Wikisource:Scriptorium/Archives/2017-10

From Wikisource
Jump to navigation Jump to search
Warning Please do not post any new comments on this page.
This is a discussion archive first created in , although the comments contained were likely posted before and after this date.
See current discussion or the archives index.

Announcements

Proposals

Bot approval requests

Repairs (and moves)

Other discussions

Cleanup of a source, but making it no longer verbatim

This original source lacks a lot of words, mostly articles. That makes reading this text tiring and difficult. However most can be unabiguously restored. Would it be Ok to create a The Long Telegram (cleaned) page for this? First as a verbatim copy, so that the changes are easily viewed. TriTachionTertiary (talk) 16:55, 27 August 2017 (UTC)

We generally don't host original editions of a work, except for original translations from other languages. At one time, we started discussing the possibility of annotated editions, but that discussion fell flat. So, at this time, the community feeling would be against creating a Wikisource edited copy of a work already in English. --EncycloPetey (talk) 17:05, 27 August 2017 (UTC)
@TriTachionTertiary: I agree with EncycloPetey. I think an emended version would be okay to host at Wikiversity though, and you could link between the two. Sam Wilson 23:47, 27 August 2017 (UTC)
Is Wikiversity the place for annotated works? I thought it was Wikibooks. Is there a consensus on this? —Beleg Tâl (talk) 15:00, 28 August 2017 (UTC)
@EncycloPetey: @Beleg Tâl: @Samwilson: So, Wikiversity, Wikibooks or maybe revive the "how much editing/annotation of pimary sources is ok?" discussions?
I've got no personal sense of whether such a work would be accepted at Wikiversity or Wikibooks, as I don't edit on those sites. My impression for Wikisource is that we'd probably be against the idea, since we never got very far on even the initial discussion. So I would recommend pursuing those other two site options first. --EncycloPetey (talk) 02:18, 26 September 2017 (UTC)
Yeah, I don't think it's suitable here on Wikisource. But I do think it'd be a good project for Wikiversity, especially if it were to also perhaps some other commentary about the work. Of course, I don't speak for the Wikiversity community (but I have done a little bit over there; never run in to any drama); I'd say give it a go. Perhaps we could have a section of Wikiversity for "annotated editions" of Wikisource works! :-) (Oh, and I don't think Wikibooks would be the place it, unless the work as created there was mostly commentary about the text; they make textbooks, not "students' editions" or whatever they're called.) Sam Wilson 07:24, 26 September 2017 (UTC)
Thanks for your input, I will try it here TriTachionTertiary (talk) 15:13, 27 September 2017 (UTC)

Category:Index - File to check

Not urgent, but does anyone want to take on page-listing the last few remaining entries (around 25 or so) in this category?

It feels like I've done an awful lot of them from when it had reached a backlog of over 100 a few years ago.

The remaining items, are mostly items I don't feel happy working with for copyright reasons, or because someone else had previously indicated that they would be working on them in due course.

It would be nice to empty the category before the end of the year, though.

ShakespeareFan00 (talk) 15:19, 19 September 2017 (UTC)

It's now down to 2 PDF files: One of which has been replaced by a DJVU and the other the wiki seems to balk at loading pages from. -Einstein95 (talk) 07:02, 22 September 2017 (UTC)
@Einstein95: For Index:ManualofPalestineanArabic.pdf: the pages do load in my computer, but it is taking more time than is usual. Hrishikes (talk) 07:50, 22 September 2017 (UTC)
@Hrishikes: Yeah, but since the file comes from IA, a DJVU is better for proofreading IMO. I have already fixed teh reason why the other PDF wasn't loading pages (20MB PDF of uncompressed files from a PDF). -Einstein95 (talk) 10:36, 22 September 2017 (UTC)

Index namespace malfunctioning

Tracked in PhabricatorTask T176393

I've had two disturbing problems in the Index namespace in the past 24 hours.

(1) Yesterday evening, Index namespace pages would not load, and kept generating errors.

(2) Just now, I have been trying to add auto header text, but adding the text blanks the entire page so that all the content is removed.

Is anyone else having similar difficulties? I have had this issue from three different computers. --EncycloPetey (talk) 17:09, 20 September 2017 (UTC)

Addendum: I haven't had any additional odd results in the past few hours. Perhaps some update was propagating and causing the odd behavior temporarily? --EncycloPetey (talk) 21:07, 20 September 2017 (UTC)

I have Index pages simply show up as pages with the raw template when editing. Also had one index "fail to submit all the data" (paraphrasing) which showed basically a blank index page on preview. -Einstein95 (talk) 12:03, 21 September 2017 (UTC)
According to a post in the IRC channel, this is caused by https://phabricator.wikimedia.org/T176393. -Einstein95 (talk) 12:07, 21 September 2017 (UTC)
Not sure if it's related, but the side-by-side editing interface disappeared for me when editing in the Page namespace; it was replaced with the regular article-editing interface. The issue went away when I defaulted my Preferences so I'll be trying to narrow down which gadget broke when I get a chance. Beleg Tâl (talk) 12:08, 21 September 2017 (UTC)
@EncycloPetey, @Beleg Tâl: I have found disabling the "Two column edit conflict" beta feature to work for me, can you try with it on and off? -Einstein95 (talk) 12:11, 21 September 2017 (UTC)
The "Two column edit conflict" beta feature is indeed breaking Page: and Index: editing interface (thank you Einstein95 (talk • contribs) for the finding). If you disable this beta feature then the editing interface should work. If not, then there is yet another bug. Tpt (talk) 13:03, 21 September 2017 (UTC)
thank you , it was also breaking the page status buttons. and showing "noinclude" wikicodeSlowking4 ‽ SvG's revenge 13:18, 21 September 2017 (UTC)

Author pages ugly in mobile view

I had a look at author pages using mobile view today, and they are not that pretty.

it indicates to me that we need to shuffle components around for our mobile view where possible, otherwise we have a bigger job to do. First thoughts are for mobile view to

  1. Get rid of the backlink to "Author Index:Xx"
  2. Consider whether we relocate the interwiki links in {{plain sister}}
  3. Cull "listen" template from display

— billinghurst sDrewth 13:58, 21 September 2017 (UTC)

My thoughts:
  1. Author index would look ok provided that the index was above or below instead of beside the author name. It's kind of a useless feature though and I wouldn't miss it if it were removed from both mobile and desktop.
  2. Plain sister looks ok when it is full width, looks terrible when it's not full width
  3. I think "listen" template looks ok, and I don't like when info available on desktop is hidden on mobile.
Just two cents. Beleg Tâl (talk) 15:57, 21 September 2017 (UTC)
have you looked at wikisurfer4wikisource app? maybe we need an app from a hackathon, or switch to MinervaNeue skin for mobile?
author template ; sister project box does not work for mobile needs a UX redesign, to drop down or pinch in. Slowking4 ‽ SvG's revenge 17:14, 21 September 2017 (UTC)
My thoughts:
  1. I find the Author backlink incredibly useful, and use it often, though we might find a better way to implement it.
  2. Can the sister interwiki links be made collapsible?
  3. The listen template is due for an overhaul. A much simpler display would suffice for Author pages. But if we think the presence of the speaker icon is explanatory in itself, then we need not explain it in the header.
--EncycloPetey (talk) 17:15, 21 September 2017 (UTC)
The speaker could have a tooltip with an explanation "This text contains a spoken word version"; and on mobile the tooltip could be a popup or an inline note. If this is technically feasible, it could be a good replacement for the note in the header. Beleg Tâl (talk) 17:39, 21 September 2017 (UTC)
Lengthy popup text becomes a problem in mobile devices, since it blocks view of the page and of scrolling, zooming, etc. --EncycloPetey (talk) 17:44, 21 September 2017 (UTC)
i know, but the problem here is a template with left align text clashing with right align text. maybe we need to change to all left align (not as elegant in desktop, but better for mobile?) a question of real estate - need some mobile designer experience. Slowking4 ‽ SvG's revenge 19:12, 21 September 2017 (UTC)
Or perhaps the speaker icon could be combined into something else so that no alignment is necessary. --EncycloPetey (talk) 20:09, 21 September 2017 (UTC)
I would argue that for mobile view that it is less necessary to identify that an author has podcast works, compared to identifying the individual works have podcasts. — billinghurst sDrewth 22:17, 21 September 2017 (UTC)
I don't buy that argument. Having an icon at the outset, letting the user know that audio versions exist, before scrolling through the entire page, is a plus. In my experience, mobile readers are more likely to make use of audio versions precisely because the screen is too small for reading long texts. --EncycloPetey (talk) 01:08, 22 September 2017 (UTC)


 Comment We should be referencing mw:ResourceLoader/Writing a MobileFrontend friendly ResourceLoader module, mw:Mobile web projects/Infoboxes, and mw:Reading/Mobile Friendly Content — billinghurst sDrewth 04:24, 22 September 2017 (UTC)

@Jdlrobson: as this is an area with which you have expertise, would you be so kind to point us towards where we should be working towards, and how to best thing about mobile design. We are definitely old school and preferring transcribing than dabbling in css. So simple and direct pointers would help us start the long haul process that we have to migrate. — billinghurst sDrewth 04:36, 22 September 2017 (UTC)

Tech News: 2017-39

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Recent changes

  • Wikitext syntax highlighting is a beta feature on Wikimedia wikis with text written from left to right. This beta feature is based on CodeMirror. You can now search through the entire article with CTRL+F or cmd+F when you edit. Before it just searched through a part of the article. The developers are also fixing a couple of other bugs. [1][2]
  • Administrators on wikis that use Structured Discussions as a beta feature or by default can now create and move Structured Discussions boards. Structured Discussions was previously called Flow. [3]

Changes later this week

  • You now get a notification when someone tries to log in to your account and fails. If they try from a device or IP address that have logged in to your account before you will be notified after five failed attempts. For security reasons you will soon get an email by default when someone tries to log in to your account and when someone logs in to your account from a new device or IP address. This can be turned off in your preferences. [4]
  • Users with extremely old browsers (for example Netscape 2–4, released from 1995–1997) which do not support Unicode will no longer be able to edit. They should try to install a new browser. [5]
  • The new version of MediaWiki will be on test wikis and MediaWiki.org from 26 September. It will be on non-Wikipedia wikis and some Wikipedias from 27 September. It will be on all wikis from 28 September. (calendar).

