Wikisource:Scriptorium/Archives/2017-05

From Wikisource
Jump to navigation Jump to search

Announcements

Proposals

Bot approval requests

Repairs (and moves)

I discovered that the scan of page 49 in this project is incorrect, with a scan of page 57 on that page. The rest of the document seems OK outside of that.

The PDF seems to be inaccessible on the DOT database where the document was sourced from, although a different instance of the report is available from the 1st reference on the wikipedia page links to a different copy of the report with that page in tact, being page 56 of that particular PDF. I've done a quick cross-reference of that scan and the separate text layer in the DOT database where the projects original scan came from and it all aligns, suggesting the page in that PDF is the correct one.

I'm fairly new here, hopefully this is the right place to post to get this looked at and fixed up :)

Thanks Nickw25 (talk) 10:54, 8 April 2017 (UTC)

PING COMMUNITY Anyone have the tools (and patience) to take 56th page (page 49) from here and plugging that into file:CAB Accident Report, Pennsylvania Central Airlines Flight 19.pdfbillinghurst sDrewth 09:28, 14 April 2017 (UTC)
@Nickw25: Done -- Hrishikes (talk) 12:36, 14 April 2017 (UTC)
This section was archived on a request by: — Mpaa (talk) 08:27, 17 April 2017 (UTC)

Index:Emanuel Swedenborg, Scientist and Mystic.djvu

Hello. Is anyone able to do a mass find and replace to replace the following ligatures with the proper text that I have missed whilst proofreading this work. They are the ligatures fi fi and fl fl. Thanks for any help.Jpez (talk) 16:09, 14 April 2017 (UTC)

Done.—? Mpaa (talk) 07:42, 16 April 2017 (UTC)
Thanks! Jpez (talk) 14:43, 16 April 2017 (UTC)
This section was archived on a request by: — Mpaa (talk) 08:27, 17 April 2017 (UTC)

Other discussions

{{nop}} , it's overloaded use and possible replacement.

This isn't (yet) an issue, but future versions of the parser could potentially "tidy" blank tag pairs out of generated output entirely, thus possibly including the "tidying" out of the implied paragraph break generated.

{{nop}} is also used in three different instances:-

  1. is the original intended use, at the end of a page to force a paragaph break
  2. in ref tags to force a paragraph break.
  3. is at the head of a continued table (to provide a place holder for the pagenumber script, as I understood it.) and to ensure the relevant table element is read correctly.

Whilst the first 2 uses are the same, the third isn't and is a workaround for certain (current) limitations in the parser/Proofread page, ( which have been discussed at least twice here before.)

I've started https://phabricator.wikimedia.org/T163073 on my own initiative, given that I felt there should be specific magic tags for overriding the default paragraph handling (like there are tags to indicate NOINDEX or page specfic TOC), in the parser, rather than it relying on a local template.

Feedback would be appreciated, as "community" support helps decide what gets fixed. ShakespeareFan00 (talk) 22:39, 16 April 2017 (UTC)

The use in ref tags is an incorrect use. This should be done with paragraph tags.

With respect to the other usages, has a future problem actually been flagged or are you trying to be preëmptive of something that may never happen? Beeswaxcandle (talk) 23:51, 16 April 2017 (UTC)

A lone <p> will generate "malformed" HTML, according to the Linter Extension's checks (having had to fix over 100 or so pages utilising this quirk, I'm not happy about using it elsewhere., and is courtesy in . and doing <p></p> per what's supposed to be done doesn't yet render nicely in references IIRC.

I wasn't aware of the other usages being flagged as a problem as such, but given that the parser is being re-developed anyway, resolving what is a LONG standing limitation (necessitating a local template for something that SHOULD be in the core parser) was felt to be entirely reasonable.

