custom language with the highlighting you require, then export as HTML to paste into your sandbox. All the highlighting would be already done. The devil's in the details. fredgandt 11:30, 17 April 2016 (UTC)[reply]
The visual editor doesn't preserve his color-coding or text formatting. The table itself is fine, but there's no "highlighter marker" available. Whatamidoing (WMF) (talk) 15:13, 17 April 2016 (UTC)[reply]
(Ah. I see it now.) Thanks for the NP++ tip, Fred. I'll have a play with it. --Anthonyhcole (talk · contribs · email) 07:25, 18 April 2016 (UTC)[reply]

@Anthonyhcole: - I have a simple (more or less) solution for part of the problem for you. A way to maintain the highlighting when you copy from Word and paste into the wiki. I'll be back after code wrangling (will require a small amount of JavaScript). fredgandt 08:36, 18 April 2016 (UTC)[reply]

Anthonyhcole (talk · contribs · email) 12:34, 18 April 2016 (UTC)[reply]
User:Fred_Gandt/contentEditTable.js <-- still a bit sketchy, but my dog needs a walk. I'll explain when I get back. Fundamentally it turns the table cells of the preview of one of the tables (per section basis) in your sandbox into a contenteditable version of itself. You can then copy the equivalent cell from the browser editing version of the Word doc, and paste it into the editable cell. The syntax highlighting is maintained. Then down by the save button there's another marked "Get raw table HTML", which when clicked, populates the textbox (you're in preview remember) with the HTML of the table (with some tweaks). Then preview that markup and the sky falls! Like I say, I have to walk my dog, so play around if you like. I'll be able to walk you through it and take feedback and solve problems when I return. fredgandt 13:45, 18 April 2016 (UTC)[reply]
Thanks Fred. Have I done this right? I went to the "Introduction" table in my sandbox, clicked "Edit source", then "Show preview". The "Get raw table HTML" button appears but I can't edit the cells. (I'm in Vector on Chrome on Windows 10.) --Anthonyhcole (talk · contribs · email) 01:56, 20 April 2016 (UTC)[reply]
@Anthonyhcole: - Sorry - for use with the normal (no Visual Editor) mode. I'll upload some screen capture later to assist. fredgandt 08:22, 20 April 2016 (UTC)[reply]
Yep, I am using the normal mode. I should be able to copy the contents of a cell from this "Word online" document into a table in my sandbox (into the preview, not the edit box), right Fred? How do I make the table cell in the preview editable? Do I have to double-click or something? --Anthonyhcole (talk · contribs · email) 11:25, 20 April 2016 (UTC)[reply]
I shall ping you with a link to a video of me using it in a while. I have to eat my lunch and wallow in the afterglow of a stuffed belly first :-) fredgandt 12:08, 20 April 2016 (UTC)[reply]
Wow. Take your time. I really appreciate your engagement. --Anthonyhcole (talk · contribs · email) 14:21, 20 April 2016 (UTC)[reply]
@Anthonyhcole: Interrupted by Wiki maintenance - upload linked at the top of User:Fred Gandt/contentEditTable#Use. Short and simple. fredgandt 15:35, 20 April 2016 (UTC)[reply]
For the record, Fred addressed my queries at my talk page, and now I can copy and paste from Word Online into the wiki table, preserving my highlighting and strikethroughs. Thank you Fred! --Anthonyhcole (talk · contribs · email) 00:05, 23 April 2016 (UTC)[reply]
 fredgandt 00:43, 23 April 2016 (UTC)[reply]
The issue concerns external review of medical articles by experts, and a discussion saying that such a review should not happen would only derail this important request. I'm hoping Anthonyhcole will get a reply from someone who knows how to handle it (perhaps try WP:Reference desk/Computing?), but if no one wants it I might have a look although I've never fiddled with review text in a doc file. Johnuniq (talk) 07:47, 17 April 2016 (UTC)[reply]
I was going to let it slide, but changed my mind; don't fob me off as some idiot passer by that's getting in the way. Read what I wrote and try again. Talk about derailing?! Just exactly where did anyone suggest that the review shouldn't happen? Thankfully Anthony was capable of reading and responding properly, so with my question (of why this can't be done normally) answered, we can move on to making his life easier. fredgandt 11:12, 17 April 2016 (UTC)[reply]
Hi Johnuniq. A discussion saying that such a review should not happen would be misconceived because it's out of Wikipedia's control. Wikipedia's rules kick in and the community exercises total control once the review is done and it arrives on the article talk page. It's entirely up to Wikipedia what it does with it. I hope some interested, intelligent editors take some notice, and engage the reviewers in polite but rigorous discussion of their proposals.
Don't worry too much about the technical question I raised. This present set-up is working (I think) and it's really not very onerous when measured against the payback. --Anthonyhcole (talk · contribs · email) 10:37, 17 April 2016 (UTC)[reply]
Export as PDF, import into Word and then you can use the Word reviewing tools (it's there in Word 2013, I don't know about previous versions). You can get the end product and submit it for review on-wiki by peers. Shouldn't that do it? --QEDK (T ☕ C) 11:19, 17 April 2016 (UTC)[reply]
That's sort of what we're doing but between the Word review and the review appearing on the article's talk page we're having an intermediate conversation, making sure the reviewers' suggestions conform to WP policy, and sorting out differences between the reviewers. Word's review format isn't ideal for that kind of extended conversation about details in a long document. --Anthonyhcole (talk · contribs · email) 07:25, 18 April 2016 (UTC)[reply]

User stats - last edit

I can use tools.wmflabs.org/xtools-ec/?user= to get the last edit date done by a user but it also dumps a bag of other stuff. Is there a way of giving a tool a list of users and getting just the corresponding last edit date, either a list or one at a time does not matter ? (I am thinking of reinvigorating a project, with a couple of hundred "registered" participants but even the lead coordinator formally retired as a Wikipedian over 12 months ago and the project page has not been updated accordingly and many others seem to have not edited at all anywhere for one or more years, so it would be good to be able to easily see who was still around.) Aoziwe (talk) 13:35, 17 April 2016 (UTC)[reply]

Give me a list of user names, or point to place, where I can find them. --Edgars2007 (talk/contribs) 14:06, 17 April 2016 (UTC)[reply]
Here - Wikipedia:WikiProject Law Enforcement#Participants and here - Wikipedia:WikiProject Law Enforcement/Participants. Thanks in advance. Aoziwe (talk) 12:42, 18 April 2016 (UTC)[reply]
Sounds like you want my useractivity tool. — Dispenser 14:40, 17 April 2016 (UTC)[reply]
Thanks I will look at that too. Aoziwe (talk) 12:42, 18 April 2016 (UTC)[reply]
Looks good. Thanks. Is it written in python? Any chance of seeing the source ? Aoziwe (talk) 12:32, 20 April 2016 (UTC)[reply]
You might also be interested in Wikipedia:WikiProject Directory/Description/WikiProject Astronomical objects (or equivalent for the project you're interested in). WhatamIdoing (talk) 15:27, 17 April 2016 (UTC)[reply]
Thanks I will look at that too. Not quite the exact same thing but will probably be very useful if I get the project going again well. Aoziwe (talk) 12:42, 18 April 2016 (UTC)[reply]
Looks very useful too. How does the bot know which page to load the list of users into ? It looks like the opt out page is shared by all projects/user list loads ? Aoziwe (talk) 11:13, 22 April 2016 (UTC)[reply]
That's probably a question for User:Harej. Whatamidoing (WMF) (talk) 00:50, 24 April 2016 (UTC)[reply]

Server switch 2016

➤ To read this in other languages, see m:Tech/Server switch 2016.