Meetings

Future changes

  • The Wikimedia Foundation Readers department has proposed changing the mobile website. You can read more on mediawiki.org. This would be a big change.
  • You can't use OCG to create PDFs after 1 October. This is because of technical problems. You can still create PDFs. Tech News 2017/37 said the function to create PDFs from books would still work. It is now clear it will not work for the next few months. The developers are working on fixing this. You can read more on mediawiki.org.

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

15:59, 25 September 2017 (UTC)

Code4Lib pre-conference workshop

hey, user:Londonjackbooks, User:Econterms, if i submit a pre-conference workshop, "Wikisource 4 Lib" for Code4Lib, will you help me out? pre-conference is Feb 13, 2018 at LOC,[6] and deadline is Sept 30. Slowking4 ‽ SvG's revenge 19:05, 26 September 2017 (UTC)

Help out in what way? Londonjackbooks (talk) 19:47, 26 September 2017 (UTC)
tutor librarians, i can do the hand waving in front of projector for a tutorial, but if there is a lot of interest, then 1 is not enough for editing help. it is a weekday, if you are free, i would be grateful. maybe we can get WMDC to buy lunch. Slowking4 ‽ SvG's revenge 00:12, 27 September 2017 (UTC)
OK. I can help. My current approach to editing is somewhat undisciplined, but there's plenty of time to coordinate & streamline my technique between now and then. Londonjackbooks (talk) 00:54, 27 September 2017 (UTC)
thanks, the main thing will be to engage librarian editors and answer their newbie questions, as they edit. i put in proposal - we will see what they say. if you have a list of works to start newbies on, let me know. Slowking4 ‽ SvG's revenge 04:15, 27 September 2017 (UTC)

LintErrors extension incompatible with {{nop}} approach used for tables..

This page https://en.wikisource.org/w/index.php?title=Page:Rothschild_Extinct_Birds.djvu/16&action=edit is showing up on the list here https://en.wikisource.org/wiki/Special:LintErrors/fostered?namespace=104

Other than the {{nop}} there is NO fostered content on the page, and in the absence of anything saying otherwise {{nop}} is how table continuations are currently implemented, the parser being unable to otherwise handle constructs like this gracefully. Is it unreasonable to ask that a long-term fix is implemented, so that something which clearly isn't an error, isn't being reported as such?

Of course it would be really, really nice if someone took an axe to the parser, and re-wrote it so that it coped more gracefully with constructs that are transcluded from multiple pages, but I don't see that happening any time soon. ShakespeareFan00 (talk) 13:40, 27 September 2017 (UTC)

Doing it this way seems to be fine. It is the approach I use here. Suzukaze-c (talk) 02:22, 28 September 2017 (UTC)
Which was how I thought I used to do it, until someone claimed that broke other stuff, like page numbering if I recall correctly... ShakespeareFan00 (talk) 10:11, 28 September 2017 (UTC)
What you did worked, but it's not to me at least clear why it didn't break, when it has in the past.ShakespeareFan00 (talk) 10:13, 28 September 2017 (UTC)
In any-case if you can resolve this in that work feel-free:- https://en.wikisource.org/wiki/Special:LintErrors/fostered?namespace=104 has plenty of tables with a simmilar issue.ShakespeareFan00 (talk) 10:14, 28 September 2017 (UTC)
That is correct they have this terminating a table (table row start then no cell start)

|-
{{nop}}

which was not the guidance until someone incorrectly changed it at some point without good reference to the community. It is exactly what breaks the page numbering and why it was not to have been used (and why we reinstated the original guidance). — billinghurst sDrewth 02:17, 29 September 2017 (UTC)
Part of the problem in the past is that many editors habitually remove blank lines like those, at the top of a page or top of a footer, because they are usually wrong to include, and so bots and AWB will likely remove them too. So how to we indicate these purposeful blank lines that are necessary for proper transclusion? --EncycloPetey (talk) 17:43, 28 September 2017 (UTC)
which is why we have the nop in place, among other reasons. — billinghurst sDrewth 02:11, 29 September 2017 (UTC)
  •  Comment That is not a correct conclusion. If people follow the table continuation guidance you will find that there is no issue.

    In the original page mentioned, it was not /16 that was the issue, it was the end of /15 which was problematic as it closed with a new table row creation. So the problem was that there was two row creations in a row without any cell creation. So just follow the guidance of no extra row creations to terminate, then start with a {{nop}} and then the row creation. — billinghurst sDrewth 02:09, 29 September 2017 (UTC)

      • Can someone therefore please impose ONE CORRECT way of doing continued tables across all works on Wikisource such that it does NOT generate LintErrors, does NOT put in spurious line-feeds and will work CONSISTENTLY until the developers break something else, Thank you? Having moved works back and forth based on apparently changing guidance is disruptive and unhelpful.

ShakespeareFan00 (talk) 10:20, 29 September 2017 (UTC) 

approach are not compatible.ShakespeareFan00 (talk) 11:38, 29 September 2017 (UTC)
  •  Comment Okay I can see that it is kicking an error for that lint filter in the Page: namespace, due to the use of the &<div>></div> and with the way that we transclude that disappears when transcluded. I wouldn't fuss it, is an inconsequential error and will break nothing where it basically takes a blank div and pokes it ahead of the table in the page namespace where we have empty space above the table. It disappears with transclusion. We can look at it in time, to see if there is a better solution now that they have changed the renderer. — billinghurst sDrewth 12:21, 29 September 2017 (UTC)

Chronological Table and Index of the Statutes (including subpages)

Persuant to the concerns raised about table continuations above, I note that this work is incomplete, and is using at least 3 different composting styles to grab part of the table. None is ideal.

Many of the issues in trying to get this to transclude and display correctly, are related to a weakness in Mediawiki markup, namley that it can't gracefully cope with constructs that are split across pages, in a manner that is consistent or compatible with the additional content the Proofread page extension adds for things like page numbering.

To allow for properly displaying a work like this nicely, using <pages> with section paramaters as it should be done, is apparently beyond the capacity of mediawiki/proofread page currently. As I've not seen any interest in tackling the long-term issues as a priority, I am considering asking for the partial transclusion to be deleted until such time as Wikisource contributors can come up with a long-term solution that doesn't break inconsistently due to minor changes.

It should not require a practical cloning of the functionality of a <pages> tag in complex template markup to do something like 

{{header}}
<pages>
{{footer}}

consturctions.

Perhaps compositing a table that needs to be split-up like in this work is something that could better be done in a Module: ?

The actual {{statute table}} template itself should ideally also be simplified (and made a module if needed)


ShakespeareFan00 (talk) 12:03, 29 September 2017 (UTC)

Global Collaboration products newsletter: 2017-09

Latest news from the Wikimedia Global Collaboration team, about Notifications, Structured Discussions, Edit Review Improvements and Content translation. Please tell other users about these changes. Not all changes will affect you.

What's new?

  • This newsletter is the first for the Global Collaboration team!
    • Global Collaboration team formed in July by merging the Language and Collaboration teams of the Wikimedia Foundation. The team is part of the Contributors department.
    • The newsletter now includes updates about Content translation. The Global Collaboration team now have an extended report (in English), with more information and data about the different projects we take care of.
  • Due to vacations and Wikimania, no newsletter has been written in August.
  • Major changes, with possibly an important impact on the wikis, are now marked with the star icon in the newsletter. Let us know if you like it.