In relation to the usage at a head of a table, this caused some pages to be listed as "fostered content" by the LinterExtension (the first instance of this being one I raised a phab ticket about as I initially thought it was a false positive.) as the lead {{nop}} is seen by the relevant checker as NOT being inside a table row or table element, and thus will get moved to the parent element before the table, automatically. This was another quirk behaviour which I understood wasn't necessarily guaranteed to behave the same in the new parser. (I've also had several rows with the parser getting it to handle complex tables( with split header/content/footer) correctly and consistently using the {{nop}} route.) Having a magic tag with a CONSISTENT, COHRENT and "documented" behaviour in all contexts, would be far far better than regularly playing 'guess the random interaction relying on Mediwiki behaviour you are supposed to know by mind-reading..' that has been used till now. ShakespeareFan00 (talk) 00:26, 17 April 2017 (UTC)

Unfortunately you do go off half-cocked. In all cases "nop" is used as a placeholder, for a variety of elements of wiki syntax. Your continued panicked run-arounds and ill-judged comments at phabricator are not helpful.

Don't let Linter drive behavioural editing. The use of a single "p" marker is as old as the hills, and while the modern syntax likes closure, simply don't fuss it. So you don't have to fix anything like that, and you can be happy or not that is solely your choice. If you are not happy doing it, then don't. When you don't know what you are talking about, sometimes you should learn to say nothing. — billinghurst sDrewth 06:03, 17 April 2017 (UTC)

I'm sorry you feel that way, but based on personal experience of trying to get things working on more than one work, I'm increasingly fed up as I said, of playing "hunt the quirk" (which probably shows in the Phabricator tickets).

Time to take another time out. ShakespeareFan00 (talk) 11:02, 17 April 2017 (UTC)

in all the parser fun, maybe we should take council from user:Tim Starling - "We'll soon be getting rid of Tidy on WMF websites in favour of a pure PHP solution called RemexHtml that I recently wrote. It should eventually become the default for new MediaWiki installations as well. It accepts either form and will initially output "<br />" for compatibility with parser tests." [1] - Slowking4SvG's revenge 14:20, 17 April 2017 (UTC)

Save/Publish page

The boilerplate at the foot of the edit window begins

By clicking the “Save page” button,...

But there is no “Save page” button. The button being referred to is labeled “Publish page”. Either the boilerplate or the button text should be changed.

The equivalency is obvious to me and probably to most editors, but most ≠ all. I am a (1) highly educated (2) native speaker of American English (3) with 12 years of Wikipedia experience, (4) a career largely spent in the software industry, and (5) a doctorate in linguistics. It may not be so obvious to, e.g., a non-native speaker, or an educated native speaker new to Wikimedia who wonders what the difference is between saving a page and publishing it.

Please {{Ping}} me to discuss. --Thnidu (talk) 05:31, 17 April 2017 (UTC)

@Thnidu: Thanks for the note. The change from "save page" to "publish page" is a change that has taken place in the last month. It is contained in the message MediaWiki:Wikimedia-copyrightwarning and we are using the default for all wikis. @Whatamidoing (WMF): the global message needs to be updated, and maybe it needs to be done in many languages. Can we poke that at you for resolution? — billinghurst sDrewth 06:10, 17 April 2017 (UTC)
Whatamidoing (WMF): Noting that this footer used to be Philippe's text, so it may be something to wave past Mdennis (WMF). — billinghurst sDrewth 07:21, 17 April 2017 (UTC)
@Billinghurst: Whee! Thanks for the quick and helpful answer. :-) --Thnidu (talk) 06:17, 17 April 2017 (UTC)
Noting to community the only overt place that needed change is Mediawiki:Newarticletext which I have just updated. We should be looking through our pages in the Help: namespace for where we may have the "save page" text. — billinghurst sDrewth 06:19, 17 April 2017 (UTC)
I wasn't making a new page. I was editing Bright's Anglo-Saxon Reader/Anglo-Saxon Versification, a couple of typo fixes. (Don't worry, I compared with the scan of the original. Typos like "teh".) What I noted would exist on all edit pages. --Thnidu (talk) 07:14, 17 April 2017 (UTC)

Tech News: 2017-16

19:32, 17 April 2017 (UTC)

Notes

13:04, 18 April 2017 (UTC)

Author pages, wikidata and family name

Hi to all who do work in Wikidata, especially in the creation and editing of author pages over there. I have been looking at the population of "family name" (Property:P734) data as that would be a really useful field for us to utilise here (it equates to our use of lastname in the Author template.)