The Wikimedia Foundation will be testing its newest data center in Dallas. This will make sure Wikipedia and the other Wikimedia wikis can stay online even after a disaster. To make sure everything is working, the Wikimedia Technology department needs to conduct a planned test. This test will show whether they can reliably switch from one data center to the other. It requires many teams to prepare for the test and to be available to fix any unexpected problems.

They will switch all traffic to the new data center on Tuesday, 19 April.
On Thursday, 21 April, they will switch back to the primary data center.

Unfortunately, because of some limitations in MediaWiki, all editing must stop during those two switches. We apologize for this disruption, and we are working to minimize it in the future.

You will be able to read, but not edit, all wikis for a short period of time.

  • You will not be able to edit for approximately 15 to 30 minutes on Tuesday, 19 April and Thursday, 21 April, starting at 14:00 UTC (15:00 BST, 16:00 CEST, 10:00 EDT, 07:00 PDT).
  • If you try to edit or save during these times, you will see an error message. We hope that no edits will be lost during these minutes, but we can't guarantee it. If you see the error message, then please wait until everything is back to normal. Then you should be able to save your edit. But, we recommend that you make a copy of your changes first, just in case.


Other effects:

  • Background jobs will be slower and some may be dropped. Red links might not be updated as quickly as normal. If you create an article that is already linked somewhere else, the link will stay red longer than usual. Some long-running scripts will have to be stopped.
  • There will be a code freeze for the week of 18 April. No non-essential code deployments will take place.

This test was originally planned to take place on March 22. April 19th and 21st are the new dates. You can read the schedule at wikitech.wikimedia.org. They will post any changes on that schedule. There will be more notifications about this. Please share this information with your community. /User:Whatamidoing (WMF) (talk) 21:07, 17 April 2016 (UTC)[reply]

Translation for NZ readers: You will not be able to edit for approximately 15 to 30 minutes on Wednesday, 20 April and Friday, 22 April, starting at 02:00 NZST. Akld guy (talk) 01:20, 18 April 2016 (UTC)[reply]
Your comment seems more likely to confuse than help readers. Many may not realize NZ refers to New Zealand and NZST is New Zealand Standard Time. There must be relatively few editors in that time zone and most of them are probably aware they are in an unusual time zone for Wikipedia and sometimes need to change the date/day when converting UTC times to New Zealand time. For the large majority of editors it takes place Tuesday, 19 April and Thursday, 21 April in their local time zone. Your earlier objection to including day of the week in UTC announcements got no support. PrimeHunter (talk) 02:10, 18 April 2016 (UTC)[reply]
Your comments come across as snarky and argumentative. NZ readers will immediately realize what NZ and NZST refer to. The translation is directed to nobody else, so why comment? Akld guy (talk) 02:36, 18 April 2016 (UTC)[reply]
Non-NZ readers will not know the post is not directed to them if they don't know what NZ refers to. And NZ readers who are helped by the comment are probably so rare it isn't worth the time everybody else spend reading the comment. Using UTC time is standard in Wikipedia and I hope people don't start posting replies just to convert to other time zones. PrimeHunter (talk) 02:58, 18 April 2016 (UTC)[reply]
But they already did. In case you didn't notice, Whatamidoing converted it to BST, CEST, EDT, and PDT. So all I did was convert the time for NZ readers too. Why pick out my post as likely to be misleading? Are readers not also just as likely to misinterpret Whatamidoing's conversions, or is this something personal? Akld guy (talk) 03:27, 18 April 2016 (UTC)[reply]
Continued on your talk page. PrimeHunter (talk) 11:18, 18 April 2016 (UTC)[reply]
Barely anyone in that country would even notice what happened. Most people there would have been asleep that time. 49.148.29.21 (talk) 00:46, 20 April 2016 (UTC)[reply]
I should have included a link to http://www.timeanddate.com/worldclock/fixedtime.html?iso=20160419T14 with this message. That provides a translation to any time zone. Whatamidoing (WMF) (talk) 19:42, 18 April 2016 (UTC)[reply]

I just went to edit a page, and realized that the "Edit" button for VisualEditor is missing. It still gives me the "Edit Source" button, which seems to work fine. Is that related to this whole issue? Because I don't find anything else about it, and the timing seems coincidental. AnnaGoFast (talk) 00:22, 22 April 2016 (UTC)[reply]

That's the mw:VisualEditor/Single edit tab. Go to Special:Preferences#mw-prefsection-editing, scroll halfway down to "Editing mode", and pick whichever you want. At a guess, your prefs are probably set to "Always give me the source editor" right now. Whatamidoing (WMF) (talk) 00:52, 24 April 2016 (UTC)[reply]

Not exactly; I checked, and it was set to "Remember last editor", so I changed it to "Show both edit tabs". I don't understand why it changed all of a sudden like that though; it's never done that in the several months I've been around, and I've used both Visual and Source editing during that time. And I'm not sure what "remember last editor" even means, or why it would make the Visual editor disappear. Oh well, fixed now. Thanks for your help. AnnaGoFast (talk) 01:13, 24 April 2016 (UTC)[reply]

That watchlist jump that makes us misclick

You know how when the watchlist loads and then at the last minute the top announcements that you dismissed vanish? The whole watchlist jumps up a few lines. That makes me accidentally hit rollback from time to time. I'm probably not the only one.

So, can't we instead have the option of a little notification somewhere linking to a page with announcements? Wouldn't that fix it?

Anna Frodesiak (talk) 05:52, 18 April 2016 (UTC)[reply]

Thanks for posting here Anna. I struggle with this too. Jytdog (talk) 06:06, 18 April 2016 (UTC)[reply]
And me. Thanks Anna for opening this discussion. --Dirk Beetstra T C 06:13, 18 April 2016 (UTC)[reply]
Me too. A very regular and troublesome problem. Martinevans123 (talk) 12:47, 18 April 2016 (UTC)[reply]
Yes, thank you. If you dismiss the announcement, it shouldn't load at all. The dismiss status needs to be checked before the page loads, not after. Akld guy (talk) 06:16, 18 April 2016 (UTC)[reply]
Stopgap solution: put #watchlist-message { display:none; } in your user css and watchlist MediaWiki:Watchlist-details. I've done something similar with the sitenotice for years. —Cryptic 06:17, 18 April 2016 (UTC)[reply]
For the watchlist messages this can be fixed by depoying this updated JS code and a new piece of CSS. Which admin wants to ? —TheDJ (talk • contribs) 09:16, 18 April 2016 (UTC)[reply]
TheDJ, I was going to ask whether you were familiar with the coding and could stand behind it, but now I see that you're a developer. Only one issue remains: what do I do with it? Instead of suggesting that I deploy it, please give me the entire code for the destination (and a link, to ensure that I go to the right place), with your new piece added. I don't trust myself to add it to the right spot on the page, but copy/pasting the entire page's code should be failproof. Nyttend (talk) 14:41, 18 April 2016 (UTC)[reply]
I got it. -- [[User:Edokter]] {{talk}} 16:14, 18 April 2016 (UTC)[reply]
Was this intended to break the ability to permanently hide all watchlist notices with #watchlist-message { display:none; }? – Steel 17:32, 18 April 2016 (UTC)[reply]
Intended, no, but that CSS rule won't work anymore since it is not strong enough. You'll need something like: #watchlist-message { display:none !important; }. —TheDJ (talk • contribs) 17:37, 18 April 2016 (UTC)[reply]
Does size of watchlist have any bearing on the prevalence of this glitch? Martinevans123 (talk) 18:20, 18 April 2016 (UTC)[reply]
The size of your watchlist has no bearing on this. At most the size of the part that you DISPLAY of your watchlist could have any bearing on it. —TheDJ (talk • contribs) 18:31, 18 April 2016 (UTC)[reply]
Thanks. It must be my imagination that it's got worse over the course of 7 years. Or I've unwittingly changed the size of the part displayed. Or the top announcements have got bigger. Or I've learned to progressively click too impatiently. Or something else. I guess. Martinevans123 (talk) 18:36, 18 April 2016 (UTC)[reply]
Thank you Anna! The struggle is real. I'm glad someone brought this up. EvergreenFir (talk) Please {{re}} 18:34, 18 April 2016 (UTC)[reply]
I have found that there are three things that cause watchlist jumping, they are largely independent. They are: (i) not clicking [hide] on all geonotices; (ii) clicking [dismiss] on any non-geonotice messages; (iii) disabling "Display pages on your watchlist that have changed since your last visit in bold (see customizing watchlists for more options)" at Preferences → Gadgets → Watchlist. There may be others. --Redrose64 (talk) 23:32, 18 April 2016 (UTC)[reply]

yeah the "Display pages on your watchlist that have changed since " issue is the largest remaining one, that is a bother even if you dismissed the messages. The case of jumping when there are unread messages is harder to fix. —TheDJ (talk • contribs) 13:11, 19 April 2016 (UTC)[reply]

