Wikisource:Administrators' noticeboard

From Wikisource
Jump to: navigation, search
Administrators' noticeboard
This is a discussion page for coordinating and discussing administrative tasks on Wikisource. Although its target audience is administrators, any user is welcome to leave a message or join the discussion here. This is also the place to report vandalism or request an administrator's help.
  • Please make your comments concise. Editors and administrators are less likely to pay attention to long diatribes.
  • This is not the place for general discussion. For that, see the community discussion page.
Report abuse of editing privileges: Admin noticeboard | Open proxies
Wikisource snapshot

No. of pages = 1,937,755
No. of articles = 576,408
No. of files = 17,094
No. of edits = 6,289,041

No. of pages in Main = 338,731
No. of pages in Page: = 1,283,234
No. validated in Page: = 259,803
No. proofread in Page: = 375,892
No. not proofread in Page: = 519,965
No. problematic in Page: = 23,638
No. of validated works = 2,036
No. of proofread only works = 1,127
No. of pages in Main
with transclusions = 126,209
% transcluded pages in Main = 37.26
Σ pages in Main

No. of users = 2,780,756
No. of active users = 320
No. of group:autopatrolled = 420
No. in group:sysop = 32
No. in group:bureaucrat = 3
No. in group:bot = 27

Checkuser requests[edit]

  • Wikisource:checkuser policy
  • At this point of time, English Wikisource has no checkusers and requests need to undertaken by stewards
    • it would be expected that requests on authentic users would be discussed on this wiki prior to progressing to stewards
    • requests by administrators for identification and blocking of IP ranges to manage spambots and longer term nuisance-only editing can be proggress directly of stewards
    • requests for checkuser

Bureaucrat requests[edit]

Page (un)protection requests[edit]


Resource Loader issue needs outside guidance[edit]

The more I read up on this RL change and the subsequent actions needed (or taken?), the more I get the feeling some of my approach to site wide & gadget .js/.css organization over the months is going to behind this week's latest problems. If that winds up to be the case, then I'm truly, truly sorry for that. Let me try to document those steps and the reasoning behind them in hopes someone (@Krinkle:) can made sense of our current state and put us on the right path post RL change(s).

Originally, we not only had a ridiculous amount of scripting and .css definitions in our primary site-wide MediaWiki files to begin with but also called a number of stand-alone .js/.css files within those primary MediaWiki files called unnecessarily in addition to calls to various sub-scripts on top of any User: selected gadgets being called -- some of which eventually became default loaded per concensus, etc..

A simple depiction of the key files mentioned minus any Gadgets basically went like this...

Over several months with help of other folks, I began to consolidate and/or eliminate as much scripting calls as I could -- creating optional Gadgets whenever possible -- and tried much the same for the .css class definitions. The rationale behind doing this can be found in several places, most importantly: Wikipedia. The premise to keep the MediaWiki site-wide files "lean" goes like this....

 * Keep code in MediaWiki:Common.js to a minimum as it is unconditionally
 * loaded for all users on every wiki page. If possible create a gadget that is
 * enabled by default instead of adding it here (since gadgets are fully
 * optimized ResourceLoader modules with possibility to add dependencies etc.)
 * Since Common.js isn't a gadget, there is no place to declare its
 * dependencies, so we have to lazy load them with mw.loader.using on demand and
 * then execute the rest in the callback. In most cases these dependencies will
 * be loaded (or loading) already and the callback will not be delayed. In case a
 * dependency hasn't arrived yet it'll make sure those are loaded before this.

The result of that effort as it stands today can be depicted basically like this....

The predominant change in order to move towards the previously cited rationale & approach is that the bulk of the scripting and class definitions now reside in the default-enabled Site gadget files, MediaWiki:Gadget-Site.js & MediaWiki:Gadget-Site.css. And by no means is the current state the desired final approach; its been a work in progress as time allowed over several months.

Obviously, now with the recent change to Gadgets and ResourceLoader, either the existing rationale or my attempts (or both) are no longer in harmony -- if they ever were. In my view, we need someone like Krinkle (or maybe the collective minds of Wikitech-l?) to take the time and attention needed to come in here and straighten all this out -- one way or the other. My gut tells me THAT will resolve the reported loss of one thing or another post-RL change(s). Again, if I'm right about my actions exacerbating problems for other, I apologize and take full responsibility. -- George Orwell III (talk) 20:54, 8 August 2015 (UTC)

I've made a few minor changes in addition to yours that hopefully make things work a bit more like you intended. I'm happy to provide further guidance but that probably works better for a more specific need or question. Perhaps bring it up on Wikitech-l or on IRC so we I can help you move forward with any unresolved issues. Krinkle (talk) 21:37, 20 August 2015 (UTC)

Bots are not capable of proofreading[edit]

I am comfortable with a bot formatting and undertaking some standardised text changes, however, I am not comfortable with the bot SHSPbot (talkcontribs) setting works to proofread. The act of proofreading is a human act where the text is visually checked against the image, and I don't see that a bot is capable of doing that work. I am surprised that this has been allowed to occur and that the community in its patrolling has not made some comment about this. — billinghurst sDrewth 23:48, 23 August 2015 (UTC)