By my searches, the data looks to be

and I think that I added about 1000 yesterday with the use of tools.

I would like to ask that anyone adding author data over there to please consider the addition of P734 data, and even creating items for surnames if they are not within the system there. I will look to create the simple WS guide to author addition at WD to assist this process.

My reason for wanting this populated is that when done, we can look to the bot creation of compilation lists, so we can autogenerate lists like Wikisource:Authors-A rather than the porous manual lists we have now. "Given name" and "family name" data populated to WD is required for this to be successful.

Thoughts? Questions? — billinghurst sDrewth 04:44, 20 April 2017 (UTC)

is there a way to mix and match against worldcat or viaf ? Slowking4SvG's revenge 20:00, 21 April 2017 (UTC)
How would this work for people like John Emerich Edward Dalberg-Acton, 1st Baron Acton, surname "Dalberg-Acton" but listed in most indices at "Ac"? Or people whose last names are not generally used for indexing, like the house names of royal families, or the former names of monks and popes? The edge-cases are the places where an automatic solution is likeliest to fail. —Beleg Tâl (talk) 20:59, 21 April 2017 (UTC)
@Beleg Tâl: In principle, it's not hard to have a default value and then add in a field for alphabetical_sort_override=. In 99% of cases, it would be fairly straight-forward to sort someone by surname and in a relatively small number of cases, it would be necessary to manually check and redefine it. —Justin (koavf)TCM 21:17, 21 April 2017 (UTC)
In that case we would need the "alphabetical_sort_override" key to be in Wikidata too, I imagine, so that the indexer-bot can access it. —Beleg Tâl (talk) 21:59, 21 April 2017 (UTC)
Noting that Wikidata has hyphenated family names, and I am sure that we can involved in that conversation there about how we wish to utilise data pulls. Noting also that there is still prefered and normal scope. It shouldn't be a blocker to us populating data, it is a fine-tuning matter. — billinghurst sDrewth 00:44, 22 April 2017 (UTC)
And noting we already have defaultsort in {{author}} and then it would how we could utilise. — billinghurst sDrewth 00:46, 22 April 2017 (UTC)

RFC: Author template firstname/lastname and the "Such and such of Placename"

Separate though related note. By my review of data for one name people eg. "John of Wendover", we still have varied use, all in firstname, all in lastname, split to both fields.

It is my belief that we should have an accepted practice and be giving specific guidance for such names. I believe that it all belongs in the firstname parameter, with the lastname parameter left empty. My reason being that when presented it should present as a string, it should default sort as that string, and based on firstname.

I know that is a provocative and bold statement, and I would like to hear other opinions about how we can organise the data, strengths and weaknesses of the approaches, so ultimately we can start to tidy up the existing data, and look to monitor the input. Thanks. — billinghurst sDrewth 04:50, 20 April 2017 (UTC)

I agree that it ought to be all in firstname with lastname blank, especially if that still sorts correctly. However, I don't see any reason for it to matter, so long as it sorts correctly. The primary difficulty in standardizing is the significant number of people for whom it isn't clear whether it should be counted as a surname or not, since the line is blurry in many cases. —Beleg Tâl (talk) 12:07, 20 April 2017 (UTC)
Yes it sorts properly, where lastname is missing it sorts on firstname. It displays properly. Richard of Wendover should appear at R lists, not O or W. We are clarifying to uses which way to push them, to firstname, not lastname. We are not removing the choice of what is a surname or not; and we will never get 100% purity, though I would like something better than our current 50% (based on users changing one or the other). — billinghurst sDrewth 15:31, 20 April 2017 (UTC)

Can the publisher be the Library of Congress?