@Edokter: I realize that I probably broke geonotices with that change to the watch list messages. Geonotices use the same div, but don't explicitly show the parent #watchlist-message element, when there are messages for display. —TheDJ (talk • contribs) 13:18, 19 April 2016 (UTC)[reply]
@TheDJ: I didn't study the code in detail. Is there anything specific that should be done to fix it? Revert? -- [[User:Edokter]] {{talk}} 15:43, 19 April 2016 (UTC)[reply]
@Edokter: Can you add .show() to line 111 of MediaWiki:Gadget-geonotice-core.js ? —TheDJ (talk • contribs) 17:06, 19 April 2016 (UTC)[reply]
 Done. -- [[User:Edokter]] {{talk}} 18:03, 19 April 2016 (UTC)[reply]

Thanks all. I don't understand much of the above, but it seems fixed now. Is that right? Anna Frodesiak (talk) 01:37, 20 April 2016 (UTC)[reply]

Restoring previous protection

India national cricket team was indefinitely semiprotected in 2009. I've just now fully protected it because of an edit-war, but once the full protection expires in a few days, the software won't automatically restore the semiprotection. Perhaps a good thing in this specific situation; we can see whether the end of semiprotection will result in more vandalism, or not. Is there any tool that tracks situations in which temporary full protection of an indef-semiprotected page causes all protection to end? Some time ago, I suggested that we have a bot to remind admins to restore semiprotection if they thought it necessary, but I don't think that went anywhere. Nyttend (talk) 13:17, 18 April 2016 (UTC)[reply]

Assuming the page is on your watchlist, you should get a watchlist entry detailing the protection change - right? If so, and also assuming that you have an epic watchlist, a script could alert you when specific changes occur i.e. use the API to keep tabs on a special user defined list of watched pages such that when they change, you get an almost instant heads up. If anything I just said makes sense, I can add it to my list of things to do. Currently sleepy. fredgandt 18:17, 18 April 2016 (UTC)[reply]
Previous protection states are not remembered, although they are logged. For articles that exist, there are two actions that can be protected - edit and move; and each of these two can have any one of five protection levels (none, semi, 30/500, template, full) set for it, they don't need to be the same for both. Any protection level (other than "none") can be either indefinite, or be given an expiry time. When that expiry time is reached, the prot level drops right back to "none", regardless of the prot history. So, if an article has indef semi prot for editing, and is then changed to have a fixed-duration full prot (for editing), then when the expiry is reached it doesn't return to semi-protected but becomes unprotected (for editing). I'm sure this has come up at VPT before, if so, there may be a phab: ticket for it. --Redrose64 (talk) 09:02, 19 April 2016 (UTC)[reply]
@Nyttend: - Would you like a User Script that reminds the user when temporary protections they've added are about to expire? A stop-gap measure, but probably helpful. fredgandt 23:30, 23 April 2016 (UTC)[reply]
By "any tool", I was envisioning either a private user script or a more generally used feature (a bot, perhaps?) that lots pages that are about to become unprotected because of a modification in protection state. It would be slightly useful to me (not hugely, because I don't do much with page protection), but I can envision it being lots more significant to folks who hang out at WP:RFPP all the time. Nyttend (talk) 04:03, 24 April 2016 (UTC)[reply]

No editing

You've probably already seen this, but just a small reminder: No editing for ~30 minutes starts in less than two hours from now. You can read more on Meta. /Johan (WMF) (talk) 12:26, 19 April 2016 (UTC)[reply]

Editing resumed after about 47 minutes, but first attempt to delete a page gave the following:

[VxZFZQrAAEcAAGB3Z8kAAAAK] 2016-04-19 14:49:11: Fatal exception of type JobQueueError

Not good... Fram (talk) 14:50, 19 April 2016 (UTC)[reply]

Me too, Johan (WMF). --Malcolmxl5 (talk) 14:58, 19 April 2016 (UTC)[reply]
It seems OK now. --Malcolmxl5 (talk) 15:26, 19 April 2016 (UTC)[reply]
I'm guessing this was probably related to the cause of phab:T133053; the job queue backs features like recent changes and the watchlist, which were missing entries from the time you encountered this error. --Dan Garry, Wikimedia Foundation (talk) 16:25, 19 April 2016 (UTC)[reply]
For the record, no editing was from Special:Diff/716040655 (14:02 UTC) to Special:Diff/716040656 (14:48 UTC). PrimeHunter (talk) 15:36, 19 April 2016 (UTC)[reply]
[barnstar of technical cleverness] for digging up those links:) DMacks (talk) 16:02, 19 April 2016 (UTC)[reply]

watchlist, recent changes list missing some entries

Resolved
 – The missing data has been backfilled. This issue is now resolved. --Dan Garry, Wikimedia Foundation (talk) 22:58, 19 April 2016 (UTC)[reply]
Tracked in Phabricator
Task T133053

The history page for Help talk:Citation Style 1, here, shows that I made edits to that page at 15:03 UTC and 15:09 UTC. This page is on my watch list but niether of those edits appear there. When I looked at Special:RecentChanges, filtered for the Help talk namespace, my edits are not listed.

It appears that someone broke something in the recent server move.

—Trappist the monk (talk) 15:21, 19 April 2016 (UTC)[reply]

Please tell me I'm not the only one on here who tried to use Listen to Wikipedia to figure out when the site would be back up, and was rather confused when edits were being allowed but that site was silent! :-) Oh well, these things happen. I'll try the same procedure out tomorrow. Graham87 07:40, 20 April 2016 (UTC)[reply]
I understand that "Listen to Wikipedia" depends upon Special:RecentChanges and Wikimon working. Special:RecentChanges had some difficulties yesterday for about 20 minutes. As of three hours ago (and subject to change), the Ops team seemed to think that there would be a short delay for Special:RecentChanges – a big improvement, but not complete success.
There is, as always, a chance that they'll decide to postpone the planned switch back (to the regular servers in Virginia). Even things outside their control, such as a power outage at the last second, could scuttle the whole thing. However, as of this moment, everything is proceeding under the assumption that the switch back will happen as planned. Whatamidoing (WMF) (talk) 19:04, 20 April 2016 (UTC)[reply]

About the "delete" function

Just wondering, is it technically possible to allow the "delete" function to be enabled for only (a) specific namespace(s)? Im thinking of a new user group that may include something like this, but all I really want to know right now is if it is technically possible. Steel1943 (talk) 20:25, 19 April 2016 (UTC)[reply]

Almost anything is technically possible. The better question is always: how much time, energy and money are you willing to feed the problem with. If the question is, is this already implemented in MediaWiki, then the answer is no. There is only one user right for page deletion and it applies to all namespaces. —TheDJ (talk • contribs) 22:04, 19 April 2016 (UTC)[reply]

Shared tabular data storage for Commons!