We might set all texts to proofread: such a solution would be a quicker and easier one, don't you think so? :-) (Don't take my words seriously, I see the proofreading activity as a human affair, not in the capacities of a bot.) --Zyephyrus (talk) 07:13, 24 August 2015 (UTC)
Despite its name is SHSPbot genuinely a bot? It does not appear in this list (not that I am sure that that is definitive.) Would it not be legitimate for Dictioneer to simply be using this as an alternate account? (I ask for information; not to express a position.) AuFCL (talk) 08:29, 24 August 2015 (UTC)
The discussion was held here: Wikisource:Scriptorium/Archives/2015-07#BOT_approval_requestsMpaa (talk) 17:27, 24 August 2015 (UTC)
I am just reading the bot user page. I have asked that question at the talk page and am awaiting a response. — billinghurst sDrewth 11:10, 24 August 2015 (UTC)
If the bot is used only to quickly save pages from a carefully scrutinised off-line file, I do not see a problem. No difference for me in accuracy whether you work directly in WS or in a text editor + djvu viewer side-by-side off-line. I think it is a matter of preference regarding your tools. Another story is an automated program that you launch and forget.— Mpaa (talk) 17:27, 24 August 2015 (UTC)
I agree: as long as the bot is a tool and not a decision-maker, it's ok for me. --Zyephyrus (talk) 09:02, 25 August 2015 (UTC)
Sorry bilinghurst and others, I've been away on vacation. I believe the confusion is caused by what is probably my inadvertent violation of another policy -- I've been logging on as SHSPbot manually and proofing the pages it's uploaded. This seemed at the time a cleaner way to collect statistics on the bot's performance. However, if as I suspect, this is a big-time violation of the rules, I will cease the activity immediately. My apologies for creating an unnecessary discussion. I will return to this page in my human entity of Dictioneer to finish commenting. SHSPbot (talk) 23:22, 27 August 2015 (UTC)
Following up on my previous post, I have left my bot persona and am back as me. I presume I broke a rule/policy by editing pages while logged on in my bot persona: while it seemed like a good idea at the time, I think I get why this is a bad idea and plead ignorance as my only defense (I'd assume there was some flag set when uploading using the pywiki software and that it was harmless to be a human using a bot account. Since I apparently fooled billinghurst I assume that it's not possible to tell the difference. My intent was simply to collect productivity statistics for the bot work + manual corrections and proofreading under the same account and not to engage in any subterfuge or other misbehavior. From now on, all manual edits will be done under my dictioneer persona. If there is anything else I should do or any administrative process I face, I am available as needed. Again, mea culpa. Dictioneer (talk) 23:33, 27 August 2015 (UTC)
I know that Dictioneer's intentions were and remain honourable. We all make mistakes, well, many of us. So give him the benefit of doubt. He has nothing to gain by doing as he did. He could just let that project sit forever unproofed and unvalidated. He came from WP with good intentions and I have worked with him in private and in the open. He is smart but doesn't know all of the rules yet but who does? Benefit of the doubt is one of those important rules and a smart one on Wikisource. —Maury (talk) 16:23, 28 August 2015 (UTC)

AutoWikiBrowser - some questions[edit]

Is there anyone here who uses AWB on Wikisource? Do I need additional permission, in addition to my WP signup? The most important question is how to hook into WS. Any help/info given is much appreciated because I couldn't find anything in the manual that would help. — Ineuw talk 01:58, 21 September 2015 (UTC)

BD2412 is a heavy AWB user here on Wikisource. Hesperian 02:08, 21 September 2015 (UTC)
  • If you can use AWB on Wikipedia, all you need to do is boot it up, go to the [Options] tab on the top (fourth tab over), select [Preferences] (first tab on that dropdown list), then select [Site] on the menu that opens (second tab) and change the project from the default Wikipedia to Wikisource. Then click [OK] and you're in. You'll need to confirm your login information again once you start making edits. So far as I recall, I did not need to seek any additional permission to use AWB here.
  • There are many wikis that require such a permission, actuated by having an a page editable only by admins where the names of approved users must be added before they can use the tool. Frankly, it would be a good idea to have that here also, for the same reasons it exists elsewhere. Cheers! BD2412 T 03:41, 21 September 2015 (UTC)
    As Ineuw is an admin, he wouldn't be restricted. @Ineuw, generally the advice of take it slowly is a good approach (typical enWS). When you are starting to be bot-operation-focused, then the normal application processes for bots applies.

    Re a permissions page, we haven't had abuse issues in all these years, so we remain mindful and aware. — billinghurst sDrewth 10:10, 21 September 2015 (UTC)

    Fair enough. BD2412 T 15:35, 21 September 2015 (UTC)
Thanks to all for the replies. I have no intention to use any bot at this point. My reason of use is simple. I want to understand the tools that are available to a user, starting with myself. — Ineuw talk 15:49, 21 September 2015 (UTC)
I echo Billinghurst's advice about just taking it slow at first; if I can utilize it there's no reason you shouldn't be able to. Also, I find it advantageous to enable the Flood Flag User: right prior to running AWB since the bit can -- for example -- bump up search / result-like tasks from 50 or 500 to 5000 (among other enhancements).
And I echo Ineuw that there is nothing to worry about. Just managed to log into Wikisource but every day I intend to take some time to study and understand the features. I didn't even reach baby steps yet, still in diapers. . . . and certainly don't intend to use any bots. I am very curious if it's usable for proofreading and that's all. — Ineuw talk 05:07, 22 September 2015 (UTC)
No images visible, so next to useless for proofreading against scans. Very useful if you move a page and need to disambiguate. Useful if you wish to undertake the same action repetitively , etc. AWB can behave in bot mode, or it can be used in full (re)view mode, and I do both. I use the latter situation where I have no interest in the image and can play with the text alone, especially as it can operate outside of the javascript restraints. It is also useful for building lists. — billinghurst sDrewth 05:22, 22 September 2015 (UTC)
This has slowly dawned on me as I read the manual. Although it's claimed that is useful for proofreading, but not here, and now I understand why BD2412 uses it. I noticed that he is cleaning a lot of text pages of PSM. — Ineuw talk 05:53, 22 September 2015 (UTC)
I find AWB useful for a number of tasks, of varying degrees of repetitiveness. For example, if a series of pages in a work have a common header, and this was not put in when the pages were initially created, I can use AWB to create the header template and the repeating portion of the text, and then add the non-repeating portion (e.g. page numbers) on the fly. BD2412 T 15:30, 22 September 2015 (UTC)

Special page listing for wanted templates broken?[edit]

Some recent change has expanded the Special:WantedTemplates' search results to now include all non-existing "calls" (similar to {{Page:Book.djvu/024}} ) in every namespace instead of just the template namespace.