This book Index:Notes of the Mexican war 1846-47-48.djvu doesn't list the publisher but the inside page linings are from the Library of Congress. Could they be the publishers? — Ineuw talk 07:18, 20 April 2017 (UTC)

I don't think they are; it's the same lining as books I know to be published by the Library of Congress, but the scan is from the LoC, and I think it more likely they just using their paper where they were rebinding it.--Prosfilaes (talk) 07:34, 20 April 2017 (UTC)
no - the metadata is frequently wrong at Internet Archive. the LOC catalog link is here https://lccn.loc.gov/02015450 or (OCLC) 2652213 this is a book by J. Jacob. Oswandel, published Philadelphia : [s.n.], 1885. i would blank publisher and add the correct metadata. worldcat works also. LOC does publish materials for their use, such as the Catalog of Copyright Entries, but it is very unlikely, more likely self-published or unknown, especially for war memoir. Slowking4SvG's revenge 19:56, 21 April 2017 (UTC)

Notice: training event on afternoon of Monday 24th April 2017

Sorry for the short notice, but I will be running a training workshop from 2pm-5pm British Summer Time (GMT +1) at the University of Cardiff, which will involve up to twenty new accounts dropping in and proofreading one page of a text that I've prepared. I will review their edits afterwards and clean up any mess. Thanks, MartinPoulter (talk) 22:08, 23 April 2017 (UTC)

@MartinPoulter: thanks for the heads-up. I see that you have enWP account creator rights, so that is good. Good luck with what you are doing, and we would appreciate any feedback that the newbies have that could assist others; and thanks! — billinghurst sDrewth 23:44, 23 April 2017 (UTC)

Tech News: 2017-17

16:40, 24 April 2017 (UTC)

Local notes

With regards to the ability to do temporary rights any administrator who is assigning themselves the bot user rights should look to make these temporary rights, and I think that the community should have that as a base position.

To note that if you wish to give yourself rights for a couple of hours with that finishing on the same day, that you can just add the UTC time hh:mm (and it does take YYYY-MM-DD hh:mm as other time if you ever want to be specific). Such temporary rights log as being temporary, and the rights do expire, that the expiry does not show afterwards; and the rights addition in still displays Special:UserRights with a tick at least for a while (purging no difference), though with the expired time. — billinghurst sDrewth 05:57, 25 April 2017 (UTC)

Requesting the deletion of a file on Wikisource

Please could someone delete File:Taking of Upton Bridge.jpg? I accidentally uploaded it on to Wikisource years ago. There is a version of the same file on Commons. -- PBS (talk) 16:15, 25 April 2017 (UTC)

Done. For future reference, feel welcome to use {{sdelete}} and say it is transwikied. As it is unused we can speedy it. — billinghurst sDrewth 16:19, 25 April 2017 (UTC)

19:49, 1 May 2017 (UTC)

Table assistance

If there is anyone who would like to fiddle with this page to make it resemble the original, I would be grateful. Text has been proofread; just needs formatting. Londonjackbooks (talk) 21:51, 2 May 2017 (UTC) @Londonjackbooks: Done. — Ineuw talk 02:59, 3 May 2017 (UTC)

@Ineuw: Thank you, that looks great. I have adopted blackletter throughout as well, although I was initially hesitant to, since headings were not quite blackletter... And I am undecided as to whether I want to set width for images... Thanks for all! Londonjackbooks (talk) 10:19, 3 May 2017 (UTC)
@Londonjackbooks: It is not possible to be 100% faithful to the originals. I often tried, but never succeeded. I did blackletter because it is similar to the original, and breaks the constant monotony of Arial. About the image width, I always consider the main namespace Display Option's narrow layouts, #2 and #4 (which is 540px), Please see this random page That is what I used as my guide for image width. — Ineuw talk 18:35, 3 May 2017 (UTC)
@Ineuw: Thank you for the explanation. I will keep it in mind. Again, thanks much for the formatting—and apparently while you were experiencing page loading issues? Appreciated, Londonjackbooks (talk) 18:46, 3 May 2017 (UTC)

19:16, 3 May 2017 (UTC)

Getting missing pages from another file

Colleagues and I have started a transcription project at Index:Plomer Dictionary of the Booksellers and Printers 1907.djvu. We've just noticed that pages 88 and 89 are missing from the IA scan we've used. Those pages are available in a scan of a different copy of the book. What is the recommended way to combine those pages with the existing transcription in progress? Thanks in advance for any help, MartinPoulter (talk) 13:06, 3 May 2017 (UTC)