CCing from commons, please participate there as this will be a cross-wiki shared feature

Tracked in Phabricator
Task T120452

During the last hackathon I created a new on-wiki tabular storage described in T120452, similar to CSV and TSV formats. It allows any user to create a page, e.g. "Data:List of interesting facts.tabular" (demo), and keep it as a table, rather than wiki text. Tabular storage allows strings, numbers, Booleans (true/false), and "localized strings" – a string that has different value depending on the language. Additionally, tabular data stores metadata, such as description (localized) and license. More metadata can be added as needed.

Tabular storage greatly simplifies storing data for lists, tables, and graphs. Graphs may directly access tabular data, and on-wiki tables and lists can be created by using simple Lua scripts. This storage is fundamentally different from Wikidata, because it works with "blobs" (batches) of data, whereas Wikidata works with tiny "facts". Wikidata technology is simply not suited for large storage such as the list of the most expensive paintings, the shoe size comparisons table, or data to plot Moscow subway growth graph.

After a long discussion, it seems Commons is the best fit for such data. Commons community already has good experience with international multi-licensed content. The current proposal is to create the data namespace on Commons, and use it from all of the wikis.

Feel free to experiment with it at http://data.wmflabs.org/wiki/Data:Sample.tabular. Note that you can view it with different languages, e.g. http://data.wmflabs.org/wiki/Data:Sample.tabular?uselang=fr

Technical notes: When storing, the data is validated and stored as JSON, so there are no delimeter problems common to the traditional CSV/TSV files. At this point, the wiki editor shows tabular data as a JSON, but very soon I hope to have a CSV/TSV editor to simplify copy/pasting, and afterwards – a full scale spreadsheet table editor. Eventually, I would also like to implement Q number support, allowing direct links to Wikidata. --Yurik (talk) 20:54, 19 April 2016 (UTC)[reply]

Very interesting, thanks. In case you have posted this elsewhere, I fixed the commons link in the first line. Johnuniq (talk) 02:00, 20 April 2016 (UTC)[reply]

More than usually buggy

Has anyone else noticed Wikipedia being more than usually buggy since the planned outage earlier today? I am getting more frequent edit-conflicts-with-self, database locked for maintenance notices, and Wikimedia down notices. DuncanHill (talk) 21:19, 19 April 2016 (UTC)[reply]

I note that the elements in my edit window are very tiny. I have a 4K monitor, so my size setting is at 200%, but the elements and the text in my edit window looks like it's running 100%. Other than that, no.—cyberpowerChat:Online 21:36, 19 April 2016 (UTC)[reply]
I'm getting some lag beyond what I usually experience every now and then, but other than that, no, no issues for me. AddWittyNameHere (talk) 21:40, 19 April 2016 (UTC)[reply]
I have had some database locked messages. Keith D (talk) 21:42, 19 April 2016 (UTC)[reply]
I seem to be getting the database notice on every page I've edited so far, Other than that It all seems fine to me. –Davey2010Talk 21:43, 19 April 2016 (UTC)[reply]

Receiving wikipedia emails 6 hours late

So I just got a Wikipedia email alerting me to a post on my talk page from another user, only the edit was made 6.5 hours ago by MarnettD.—cyberpowerChat:Online 22:18, 19 April 2016 (UTC)[reply]

I think that some of the logs entries were lost and re-added. For example, my notifications show two entries "Pegship‬ thanked you for your edit on Wikipedia talk:WikiProject Stub sorting‬", one a few minutes ago, the other eight hours ago (which I noticed this morning). They both relate to this edit, but there is only one entry in the thanks log - the one that should be there from eight hours ago is missing. --Redrose64 (talk) 23:04, 19 April 2016 (UTC)[reply]
Got two more emails. 3 hours gap.—cyberpowerChat:Online 23:26, 19 April 2016 (UTC)[reply]
The mail server in Carrollton wasn't set up quite right, so email broke for a few hours after the datacentre switch. You can see the server's queue graph at [1]. The peak of the graph is roughly when someone reported it and ops fixed it. It should have been working normally since that went back down to 0. --Krenair (talk • contribs) 14:17, 20 April 2016 (UTC)[reply]

Unable to download PDF files of articles

Tracked in Phabricator
Task T133136
Resolved

Following this on the help desk, I attempted to download a PDF of an article myself, only to get an error that states that the file cannot be found. This also appears to line up with the recent updates, so I decided to list this here as well. -- The Voidwalker Discuss 00:23, 20 April 2016 (UTC)[reply]

Hello, thanks for reporting this issue. I could confirm it and created T133136 on Phabricator. The issue is being investigated by Operations right now. DZahn (WMF) (talk) 05:22, 20 April 2016 (UTC)[reply]

Update: Also see this and this. DZahn (WMF) (talk) 05:32, 20 April 2016 (UTC)[reply]

Update 2: Please try again. Issue should be fixed now. DZahn (WMF) (talk) 05:54, 20 April 2016 (UTC)[reply]

Thanks, I've left a note at the Help desk that this is fixed. -- The Voidwalker Discuss 20:18, 20 April 2016 (UTC)[reply]

Server transfer went too long

It was originally announced that the migration of servers to Dallas would take only 15 to 30 minutes, however, it took longer than expected, and it started a bit late. Rather precise, the transfer began at 15:03 BST and would end around more than 40 minutes later. In addition, there were cases when there was a server difficulty and we were shown an error page which can allow us to search for archived mirrors by using Google. I wonder what went wrong during the lost time? Has WMF issued any statement regarding what happened? 49.148.29.21 (talk) 00:33, 20 April 2016 (UTC)[reply]

Things never go to plan. It started 3 minutes late, and ended 10 minutes late. For a server move that's pretty good. Nothing to worry about here. Also since the wiki is being moved it's reasonable to assume that site errors where going to crop up as the domain points to a new physical location.—cyberpowerChat:Online 01:06, 20 April 2016 (UTC)[reply]
@49.148.29.21: Sounds like you're just curious about how server administration works here. Fortunately, #wikimedia-operations is a public channel, so you can look back on all the fun stuff that happened during the switchover. Starts at 1400. Mamyles (talk) 19:13, 20 April 2016 (UTC)[reply]
Usually, Ops posts reports on outages at wikitech:Incident documentation. In this case, they will probably do something bigger. But I wouldn't expect them to write up anything until at least next week. They're busy getting ready to switch back to the Virginia servers right now (same time; watch wikitech:Switch Datacenter#Schedule for Q3 FY2015-2016 rollout for possible changes). Whatamidoing (WMF) (talk) 23:22, 20 April 2016 (UTC)[reply]

Strange problem with Template:Fb rs

See the discussion at Wikipedia talk:WikiProject Football#{{Fb rs}}. For some reason, {{Fb rs}} now displays positive overall and away goal differentials above the wikitable. I have tested a solution which appears to work in my sandbox (code at User:Jkudlick/sandbox/Fb rs), but when I applied it, it didn't fix the error and actually started making negative goal differentials in some articles disappear. I reverted all changes I made to {{Fb rs}} and {{Fb gd}}, and we need someone with more template experience than I have to take a look and see if there is an explanation. — Jkudlick • t • c • s 04:18, 20 April 2016 (UTC)[reply]

The error was solved by another editor. The removal of {{Unicode}} removed a leading non-breaking space from {{Fb gd}} which was required for proper functionality. A &nbsp; was hardcoded and the error was fixed. — Jkudlick • t • c • s 04:25, 22 April 2016 (UTC)[reply]

Can't save edits from the default editor?

I originally posted this to the help desk, it was recommended that I ask here.

I can't save changes made in the default editor. When I click the "Save page" button it just reloads the edit page with a message to sign in (I have not tried signing in, this should be unnecessary). My IP does not appear to be blocked, and if I use the visual editor my changes save correctly. This is not a new problem, it's affected me for at least a few months.