Edit Review Improvements [More information • Help pages]

Recent changes

  • Filters for Recent Changes will be deployed by default on all wikis. [7]
    • Some features for Recent Changes remain on Beta. See the list.
    • The initial deployment was scheduled for September 26. It has been postponed after October 3 for performance issues. [8]
  • The filters for Watchlists, as a Beta feature, have been deployed as a Beta feature on Sept. 19. [9]
  • "User registration" and "experience level" groups have been unified. It is now possible to select only actions from logged-out users or from logged-in users. [10]

Content translation [More information • Help pages]

Recent changes

  • Delete action on the list of translations is now more reachable. Now it's possible to open it on click, and easier as well, because active area is bigger. [11]
  • Content Translation language selector is restyled to be more compact. [12]

Future changes

  • The translation editor of Content Translation will be replaced by the visual editor. Work is ongoing. [13]

Notifications [More information • Help pages]

Recent changes

Structured Discussions [More information • Help pages]

Recent changes

Collaboration team's newsletter prepared by the Global Collaboration team and posted by bot • Give feedback • Subscribe or unsubscribe.

17:10, 29 September 2017 (UTC)

Wikimedia Movement Strategy phase 2, and a goodbye

Hello,

As phase one of the Wikimedia movement strategy process nears its close with the strategic direction being finalized, my contractor role as a coordinator is ending too. I am returning to my normal role as a volunteer (Tar Lócesilion) and wanted to thank you all for your participation in the process.

The strategic direction should be finalized on Meta late this weekend. The planning and designing of phase 2 of the strategy process will start in November. The next phase will again offer many opportunities to participate and discuss the future of our movement, and will focus on roles, resources, and responsibilities.

Thank you, SGrabarczuk (WMF) (talk) 21:55, 30 September 2017 (UTC)

verse template question

I plan on working on this Page:Leeser-Bible-1853.djvu/13. Notice how the verse numbers aren't superscripted as they usually are in other bibles. I'd like to be able to use the formatting as it is in the original. Is it possible for the {{verse}} template to be edited to be able to get rid of the sup tags on demand? I think it would be a nice addition to the template to be able to do this. Jpez (talk) 12:21, 1 October 2017 (UTC)

How about the method in Page:The Holy Bible Vol 1 (Thomson).djvu/11? Hrishikes (talk) 16:29, 1 October 2017 (UTC)
{{verse|1|align=left}} seems to replicate what you want also. Slowking4 ‽ SvG's revenge 00:10, 2 October 2017 (UTC)
I needed something more specific so I ended up making a new template called {{versell}} which is basically {{verse}} without the superscript and with and a little added gap after the verse number. thanks Jpez (talk) 04:09, 2 October 2017 (UTC)
ok, but please add some documentation and categories. Slowking4 ‽ SvG's revenge 13:13, 2 October 2017 (UTC)

Guru Granth Sahib

Was this meant to be an Author page? There is no license information either way. --EncycloPetey (talk) 17:09, 3 October 2017 (UTC)

@EncycloPetey: It is a book, not an author. See w:Guru Granth Sahib. Hrishikes (talk) 00:28, 4 October 2017 (UTC)
OK, thanks. Then it just needs a source and a license tag. --EncycloPetey (talk) 00:45, 4 October 2017 (UTC)
The first complete translation is of 1960 vintage (see under Translations in the WP article) and therefore would come under copyright. It is in IA, all 4 vols: 1, 2, 3, 4. Hrishikes (talk) 01:11, 4 October 2017 (UTC)
This one is the Khalsa Consensus Translation (2000) by Sant Singh Khalsa and is definitely copyrighted. Beleg Tâl (talk) 02:06, 4 October 2017 (UTC)

Volume 3 of the DNB

Just needs one page to be validated before it is nominally all validated. Can someone do the honours? See here Victuallers (talk) 09:12, 4 October 2017 (UTC)

Done — billinghurst sDrewth 11:06, 4 October 2017 (UTC)
Thanks - nice to see it done Victuallers (talk) 09:16, 5 October 2017 (UTC)

Could this a be another bug, due to the latest wmf software update?

I was checking some pages of PSM in the main namespace, and it seems that the page numbering is not showing up at the left margin or along the text. (I am aware of the page number display controls in the left margin menu.) Could someone confirm this by looking at some randomly selected main namespace work? Thank you. — Ineuw talk 06:10, 8 October 2017 (UTC) (aka: The canary in the word mine) Sorry for the post, I got it functioning. — Ineuw talk 06:12, 8 October 2017 (UTC)

Index:Ch'un Ts'ew Pt I.pdf and related volumes

use {{abbr}} instead of {{tooltip}} (an affront to semantic HTML) and {{nowrap|[ ''Chinese text'' ]}} instead of pre-existing templates such as {{Chinese missing}}. Maybe more problems but I didn't look too closely.

If you have solutions to problems then suggest them. Remember that many here are just transcribers and never bury themselves into the arcane black magic of html, semantic or otherwise. — billinghurst sDrewth 12:55, 9 October 2017 (UTC)

(A more minor issue is that I disagree with the naming of the template {{lzh}}; lzh is the ISO code for Literary Chinese, but here it is being used as an abbreviation for {{Legge Chinese}}, a template with an overly specific name that does little more than merely link to Wiktionary) Suzukaze-c (talk) 02:59, 21 September 2017 (UTC)

if it is problematic we can fix it and then get a bot to go through the requisite pages. Personal choices are sometimes not perfect in retrospect. — billinghurst sDrewth 12:50, 9 October 2017 (UTC)

Tech News: 2017-40

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Recent changes

  • You can now use preload features in the visual editor. This means you can create links to create new pages and some text will already be in the edit window when someone clicks on it. [18]

Changes later this week

  • If you have turned on "Automatically enable most beta features" you will now get the 2017 wikitext editor. Before this you had to turn it on manually even though it is a beta feature. [19]
  • Special:Block and Special:Unblock will get the OOUI look. [20]
  • The jQuery library will be upgraded to version 3 on all Wiktionary wikis from 2 October. It will be on all non-Wikipedia wikis from 4 October. It will be on all wikis from 5 October. [21]
  • The new version of MediaWiki will be on test wikis and MediaWiki.org from 3 October. It will be on non-Wikipedia wikis and some Wikipedias from 4 October. It will be on all wikis from 5 October (calendar).

Meetings

  • You can join the next meeting with the Editing team. During the meeting, you can tell developers which bugs you think are the most important. The meeting will be on 3 October at 19:00 (UTC). See how to join.
  • You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 4 October at 15:00 (UTC). See how to join.

Future changes

  • A new feature called Reading Lists are coming to the Wikipedia mobile apps. First it will be in the Android version. Reading Lists are a private user preference where you can see lists of articles on multiple devices. You can give feedback on this feature. [22]
  • The search function has used fallback languages for language analysis. This means a language could be analysed with the grammar of a completely unrelated language to find related words. This will be removed from most wikis the week starting with 9 October. You can read more and give feedback.

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

23:25, 2 October 2017 (UTC)

Filter changes

The latest update has filled most of the first screen of Recent Changes with filter selection options. Is there any way that all this could be minimized, hidden, or set in a drop-down menu, so that the initial screen view will show actual changes instead of just links and filter settings? --EncycloPetey (talk) 20:38, 5 October 2017 (UTC)

@EncycloPetey: I would suggest that you add a request to phabricator: so that these options can be collapsed. — billinghurst sDrewth 11:54, 9 October 2017 (UTC)

On a related note: The link under Utilities that used to display "IP's contributions" is no longer functioning. --EncycloPetey (talk) 13:45, 6 October 2017 (UTC)

??? where? which skin? which page? — billinghurst sDrewth 11:57, 9 October 2017 (UTC)

Page numbers vanish

I'm suddenly getting odd behavior with page numbers in the Main namespace. When I visit a chapter of a work, no page numbers are displayed, even though I have it set to "Page links beside text". Neither refreshing the page nor purging has any effect; still no page numbers are displayed. However, if I toggle to "Page links within text" the page numbers appear, and if I toggle back to "Page links beside text" the page numbers appear as well. But the next time I visit the page (or go to another page) the page numbers disappear again, and I have to toggle all over again. This behavior has begun within the past hour. --EncycloPetey (talk) 21:38, 4 October 2017 (UTC)

Addendum: I see from this conversation that other users are experiencing the same difficulties. --EncycloPetey (talk) 21:03, 5 October 2017 (UTC)

@EncycloPetey: Toggling "Page links beside text" will get them back on the pages in question, though it isn't retained when page is reloaded. No idea why, though it is evidently a javascript issue. — billinghurst sDrewth 03:24, 6 October 2017 (UTC)
In my case, it is retained (with both Page links displayed and Page links within text options) when I reload or even move to another chapter. But it disappears on reloading with Page links beside text option. Hrishikes (talk) 11:38, 6 October 2017 (UTC)
I wonder if this is a local problem, or if the issue is occuring cross-wiki (i.e. multilingual, DE-WS, FR-WS)? Determining that could be important. Inatan (talk) 19:00, 7 October 2017 (UTC)
It looks like in our local MediaWiki:PageNumbers.js the function pagenumbers.init() is being called before layout.init() when it should be the other way around, maybe due to a change in mw.loader jQuery? I've never touched it before and don't want to start now by breaking anything but maybe they could be made to be called sequentially after the asynchronous thingy rather than separately asynchronous? Prosody (talk) 00:12, 9 October 2017 (UTC) (modified 00:26, 9 October 2017 (UTC))
@Samwilson: are you around and have the jquery skills? — billinghurst sDrewth 12:04, 9 October 2017 (UTC)
I've removed the layout promise, and it seems to work. I'm not sure if that's a unstable way... the whole module could do with a bit more rigour I think! But seems to be working now? Sam Wilson 12:45, 9 October 2017 (UTC)
Thanks. Fixed for me. — billinghurst sDrewth 13:06, 9 October 2017 (UTC)