If memory serves, @Mpaa: has helped me out with similar issues in the past. But please don't do any more proofreading of the index until it is taken care of, due to potential page moves that will need to be made. Londonjackbooks (talk) 18:49, 3 May 2017 (UTC)
Fixed.— Mpaa (talk) 21:24, 3 May 2017 (UTC)
That's brilliant. Thanks very much for your prompt and helpful action, Mpaa. MartinPoulter (talk) 13:23, 4 May 2017 (UTC)

Page loading for editing is painstakingly slow

For the past week, opening and displaying a page for edit has been very slow - to the point that I must wait minutes for a page to render, if they render at all. I open and edit consecutively, several pages at a time but that shouldn't be a problem because I have sufficient RAM memory and other websites load normally.

  • Cleared the Index caches, with all cache controls available to users.
  • Yesterday, monitored both my internet down and upload speeds every hour, and it's always normal and the same.
  • Requested my ISP to do the same, and clear any cache that would affect performance, and he confirmed that all is normal.
  • Checked the Wikimedia servers and they were all operating normally.
  • Scanned for viruses and adware and everything is clean. (Also I am going through OpenDNS for added protection.
  • Tested both Chrome and Firefox and the problems are the same.

Wikisource page loading from the Commons is painstakingly slow. Any ideas? — Ineuw talk 04:43, 3 May 2017 (UTC)

Possibly the same issue as #Problems with page loading in the Page: namespace above? —Beleg Tâl (talk) 12:19, 3 May 2017 (UTC)
test with safemode (instructions above) — billinghurst sDrewth 13:41, 3 May 2017 (UTC)
Check your console for errors. — billinghurst sDrewth 13:42, 3 May 2017 (UTC)
As usual, completely missed that post. I knew that there are loading errors with load.php from looking at the console at other times, but I wasn't sure if it's because of my gadgets or scripts. I will repeat the actions with the console. — Ineuw talk 18:51, 3 May 2017 (UTC)
I tested "safemode" and it works, and generated the following error:
Gadget "CollapsibleNav" styles loaded twice. Migrate to type=general. See <https://www.mediawiki.org/wiki/RL/MGU#Gadget_type>.
Also, this is not a productive solution. So, I created a phabricator ticket T164443Ineuw talk 00:44, 4 May 2017 (UTC)
@Ineuw: If the page loaded in Page: nswithout delay using safemode, then the problem that relates to your use of gadgets and/or js scripts, so poking it to phabricator isn't easily going to resolve your issue. You need to identify where your issue is situated. That will be either turning off gadgets and/or blanking scripts while the source is identified; OR via the use of the console and developer tools.

I have resolved issues that I see with the gadgets that I have selected. Each persons choice of gadgets will be different, so each person can check to see if they are getting errors, or resolving issues. You alone can test your chosen environment. — billinghurst sDrewth 01:15, 4 May 2017 (UTC)

  •  Comment collapsiblenav should be resolved, I amended how the gadget loads. — billinghurst sDrewth 00:57, 4 May 2017 (UTC)
Got your point about finding the error, so I must disable all, and re-enable them one by one. — Ineuw talk 01:23, 4 May 2017 (UTC)
  • Disabled all gadgets and tested page opening, and nothing changed. I am sure it's not the gadgets.
  • Removed all scripts, (but left the common.css untouched) and emptied and disabled browser cache. The load was still slow but faster than with the scripts. I also got the following console messages.
This page is using the deprecated ResourceLoader module "jquery.ui.widget".  load.php:31:289
This page is using the deprecated ResourceLoader module "jquery.ui.core".
Please use "mediawiki.ui.button" or "oojs-ui" instead.
Ineuw talk 02:28, 4 May 2017 (UTC)


Okay, and that confirms that it is in your gadgets or user scripts. Safemode also turns off our enWS forced scripts. All of those are warnings, not faults/errors, so they can be ignored.