I'm using Chrome on Windows 7. I have tried with IE11 and it works fine. I can also edit at work, from Chrome on Linux (what I'm doing right now). — Preceding unsigned comment added by 2620:0:1019:13:7009:5270:E0D7:8D0D (talk) 04:28, 20 April 2016 (UTC)[reply]


AFD link on article talk page fails

At Talk:Emoji there is a link which should lead to an AFD debate regarding an article.It says "💮 was nominated for deletion. The debate was closed on 05 March 2013 with a consensus to merge. Its contents were merged into Emoji. The original page is now a redirect to here. For the contribution history and old versions of the redirected article, please see its history; for its talk page, see here." It is part of a template. Instead of leading to the old AFD debate, clicking on "the debate" directs the reader to the current set of AFDs being debated. The template is there, and the (former) name of the article (just an emoji symbol, since redirected to Emoji) is included. Is there a problem with the template working in general? Does it not work when the article title was just one emoji symbol?Edison (talk) 13:58, 20 April 2016 (UTC)[reply]

A required parameter was removed in [2]. I have restored it.[3] PrimeHunter (talk) 19:01, 20 April 2016 (UTC)[reply]
Thanks for the prompt corrective action. Edison (talk) 04:00, 21 April 2016 (UTC)[reply]

Strange file usage problem

If you take a look at File:Pulling Strings movie poster.jpg, you will find that the file usage section reports that the file is in use on the page Pulling Strings (film), and only on that page. On the other hand, if you ask the database where the file is used (see quarry:query/9163), then the database reports that the file is used on page #50035842 (Pulling Strings (film)) and additionally on page #50084289 (doesn't exist, the query reports it as being in the draft namespace), and as a result, it looks as if the file is violating WP:NFCC#9. What's causing this and what exactly is page #50084289? --Stefan2 (talk) 15:00, 20 April 2016 (UTC)[reply]

Server changeover - API delays - user contributions

Since yesterday's server changeover I've been suffering intermittent delays (of many seconds in some cases) when requesting lists of user contributions via the API – for instance when using Popups and hovering over a user's contribs entry in a page history (try it on this page's). The delays particularly affect accounts with many contribs. Until yesterday the lists always used to appear pretty much instantaneously. Is anyone else affected?  —SMALLJIM  15:33, 20 April 2016 (UTC)[reply]

I don't know if it's related, but I'll pass your observation along. Thanks. Whatamidoing (WMF) (talk) 23:23, 20 April 2016 (UTC)[reply]
Thanks Whatamidoing (WMF). I've tried Popups on different PCs and using two different accounts, and also calling the API via a script, and I still get the same problem – a long delay the first time I call API:usercontribs on a user with a long contribs record. The delays are variable, but often more than 30 seconds. It doesn't happen on other API calls (API:Revisions is fine, for example). I can't see how this could be a problem at this end.
I note that since the changeover to the backup server, ClueBot NG has been up and down too – this may be related because I expect that bot would look up usercontribs via API too: I reported this here. I'd expect it to affect Huggle and other AV tools too.
Will someone else check to see if they see the same delays please (try Popups, see above); if it's not just me this may indicate something that needs fixing on the new server. And if it is just me, then...why?! Thanks.  —SMALLJIM  10:00, 21 April 2016 (UTC)[reply]

I get a probably related issue when I refresh my watchlist: "Changes newer than 13 seconds may not appear in this list." The last time I saw such a message here is years ago. Fram (talk) 10:07, 21 April 2016 (UTC)[reply]

I get that every few weeks. 13 seconds is insignificant; delays like these are nothing to worry about until they exceed 10-20 minutes. --Redrose64 (talk) 12:36, 21 April 2016 (UTC)[reply]
I'm not sure that would be related, Fram & Redrose64. API delays of a few seconds can be pretty disastrous. Anyway, as of now it's back to responding almost instantaneously, so thanks to whoever fixed it :)  —SMALLJIM  15:57, 21 April 2016 (UTC)[reply]
Ah, I see we've been transferred back to the old server (I really shouldn't leave the computer and have a life!). So whatever was causing the delays may not have been fixed after all.  —SMALLJIM  16:05, 21 April 2016 (UTC)[reply]
@Smalljim - Could you provide the full URL/API call options where you had the issues and the exact times in UTC at which those happened? There are 2 possibilities here- some calls are known to be more expensive than others, and workarounds on the new servers may not yet exist on the current ones. Another possibility is that it is a transitional problem- as this was the first time the new databases were used, despite the efforts to warmup the database, more rows than usual may have to be read from disk, which, combined with more than usual stress, they can lead to more error rates than usual. Time fixes the second option, but the first must be researched. If you could provide the full URL and the time, we will be able to improve the performance of those calls and hopefully fix them. Thank you. --jynus (talk) 10:45, 23 April 2016 (UTC)[reply]
@Jynus and Whatamidoing (WMF): I think it's a shame that no-one else here was able to take a minute to check if they were also affected. I can't provide the level of detail that you request, but I know that I used Popups to look up the contribs of numerous recent editors of this page (including myself) around the time that I made the above reports. Further checks on the same editor made soon after the first were typically much faster, but after some time (or additional edits made?) were slow again. It may be worth talking to User:DamianZaremba, because ClueBot NG was clearly unhappy running through the backup server too (see User talk:DamianZaremba).  —SMALLJIM  15:33, 23 April 2016 (UTC)[reply]

Strange image behavior

Pilsner glass from Brauerei Schloss Eggenberg
Pilsner glass from Brauerei Schloss Eggenberg

I have never seen this one before. These two images are identical except that the one on the right has an "upright" parameter. (The one on the left has a "left" parameter but that makes no difference.) On my screen anyway (Firefox) the one on the right renders sideways. How did that happen? Kendall-K1 (talk) 19:19, 20 April 2016 (UTC)[reply]

The 220px one is rotated for me, while the 170px one looks correct. This is a problem which sometimes happens. Image thumbnails are cached and sometimes the server forgets to update old thumbnails when the file is overwritten, but the problem tends to go away if you wait for a few days or months. The solution is to use a different file size. 220px is a very common thumbnail size, and many images are therefore cached in that thumbnail size. Try using [[File:Schloss Eggenberg Hopfenkönig.jpg|221px|thumb]] or something to force a different size, and remember to remove the page size whenever MediaWiki has corrected the 220px thumbnail. --Stefan2 (talk) 21:54, 20 April 2016 (UTC)[reply]

Regarding connecting to wiki in other languages

I assume the content in wiki is free, so I'm building an android app that shows wiki content.

I show random pages by using Java to connect to "https://en.wikipedia.org/wiki/Special:Random"

However, going to "https://he.wikipedia.org/wiki/Special:Random" doesn't work!

I only use a simple code to connect: 

  URL url = new URL(site);
  HttpURLConnection connection = (HttpURLConnection) url.openConnection();
  connection.setRequestMethod("GET");
  connection.setRequestProperty("Accept", "application/xml");
  (InputStream) connection.getContent()

but connection.getContent() is null in Hebrew, but not in English. Why?