Can somebody please confirm they are experiencing the same results for starters? -- George Orwell III (talk) 00:02, 29 September 2015 (UTC)

Yep, I see that it is listing missing transclusions, rather than missing templates. Now is that good or bad? Do we really want empty red links for missing pages in the main ns? — billinghurst sDrewth 03:12, 29 September 2015 (UTC)
That's beside the point - the page is called Special:WantedTemplates for a reason. It's broke plain and simple.

If anything, Special:WantedPages would be where I'd expect missing main ns pages to come up but now that I've checked it, that seems screwed up too (though its been so long that I've used it I can't be sure).

I get the feeling somebody's been dickin' around with the internal search mechanism's content-type settings or something - I'll check Phab when I get some more time. -- George Orwell III (talk) 03:25, 29 September 2015 (UTC)

Fair points, I was thinking aloud, not making recommendations, or reaching conclusions. — billinghurst sDrewth 03:53, 29 September 2015 (UTC)
Well isn't that just peachy? (Fault seen here as well. And stop me if I'm wrong, don't I recall there used to be three links for each :WantedTemplates entry ala Special:WhatLinksHere i.e. page (←links|edit)? Whatever happened to (|edit)?) AuFCL (talk) 05:10, 29 September 2015 (UTC)

Note: It has been recorded that the change came about due to wishes to view Module: ns, and presumably it was unknown/forgotten by the editor and those checking that other components are transcluded. — billinghurst sDrewth 06:17, 30 September 2015 (UTC)

Yep. Saw that. Out and out "Works for wikipedia so why are you whinging?" mindset. AuFCL (talk) 06:41, 30 September 2015 (UTC)
I may well have rediscovered Billinghurst's "it has been recorded" above but Requesting "Special:WantedModules" page looks to me to be a likely precursor to the problem. AuFCL (talk) 00:40, 2 October 2015 (UTC)
After getting no traction on the fix/amendment, I have reverted it in gerrit. I am hoping that it will be incorporated, though probably won't show up here for at least a week. — billinghurst sDrewth 10:57, 13 October 2015 (UTC)

Template Code in Module: name space[edit]

I have been noticing a number of entries under the module: name space which contain no LUA code at all. Most seem to have been imported (with revision history) presumably from wikipedia and all by user:AdamBMorgan just under one year ago. Some examples:

Is there any point in keeping these—as obviously nobody has noticed that they are unused—or alternatively should they be moved into template: space where at least they might become useful? AuFCL (talk) 08:10, 8 January 2016 (UTC)

Stray Babel categories[edit]

Just wanting to do a reality check before I delete a pile of categories. Special:UncategorizedCategories has some empty Babel categories that were autocreated on 17 Sept. They look to be wrongly named to me, given that the usual format is Category:User en-N. Can anyone think of a reason to keep the new ones? Beeswaxcandle (talk) 06:41, 8 November 2015 (UTC)

Put it this way: as autocreated if they are necessary and you delete them they will be populated again anyway; if unnecessary (and empty) it is not as if you have lost any content. My opinion is this: go ahead and delete them. AuFCL (talk) 06:54, 8 November 2015 (UTC)


Clearly a vandalizer, please block.

Yes check.svg Done and all contributions have been either deleted or reverted. Thanks, Beeswaxcandle (talk) 05:34, 24 November 2015 (UTC)

Wave of spam accounts[edit]

The start of the holiday season in the US is bringing us the joy of SPAM.

Spam, Spam, Spam, Spam, Spam.

. . .and spam. --EncycloPetey (talk) 18:39, 26 November 2015 (UTC)

Given the persistence of the spammer(s), should we consider temporarily protecting the Scriptorium? --EncycloPetey (talk) 21:01, 28 November 2015 (UTC)
Is it worth bothering since they are creating almost as many main-space pages as edits there? (Also I am in two minds about your deleting the revisions: the effort you are putting is appreciated but at the same time will a normal user be equally put off encountering so many "You cannot view this diff because..." It sort of rubs their noses in an unnecessary 'us vs them' situation; and perhaps that very annoyance is something the spammers are enjoying?) Fimus equinus (talk) 21:27, 28 November 2015 (UTC)
It's worth hiding them, or the spam information is visible in the logs and through permalinks. --EncycloPetey (talk) 21:31, 28 November 2015 (UTC)
I do get your point. It is just that the spammers may be getting a certain amount of perverse joy out of watching you expend all that effort too. It is always a trade-off. Fimus equinus (talk) 21:52, 28 November 2015 (UTC)
Yes, but then they're trolls. There is little we can do about the perverse joy of trolls, but we can prevent spam from showing up on our site. --EncycloPetey (talk) 21:59, 28 November 2015 (UTC)


Either of you gents taken a crack at the abuse filter or log yet? I tried to reinforce some existing filters for a quick fix yesterday but that "help number" one keeps getting through it seems. Don't worry if its all above your pay grade but I do recommend taking a look if not just to isolate then block some of the IP addresses that make attempts but "fail" thanks to a in-place in filter. Rememeber to check your User: rights settings too.

I just don't have the free time at the momement; @Billinghurst, @Mpaa:? -- George Orwell III (talk) 23:13, 28 November 2015 (UTC)

<chuckle> Can we please leave just one "help line" type-of-spam in place the next time? Its kind of hard to build a filter without an example to test it against :) -- George Orwell III (talk) 23:18, 28 November 2015 (UTC)
Fwiw... I've modified Filter 4 to [hopefully] "address" the help line type of spamming. -- George Orwell III (talk) 00:28, 29 November 2015 (UTC)

No longer on OTRS mail[edit]

For the community to note that I am no longer handling OTRS mail. — billinghurst sDrewth 01:13, 15 December 2015 (UTC)

Moved page text gone awry[edit]

I am not sure whether it is just me or whether others can reproduce.