What constitutes source as per {{No source}}

I have had some disagreement with another editor about the definition of source, as envisaged in the No-source template, in case of a piece extracted from a bigger work. As I see it source encompasses: 1. Scan 2. url to outside copy 3. Mention that it was manually taken from a hard copy 4. mention that it was taken from a later compilation, with, if possible, url to that compilation. The other editor thinks that mentioning the parent work is equivalent to giving source, although, the template gives this definition of source: "Source" means a location at which other users can find a copy of this work. The community can see the discussion at my talk page (also edit summary of the article). An opinion is requested. Hrishikes (talk) 04:30, 8 October 2017 (UTC)

@Hrishikes: and I are having a disagreement over what constitutes a "source" and the meaning of the word "location" in {{no source}}. Central to this disagreement is the following statement of Usage in the template's documentation:
This template is used to mark works which do not have a declared source. In this case, "source" means a place where a contributor new to the work can find a reliable copy of the work. Generally this is a URL. If the data is known, please enter it to the talk page, and more usefully using the {{textinfo}} template.
I maintain that the template documentation demonstrates it is not needed on a work such as the Letter to Thomas Auld by Frederick Douglass, because the source is given in the header and on the associated talk page as "first published in The Liberator, September 22, 1848." This statement constitutes an identification of a place where a contributor can find a reliable copy of the work. I support this view with the final paragraph of Help:Beginner's guide to sources, which explains how users may add works from a hard copy of a printed work.
Hrishikes maintains that a "source" is limited to only two options: scanned copy and url. If there is no scan nor a url, then the {{no source}} template should be used. I do not follow his reasoning, but he has written at length in response on his talk page, and may choose to add some (or all) of that response here. --EncycloPetey (talk) 04:34, 8 October 2017 (UTC)
requiring an online source (or scanned) is a commons interpretation of sources, not a bibliographic one. it is normally a prelude for deletion there. we need a "source needs improvement" for these cases; we are migrating "copy paste from the web" to scan based page transcriptions; but the cases are different between "we do not know where this is from", rather than "we know where it is from but do not have a scan"; and the cases require different actions. Slowking4 ‽ SvG's revenge 16:14, 8 October 2017 (UTC)
  •  Comment [It is an old template, so its meaning should be considered to be broad.] I think if we are seeing someone doing a copy and paste, we are expecting a url. If we are seeing a slowly developing typed work then we know that it will be a transcription. So we are asking for a source that best represents where the contributor got the work. Our desire is for a source that allows us to determine the veracity and accuracy of the contribution, and to also get it proofread, so we aim for as high as we can get.

    On a personal note I would hope that nobody would be pasting "no source" for my typed transcriptions from The Times, I won't be uploading all the little images which are of poor quality. — billinghurst sDrewth 11:51, 9 October 2017 (UTC)

Old magazines are difficult to come by in comparison to old books. So I believe that the contributor might have got the piece not from the magazine where it was first published, but from a later compilation, for example, from a volume subsequent to this one. Only first publication data is mentioned in the article, nothing has been stated specifically about wherefrom the item was taken. The reader has no scope of verification. First publication data is not equivalent to giving source, as I see it. (I usually take care to give exact source, e.g. here, so that verifiability is enabled to the reader.) Hrishikes (talk) 12:19, 9 October 2017 (UTC)
I was talking at the theoretical level, and my experiences, not on the circumstances in question. We have to make the assessment of where the contributor got the work, and push for maximum detail. — billinghurst sDrewth 12:47, 9 October 2017 (UTC)
an intermediate compilation may well be true, but an off line source is a source, and gives a verifiable reference. this is the way scholarship was done. we need to model the better practice of taking photos/scans as backup, but we should have a separate work flow to scan for offline sources, as opposed to researching where unsourced texts came from. Slowking4 ‽ SvG's revenge 15:16, 9 October 2017 (UTC)

Tech News: 2017-41

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Recent changes

Problems

  • You will not be able to edit or upload files to Commons for around 30 minutes on 11 October. This will start at 06:00 UTC. This is because of work to let Commons show 3D models. [25]

Changes later this week

  • The font in the edit window is monospace by default. Today you can change preferences to another font family or use "the browser default" instead of monospace. Switching to the browser default has only changed the font on Macs. The option to use browser default will be removed this week. [26]
  • If you want to print a page or make a PDF of a page you can do so in the side menu. The way the pages look when you do this will change. You can read more on mediawiki.org.
  • The new version of MediaWiki will be on test wikis and MediaWiki.org from 10 October. It will be on non-Wikipedia wikis and some Wikipedias from 11 October. It will be on all wikis from 12 October (calendar).

Meetings

  • You can join the next meeting with the Editing team. During the meeting, you can tell developers which bugs you think are the most important. The meeting will be on 10 October at 19:00 (UTC). See how to join.
  • You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 11 October at 15:00 (UTC). See how to join.

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

14:20, 9 October 2017 (UTC)

Explaining source for an alternative word indicated by Template:SIC

Sometimes when proofreading An Exposition of the Old and New Testament (1828) I come across a word that appears to be a typo in the original, but I don't know immediately what it should have been. In that case I look at an older edition and a newer edition which I have to hand, and, making sure they agree, add a SIC. But I am disappointed that there seems to be nothing in the SIC syntax that enables me to state why I have done this. Am I missing something? Is there some way of adding hidden text that would be appropriate?--PeterR2 (talk) 08:55, 10 October 2017 (UTC)

No, the template does not have provision for an explanation, but the correct word is hover text only. So if you don't know the correct word but want to give an explanation, you can put the explanation in parameter 2, in place of the correct word, which will appear as a hover text. Although the template is not meant for this function, {{Tooltip}} would be better. If you want truly hidden text, beside the word you can use <!-- explanation --> Hrishikes (talk) 10:11, 10 October 2017 (UTC)
+1 to this, use <!-- HTML comments --> if you want to leave an quick explanation for the reference of future editors. If you need a longer explanation, put <!-- see talk page --> and put your explanation on the talk page. Beleg Tâl (talk) 12:48, 10 October 2017 (UTC)
If you want the readers to see your explanation, one other way is to footnote it with a user annotation:

<ref>{{user annotation|your explanation here.}}</ref>

That's far from being a preferred method on Wikisource but I think it has its place. Mudbringer (talk) 00:14, 11 October 2017 (UTC)
I don't agree that user annotation has its place in this case—any more than my taking a pen to a typo or making margin/footer notes in a local physical library book. It would become a "user annotated text" by definition, and proposed policy/guidelines state that "User annotated versions of works on Wikisource must ... have an unannotated 'clean text' elsewhere on Wikisource". Londonjackbooks (talk) 00:34, 11 October 2017 (UTC)

Archive.org found a legal loophole

Please see https://blog.archive.org/2017/10/10/books-from-1923-to-1941-now-liberated/ . I wonder what this means, if anything, for Wikisource. This *is* a library, after all. NMaia (talk) 11:59, 11 October 2017 (UTC)

There's already a thread up above on this issue, under Proposals. --EncycloPetey (talk) 13:17, 11 October 2017 (UTC)
Link: #Consider Wikisource a library (for U.S. copyrite law) —Beleg Tâl (talk) 13:44, 11 October 2017 (UTC)

Template:Author bug

On pages such as Author:G. R. O'Reilly, authors with no death year listed here or at d: show up as living. In this case, someone published in 1891 is very likely not alive, so we should probably pages such as this added to a tracking category such as Category:Authors with missing death dates rather than default to be alive. —Justin (koavf)❤T☮C☺M☯ 08:18, 13 October 2017 (UTC)

Also, I filed an external bug at phab:T178143 for those who spotted that the page's title is malformed. —Justin (koavf)❤T☮C☺M☯ 08:24, 13 October 2017 (UTC)
  • It's true! But sadly there isn't any way to determine from an Author page what the dates of their published works are. In this example, we can add birth and death dates to G. R. O'Reilly on Wikidata (d:Q18911925), and they are no longer in Category:Living authors (I've just done it). We can add birth and death dates on Wikidata even if we have no idea of when they died — well, we usually know within a century, but even if we don't know that we can give them the special Wikidata value of 'unknown'. And as for categories here, we do already try to extract a year from the birth date, and if it's longer than 110 years ago we don't class them as living. So this currently certainly errs on the side of more living authors than is correct but I think this is somewhat due to the scary legal stuff around data about living people. Perhaps we need to do more to direct editors here to add birth and death dates to Wikidata. A link with instructions or something (on the Author header itself, visible to anyone with the page on their watchlist or something)? I dunno what the best fix is. Sam Wilson 08:40, 13 October 2017 (UTC)
The correct way to handle this is to utilise dates = component of the header and add flourishing data. @Samwilson: maybe we can look to adjust by where no date of birth or death that we can utilise either floruit or the pair work period start and work period end—whichever is used at WD—and plug them into dates parameter. It is my opinion that is butt ugly and unhelpful to just plug in vague birth and death dates of 19thC and 20thC. If we are looking to fill in dates of life having them filled with vague dates is no longer an indicator locally. I much prefer that we identify them as missing.— billinghurst sDrewth 04:16, 14 October 2017 (UTC)
Hm, yes you're quite right! I'm sorry. And I totally forgot that flourit dates are supported (they already work in the Author module). Sam Wilson 12:44, 14 October 2017 (UTC)

Retroactive automated autopatrolling?