Also, I have many other connection problems when I try to get data. Do you have a mechanism that says "Person 'X' can get 'Y' bytes of data, and other requests are rejected"? רן כהן (talk) 07:16, 21 April 2016 (UTC)[reply]

Going to https://he.wikipedia.org/wiki/Special:Random works for me, as does he:Special:Random. Using those links, I've gone to about six different articles; I can't read Hebrew, so I don't know what they were about, but the pictures were different so they must have been different pages. --Redrose64 (talk) 12:39, 21 April 2016 (UTC)[reply]
Is there any limit on the number of calls I can make to the wiki server? רן כהן (talk) 12:59, 21 April 2016 (UTC)[reply]
Please always set a user-agent. Also, why are you requesting xml and not "*/*", as to why you have connection problems.. That I'm not sure.. —TheDJ (talk • contribs) 14:59, 21 April 2016 (UTC)[reply]
I have a small java class that does what is required... I ask xml because its easy to get...
So tell me - what is my limit? How many bytes per second? רן כהן (talk) 06:43, 22 April 2016 (UTC)[reply]
Use the API. As long as it is single threaded, you should be fine throttle-wise. MER-C 08:18, 23 April 2016 (UTC)[reply]

Template:Col-2

There is bug in where {{Col-2}} is used. Bare code | style="width:50%; text-align:left; vertical-align:top;" | displays in articles. Eg: River Cam or style="width:50%; text-align:left; vertical-align:top;". --AntanO 11:10, 21 April 2016 (UTC)[reply]

To use {{Col-2}}, you need to use {{Col-begin}} and {{Col-end}} as explained on the documentation page. On that page, however, I think you actually want {{Refbegin|30em}}. Relentlessly (talk) 11:27, 21 April 2016 (UTC)[reply]
@Relentlessly: They're not references. @AntanO: It's best to avoid manual column breaks, for accessibility reasons. In this case you want a normal unordered (bulleted) list to show as multiple columns, for which the {{div col}}/{{div col end}} construct is eminently suitable, like this. --Redrose64 (talk) 12:59, 21 April 2016 (UTC)[reply]
Thanks @Redrose64:. Usage of {{div col|small=yes|colwidth=35em}} and {{div col end}} look good. I think we need to cleanup these style="width:50%; text-align:left; vertical-align:top;". --AntanO 13:11, 21 April 2016 (UTC)[reply]
Redrose64, I suggested it only because it was already being half-used. Relentlessly (talk) 15:54, 21 April 2016 (UTC)[reply]

Category:Wales railway station stubs

Hi, can somebody create me a list organized by county? ♦ Dr. Blofeld 11:23, 21 April 2016 (UTC)[reply]

Which flavour of county, Dr. Blofeld? PetScan looks like it may cooperate, to the extent that the stubs have sane cats. (And, note, many of the Welsh railway stubs are by now far beyond stubs & need de-stubbing. I'll not touch them until after the 1st May.) --Tagishsimon (talk) 13:33, 21 April 2016 (UTC)[reply]
By modern county I think, they can then be sifted into preserved counties. Yes, some of them meet length requirements but destubbing is more than just about length, the information has to all be sourced. Not a problem, plenty of SSSI stubs to expand ;-)♦ Dr. Blofeld 14:21, 21 April 2016 (UTC)[reply]
Yes. I was going to slip them in. I'll have a play with rail stations tonight - seven or so hours hence. --Tagishsimon (talk) 14:51, 21 April 2016 (UTC)[reply]
Dr. Blofeld Full list on my talk page. --Tagishsimon (talk) 23:56, 21 April 2016 (UTC)[reply]
@Tagishsimon: Your problem was that you included level 2 headings in the collapsed section. I would have done it as nested lists thus:
I was also preparing a list, it's not yet finished, but the above shows how I was doing it. --Redrose64 (talk) 00:41, 22 April 2016 (UTC)[reply]
Yeah, that'd work, thanks. Bit new to this collapsing lark. --Tagishsimon (talk) 01:00, 22 April 2016 (UTC)[reply]
Full list

Finding the most viewed Wikipedia articles on education

Hi all

I'm trying to use TreeViews to get information on what are the most viewed articles in Category:Education, unfortunately such large categories just crash my browser, it means I will have to split the query up into at least 50-100 smaller queries.

Does anyone know of a less manual way around this? Ideally the output would be spreadsheet of the article title and the number of page views of the article for a 30, 60 or 90 period in the recent past. I will use Treeviews if it is the only way but I'd really love to save myself from half a day of data entry. I imagine this would also be useful for people working with other organisations for other subjects.

Thanks

John Cummings (talk) 12:12, 21 April 2016 (UTC)[reply]

@John Cummings: If the set of articles bearing the {{WikiProject Education}} template is a close enough approximation to the set of articles in Category:Education, then you might find Wikipedia:WikiProject Education/Popular pages useful. --Worldbruce (talk) 01:41, 22 April 2016 (UTC)[reply]
@Worldbruce:, thanks very much, its not exactly what I was looking for but is definitely a good proxy, thanks again. John Cummings (talk) 08:17, 22 April 2016 (UTC)[reply]

Earth to JDForrester

Wikipedia:Village pump (technical)/Archive 145#Do you want one Edit tab, or two? It's your choice, and especially the subsection "VE was imposed as primary editor", have been archived. User:Alsee was told off last week because he wasn't patient enough and User:Jdforrester (WMF) had no time then to address this, but his interpretation that nothing would be forthcoming the week after that comment (the 14th) was completely misguided.

Sure enough, it's the 21st, and the product manager, who promised before the implementation that VE would not be the primary editor on enwiki, has still not responded anywhere about this. He has edited a lot of other things, and must have seen the messages on his talk page, but apparently acknowledging that something seems to have gone wrong and that you will look into it (preferably with some timeline) is too hard. Easier to just let it archive and forget about it. If only we had some people at the WMF who could drop you a note at your talk page there, some WMF-community intermediaries or something similar. Oh well, we can always dream... Fram (talk) 12:12, 21 April 2016 (UTC)[reply]

I left a post for the Executive Director noting that we were assured that this was not going to happen, and that we have been unable to get any response on it. Alsee (talk) 01:51, 24 April 2016 (UTC)[reply]

Going back again

It's server transfer again, but this time around it's going to be a lot more overloaded, as not only MediaWiki will be returning from Dallas, but also everything else that went ahead of it—all to go side by side. I'm rather anxious, as I'm expecting a longer transfer this time around. (Will it?) I'm not also exactly sure whether the technical staff has done anything about Tuesday's issues, or if they will work on them after the transfer. IRC looks rather too technical to non-programmers like me, but I'll try to take another look there. 49.148.27.180 (talk) 13:15, 21 April 2016 (UTC)[reply]

Today's transfer took much faster, around 20 minutes to be precise. I also like the banner on technical maintenance, but it seems like it's appearing even after it's done. I think I thought too much on the transfer time. 49.148.27.180 (talk) 14:25, 21 April 2016 (UTC)[reply]
They actually were talking about a plan that would be slower, but which might reduce a couple of cacheing problems and maybe Special:RecentChanges. But it looks like they found faster ways to address those problems.
The banners get turned off manually, so we expect them to run for a few minutes longer than necessary. I'm glad that you liked them. Whatamidoing (WMF) (talk) 01:04, 24 April 2016 (UTC)[reply]

Wikipedia to the Moon

Hello! Sorry that this is in English only, but we are using village pump messaging in order to reach as many language communities as possible. Wrong page? Please fix it here.

This is an invitation to all Wikipedians: Wikimedia Deutschland has been given data space to include Wikipedia content in an upcoming mission to the Moon. (No joke!) We have launched a community discussion about how to do that, because we feel that this is for the global community of editors. Please, join the discussion on Meta-Wiki (and translate this invitation to your language community)! Best, Moon team at Wikimedia Deutschland 15:35, 21 April 2016 (UTC)

Category:Song articles with missing songwriters

There appears to be a bot adding articles to this category without reason. For example A Day in the Life and Billie Jean. Should be less than 400 members in the cat. --Richhoncho (talk) 17:19, 21 April 2016 (UTC)[reply]

@Richhoncho: See Template talk:Infobox song#Missing songwriters. There is no bot. --Redrose64 (talk) 17:58, 21 April 2016 (UTC)[reply]
@Redrose64:. Trouble is the two examples I gave above shouldn't have been added and there is way of removing the additions, they are not there, and, more strangely, appear in the history of the articles before the cat was created! --Richhoncho (talk) 18:01, 21 April 2016 (UTC)[reply]
Yes, they will do, that is because the change was in {{Infobox song}}. Previous versions of articles always transclude the current version of a template. --Redrose64 (talk) 18:04, 21 April 2016 (UTC)[reply]
OK, there are articles being added when songwriters are in the infobox, (i.e. A Day in the Life and Billie Jean, as mentioned above) but I have taken that back to Missing Songwriters. --Richhoncho (talk) 18:17, 21 April 2016 (UTC)[reply]
A Day in the Life has two infoboxes, {{Infobox song}} and {{Infobox single}}. Both of them have the ability to put pages in Category:Song articles with missing songwriters, the relevant change for the latter infobox is this. Both changes are covered by Template talk:Infobox song#Missing songwriters, which I mentioned earlier. --Redrose64 (talk) 18:54, 21 April 2016 (UTC)[reply]

Page viewer overload

I was just wondering: I tried to read Prince (musician) a few minutes ago, and ran across this error text:

Sorry, the servers are overloaded at the moment.