When we move a page the text from MediaWiki:Movepage-moved should be used as the template to reproduce. When moving pages today, I am just seeing the default templateas show in AllMessages, not our updated page. Anyone able to report on what they are seeing before I seek the help of Phabricator. Anyone been in and made any changes? FWIW I test at Meta, and it seems to be functioning as expected there.

If you want something to test, you can use Author:testpage. — billinghurst sDrewth 13:39, 21 December 2015 (UTC)

@George Orwell III, @Beeswaxcandle, @Mpaa, @Prosody, @Hesperian, @Ineuw, @EncycloPetey: anyone else. Would one of you please confirm or deny my thoughts above. Makes it hard to report a bug if no-one confirms it. Thanks. — billinghurst sDrewth 00:24, 1 January 2016 (UTC)
@Billinghurst: just moved the Author:Testpage selecting to create a redir and the message that came up upon success is {{MediaWiki:Movepage-moved}}

The page "$1" (links, delete, edit) has been moved to "$2" (links, delete, edit, revert).


"$1" has been moved to "$2"

So I'm not seeing anything "odd" or like what you are reporting. George Orwell III (talk) 00:34, 1 January 2016 (UTC)
Likewise, I'm seeing nothing out of the ordinary. --EncycloPetey (talk) 00:35, 1 January 2016 (UTC)
I assume you mean the extra text about creating soft redirects and checking "what links here". These are missing. Beeswaxcandle (talk) 00:41, 1 January 2016 (UTC)
Based on that tidbit from BWC, i tried moving the test-page again under a User: preference language setting of en-gb (English-Goofy British) and now see the default message as well, so I bet you folks are not set to "straight" en in your settings too.

FYI... The recommendation has long been to avoid selecting language subsets just for this kind of reason (html input form recently redesigned to use OOUI layout with mw-messaging). -- George Orwell III (talk) 00:53, 1 January 2016 (UTC)

If anybody decides to "fix" this remember Ineuw: MediaWiki:Movepage-moved/en-ca AuFCL (talk) 01:06, 1 January 2016 (UTC)
English to Catalan? Or is that "Canadian English"? --EncycloPetey (talk) 01:13, 1 January 2016 (UTC)
<chuckle> He meant English (Canadian).

I'm hesitant to go about the "usual fixin" since this is a new phenomenon (its not like the sub-page, lang-specific messages were recently deleted or similar as was usually the cause before). -- George Orwell III (talk) 01:17, 1 January 2016 (UTC)

It was meant both as a dig at the fact mediawiki makes a great show of being all-international-linguistics-friendly and then drops the ball like this; as well as a subtle reminder that (lots of) MediaWiki:-space pages have a /en-gb, /en-ca etc. sub-page for language variants which are easily forgotten about. However you guys just let the "magic smoke" out… AuFCL (talk) 03:16, 1 January 2016 (UTC)
Back-up a sec. I can only find a dozen or so sub-English pages; where are you getting "lots of" from?

In fact there's more [crap] message sub-pages in other languages than English (not for long though!:) -- George Orwell III (talk) 03:37, 1 January 2016 (UTC)

I stand corrected [1]. Far fewer than I had expected. AuFCL (talk) 04:53, 1 January 2016 (UTC)
Thanks to all, that would definitely seem to be the issue, though I haven't been in and changed my preference for a very long time. Seems a problematic setting, and not sure when that came into play in that fashion. — billinghurst sDrewth 02:01, 1 January 2016 (UTC)
It seems every time a Special: page or other core component is redesigned to the oojs-ui "format", there is no guarantee any previous behavior/feature that were due to "happy mistakes" that went un-addresses before will be automatically "carried over" in the switch. Drop offs are usually because "bits" are being deprecated at the same time from what I've read on various Village Pumps. It may be time to truly avoid using monobook from now on as well (from what User:s have stated in those threads that is). -- George Orwell III (talk) 02:39, 1 January 2016 (UTC)

There was comment in the phabricator that some of our suspicions of cause were unlikely, so it is on hold. I may escalate it to the languages team as the system should be able to default to /en when there is no /en-gb or /en-ca version. In the interim for this case I have created the pages and inserted {{Mediawiki:Movepage-moved}}. — billinghurst sDrewth 00:35, 8 January 2016 (UTC)

Nothing else has changed except for the layout of the Special: move page itself. The MediaWiki:Movepage-moved message hasn't been touched since 2008 until you "bumped it" in December 2015 in search of a solution -- the message content was still unchanged regardless. Nobody deleted .../en-gb or .../en-ca; neither ever existed. Left with no other rationale explanation, the layout change had to trigger the new behavior; most other reporters believed that the "new behavior" should have been the default behavior all along (e.g. the absence of .../en-ca, .../en-gb messages should have never defaulted to the en message). It was only happenstance that it worked the way it did thanks in large part to not being ""touched" or "altered" over the course of many years.