I'm looking a large number of similar unpatrolled edits that I'd've added to User:Wikisource-bot/patrol whitelist if I had known in advance. Anything to do with them besides going through them one by one? Prosody (talk) 22:00, 15 October 2017 (UTC)

Not sure that I understand the question @Prosody:. If you are looking for patrolled edits by the bot, this should work special:log/patrol/Wikisource-bot, if you are wanting to see who did those patrolled edits, then click the number to see the edit. Skip anything marked "automatically ..." as they will be the bots own edits. — billinghurst sDrewth 12:37, 16 October 2017 (UTC)
I was unclear, sorry. Wikisource-bot takes care of patrolling edits matching whitelisted filters as the edits are made, right? Is there any way to have edits matching a whitelisted filter patrolled after the fact? Prosody (talk) 01:01, 17 October 2017 (UTC)

en.wikisource vs wikisource

whats the difference? Artix Kreiger (talk) 13:53, 16 October 2017 (UTC)

English Wikisource (en.wikisource.org), is for works in English and similar languages like Anglo-Saxon and Scots. Multilingual Wikisource (wikisource.org) is for works that do not belong to a specific-language Wikisource, usually because no specific Wikisource exists for that language, or because the work is itself multilingual, or because of copyright restrictions on the language-specific Wikisource. Multilingual Wikisource also provides a portal to the various language-specific Wikisources. —Beleg Tâl (talk) 14:22, 16 October 2017 (UTC)

Tech News: 2017-42

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Recent changes

  • When you edit a template with the visual editor it follows the template's format configuration on how to save it. You can now change the format in more powerful ways if your wiki wants it. [27][28]
  • The TwoColConflict extension is a new way to solve edit conflicts. It makes it easier to copy and paste the relevant text to the text field. It has been a beta feature for the last five months. You can now try out the new interface on a page where you will actually get simulated edit conflicts. You can give feedback to the developers.

Problems

  • There is a problem with recent changes pages and your watchlist that show a large number of pages. Until it is fixed, Wikidata edits will not be shown on recent changes or in your watchlist on Commons or Russian Wikipedia. If necessary, Wikidata edits will be removed from all wikis. [29]
  • Notifications are not working for some actions. For example some users don't get a notification when they are mentioned. The developers are working on fixing this. [30]
  • Users on some wikis could not change their preferences. This has now been fixed. [31]

Changes later this week

  • Editors and readers who still use Internet Explorer 8 on Windows XP will not be able to use Wikipedia from 17 October. Internet Explorer 8 on Windows XP can't connect securely to the wikis. When we allow them to do so it means that we get less security for everyone else. If you use Internet Explorer 8 on Windows XP you can install Firefox 52 ESR instead. Around 0.1% of the traffic to the Wikimedia wikis comes from Internet Explorer 8 on Windows XP. [32]
  • The new version of MediaWiki will be on test wikis and MediaWiki.org from 17 October. It will be on non-Wikipedia wikis and some Wikipedias from 18 October. It will be on all wikis from 19 October (calendar).

Meetings

  • You can join the next meeting with the Editing team. During the meeting, you can tell developers which bugs you think are the most important. The meeting will be on 17 October at 19:00 (UTC). See how to join.
  • You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 18 October at 15:00 (UTC). See how to join.

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

15:31, 16 October 2017 (UTC)

Open access papers and annotation

A technical proposal is posted for discussion at d:Wikidata:WikiFactMine/Annotation for fact mining. For more general context, see w:User:Charles Matthews/Facto Post#Editorial annotations.

Wikisource rightly comes up in these discussions: see the proposal's talk page. Please contribute there, if this area interests you. Debate on open access papers is nothing new here: this time I'm hoping for some progress. Charles Matthews (talk) 12:28, 18 October 2017 (UTC)

Proposal for the addition of an OPDS catalog for our works

While there are certainly issues with how we convert our works here to ebook files, there are very definite advantages to reading books locally rather than on a website. Mainly there are facilities that are provided by various ebook readers that can't be implemented on a website and the fact that a large portion of the world does not have access to a reliable internet connection yet. An OPDS catalog will allow people to easily search for and download the works that they may want to read later. What do you think? Can we implement this? How much work would it require? And if it can be implemented, which works do we make available first? Ciridae (talk) 05:56, 16 October 2017 (UTC)

It's sort of supported in toolforge:wsexport, I think, but I'm not sure it works. I've also experimented with adding it to toolforge:ws-search but it's certainly not working yet—I don't think it'd take much to improve it though. Do you have experience with using OPDS? I've only ever used it with FBreader on android. I know that Standard Ebooks supports it for their epubs. Sam Wilson 07:34, 16 October 2017 (UTC)
I've only really worked on Atom, never OPDS directly, but on a technical level OPDS itself doesn't seem too complicated. I, however, have no experience working with Wikisource on a technical level. We could, perhaps, build a page with the required XML and have a bot update it regularly based on some criteria. Maybe a template added to the Talk or Index Talk page of the work? Ciridae (talk) 08:26, 16 October 2017 (UTC)
@Ciridae: Yeah, I reckon we can figure it out! :-) I've found a validator, e.g. here's the OPDS of all of Kipling's works: http://opds-validator.appspot.com/?uri=https://tools.wmflabs.org/ws-search/?author=Rudyard+Kipling&lang=en&output_format=opds and so can start working through what data is required. The first roadbump is the 'updated' time of a work... I guess we'd say that this is the most recent modification time of any of a work's pages? I'll add such a thing to wikisource/api. Sam Wilson 09:04, 16 October 2017 (UTC)
@Samwilson: We would be using the wmflabs tools to provide the ebook files, wouldn't we? I believe the creation of the files in this case would be on-demand so I don't think we'd need to change the catalog whenever any page is modified. We'd only need to track the last modified date-time for wherever we're storing the OPDS data. And about this part, how do you think should we go about it? Ciridae (talk) 11:30, 16 October 2017 (UTC)
@Samwilson: I finally had a look at the OPDS specification today and yes, you’re exactly right. I'm wondering where this catalog would be hosted (and how) Ciridae (talk) 15:02, 19 October 2017 (UTC)
@Ciridae: My current thinking is that we would have a database on Toolforge (i.e. wmflabs.org, which is where ws-search is) that is built by collecting all the metadata from all the works on all Wikisources, and it would then serve up the OPDS on demand. The OPDS entries would then link to wsexport for delivery of the actual epubs. The metadata is the problem, but solvable with a combination of scraping HTML from wiki pages and looking things up on Wikidata. The scraping especially is a slow operation, hence the need for a database.

My experiments so far are sort of looking okay — the current bug is that the scraper hits big works like 1911 Encyclopædia Britannica and attempts to treat it like a single item (e.g. how to find the last-modified date of that work? Let alone export it to an epub).

I'm also not sure how we'd divide up the ODPS catalogue: by genre perhaps? With the data coming from Wikidata. I think there must be some upper limit on the number of items in a single ODPS feed. Sam Wilson 23:48, 19 October 2017 (UTC)

@Samwilson: But do we want to provide all works on Wikisource through the catalog? I personally think that might be a bit too ambitious considering the quality of epubs generated by wsexport isn't great. (We should also provide mobipocket and PDF files.) How about we start with a few select categories only and expand from there? We can divide the catalog along various criteria (genre, topic etc) rather than having it as a single feed (which would make it easy to browse, for starters). How about all Featured Works to get started?

Oh, and if you have the time I'd love to have a few pointers on how I could help out technically. I haven't contributed code to any Wikimedia project and I'd like to help. Ciridae (talk) 16:00, 20 October 2017 (UTC)

┌───────────────────┘

@Ciridae: Good point! I wonder if we a) limited it to validated categorized works; and b) made individual OPDS files for each point in the category hierarchy that had under say 100 works. Something like toolforge:ws-cat-browser with the addition that at each tree node where there are fewer than 100 child works we create a (static-built, fast-to-serve) ODPS file? I've never used more than a single OPDS file in e.g. FBReader, so I'm not sure what the optimum browsing system is, or what most people want from such a feed. It seems that this sort of thing is common for the top-level, which isn't a tree structure (but we could replicate our category tree breadcrubs in the section titles perhaps).

And as for contributing code: hurrah! Yes, I'd love to help you get started. What languages do you work in? Have you read mw:How to become a MediaWiki hacker? Wikisource also has its own Github org: https://github.com/wikisource . There's a million Lua things that could be worked on, on-wiki, too, with bringing more data in from Wikidata. The best project to start with is one that fixes a problem for you. :-) Join ##wikisource any time, if you want to chat.