Too many users are trying to view this page. Please wait a while before you try to access this page again.

Pool queue is full

I completely understand the reason why this is happening: A lot of readers are trying to view this page since the subject of the article was just reported to pass away. But my question is: Does anyone here know how many viewers are "too many users" when it comes to the amount of people trying to view a page at once? Just wonderful since I've never seen this error message before. Steel1943 (talk) 17:19, 21 April 2016 (UTC)[reply]

  • I think it happened with Michael Jackson too, a few years ago. I sense a pattern. --Redrose64 (talk) 18:01, 21 April 2016 (UTC)[reply]
    • @Redrose64: The coverage of Michael Jackson's death took down all WMF wikis for a few hours. After that incident, PoolCounter (described by Deskana below) was put in place to prevent such an incident from taking down the entire site. What User:Steel1943 saw was PoolCounter doing what it was designed to do: breaking the Prince article to protect the rest of the site. The problem is caused specifically by lots of people trying to read the article right after it's been edited, and so PoolCounter only limits the number of viewers right after an edit. But when there's an edit every 10-20 seconds, "right after an edit" is all the time. Now that the edit activity has calmed down a bit (one edit every few minutes now, compared to 4-6 edits per minute when you posted your question), you shouldn't see this issue any more. The servers are able to handle many people viewing the same article (Main Page gets that every day), as long as it's not being edited at a frantic pace. --Roan Kattouw (WMF) (talk) 18:54, 21 April 2016 (UTC)[reply]
  • @Steel1943: Good question! There's no single number for how many people is too many; it'll depend on a variety of different, complex factors. The error you ran in to was caused by the PoolCounter, which attempts to mitigate server strain when an article is edited faster than the servers can handle. Now that editing has slowed down, this problem should hopefully not happen again. Hope that helps. --Dan Garry, Wikimedia Foundation (talk) 18:21, 21 April 2016 (UTC)[reply]

Community input requested for development of an informational template

I just created {{subst:wiki tech info}} (documentation to follow development) to make the job of passing on a basic where-to-start guide to the technical aspects of MediaWiki sites easier.

I'd like to actively invite you (yes you) to assist in fleshing it out, so the next time you'd like to help someone get started with this stuff, you can rest assured that this template provides a good basic roundup of key links and info.

I hope you'll see by the current content, the aim is not to overwhelm or baffle, but simply to collect together relevant resources across the sister sites in a non-scary overview of possibilities.

Thanks. fredgandt 22:43, 21 April 2016 (UTC)[reply]

Wouldn't Help:Wiki_markup fulfill that need?--v/r - TP 07:05, 22 April 2016 (UTC)[reply]
Is that some kind of user template (=presumably only you need it) or Wikipedia essay in the template namespace? I'm a fan of {{Sofixit}} (as template, and as "not yet" policy), but you could also transclude a project page or subst: a user subpage.
Maybe start with the spec. (/doc or inline), depending on the spec. it could be implemented in other ways, e.g., as {{welcome/tech}} subpage. – Be..anyone 💩 07:18, 22 April 2016 (UTC)[reply]
TParis - Wiki markup is one of the linked resources. The idea is to be able to quickly offer a linked overview of technical capabilities and where to find further info about them, not only a how to compose pages.
Be..anyone - Your presumption is hopefully wrong; standardized messages assist editors by simplifying the process of passing information to others. There's no reason to assume that I am the only editor willing to help others find their feet. No it's not an essay or anything to do with policy, and although I could indeed subst the information from a subpage of my user space, this creation is not only for me to use. Why implement it in other ways? A standardized message in the template space is entirely proper. fredgandt 09:34, 22 April 2016 (UTC)[reply]
That was a question, not an assumption. I wouldn't use it as is, or presumably never. Sometimes I link (not transclude) sofixit, sometimes I use welcome (subst by bot), and for a tech info I suggest to add m:Tech/News and/or VP/T. –Be..anyone 💩 09:44, 22 April 2016 (UTC)[reply]

Twinkle AFD and redirects issue

I used Twinkle to nominate Reactions to the death of Prince for deletion. While I was typing up my rationale, the article was turned into a redirect to Prince (musician) ([4]). This caused Twinkle to put the AFD notice on Prince's BLP ([5]). Not sure if this qualifies as a "bug", but it certainly could use fixing. Not sure if this is the right place (and if not, please direct me to where I should raise this issue). Please ping me if you reply. Thanks! EvergreenFir (talk) Please {{re}} 05:34, 22 April 2016 (UTC)[reply]

Drumpfinator extension for chrome and unintentional vandalism

Not sure if this should go here. It's not a technical problem--a heads up really. John Oliver popularized a chrome extensions which changes Trump to Drumpf if viewed on a website. https://chrome.google.com/webstore/detail/drumpfinator/hcimhbfpiofdihhdnofbdlhjcmjopilp?hl=en I made the mistake of having this on while I asked a question on the main help desk page. Evidently when I clicked "edit source" Drumpf was brought in rather than Trump because that was what my browser was displaying at the time. Oops. I have since removed the extension. — Preceding unsigned comment added by AugusteBlanqui (talk • contribs) 09:56, 22 April 2016 (UTC)[reply]

I've come across this, and there has been an ANI thread about it. The scary thing is that Trump is changed to Drumpf on the entire page you're innocently editing. The whole of ANI, for instance. Bishonen | talk 20:16, 22 April 2016 (UTC).[reply]

Range contributions for IPv6?

X!s Tools range contributions used to give range contributions for both IPv4 and IPv6 ranges, but the last few weeks it hasn't been doing IPv6 — I never get any results (even though I know of one or two edits from the range I'm asking about). Does anybody know of another tool? Or can I report the problem anywhere? I'm getting quite frustrated, and IPv6 vandals are getting away with murder. (Please don't tell me to use Phabricator, I can't handle that. I'm only a little old lady. Use words of one syllable.) Bishonen | talk 20:16, 22 April 2016 (UTC).[reply]