I recall the same thing happened on the article history page and the watchlist page when the collapsible Legend boxes where added after years of being "static" too; .../en-ca and .../en-gb messages had to be added manually in order to restore the previous behavior. The key is that a "change" breaks the previous happenstance - what the specific change entails seems to be irrelevant here. -- George Orwell III (talk) 02:29, 8 January 2016 (UTC)

Maybe it had been cached for a long period, but it not been working for me for over a month prior to that date, so the change isn't from anything that we have done locally, as best as I can tell. I have waved the phabricator under the nose of the language team and maybe they can explain how things are meant to work. One would hope that in English language wikis that they would take the existing (customised) /en rather than the default /en-gb. Asking us to have every message as /en, /en-ca and /en-gb is silly, having the users to know that /en-xx variations will cause that level of change and not to change it, makes a mockery of having the setting. So if that is the new expected behaviour, I am hoping for a solution from them. — billinghurst sDrewth 04:11, 8 January 2016 (UTC)
I know its not anything "we" did locally; the meaningful change(s) to the Special:MovePage layout started to be merged around September 15 (see T86865) and continued to take refinements all the way through the end of November thanks to various other Tasks.

As best as I can recall, the only time a MW message "sub-page" is suppose to inherit the "root-page" message is when that "root-page" is an unaltered default populated by the "list" in the wiki code on the servers. At some point in time, inheritance might have worked with a manually modified "root-page" but that changed. The behavior, however, remained the same until something else came along and altered that status quo (like a format redesign in this case). So I suppose one option for us is to lobby to make our local message the default message. Of course that would most likely require some editing to make the various points more "project friendly".

My point all along has been not to use the MediaWiki namespace to host "complex" alterations of the default "root-page" message; that sort of complexity belongs in the Template: namespace and said template should then be called upon by the corresponding mw message space. Any sub-pages of that root-message should also call that same template if applicable. -- George Orwell III (talk) 01:59, 9 January 2016 (UTC)

Mediawiki namespace is purposeful for editinterface and the rights and permissions, and is designed to take complex alterations, so moving such to the Template is your decision, not the communities, and not a design aspect of the namespace. The namespace is set up with very specific restrictions by design, and moving such pages to the Template means that you then have to separately manage (and replicate) the security, and that is less than ideal. Years ago we faced issues with templates used being vandalised and had to implement security on those, it is clearly should not be the default choice to use template: ns. over the more secure mediawiki: ns. Beyond that, the basic principle that en-gb settings should not change the move message on English WS, and we should not have to design components to fix such. unsigned comment by Billinghurst (talk) .
This member of the community would feel a bit better represented by someone who took the effort to understand the software design of the system they are speaking such bollocks about. It seems your interpretation of what you (personally) want the software to do is exactly the opposite of how it has always fundamentally (vs the occasional bug behaviour) operated. Now can this conversation please be dragged back out of fantasy to some kind of reality? If you want change there is a process, and it just may be the first step of that process is to stop making yourself look foolish to the developers on Phabricator? AuFCL (talk) 03:57, 9 January 2016 (UTC)
Unsupported nonsense. That is the definition "imposed" upon folks by the ThomasV-types who wanted you to believe that was true in order to accept their vision of one thing or another. I know its hard for folks to accept they were sold a bunch fallacies early on in Wikisource's history [admittedly nobody knew any better at the time] but just because some things have managed to remain operational all this time doesn't mean it was done correctly to begin with. Nor does not change the facts at hand.

A system message is a snippet of plain text, wiki text, CSS, or Javascript that can be used to customize the behavior of MediaWiki and its appearance for each language and locale. MediaWiki uses messages for any user-facing part of the interface, allowing for internationalization and localization of the MediaWiki UI, for both core and extensions.[2].

The key phrase there is a snippet of plain text [or] wiki text.. So I'm not saying we can't have "complex" messages in the MediaWiki namespace; just that need to be mostly comprised of other MediWiki messages. When the more "complex" message cannot be comprised of other mw-messages or begins to introduce new functionality or features that can't be easily "translated" at the end of the day, they should be templates and called from the Template: namespace instead. And any protection issues are minor ones at worst. -- George Orwell III (talk) 05:33, 9 January 2016 (UTC)

I am frankly concerned that this demonstrates either a lack of understanding or of consistency between the base message and its language variants. Surely the trio MediaWiki:Movepage-moved, MediaWiki:Movepage-moved/en-ca and MediaWiki:Movepage-moved/en-gb should be either all deleted (to return to software default) or all equivalently customised; but two have two-up one down like this seems, well, a bit cock-eyed? AuFCL (talk) 06:21, 11 January 2016 (UTC)

PSM watch maintenance[edit]

I would gladly take on the maintenance of the PSM filter if it would help to reduce some workload. Like removing old entries to reduce the size of the data, etc. I would also reduce the watch days to be no more than 10 days max.— Ineuw talk 06:55, 10 January 2016 (UTC)

EasyLST gadget: a possible improvement[edit]

By chance I uncovered a few cases recently where the very act of attempting to validate a page in fact damaged it instead as a consequence of the validator having the above gadget enabled (whereas the proofreader presumably had not) and not noticing undesirable changes. I have made a small proposal for tweaking the gadget code and invite review, correction or action as appropriate. AuFCL (talk) 03:51, 13 January 2016 (UTC) Withdrawn. AuFCL (talk) 03:24, 19 January 2016 (UTC)


Vandalism run amok. [3] Not familiar enough yet with blocking policy or I would take it up myself. If no one is around and they keep up the behavior, I will give it a shot. Londonjackbooks (talk) 11:52, 4 February 2016 (UTC)

Repeated vandalism is always blockable, if uncertain, especially with IP addresses, you can just make it for a couple of hours or a day, which stops the there and then while you start a conversation on their user talk page. — billinghurst sDrewth 21:43, 4 February 2016 (UTC)
How does one determine whether the block will be indefinite, three months, or just an hour or so? Londonjackbooks (talk) 22:06, 4 February 2016 (UTC)
When the account is used for nothing but disruption, spam, or vandalism, we tend towards indefinite blocks. There is no reason for such an account to be active. --EncycloPetey (talk) 22:09, 4 February 2016 (UTC)
Agree that for named accounts at enWS, it is either short term, or it is infinite. For IP addresses it can fill the range, except not infinite, which can be problematic. Longer term IP blocks would tend to be used for those managing spambots, or persistent vandals, and something that best comes from watching what others. — billinghurst sDrewth 04:23, 5 February 2016 (UTC)