—Sam Wilson 00:36, 21 October 2017 (UTC)

File:Dick Sands the Boy Captain.djvu

The scan File:Dick Sands the Boy Captain.djvu on Commons needs to have the Google notice stripped from the front of the file. --EncycloPetey (talk) 05:14, 21 October 2017 (UTC)

Done —C. F. 06:49, 21 October 2017 (UTC)

Life of Jesus

Can this page (and all its subpages) be moved to Life of Jesus (Renan), and any internal links repaired (if necessary)? There are other works with this title (including Strauss, transl. George Eliot), and so the original location is needed for disambiguation. --EncycloPetey (talk) 20:09, 21 October 2017 (UTC)

Done —Beleg Tâl (talk) 02:45, 22 October 2017 (UTC)

Indexed works containing multiple titles

Can someone please point me to cases of transcluded indexed works here which contain more than one whole major work (by the same author) within the text? I am trying to figure out how to best transclude Index:Sartor resartus; and, On heroes, hero-worship and the heroic in history.djvu, being that it contains two major works. Both works are also already present separately here at WS, albeit unindexed. Thanks, Londonjackbooks (talk) 16:15, 16 October 2017 (UTC)

Index:Austen - Northanger Abbey. Persuasion, vol. I, 1818.djvu is one, where each work has been transcluded into a top-level mainspace set of pages. I'd say in general you can create separate top-level mainspace pages for each work in this sort of situation. Or, in the case of anthologies and similar, two structures, one for the actual published structure and one for the logical separate-work one (like we do for serial works). Sam Wilson 22:25, 16 October 2017 (UTC)
@Samwilson: So perhaps Sartor Resartus and On Heroes, Hero-Worship and the Heroic in History (Macmillan) as one structure (if I understand) that lists both works, and Sartor Resartus (Macmillan) & On Heroes, Hero-Worship and the Heroic in History (Macmillan) as separate-work structures? with "Macmillan" serving the purpose for disambiguation? Londonjackbooks (talk) 23:26, 16 October 2017 (UTC)
@Londonjackbooks: Yeah, sounds about right. I wouldn't worry about disambiguating unless there's a clash, so would go with e.g.:
Sam Wilson 23:39, 16 October 2017 (UTC)
I've usually just done one super-page with two sub-pages, but I've never done ones where the two works are so extensive. Mine are more along the lines of Herald Angels, and The Holly & Ivy or The Holly & the Ivy, and Twelve Articles. —Beleg Tâl (talk) 23:45, 16 October 2017 (UTC)
Me too, but I'm also keen about them having their own top-level pages as they are actually stand-alone works and doing it that way makes a bunch of things more fun, e.g. they get their own wikidata items and so can be catalogued separately (as well as together), and so it hepls with discoverability. Moll Flanders and Roxana is another example (although, I think there's a slight mess here too because there are other editions of those books on Wikisource as well). Oh, and in Londonjackbooks' case above, would we end up with e.g. Sartor Resartus and On Heroes, Hero-Worship and the Heroic in History/On Heroes, Hero-Worship and the Heroic in History/Chapter 1... that's a long title (but probably the correct one)! Sam Wilson 23:56, 16 October 2017 (UTC)
Thanks all! I will wait until my second cup of coffee before I give it a go :) Londonjackbooks (talk) 11:11, 17 October 2017 (UTC)
  •  Comment I would have done them as subpages of the printed edition, then done redirects from the actual titles to the subpages. It stays true to the published work, and it allows for future editions of the work. It is how we have handled multiple poetic works. — billinghurst sDrewth 11:01, 21 October 2017 (UTC)
I think that is likely the correct way. Title length shouldn't really be an issue. Thanks, Londonjackbooks (talk) 11:42, 21 October 2017 (UTC)

Tech News: 2017-43

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Recent changes

  • You can use ccnorm_contains_any when you create an abuse filter. This can be used to look for multiple words or phrases within a string. It will find words where some characters have been replaced. You can read more in the documentation. [33]

Changes later this week

  • When you search in some languages the search function could use specific other languages if the first language didn't work. This is called a fallback language. This didn't work properly and created bad searches. The search index is being fixed to work better. [34]
  • The new version of MediaWiki will be on test wikis and MediaWiki.org from 24 October. It will be on non-Wikipedia wikis and some Wikipedias from 25 October. It will be on all wikis from 26 October (calendar).

Meetings

  • You can join the next meeting with the Editing team. During the meeting, you can tell developers which bugs you think are the most important. The meeting will be on 24 October at 18:30 (UTC). See how to join.
  • You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 25 October at 15:00 (UTC). See how to join.

Future changes

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

18:18, 23 October 2017 (UTC)

Internet archive in collaboration with mentalfloss.com is publishing some obscure books

This article provides additional information in greater detail. — Ineuw talk 07:38, 24 October 2017 (UTC)

More info on the availability of books out of print — Ineuw talk 08:24, 24 October 2017 (UTC)

This ties into the discussion below, #Consider Wikisource a library (for U.S. copyrite law) —Beleg Tâl (talk) 12:30, 24 October 2017 (UTC)
@Beleg Tâl: Thanks for the link. It goes to show how much can be learned by scrolling further down. Unfortunately, I get lost in the variety of topics, especially dealing with copyright, so I stick with public domain. — Ineuw talk 18:01, 24 October 2017 (UTC)

Looking for small technical tasks+mentors for new contributors - got something in mind?

Hi everybody! Google Code-in (GCI) will soon take place again - a seven week long contest for 13-17 year old students to contribute to free software projects. Tasks should take an experienced contributed about two-three hours and can be of the categories Code, Documentation/Training, Outreach/Research, Quality Assurance, and User Interface/Design. Do you have an idea for a task and could you imagine mentoring that task? For example, do you have something on mind that needs documentation, research, some gadget or template issues on your "To do" list but you never had the time, and can imagine enjoying mentoring such a task to help a new contributor? If yes, please check out mw:Google Code-in/2017 and become a mentor! Thanks in advance! --AKlapper (WMF) (talk) 19:55, 16 October 2017 (UTC)

I've signed up as a mentor for the Code-in, so if anyone knows of Wikisource-related little things I'd be very happy to help students work on these! :-) Sam Wilson 23:42, 16 October 2017 (UTC)
Do not know how easy: https://phabricator.wikimedia.org/T173385 .— Mpaa (talk) 17:35, 18 October 2017 (UTC)
Musical Notation editor would be nice. user:Beeswaxcandle is the expert there; do not know how easy it could be. Slowking4 ‽ SvG's revenge 02:56, 19 October 2017 (UTC)
Agreed, musical notation editor would be nice. I think the simplest and most feasible solution would be to write a script that can take the LilyPond output from an established WYSIWYG music editor and convert it to be compatible with the Score plugin. —Beleg Tâl (talk) 12:28, 24 October 2017 (UTC)
Signed up as a mentor for table layout analysis and design. — Ineuw talk 19:42, 23 October 2017 (UTC)

Wikipedia/Wikisource keyboard hotkey assignments

Wikipedia has a set of keyboard key assignments using the combinations of the Alt+Shift+character Keys. Is there an Alt+Shift key combination in Wikisource for opening/closing the header/footer in edit mode? If not, can this be assigned to the combination of Alt+Shift+h somehow ? — Ineuw talk 00:51, 23 October 2017 (UTC)

That would need to be separate javascript, and it would probably best to be added into the proofread page extension rather than a separate overlay. I suggest that put in a phabricator request. — billinghurst sDrewth 13:24, 23 October 2017 (UTC)
Thanks, will do it later today. — Ineuw talk 19:39, 23 October 2017 (UTC)
Alt+Shift+h already invokes page history throughout all the sisters and us, so it would be unacceptable to use it for something else. Beeswaxcandle (talk) 05:36, 24 October 2017 (UTC)
The key combination was just an idea without checking the existing key assignments. I leave the actual key assignment up to the programmers. — Ineuw talk 07:45, 24 October 2017 (UTC)

WikiProject DNB: Hyphens, en dashes and redirects

I had a look in the archives of Scriptorium and could not find a discussion on this issue.

Some editors on Wikipedia are concerned that the DNB templates that link to articles on Wikisource display dash rather than ndash in article titles. They want them changed to ndash. In the case of the DNB articles on Wikisource these are disambiguation extensions and do not form part of the original DNB text. A work around for the differences in style between Wikipedia and Wikisource over the use of dash and ndash would be to provide redirects on Wikipedia for those DNB articles that have date range extensions.