We need for you to determine which script or scripts cause the problem. Either start with all on, or all off, and singularly increment until the erroneous script is identified. Or start half way with one of gadgets or user scripts running and work up or down from there. — billinghurst sDrewth 02:51, 4 May 2017 (UTC)

Still not 100% sure what causes the problem but the first thing I noticed that after removing all code, the CharInsert bar remained on top of the edit window, thus making the following snippet of code unnecessary, and possibly the cause of the problem. window.charinsertMoveTop = true;. Restored my user defined Charinsert and a page is loading faster but still not as it was in the "old days".
I also suspect that the Common.js page is corrupt because when in edit mode, it didn't bring up the normal .js editor features and remained an ordinary text page. I managed to bring it back by clearing the page cache, but it took some three minutes do display correctly. Can I delete this page and recreate a new common.js? Or, should I move my code into the vector.js? Vector is what I use always. — Ineuw talk 07:12, 4 May 2017 (UTC)
You are running nopinserter as a pushed script rather than the gadget, and it has dependencies, so that will clearly have been one issue. I have wrapped it in a means to utilise the dependent script. I would have thought that you could have just used the gadget, but that is your choice. Your other two loaded scripts may or may not cause issues, if they use mw.util, then they will cause issues and we will need to amend their actions too.

re .js files. Common.js is "common" to all skins, and is the place that you should run scripts. You would only put vector skin only scripts in vector.js. — billinghurst sDrewth 10:05, 4 May 2017 (UTC)

┌─────────────────────────────────┘
@Billinghurst: With all due respect, I challenge your assessment of the problem, and you ignore my concerns. I say this because User:Beeswaxcandle has the same problem and I doubt we use the same setup and editing tools.

I don't expect you to know the answers and solutions to the issues, but I was hoping that you give more consideration to my concerns. I tested the loading problem for the past week before I posted in the Scriptorium.

I am now logged in as User:IneuwPublic, where the setup mirrors that of User:Ineuw, including the gadget selections. In fact, this account is linked to the Ineuw account's scripts because they are don't exist in this account. With this account everything works amazingly fast.

The problem must originate with the User:Ineuw account overall. An hour ago, I deleted ALL cookies of the Wikimedia family from the browser and logged in anew, but it still hangs when attempting to load a page.

My modified copy of the nopinserter has worked for over a year without any problems. It was modified to omit the constant popup message, since I insert nop en masse after proofreading at least 100 pages and this was the original intent of the script designer.

I also use a modified copy of pathoschild's proofreading script as User:Ineuw/common.js/proofreading.js because the standard script surrounds mdashes with spaces, (the French Wikisource requirement), whereas we don't.

Did the wmf software updates remove the need for window.charinsertMoveTop = true;.? I asked because now the Charinsert is on top of the text area box by default.

It occurred to me that the problem may be a disk error on the server where my settings are stored and that is why I moved and recreated common.js and common.cs and then I deleted the originals. Suggestion the use of vector.js and vector.css was an alternative to recreating the two modules, because I only use the vector skin. — IneuwPublic talk 18:09, 4 May 2017 (UTC)

@Ineuw: May I double your "with all due respect." I have fixed four gadget issues,[28] and mimicked one of those fixes in your common.js file each of these could have contributed to the issue that you describe, and I have spent numbers of hours identifying issues of the community concern, and looking at the broader WMF environment (rather than doing the things that I would have liked to have done).

Your commentary is facile, and it ignores facts that it is only happening in limited situations (some people and at this wiki) which indicates local settings issues, not systemic problems.

You have not looked at the broader wikimedia discussions that have been occurring about mw.util and where people are using it that they need to fix local settings. You simply are busy reacting and talking, not listening and providing useful information that allows some to diagnose your issue. Deleting files, moving things out of common.js to your skin, rather than simply commenting out lines of code reinforce this. The simple fact is that you need to identify where the problem is occurring (which script, which gadget, which setting) and not yammer about what has or hasn't worked over periods of time, it is all irrelevant in wikis where the underlying code develops on a weekly cycle

Actually stuff it all. I am giving up explaining where people aren't listening. Read the following, the information is there and you get to join the dots.