No bot. -- 12:09, 4 February 2016 (UTC)

Still vandalism and disruptive edits and nothing else. --EncycloPetey (talk) 22:09, 4 February 2016 (UTC)

AbuseLog issues dealt with globally[edit]

To let people know that the recent persistent spam attempts for payday loans, I dealt with through the global title blacklist, rather than local blacklist. I will presume that they will be back with another variation, however, the recent mods seem to be holding for the moment. There was some value in not letting go of all my rights when I retired my xwiki role. :-) — billinghurst sDrewth 11:33, 5 February 2016 (UTC)

Repeated spamming[edit]

Please have a look at this. Hrishikes (talk) 14:10, 9 February 2016 (UTC)

Yes check.svg Donebillinghurst sDrewth 02:00, 17 February 2016 (UTC)

PDF generation issues; do we (temporarily) amend link?[edit]

A second report of PDF generation issues has been received. There is another at WS:Scriptorium/Help. I haven't had time to play with the print output to see if we have generated the problem, or it is more underlying in the extension. If we cannot get an immediate solution, then maybe we should be looking to change the link to use wsexport tool, with a reconfigured link. Thoughts? — billinghurst sDrewth 02:00, 17 February 2016 (UTC)

Reached the point which is above my pay-grade.[edit]

Duplicated the User:Ineuw/common.js/Gadget-NopInserter.js and neutralized lines 55 to 59 which eliminated the unwanted message. Then, added this code at the bottom of my common.js mw.loader.load('//');. The label shows up on the left panel but nothing happens. Can someone with the know please taste my cooking and adjust the flavour? Thanks. — Ineuw talk 15:42, 23 February 2016 (UTC)

Nothing at all, I mean no alert even when a nop is there on the previous page? BTW you have decided to never add nops, not only to suppress the message.— Mpaa (talk) 19:02, 23 February 2016 (UTC)
Thanks again uncle Mpaa. Your post on my talk page was most helpful.
Often when nothing happens at all, there is a helpful javascript error message in your javascript console. Challenge: to figure out how to view the javascript console in your browser. There may be a "developer tools" menu somewhere? Hesperian 01:17, 24 February 2016 (UTC)
H., Thanks for the reminder. There is host of developer tools, and a javascript console but rarely look at it, let alone use it. — Ineuw talk 03:06, 24 February 2016 (UTC)

Userpage request[edit]

Hello, would an administrator please delete my userpage on Wikisource so that my userpage from Meta will transclude? Thank you, --Pine (talk) 07:37, 24 February 2016 (UTC)

Done. Hesperian 08:23, 24 February 2016 (UTC)

Ensuring transclusions of proofread and validated works[edit]

Has anyone thought of a systematic means to ensure that our proofread and validated works are actually transcluded? We have our checking tools that can be manually run, though I don't know that they are necessarily are being run when a work has been validated, or a work has been proofread. Either way, having it done manually is a tedious task, and I am wondering if people have ideas on how we can do this systematically. I have a sneaking feeling that there are works that are sneaking through 1) untranscluded, or 2) untranscluded completely.

Once we have this information, we may be able to use this to look to manage the population of data to WD for the works, which is another task that we need to address. If we cannot work on this, then I suggest that we look to do some phabricator tasks to determine out our requirements with our colleagues through the Wikisources. — billinghurst sDrewth 01:45, 29 February 2016 (UTC)