To centralise the discussion please add any comments to "Wikisource talk:WikiProject DNB#Hyphens, en dashes and redirects" on the wisdom or otherwise of providing such redirects. -- PBS (talk) 10:14, 23 October 2017 (UTC)

@PBS: The issue that you raise is not DNB specific. The use of a hyphen is for all separation as part of our manual style. So bang away at that page if you choose, however, the issue is bigger and broader. It is author pages, and many pages. I think that enWP has to get over the fact that we don't use an en dash and move on. — billinghurst sDrewth 13:26, 23 October 2017 (UTC)
The guidance about hyphens is in Help:Disambiguation, the discussion will be in the archives of this place, and it will be long ago. I would guess it took place in 2009, 2010 and possibly more in 2011. — billinghurst sDrewth 13:40, 23 October 2017 (UTC)
more drama on english = meh. not a reason to change a thing. Slowking4 ‽ SvG's revenge 16:26, 23 October 2017 (UTC)

Lua can help, surely? I'm adding to the linked talk page. Charles Matthews (talk) 11:58, 30 October 2017 (UTC)

Use of Wikidata

Hi, Is there any plan to use Wikidata, at least for authors? On the French WS, we already use it for authors, and we are in the process to use it for works. Regards, Yann (talk) 17:45, 29 October 2017 (UTC)

We already use it for authors. Or is there a more specific use you are asking about? —Beleg Tâl (talk) 18:23, 29 October 2017 (UTC)
@Yann: Currently used for birth, death, image, and authority control, where not overridden with local data. Insufficient given and surname data at WD to be utilised for data pull, though I have populated a lot. With regard to works, we are waaaaaay behind the curve of getting data in place, and we pretty much await good editing tools that WD is looking to implement. There was discussion in the Wikisource-L mailing list last week about the upcoming conference and what we thought could be pushed.

I would love to be able to utilise it for licence tags one day too. We still have plenty of local override fields in play and not many people remove it from old author pages, and check the pull from WD — noting that the community requested that all data be checked prior to removal. There are plenty of maintenance categories to identify where data is surplus to requirements. — billinghurst sDrewth 22:12, 29 October 2017 (UTC)

I agree that it's a bit of a pain at the moment to enter work/edition data at Wikidata, but I also think that we'd get better at it if we were using it more. I'd love to do more work on bringing bits of edition data back here into the header template. Perhaps we could start with some data that we don't currently have here on a large scale, such as publisher? We could populate subcats of Category:Works by publisher pretty easily, and I think making 'publisher' clearer would help with reinforcing the point that what we host here are specific editions that relate back to an existing publication. We could even do it without jumping the editing-to-work link on WD if we wanted to keep it simple to start with. —Sam Wilson 02:07, 30 October 2017 (UTC)
Could you elaborate on the work flow you expect? Because right now, most copies on Wikisource don't even have a data item in existence on Wikidata for that edition yet. Wikidata has the work data item, if it has anything. So data items would first have to be created for the editions we have. We can't edit what doesn't exist. --EncycloPetey (talk) 02:22, 30 October 2017 (UTC)
@Samwilson: I actually feel that we have a better scope and chance of getting metadata input from Commons and working with the structured data project. That enables us to push from there, and more able to have a workable input form (well I hope that is happening). From there we know that we have "scan image" property link back to an edition, and we know/see that an index: page is created, so able to link through the wikisource index property. Commons sees and records that triangle. — billinghurst sDrewth 03:10, 30 October 2017 (UTC)
Yes, good point about Commons' structured data. I guess I figure that ultimately, the metadata for our editions' scan on Commons will also come from Wikidata (as it's different from most metadata on Commons, in that it is welcome on Wikidata). I don't want to wait though. :-) Although there's lots of items missing from Wikidata, it's not that hard to add them for editions, especially when adding new works here when it would mean that one needn't add that same data here at all. Not that I'm against someone working on pulling the data from Commons of course! :) I should like to have another little Lua project here, if it were seen to be useful (e.g. the publisher categorizing I mentioned). Sam Wilson 04:53, 30 October 2017 (UTC)

Wikidata for Proofreadpage index template

I meant to create a new topic but coincidentally this related topic was already started, so I'm adding it as a subheader. Please check the discussion I started at MediaWiki_talk:Proofreadpage_index_template#Wikidata. NMaia (talk) 00:29, 30 October 2017 (UTC)

Responded there. Short answer: it's impractical and could't be used unless the data item for that edition was first created and filled at Wikidata. And that wouldn't actually save on any work. It would be more useful to be able to go the other way. --EncycloPetey (talk) 00:43, 30 October 2017 (UTC)

Tech News: 2017-44

Latest tech news from the Wikimedia technical community. Please tell other users about these changes. Not all changes will affect you. Translations are available.

Recent changes

  • You can now block users from emailing you through the Wikimedia wikis. [36]

Changes later this week

  • The new version of MediaWiki will be on test wikis and MediaWiki.org from 31 October. It will be on non-Wikipedia wikis and some Wikipedias from 1 November. It will be on all wikis from 2 November (calendar).

Meetings

  • You can join the next meeting with the Editing team. During the meeting, you can tell developers which bugs you think are the most important. The meeting will be on 31 October at 18:30 (UTC). See how to join.
  • You can join the technical advice meeting on IRC. During the meeting, volunteer developers can ask for advice. The meeting will be on 1 November at 16:00 (UTC). See how to join.

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

00:20, 31 October 2017 (UTC)

Consider Wikisource a library (for U.S. copyright law)

Since the Internet Archive is the first institution to exploit this feature of America's arcane and backwards copyright law, I suggest we be the second: Section 108h of the U.S. Code allows libraries to scan and make available materials published 1923 to 1941 if they are not being actively sold. One immediate objection I see is that it would introduce overhead on our part to determine if a work is actively being sold. On the contrary, I would suggest that this is no different than a DMCA request: assume that a work is not (most aren't), use basic common sense for due diligence, and then let someone else complain if he thinks we are hosting something we shouldn't be. Thoughts? —Justin (koavf)❤T☮C☺M☯ 01:30, 11 October 2017 (UTC)

