Only not just about VE if you please.

License info: PD-text/PD-ineligible.

Nice. I sign up as interested in being an attendee of the session :) I already have experience editing modules, but having some systematic learning would be much appreciated.

Oppose. It is much nicer with having ISO codes shown somewhere. E.g. a person may be viewing from a device which does not have fonts for all the scripts -- then without the code it would be impossible to distinguish several names rendered as squares.

What I do — I just reuse the queries which better'd be deleted or which were just those empty ones. Once you've renamed and changed it nobody knows what it was before that. Perhaps some process for marking queries for reuse could be added (like a separate section where you can move queries to and from so that you can keep queries to be reused in there together, or any other way)

The tables in the desc looks like indeed the most useful ones, but I would like all the extension's tables to be replicated.

Thanks! Updated https://translatewiki.net/wiki/Wikimedia:Wikiedudashboard-courses.student_editors/qqq with this clarification, I guess the task can now be closed thus.

[[Wikimedia:Wikiedudashboard-metrics.upload usages count/en]] the same

@Aklapper the message contains just one word. For any translation related person it would be absolutely clear what it is about. For instance there were never problems with understanding such questions back when the i18n support was done on twn's support page. It should not be more difficult to report things via phab. At least in my case I am usually either in the mood for translation or in the mood for reporting stuff in details.

We have a problem on uk.wikiquote with charinsert in MediaWiki:Wikimedia-copyrightwarning not working now. Is it related?

The campaign shows total number of pics uploaded using it. In this case that includes 2014's edition uploads. The images themselves go to the correct place.

Hi. Fixed the years in the template. The category was already ok. Is there anything (else) wrong?

Another example of probably this I encountered.

https://test.wikipedia.org/wiki/User:Base/T136515 (despite it being in my userspace feel free to modify)

And not just magic words. Just created [[user:Base/test]] with "test", but {{{{User:Base/test}}}} is shown unparsed the same way.

And {{{{TALKPAGENAME}}}}

Same thing with {{{{FULLPAGENAME}}}}

{{ {{PAGENAME}} }} seems to work. {{ {{PAGENAME}}}} and {{{{PAGENAME}} }} work too.

WMUK?

In other messages «title blacklist» is spelt as «TitleBlackList» or whatever. Needs to be unified.

If the reasons are just purely linguistical and that and only that is why this task is to not be done, then T109033 should be done: the linguistical similarity and mutual intelligibility works both ways. If the task mentioned is not done, then as much as I want to assume good faith, the reasons applied are not linguistical, or not only linguistical.

How about you just set up the phrase for translation somewhere?

I note that MovePage (the utility class for moving pages) is a bit too low-level for the purpose, since it handles moving exactly one page while the rate limit should probably continue to apply to "submissions" that might move a page, its talk page, and its subpages all at once.

Sorry for possibly irrelevant comment, but could that mean that if the work as described is implemented, if one moves a page with like 51 subpage and his limit allows him to do just 50 moves at once, just the root and 49 subpages (if we do not have talks) will be moved and the rest would stay unmoved? If so, it's probably better to throw an error right away.

Then I guess T100801 should be closed.

In tux=0 I summarize translations in target language usually, but sometimes in English (e.g. if I need to complain about something bad in the source or its tagging, or if I am providing attribution for something copypasted).

Well even on smaller wikis there is this "oh, that must be done here, but I'm too damn tired/busy/lazy/whatever to do it right now" thing. And even if there are just 1000 pages in the wiki it could be difficult to recollect your plans about particular pages, and for the start which particular pages, after like half a year or sometimes even less.

That education programme thing allows to chose from preselected list? (This task is useless as it is hard to see the actual progress from it.)

I think it is blocked by setting up mailing server for WMUA (no idea if there is a task for it). It is past time we finish creating those lists without proper membership management and archives. Something like mailman should be used.

@Nikerabbit, It was you who broke it in rETRAb20d24777f293fc3f8630b448acd497ae0c15040 . Do spend time to clean after your own messing around.

In T131132#2270479, @Whatamidoing-WMF wrote:

Such messages already use a variable, so they will require no extra translation effort.