I came across an untranscluded work this weekend, but for which the entire text had been proofread. But I think this is only the second time I've seen one of those. No ideas that I've had for checking transclusion by bot seem properly feasible to me. The best I can suggest is generating a list of Index files (on suitably sized pages) and use MotM to go through as many as possible.
What about checking, for a given Index, that each page in Page ns is linked to a page in Main ns whose title starts with the same prefix specified in Index Title? I think this should cover most cases. (I was just simplify, rule needs to exclude empty pages etc. but just to give the gist of it)— Mpaa (talk) 18:42, 29 February 2016 (UTC)
Sure, nice, though maybe we don't need to be so neat? As we are wanting an indicator to check a work, if we can even identify a number of pages in a work, that is enough to get us to investigate … so where index:PAGENAME is in Category:Index Validated or Category:Index Proofread but not in Category:Fully transcluded and a Page:PAGENAME/{d}+ of the index is in Category:Proofread or Category:Validated is not marked as transcluded, tag the Index: to be investigated. Once investigated it is resolved and marked as being fully transcluded. How complex is that as a query? — billinghurst sDrewth 06:31, 1 March 2016 (UTC)
A pywikibot script is not complex (it requires many API calls per index though). I should see if it is possible to solve this with one or two SQL queries on a given index. Maybe someone with more SQL experience can find a solution meanwhile.— Mpaa (talk) 22:14, 1 March 2016 (UTC)
I (think) I might have found a solution based on two queries (Page:xxx/\d+ transcluded in Main ns) and (Page:xxx/\d+ in Cat:Pr, Val); and then compare the two sets of pages. Needs to run on tool-labs. I need to check this a bit more and then we might scan all the indexes as WS-bot and tag the suspects. Will elaborate more soon. I can post the skeleton of the code somewhere if someone is interested.— Mpaa (talk) 23:29, 1 March 2016 (UTC)
@Billinghurst: I tested this a bit, it should be OK. There are cases where a validated page (e.g. a cover or and Adv) is not transcluded so this will give a fails (e.g. Page:A Christmas Carol.djvu/198). If this logic is good enough, I can port this into a pywikibot-like bot and then we run it over the Proofread/Validated Indexes. Can you specify a bit better how you would like to tag pass/fail and with which template?— Mpaa (talk) 21:41, 3 March 2016 (UTC)
@Billinghurst:, I think I am good to go (made a few trials, see [4])— Mpaa (talk) 20:44, 5 March 2016 (UTC)
@Billinghurst:, I am currently marking as "Fully Transcluded" Indexes where not trascluded pages are 'Without text' only. Later on, I will make a second round for more complex cases (BTW I have revised the queries above, as I found several other cases to be handled). In the next step, I will mark as "Fully Transcluded" also in case not transcluded pages are 'Not transcluded'/'Without text'.
Would be nice to have a gadget to quickly add Category:Not transcluded at the end of the footer, so manual checking with 'checker' will be a lot faster. After marking the pages, the bot can be run again to take care of the indexes.— Mpaa (talk) 21:05, 14 March 2016 (UTC)
It was a maintenance tasks started years ago, where I added things manually to Category:Fully transcluded, though I have now migrated that check to {{index validated date}} with the parameter transcluded=yes and have created a negative flag category. This will cover validated works, and I will look at proofread at some point, though we could just do a standalone template. — billinghurst sDrewth 13:36, 29 February 2016 (UTC)
To close the loop, I have also created {{index transcluded}} which if says YES, does as above, if you say no, or leave empty it populates Category:Transclusion check required. If it is useful, we can bot apply it, and also add it to the Index: template as a default. I will leave the community to ponder. — billinghurst sDrewth 06:31, 1 March 2016 (UTC)
As far as WD, the little that I've worked on has proven to me that this is a very tricky thing that cannot be easily automated (if at all). Even for the small set of works I've dealt with, the amount of preparatory work is enormous. Much of it involved disambiguation (sometimes creating more than one disambig. page per title), matching the correct targets at both ends (if they even existed), setting meaningful descriptions (which isn't as straightforward as I expected), and creating version / edition / translation data items there. Pushing information on our works through to WD is going to be a very, very big project. As I say, the work I've been doing is taking much more effort than expected, and I'm dealing with only 44 plays and 4 authors (the entire corpus of surviving ancient Greek drama). I can't begin to imagine what it would be like to tackle a larger chunk of this stuff. We really need a way to (1) chunk the project suitably as smaller tasks, (2) decide what exactly we want to send over, and (3) ensure beforehand that our side doesn't include errors / inconsistencies, such as different dates on different pages of a work, or (worse) incorrect titles such I have come across several times while doing the Greek drama data. --EncycloPetey (talk) 02:36, 29 February 2016 (UTC)
I pretty much agree with you re the complexity, especially with no feedback loop. I think that we need a push tool (a gadget?) that allows the form entry of data from the Index: page with a feedback loop to what is in and WD, and/or what is not. Ideally I would like to have the data in WD first, then create a book template, like the creator tool, then create the Index page and continue to populate from the book template, or directly from WD. <shrug> I have been working my way through the works that I have done, and that is my labelling of {{wikidata edition}} so I know the works where I have more fully populated the edition's WD item.
In my limited data set (44 plays), I've come across (a) incorrect titles, (b) incorrect years of publication, (c) missing edition information, and all of that at our end. Now while it is true that wrong data is wrong data, I wouldn't want to be pushing all that information over to Wikidata without a means of ensuring that corrections in one place prompt a correction at the other project. We'd also need some mechanism for distinguishing original first editions of English-native works from all those wroks which are originally non-English and have been translated, or those works which exist here in a later edition than their first. Such works have a different data structure at Wikidata, where there is a central data item for the "original", and edition data items for all other editions, including translations. --EncycloPetey (talk) 15:38, 11 March 2016 (UTC)


User:Kolz2016, User talk:Kolz2016 -- explicite spam. Blocked indef. in ru.Wikisource. Hinote (talk) 01:15, 9 March 2016 (UTC)

Thanks for bringing it to our attention. Prosody (talk) 03:18, 9 March 2016 (UTC)

Attention all admins: Password policy for users with advanced permissions[edit]

Next Tuesday, March 15th, we will begin enforcing the result of the Password policy RFC that took place on meta and closed in January. After Tuesday, users with access to sensitive information, or the ability to change the mediawiki interface will be required to use an 8-byte long password. This affects all users in the Administrator, Bureaucrat, Checkusers, and Oversight groups on all wikis, along with a few other groups on particular wikis.

Users covered by the new policy, whose password is not currently compliant, will be prompted to change their password the next time they login.


Chris (Wikimedia Security Team)

User:CSteipp (WMF), via Wikitech-ambassadors mailing list

notified by — billinghurst sDrewth 00:18, 11 March 2016 (UTC)

Transclusion display change — right margin[edit]

I am looking at some transcluded chapters, eg. Chartism/Chapter 10 and I see that we now have an indented right margin (~90%???) and full formatting rather than left formatting (jagged right edge). Has someone made a local change or have we inherited something along the way? — billinghurst sDrewth 21:56, 14 March 2016 (UTC)