Sadly the author (or one of the authors) was blocked here, and therefore his RfC on Meta about abandoned tools is doomed. Here's a good place to report the issue. –Be..anyone 💩 20:34, 22 April 2016 (UTC)[reply]
@Bishonen: Are you sure it was X!'s tools which I seem to recall only ever handled IPv4. Someone pointed out that enabling the "Allow /16, /24 and /27 – /32 CIDR ranges on Special:Contributions forms..." gadget in preferences allowed a workaround where you can enter * to cover a range of IPv6 addresses. Does that work? I tried it once and it worked well, and I'll have another look if you like. Johnuniq (talk) 23:35, 22 April 2016 (UTC)[reply]
It's late for me, so I'm a bit blurry and request you bare that in mind if I'm missing the point, but just tried two IPv6 chosen randomly from recent changes. The first came up empty, but its only contribs were only minutes old. I tried another with contribs hours old, and the results came through. At the top of the tool's page, is a note stating "Caution: Replication lag is high, changes newer than [variable amount of time] may not be shown.". Could this be something to do with the problem you're having? fredgandt 00:14, 23 April 2016 (UTC)[reply]
I think I eventually got the range format correct (referring to mw:Help:Range blocks/IPv6#How to calculate an IPv6 CIDR range?), and sure enough, no results for ranges with edits that do get returned if adding the full IPv6. For a little old lady you're doing some very technical things!  fredgandt 00:50, 23 April 2016 (UTC)[reply]
I found 2605:A000:1200:600F:BDC2:282A:6C52:766B as an example. As you say, the full IP works, but 2605:A000:1200:600F:0:0:0:0/64 shows nothing. However, if scripting is enabled in the browser, and if the gadget I mentioned is enabled in Special:Preferences, the normal contribs page works after entering 2605:A000:1200:600F* Johnuniq (talk) 00:59, 23 April 2016 (UTC)[reply]
John, I do believe you're right. It was probably the prefs gadget that I used before, not X!'s tool. Sorry, confused little old lady. And then I probably repressed the gadget thing because it's so unfriendly and I'd so much rather use something simpler. (What does it even mean by talking about a "prefix", where's the prefix? Pre what?) Anyway. Yes, as you say, chopping off the IP you have after the first four groups and adding an asterisk works — after a fashion — it worked the second time I tried it with my own vandal, 2607:FB90:A048:8865:CE4C:BE49:E71E:EB4C — showing only the one princely contribution I already knew about, but still, that's working. The other suggested alternative, "enter a CIDR range", has never worked for me. Presumably it requires some exotic form of the range, such as what you get from the very frightening help page you linked to, Fred Gandt, and not the kind of barefoot range I know to make, without any multiplication etc, which is in this case 2607:FB90:A048:8865::/64. RexxS told me about that, and my programmer son (actually my great grandson… just kidding), and that form seems to work allright for blocking purposes. Sigh. Thank you, guys. I'll stumble on. Bishonen | talk 10:20, 23 April 2016 (UTC).[reply]
@Bishonen: You need a CIDR prefix like 2607:FB90:A048:8865::/64 when blocking a range, but such a CIDR prefix does not work in the Special:Contributions gadget for IPv6 (the gadget accepts CIDR prefixes for IPv4 only). I will update {{blockcalc}} so it produces links to IPv6 contributions as best it can, assuming the person using blockcalc has the gadget enabled. I'll let you know when/if I get it done (should be easy). Johnuniq (talk) 10:30, 23 April 2016 (UTC)[reply]
@Johnuniq: But where's the prefix? A prefix is something that goes before something. Does it have an extended definition in IT world — sometimes used to mean postfix or suffix, such as the "/64"? Of course it doesn't matter, it just confused me when the gadget talked like that. It would be great if you got the calculator to understand IPv6 ranges, thank you. ("Easy"? How can it be easy?) Bishonen | talk 10:54, 23 April 2016 (UTC).[reply]
@Bishonen: I know you're not really wanting a rant on CIDR but I can't resist something brief. Consider the IPv4 address 11.22.33.44. If that address and others nearby (say 11.22.33.49 and 11.22.33.56 and so on) needed to be blocked, you might block 11.22.33.0/24. That is called a prefix because it is the stuff at the front of each IP. The /24 indicates how many bits are in the prefix. Since an IPv4 address is 32 bits, a prefix of /24 means there are 8 bits remaining which have to be zero for a valid CIDR prefix. The remaining bits might be called the suffix but CIDR instead refers to the right-most 8 bits as the "host bits" because they identify a particular computer or "host" on a network. In this example, host number 44 is on network 11.22.33.0/24. Johnuniq (talk) 11:08, 23 April 2016 (UTC)[reply]
It's at times like this that my mind turns to pudding. I like pudding. fredgandt 12:06, 23 April 2016 (UTC)[reply]

Johann Sebastian Bach

When I save an edit at Johann Sebastian Bach the page jumps around all weird, and I wondered if it was something that needs fixed. I asked about this at the help desk, where someone suggested it might be my internet connection, but mine's among the faster ones available, so I don't think the issue is on my end. Kirk Leonard (talk) 20:32, 22 April 2016 (UTC)[reply]

Chromatic Fantasia and Fugue BWV 903 performed by Kevin MacLeod
1. Fantasia
2. Fugue
Bach re-interpreting older genres tied to the modal system
I very much think that it is the JavaScript for the {{listen}} templates kicking in one by one. It causes the enclosing boxes to expand and contract until they finally settle down. --Redrose64 (talk) 22:38, 22 April 2016 (UTC)[reply]
That's exactly what happens to me. Is there any fixing it, or is it just the way that template acts? Kirk Leonard (talk) 22:45, 22 April 2016 (UTC)[reply]
The template code is at Module:Listen but that is not going to be the whole story by a long way. There is JavaScript as well, which I cannot even begin to track down. --Redrose64 (talk) 22:50, 22 April 2016 (UTC)[reply]
The module's role in this is to create the file links; the effect itself happens whether you use the template or whether you include the file link manually. For example, [[File:Chromatic Fantasia (Bach BWV 903).ogg]] comes out like this:

It's easiest to see the effect if you edit this section and then click "Show preview". — Mr. Stradivarius ♪ talk ♪ 00:13, 23 April 2016 (UTC)[reply]
Just to provide another editor's experience; the only jump I see is a very minor reskinning of Chrome's default file GUI. The size and shapes of the two examples don't change, and the buffering .gif in the {{listen}} wrapper lasts only a second or so. I can imagine a slow connection being an issue, but would suggest that if the connection is not slow, another possible problem could be a bunged-up browser. If this is the case, rebooting it (close (not minimise) and reopen) might help. A bunged-up system can have similar slowing effects, and rebooting it could also help.
@Kirk Leonard: - what browser are you using (on what operating system can also sometimes matter)? fredgandt 11:56, 23 April 2016 (UTC)[reply]
It's a problem with the current player. It's part of the reason why i'm trying to replace it, which is an endeavor that I have already spent a year on. That whole part of the software had not been touched in 8 years, and it's a very challenging project to turn such old code around into a more modern extension, whilst not breaking existing stuff. Bottomline, it's not likely going to be fixed in this version of the player, but it's on the radar. —TheDJ (talk • contribs) 13:03, 23 April 2016 (UTC)[reply]

Missing edit summary on desktop view from iPad

I don't know if this is the place to do it but there is a bug with the Desktop view from an iPad. When viewing the history of a page the edit summaries are all missing. On PC: http://imgur.com/egg9tyC On iPad: http://imgur.com/MmSyVkh --IngenieroLoco (talk) 23:01, 22 April 2016 (UTC)[reply]

Never mind. The problem was that I had Block comments enabled in 1Blocker. I disabled that and the edit summaries are back. --IngenieroLoco (talk) 11:11, 23 April 2016 (UTC)[reply]

My comment messed up the chronology of the Talk Page at the "Gottlob Frege" article

Sry! T 88.89.219.147 (talk) 04:36, 24 April 2016 (UTC)[reply]

And the same on the "Truth" article Talk Page - sry sry sry! T 88.89.219.147 (talk) 05:15, 24 April 2016 (UTC)[reply]
I assume you're talking about the fact that there were some references below your comments. That is because there were ref tags on the page but no reference list template, so MediaWiki placed the refsat the bottom of the page, below your comments. I've added {{reflist talk}} in appropriate places to both the talk pages you've referred to. Graham87 06:15, 24 April 2016 (UTC)[reply]
Thx a lot :) T 88.89.219.147 (talk) 06:21, 24 April 2016 (UTC)[reply]
Retrieved from "https://en.wikipedia.org/w/index.php?title=Wikipedia:Village_pump_(technical)&oldid=716839691"