@Elitre

• Publish is an action you do with posts in blogs and sometimes in social media sites. Wikipedia is not a blog, nor it is a social media site: https://en.wikipedia.org/wiki/Wikipedia:WWIN#Wikipedia_is_not_a_blog.2C_web_hosting_service.2C_social_networking_service.2C_or_memorial_site . Neither are other Wikimedia projects and prevailing majority of other MediaWiki based wikis. It also from its newspapers meaning (which would be the meaning most older generation of editors to be would apply) suggests something you do once and for good, while you can save the page many times.
• Wikinews have Publishing process and it is a process preceded by article being reviewed. It is done just once. Changing this may also apply to third party sites which try to build news functionality on Mediawiki.
• It might be not too relevant in large scale, but there were a bunch of printed brochures (from cheat-sheets to rather big editions) on editing Wikipedia and other MediaWiki based sites published all around the globe. They all at once will become rather outdated as the button in question is surely referred to in most of them. (and for a newbie everything is confusing enough without trying to guess that what is referred to is the same button he sees but under its old name) It also applies to online help pages which use hardcoded button label inserted rather than transluded via an {{int: message. The latter are more easy to update (or more precisely updatable, as it is better to say that there is no way to update materials already printed at all, especially if they are already distributed) but practice says that it takes years for all of such pages to be updated.
• As somewhat a subcase of the previous it will also require interface messages translations update (both the very button label and all the rest referring to saving page in many different contexts). While for those with many very active translators it will not take long, for others it make take years if not decades to update and all the time before it people will see either English (I assure you that not every person in the world knows it) a fallback (which sometimes is a really annoying factor, e.g. many Ukrainian users are angry because of MediaWiki having Russian fallback for the language and have requested it removed many times) or old text saying "save" which will once again just complicate things.
• Written materials aside it will lead to confusion for existing editors and it is a great number of people. Surely all of them are likely to have no problems understanding what has happened and finding the button (unless you also move it somewhere) but it may cause non welcome feelings of emotional rather rational nature which are hard to describe in words.

In T104762#2270182, @Multichill wrote:

In T104762#2269919, @Base wrote:

A stupid question by a person with bad SQL (progressed from almost zero mostly thanks to quarry) and almost no SPARQL knowledge: would it be possible to have part of a request being made in SQL and part in SPARQL and have it all output as one table? (Like select a list of articles on some wiki by some WD based criteria and then fetch its sizes, creators and stuff like this via SQL and just have it all in one table)

Quite off topic on this bug. Another forum like the wikidata mailinglist is probably more suitable. Have a look at https://petscan.wmflabs.org/ . With that tool you can combine queries from different sources.

An enhancement request (I believed it would be clear from the description). Ok will rename.

A stupid question by a person with bad SQL (progressed from almost zero mostly thanks to quarry) and almost no SPARQL knowledge: would it be possible to have part of a request being made in SQL and part in SPARQL and have it all output as one table? (Like select a list of articles on some wiki by some WD based criteria and then fetch its sizes, creators and stuff like this via SQL and just have it all in one table)

I do not know hot it is build so do not know if that might occur (well hopefully it is based on IDs and not on usernames behind the scenes so it might not), but I hope that those renames if you made them will not just create a second account for new username of a user and thus leave the first one orphaned. As it would be unfortunate to stack orphaned accounts.

Would be nice if it is multilevelled limitation with basing on perhaps users rights on wikis or something like this (perhaps also some additional whitelist for users who are not active wikimedians but have some reasoned research being made).

Please no.

Erm, the title says about the left side and the description about the right side. So which side it shall be? :)

In T129627#2126230, @TJones wrote:

One last consideration: there is a non-zero number of nerds (there's me—that's 1!) who are very happy to search for "wrong-language" queries on a wiki, especially names of people and places, because it works pretty well! e.g., search enwiki for ประเทศไทย or Москва́.

In T119736#2152031, @Tgr wrote:

These scripts are very slow and I figured it's not a great idea to run some scripts the implications of which I only vaguely understand so I just ran checkLocalUser.php and checkLocalNames.php on the wikis which threw errors for the two users above - dewikiquote, jawiktionary, plwikiquote, jawiki. That should fix problems for the two users who reported them; the scripts can be run in full next week.

Is it possible instead of de-urlencoding them on output just not urlencode them to begin with when writing to db?