If you come off the default settings, you own your problems and solutions. — billinghurst sDrewth 23:10, 4 May 2017 (UTC)

Read it and got it. The yammerer. — Ineuw talk 18:42, 5 May 2017 (UTC)

Gesta Romanorum sanity check

I've finally got back to work on this somewhat complex project and think I've done enough that it's apparent where it's headed. I also want to provide more complete Wikisource translations of at least some of the tales, as the original explications of the tales are mostly abbreviated or omitted in the translation. Plus there's this 1600 edition at Internet Archive that would be very challenging to transcribe. So ... could someone please check this out so that lots of pages won't need renaming or fixing in the future? Mudbringer (talk) 07:10, 7 May 2017 (UTC)

Our guidance for different translations of the same work is to utilise the name of the translator(s) as part of the disambiguation. (Wikisource:naming conventions) I am presume that is question you are asking. — billinghurst sDrewth 11:08, 7 May 2017 (UTC)
Thank you very much for the feedback. I don't see any mention of translations on the page you linked, or any mention of disambiguation on Wikisource:Translations, but the guidance you mentioned seems reasonable. In this case, however, the two editions I have index pages for are of the same translation, by Charles Swan, which I believe is the only PD translation in nearly modern English. The 2-volume edition I'm working on right now is almost identical to the original edition of the Swan translation, merely provided with an extra introduction. The other edition, by Hooper, has been lightly revised, but is mostly important for Hooper's introduction, which has one of the best explanations of the history of the work.
Duh! See Wikisource talk:naming conventions. We are atrocious at good help and policy pages. Transcription rulez! <wry smile> — billinghurst sDrewth 22:11, 8 May 2017 (UTC)
My original intention in posting to the Scriptorium was to ask in general if anyone could look over the linked indexes pages, and the the pages in the main namespace that I've begun, to alert me to any problems that I have been overlooking. I apologise for the vagueness. As for specific concerns, I would like to hear anyone's recommendations on these issues:
  • I have the two volumes for the 1871 edition linked from the translation/disambiguation page (to which I've added more detailed information on these and other editions) and I would prefer to keep it this way rather than require the reader to click through a page listing just those two volumes. On the other hand, it could be useful to have a single page listing the tables of contents of both volumes.
  • On Latin Wikisource I have only placed one edition. Ideally I'd have a page listing various Latin editions, but I'm hesitant to try that with my poor Latin skills. Given these constraints, I'm trying to provide interlanguage links that will be helpful and logical, but I'm sure there's room for improvement.
  • There are two books that contain well-known discussions of Gesta Romanorum. I've just read on Help:Disambiguation that to include these books it is recommended to make a portal page. To do that, could I just rename the Gesta Romanorum page to Portal:Gesta Romanorum, modify the contents as needed, and leave the resulting redirect as is?
  • As I said, the volume I'm working on now is rather complex. For example Of the Life of Alexius, son of the Senator Eufemian includes an endnote longer than the translation itself, and both the translation and endnote have footnotes. I now have the footnotes appear immediately following their sections, but would it be better to have them all at the bottom of the page?
I'm afraid my queries are still too vague, but it's the best I can do right now. Mudbringer (talk) 06:31, 8 May 2017 (UTC)
My thoughts:
  • Linking directly to both volumes of the Wright translation from Gesta Romanorum is fine, but it would also be desirable to additionally create Gesta Romanorum (Swan 1824) (or whatever) as a main page for the two-volume translation work.
  • I'd leave LAwikisource as it is unless you plan to add more Latin versions to LAwikisource yourself.
  • What you'd need to do is leave Gesta Romanorum as a list of translations of the text, and create a separate Portal:Gesta Romanorum for works related to the text.
  • Help:Footnotes and endnotes can be your guide for footnotes and endnotes.
And finally, your efforts look to be quite good, keep up the good work! —Beleg Tâl (talk) 12:13, 8 May 2017 (UTC)
Thanks for the helpful and encouraging comments! Mudbringer (talk) 13:50, 8 May 2017 (UTC)