@George Orwell III: do you know where this has occurred? I cannot see local change, and then working out where the formatting occurs is unknown to me. — billinghurst sDrewth 21:59, 17 March 2016 (UTC)
Perhaps something similar with this left margin which was not on my user's page and appeared recently without my knowing how or why? I've had a similar surprise on the French wikisource too. --Zyephyrus (talk) 01:27, 18 March 2016 (UTC)
For whatsoever this is worth (probably not much) this "change" in the right margin is most definitely (Java)Script driven rather than a CSS change (load the page with javascript disabled: result renders fine without restriction on right margin.) I think I may have gotten to the bottom of this although I'll leave others to add the reasoning for why the changes were made:
  • MediaWiki:Gadget-Site.css has specified a right-margin of 3em forever (well since at least May, 2015 which is as far back as I have checked)—per
    body.ns-0 div#regionContainer {
    	position: relative;
    	display: block;
    	box-sizing: border-box;
    	margin-right: 3.00em;
    	margin-left: 3.00em;
  • MediaWiki:PageNumbers.js has attempted to enable the above since forever but using buggy code which appears to have been "fixed" in this change on 1st January, 2016.
The net result is as observed, at least as far as English wikisource is concerned. AuFCL (talk) 04:20, 18 March 2016 (UTC)
Thanks. The issue for me is the header template now is wider than the body component, so when items are centred in both, there is no alignment it looks buggy to my eyes. Does that also explain the fully justified page formatting, rather than the left alignment? — billinghurst sDrewth 06:13, 18 March 2016 (UTC)
Pardon. I understood the "okay with frown" part: I was only reporting the status quo as I observe it: with the expectation it might give the appropriate users the sensitive points at which further change might best be implemented should they (you?) so choose. However I completely fail to understand your comment and thus cannot help further. AuFCL (talk) 06:44, 18 March 2016 (UTC)

PotM administration[edit]

RL has become somewhat busy recently and doesn't look like reducing for at least the next couple of months (it's nothing to do with my health, for those who might panic about this, rather music, work and the enWS widow all need attention). This means that I'm unable to provide administration of the PotM and awards for a while. Could someone else please take over for a bit? I'm happy to answer questions and provide some guidance. Administration of PotM entails making the final decision on which work(s) has been selected for the month, putting the work into the PotM and Collaboration templates, maintaining WS:PotM, welcoming new contributors, and making the monthly participation awards. For this last, I've been keeping an offline list of participants in a Word document. None of the tasks require a sysop level of access, but they seem to fall more naturally to us because we tend to be higher contributors anyway.

I'm yet to do the February awards, but should have time over the Easter break to do so. I have made no record of the March contributions to the Prime Minister's Wives, which started late. Beeswaxcandle (talk) 08:22, 18 March 2016 (UTC)

@Beeswaxcandle: For me, the most important question is, how the monthly work is selected? Is there a pool of works from which you select? Is it based on interest of the community? — Ineuw talk 14:24, 19 March 2016 (UTC)
Selecting is a simpler matter, and isn't at issue. We have a whole discussion page for that, and works are usually discussed well in advance. Beeswaxcandle is asking for someone to handle posting of awards to participants, which is a process he's mostly done himself. This happens after a PotM has concluded, and it requires keeping track of the PotM participants. --EncycloPetey (talk) 16:30, 19 March 2016 (UTC)
BWC's request was understood. I am concerned about the process of selection. The rest is administrative work and that doesn't worry me. — Ineuw talk 16:37, 19 March 2016 (UTC)
Etiquette would be to start a separate thread on that topic, instead of holding that discussion here. BWC is seeking help on a particular issue. --EncycloPetey (talk) 17:15, 19 March 2016 (UTC)
Selection is based on assessing the suggestions and discussions at the PotM talk page. If there is no clear "winner", then I've selected the one that I think will appeal to the widest group of contributors. I've also tried to ensure a variety of topics, style and region of origin, based on the interests of our contributors. Beeswaxcandle (talk) 18:27, 19 March 2016 (UTC)

MediaWiki talk:InterWikiTransclusion.js#update[edit]

Thanks! -Aleator (talk) 22:27, 23 March 2016 (UTC)

Yes check.svg Done Prosody (talk) 06:08, 16 May 2016 (UTC)


The latest 70 category creations of this bot should be deleted as they are malformed (m:Special:Permalink/15447542#Badly_named_Babel_AutoCreate_categories). I don’t want to flood by tagging them for SD so I request it here. If it would be convinient to tag them first, I’m happy to proceed.--Ah3kal (talk) 18:06, 9 April 2016 (UTC)

Yes check.svg Donebillinghurst sDrewth 11:18, 10 April 2016 (UTC)

Proposal to globally ban WayneRay from Wikimedia[edit]

Per Wikimedia's Global bans policy, I'm alerting all communities in which WayneRay participated in that there's a proposal to globally ban his account from all of Wikimedia. Members of the Wikisource community are welcome in participate in the discussion. --Michaeldsuarez (talk) 14:48, 18 April 2016 (UTC)

It would appear as if Cirt had already anticipated this request. AuFCL (talk)
Cirt did block WayneRay, though not for actions undertaken on this wiki. Our policy has been to only block real users for local actions, and the issue of extending blocks from one wiki to another has been discussed, and not to be seen as the expected practice. I commented at the discussion that the local blocking should be discounted in their discussion. — billinghurst sDrewth 04:28, 19 April 2016 (UTC)
I see. Noted in the wrong order. Mea culpa. AuFCL (talk) 04:40, 19 April 2016 (UTC)

Page:Memorial written by Josephine Griffing asking that women be commissioned to assist with the care and education of the freedmen.tif/1[edit]

This page was moved from main namespace to Page namespace ans has wrong content model. Could an admin, please, change its content model using Special:ChangeContentModel? (likely to "Page: page")

Done— Mpaa (talk) 17:57, 22 April 2016 (UTC)

Mpaa confirmation deferred[edit]

Hi folks,

When doing the admin confirmations this morning, I started to put Mpaa up for confirmation, one year having passed since their previous confirmation, but then I realised that they were appointed bureaucrat in October 2015. I figured that the bureaucrat appointment process surely counts as sufficient opportunity for community scrutiny, and therefore I reset their confirmation date to November 2016. Please speak up if you disagree with this.

Cheers, Hesperian 02:00, 1 May 2016 (UTC)

Heartily concur with this approach Beeswaxcandle (talk) 05:36, 1 May 2016 (UTC)
+1 — billinghurst sDrewth 09:56, 1 May 2016 (UTC)
That sounds entirely reasonable. BD2412 T 00:57, 2 May 2016 (UTC)