In T131731#2177405, @Nikerabbit wrote:

The default translatable page source language is the content language of the wiki and that should work without any problems. Although, it does fall back directly to the source language for missing translations instead of following the fallback chain. Is this what you mean and want?

If you want to have a different source language (other than the wiki's default content language), Special:PageLanguage can be used if it enabled. The status of this is bit difficult to understand from T69223. If it is available, I suggest you change the page source language before marking it for translation. As this feature is not as much tested as general page translation, please do report issues if you see anything odd.

@Nikerabbit
It sounds that it indeed is what I mean and want, though my general unfamiliarity with the unstructured element translation blocks me from being certain. I believe the feature could be enabled then, though it won't be used probably for some time in practice.

@Urbanecm, oh I always thought that the notifications are part of the Translate extension too. Yes, please enable TranslationNotifications as well.

@Nemo_bis, well as to the unstructured element translation, I am not really familiar with it, neither, I think, any other user in the wiki is, so it is unsubstantial whether we have it now. Yes, surely I will ask it being enabled when we need it, thanks. Though, does the thing work fine when the source language is different than English (while the interface still falls back to en and stuff)?

Could you clarify the question please? Suggestion to close the task as invalid when the problem is obvious with bare eye appals me a little bit. In brief words the link should be rendered as it was back in 2014 or so when it still was ok. The urls parts which are not seen are oldid and useskin=monobook. The latter part is not important while the oldid (though in my opinion rcid would make more sense) is crucial. The title part should either be removed (oldid works by itself) or not percent encoded -- this should solve the problem of url taking more space than the database allows. Alternately perhaps the db could be altered to allow bigger string maximum length. The links target is a Project namespace page in ukwiki with the list in which the monument queried could be found. If you need more details -- please ask, what we want is the problem being fixed, rather than clean backlog here while the issue persists.

Ah and here's the causation for that it seems, it's the Q502560238080 element which got into the query. When it's just 5 items the behaviour is normal, just missting="" attribute is added:
https://www.wikidata.org/w/api.php?format=xml&maxlag=5&action=query&action=wbgetentities&redirects=yes&props=&ids=Q15712437|Q4477434|Q502560238080|Q1963971|Q1963253

It also looks that this is personalised to me somehow, as @Ata just didn't get the error on opening it, though I did after that:
<api servedby="mw1121"><error code="internal_api_error_DBQueryError" info="[55ac5ebfa3166a0165828f44] Database query error"/></api>

https://www.wikidata.org/w/api.php?format=xml&rawcontinue=1&maxlag=5&action=query&action=wbgetentities&redirects=yes&props=&ids=Q261243|Q1789505|Q1076875|Q770336|Q1987828|Q2040044|Q932901|Q1011047|Q4477434|Q502560238080|Q1048890|Q1963253|Q174885|Q340566|Q1010960|Q1972264|Q2657626|Q15712437|Q1011003|Q16843187|Q163384|Q956932|Q946143|Q2053267|Q956954
A shorter query one so that you can "enjoy" beholding the error from your non-bot account.

<api servedby="mw1116"><error code="internal_api_error_DBQueryError" info="[a46e3c9a140c5904c6d8aa5c] Database query error"/></api>
<api servedby="mw1208"><error code="internal_api_error_DBQueryError" info="[5079872defc7bce8a66c970b] Database query error"/></api>
just that you have more...

Is there a way to run a check against all users instead of waiting for the affected to trigger the error in order to look for it in logs?

Why Jimmy Wales? Why not Katherine Maher (interim Executive Director) or Patricio Lorente (WMF BoT Chair)? Jimbo looks like illogical option to me.

Erm, devs break things and then say that they are too lazy to fix them?

T128771

@mmodell, per Wikimedia Licensing Policy a project has to have an EDP in order to allow fair use in it. Besides I doubt you can make a proper FUR for using a meme outside of educational context.
@Aklapper, EDP stands for exemption doctrine policy, as per the Wikimedia Licensing Policy.

Doesn't @Southparkfan's image violate copyright? Do we have an EDP on Phab?

Then it covers the whole site which is completely unacceptible for any special page. I have tried it before filing the task. I thus do insist on the request.