Wikisource does far more than just scan works. --EncycloPetey (talk) 01:39, 11 October 2017 (UTC)
Books from 1923 to 1941 Now Liberated! not too hard to determine if in print. do a alibris / amazon / worldcat. and a search for non-renewal is not too hard. and internet archive is doing the search and hosting. however, this community would never agree to such a librarian standard of practice. in 2 years we will start counting up anyway. do you have any orphans before 1941 of interest? Slowking4 ‽ SvG's revenge 02:15, 11 October 2017 (UTC)
  •  Comment the proposal is changing the underlying predication of Wikisource in that we would be moving to a non-commercial type license, something akin to PD-1941-NC. These works would be unable to be taken from our site and reproduced as all our existing works can be. How would you differentiate between those works that can and those that cannot be commercialised? — billinghurst sDrewth
    As a follow-up, I am not opposed to the exploration of this matter, I just think that it needs a reasoned proposal, not a "dump and run". If it is going to be a dump and run, then I propose that it is moved to the bottom of this page. @Koavf: if you are going to put together something which we can explore and look through nuanced argument, then I look forward to your proposal. — billinghurst sDrewth 21:02, 11 October 2017 (UTC)
    @Billinghurst: Not sure what you mean--I left this here for feedback. I don't know what more you want. I could respond to every person the moment he posts but I wanted to elicit some discussion. —Justin (koavf)❤T☮C☺M☯ 22:32, 11 October 2017 (UTC)
    @Koavf: But you left it in the "Proposals" section. If you just wanted feedback and a discussion, then this isn't a Proposal but a discussion topic. --EncycloPetey (talk) 22:37, 11 October 2017 (UTC)
    @EncycloPetey: Because it would mean a pretty fundamental change to our approach to works here. It isn't just an idle chat about issues tangentially related to Wikisource but a way to refactor some of what we do and which would require some broad consensus, re-writing policy pages, etc. If other users think it's a non-starter (and clearly, several do), then the community is rejecting my proposed changes and it's just food for thought. —Justin (koavf)❤T☮C☺M☯ 23:03, 11 October 2017 (UTC)
    @Koavf: I think you've missed what first billinghurst meant, and my response to your question said. You made a "suggestion" or "comment", whereas a "proposal" is usually a more formal sort of presentation than what you posted. So billinghurst was pointing out that it didn't seem appropriate to post in the Proposals section (and I agree) because it's more a passing thought or idea than a formal proposal. So what you asked "not sure what you mean", I was trying to help answer that question. Yes, your "suggestion" involves a fundamental change, but that doesn't make it a "proposal". A proposal would be a formal well thought out and fully reasoned presentation for the community, rather than a passing thought about a big change. --EncycloPetey (talk) 01:52, 12 October 2017 (UTC)
Oppose - Proposed license is NOT compatible with 'free' licensing terms which permit commerical use. Any works uploaded would have to be locally hosted in any event, as the above would be a non-starter on Commons. ShakespeareFan00 (talk) 16:36, 11 October 2017 (UTC)
That is an interesting approach, though I am not sure that it is a reasoned approach. Some thoughts to consider:
  • We host works that are not copyright in the US, yet some of these are still copyright in their home country, and we have both text and image as they cannot be at Commons.
  • We do not host some works as they have copyright in the US, even though they are out of copyright in their home country.
  • WMF has a broad scope to copyright and licensing and how they see that it applies and give latitude to how wikis can apply. It is a range, and up and down the range different conditions apply.
  • We license all of our works with the conditions that apply to their hosting, and their re-use. It is our rule about not allowing "non-commercial" or not having "fair use", it is not WMF's.
  • There are ways that we could differentiate non-commercial works from commercial works if we chose a different approach.
So how about a reasoned and logical debate, not an emotional one, or one that hinges on a dogma. Wikisource should develop, and that development should be in line with the scope of the WMF and its development. We should not be frozen in time. — billinghurst sDrewth 21:19, 11 October 2017 (UTC)
Actually, it is WMF's rule about not allowing non-commercial. wmf:Resolution:Licensing policy says "All projects are expected to host only content which is under a Free Content License, or which is otherwise free as recognized by the 'Definition of Free Cultural Works' as referenced above." And any emphasis on a logical debate is deceptive; the question is about deciding what our ultimate goals are, and logic can't advance that question.--Prosfilaes (talk) 01:38, 12 October 2017 (UTC)
actually WMF does allow non-commercial works per "Exemption Doctrine Policy (EDP) A project-specific policy, in accordance with United States law and the law of countries where the project content is predominantly accessed (if any), that recognizes the limitations of copyright law (including case law) as applicable to the project, and permits the upload of copyrighted materials that can be legally used in the context of the project, regardless of their licensing status." maybe we could have a proposal for pre-1941 works not in print?
thank-you for being honest about the appeal to emotion, rather than appeal to reasom. Slowking4 ‽ SvG's revenge 10:19, 12 October 2017 (UTC)
The page describes how EDPs should be used as:
"3. Such EDPs must be minimal. Their use, with limited exception, should be to illustrate historically significant events, to include identifying protected works such as logos, or to complement (within narrow limits) articles about copyrighted contemporary works. ... Any content used under an EDP must be replaced with a freely licensed work whenever one is available which will serve the same educational purpose.
4. ... They must be used only in the context of other freely licensed content."
Yes, your appeal that we should maximize the volume of works we can work on is no more an appeal to reason than my appeal to staying with free works. Rationally we can speak of the value of a small set of works that we may be forced to take down if they come back in print, versus the huge universe of pre-1923 work that is untouchable.--Prosfilaes (talk) 00:54, 13 October 2017 (UTC)
you can call a fair use of the "orphan work out of print before 1941" an ideology if you want, but it is an ideology shared by the hathi trust and internet archive. they will do the work of selection, and we could support them. these are low risk items, that we can make available to the public, as a part of the sum of all knowledge. - they are partners i can collaborate with, unlike the FSF. Slowking4 ‽ SvG's revenge 23:53, 13 October 2017 (UTC)
A couple of examples of works that could fit within the EDP doctine, and I do preface that it is a little opening and possibly one that would take too much explaining to make it useful and sustainable.
  • works that are out of copyright in their home country, and that are out of print;
  • compiled works that are not copyrighted for parent work, though may contain work that is within copyright within US; traditionally we have blanked those components in our transcription, be they chapters or images.
As a question, does anyone know why there is an 1941 cutoff? I haven't seen mention of why the 75 years is pertinent. — billinghurst sDrewth 04:01, 14 October 2017 (UTC)
Because, books published upto 1941, with 95 years' copyright, are within last 20 years of their copyright; and thus covered under 17 U.S.C. Section 108(h). Hrishikes (talk) 04:39, 14 October 2017 (UTC)
  • a lot of the works will be US home country, but out of print, and orphaned, so we do not know who the copyright holder is (although hathi trust found some subsequently) see also w:Orphan works in the United States
  • are you agreeing to a fair use of the lesser term? the Canadians and Chinese would be happy to agree with you.
  • compilations are rare compared to the orphan ocean. we can also do a copyright search for non-renewal, but this is not "untouchable", the rules are too complicated for bright lines, but we can show our work as a standard of practice. Slowking4 ‽ SvG's revenge 00:56, 15 October 2017 (UTC)
Looking at the EDP, the only WS project I could find with a more-or-less clear stance is the French Wikisource, which allows both local uploads (like most except Japanese and Dutch) and, unusually, non-free content (which is discouraged, but some content is fair-use in French law; see this and this if you read French). Perhaps someone with a good command of French could research how they operate regarding these matters for ideas? Inatan (talk) 12:27, 15 October 2017 (UTC)

┌───────────────────────┘
 Other than these "Last 20" books, I would like to draw attention to another class of books, already mentioned by @Billinghurst:, which are PD-home country but not PD-URAA. Such books are now allowed in Commons. Internet Archive has a good number, especially after the large-scale addition of DLI books. In case of Indian works, two types are now allowed here (exc. Govt woks & CC): Books published before 1923 and books by authors who died before 1941. If we allow PD-home country, then books of authors who died in 1941-1956 can be allowed, which is a huge number of books. For countries that are 70 pma (like UK), books by authors who died before 1947 can be allowed. This will considerably enrich the English Wikisource (e.g., by having the post-1923 works of Rabindranath Tagore and the books of hunting by Jim Corbett, among others). These are already allowed in Commons, so we may also consider. Hrishikes (talk) 04:58, 16 October 2017 (UTC)

 Support conditionally when considering m:United States non-acceptance of the rule of the shorter term#Orphan works for non-American works only.  Oppose this for orphan American works.--Jusjih (talk) 02:53, 29 October 2017 (UTC)
 Support Of course, I support any extension of our scope. Yann (talk) 17:41, 29 October 2017 (UTC)
 Oppose per ShakespeareFan00. Perhaps this proposal would be more appropriate over at Wikilivres. NMaia (talk) 11:43, 6 November 2017 (UTC)
 Comment @NMaia: Wikilivres is in Canada. —Justin (koavf)❤T☮C☺M☯ 18:22, 6 November 2017 (UTC)
 Comment wikilivres is no more; it is now https://biblio.wiki/wiki/Main_Page - it could be appropriate here, but you choose to wall yourself off from the decisions of hathi trust, and commons. Slowking4 ‽ SvG's revenge 15:10, 6 November 2017 (UTC)
 Neutral; looks like a lot of bother and confusion for a handful of obscure works, but if people want to establish a clear policy or EDP that works with our existing policies and frameworks then go for it. —Beleg Tâl (talk) 15:25, 6 November 2017 (UTC)
 Comment you should expect a periodic questioning of why no EDP, as naive people see work that very well could be done, but is not, for a lack of it. i am not confused. Slowking4 ‽ SvG's revenge 16:30, 6 November 2017 (UTC)
If we pass this proposal, I would like to rewrite Template:Not-PD-US-URAA and possibly rename it. Chinese Wikisource is about to accept similar proposal.--Jusjih (talk) 03:04, 21 November 2017 (UTC)
We should delete Not-PD-US-URAA; it's not in use anywhere, and it's not relevant. Even if we accept this proposal, it's still not relevant; this applies to books published more than 75 years ago that have living authors as much as those with long dead authors.--Prosfilaes (talk) 08:04, 22 November 2017 (UTC)
i agree, we should delete not-pd-us-uraa - that is a commons drama, that no one cares about here enough to upload a work. chinese wikisource is going for "works of the lesser term" contrary to the "take it to wikilivres" above. but why there should be an ideological opposition to an EDP for a few works is interesting. does not add value. partnering with IA and hathi adds value. Slowking4 ‽ SvG's revenge 01:27, 23 November 2017 (UTC)
apparently you care enough to import the URAA drama, so we need the tag. how amusing. Slowking4 ‽ SvG's revenge 02:42, 6 January 2018 (UTC)
Chinese Wikisource is considering this proposal here, so Template:Bibliowiki page now corresponds to zh:template:Not-PD-US-old. Undeleting Not-PD-US-URAA is provisional while considering m:Legal/Wikimedia_Server_Location_and_Free_Knowledge. Depending on whether this proposal goes, I may propose merging relevant contents of Not-PD-US-URAA into Template:Bibliowiki page, possibly renaming Template:Bibliowiki.--Jusjih (talk) 05:09, 28 November 2017 (UTC)
As I understand from an ongoing discussion by experts at the Wikipedia Weekly Facebook Group, we cannot consider Wikisource a library (for U.S. copyright law), for hosting the "Last 20" books. Internet Archive has official recognition as a library under US law in the state of California. That's why they can go ahead with this provision of 17 U.S.C. Section 108(h). Wikisource has no such recognition; so we cannot do it. People participating in that Facebook discussion were commenting on this discussion of ours, that Wikisource is not getting the "nuance" of the matter. Hrishikes (talk) 00:56, 4 December 2017 (UTC)
"library status" is a loophole in the code, to host the pdf’s. we can still claim fair use of orphan works out of print, regardless of theories of who we are. if it is an orphan there is no one to claim copyright. i often lament the lack of nuance of what passes for consensus, in many forums.Slowking4 ‽ SvG's revenge 03:46, 6 December 2017 (UTC)
Concerning the deprecated template, why not mark it as "Historic", deleting it will benefit exactly no-one. -- DonTrung (徵國單)  (討論 🤙🏻) (方孔錢 ☯) 11:47, 16 January 2018 (UTC)
Then are we considering m:Legal/Wikimedia Server Location and Free Knowledge to conditionally tolerate works affected by the m:United States non-acceptance of the rule of the shorter term? I am ready to forget the "Last 20" books not affected by the US non-acceptance of the rule of the shorter term.--Jusjih (talk) 05:51, 11 December 2017 (UTC)
Retrieved from "https://en.wikisource.org/w/index.php?title=Wikisource:Scriptorium/Archives/2017-10&oldid=7359688"