02:24, 9 May 2017 (UTC)

Beta Feature Two Column Edit Conflict View

Birgit Müller (WMDE) 14:28, 8 May 2017 (UTC)

@Birgit Müller (WMDE): Is this expected to work in the Page namespace here? Because it doesn't appear to - I get the default edit conflict interface with the Beta enabled. Sam Walton (talk) 09:37, 11 May 2017 (UTC)
Hi Sam Walton, it doesn't work in the page namespace, because Wikisource uses a different content model for the page namespace that is not wikitext (same goes for Wikidata). But good point: If the old edit conflict resolution page does work in the page namespace, the new one should ideally work there, too. We'll look into it & check what would be needed to achieve this & I'll let you know. Thanks for the heads up! --Birgit Müller (WMDE) (talk) 14:17, 11 May 2017 (UTC)

Templates {{FI}} and {{FIS}} are broken

is template:FIS now broken? and flat sidebar menu? Slowking4SvG's revenge 01:37, 11 May 2017 (UTC)

Seems to be. Haz talk 03:04, 11 May 2017 (UTC)
Hopeful this can be remedied. I appreciate how the templates behave/render on various devices. Londonjackbooks (talk) 11:34, 11 May 2017 (UTC)
appears to work now. Slowking4SvG's revenge 11:34, 12 May 2017 (UTC)

index page problem

Hi. I just uploaded a book to the Commons but when I made the index page in the Persian Wikisource, it didn't show the page numbers. Can anyone help me with this? --Yousef (talk) 08:11, 11 May 2017 (UTC)

Try now. It looked like the pagelist command was missing for some reason. Beeswaxcandle (talk) 08:20, 11 May 2017 (UTC)
Thank you very much. --Yousef (talk) 08:23, 11 May 2017 (UTC)

@Yoosef Pooranvary: Can you please help with the Persian writing in djvu pages 6, 7, 9 of Index:Wanderings of a Pilgrim Vol 1.djvu? The author's name is always in Persian in this work. Hrishikes (talk) 00:39, 12 May 2017 (UTC)

Done. Although it is written in Arabic alphabet and it seems to be Persian, the name is not Persian. It’s transliteration to English is "Funny Parx". --Yousef (talk) 06:51, 12 May 2017 (UTC)
@Yoosef Pooranvary: Many thanks. If you can please transcribe the Persian at the lower right of the image on page 6, I have plan to crop the image (and de-color the background) and transcribe the writing. Hrishikes (talk) 07:39, 12 May 2017 (UTC)

Layout 3 skewy for me

Looking at The Strand Magazine/Volume 1/Issue 1/The Metropolitan Fire Brigade the header for layout 3 appears off to the right of the page and I need to scroll the page. Prior to my editing I though that I would confirm that it is a problem wider than me. — billinghurst sDrewth 03:41, 15 May 2017 (UTC)

I see the same thing, and likewise on other, simpler pages. Mudbringer (talk) 05:28, 15 May 2017 (UTC)

Tech News: 2017-20

21:48, 15 May 2017 (UTC)

A note about a future change

Hi everyone. Some of you have already noticed this in m:Tech News above and as it's been already mentioned twice, but I wanted to flag again that because of an upcoming change, there's some code that may need to be fixed on this wiki:

You can find the list of the pages you'll want to check in a table on mw.org. In most cases those are false positives (I have looked at some of those pages and looks like there are just several cases of bad OCR), and you only really need to fix as explained on mw.org when the "defective" syntax is inside a template or a wikilink, basically. When you have completed the checks, please edit the Notes field of the table accordingly! If you have questions, please reach the team on the talk page. Hope this helps, and thanks for your attention and your help, --Elitre (WMF) (talk) 10:35, 16 May 2017 (UTC)

Hi,
As most of these problem are bad OCR typo in 'not proofread' pages (especially in the United States Statutes at Large), I've launched WS:AWB to replace them by a simple hyphen.
Cdlt, VIGNERON (talk) 12:18, 16 May 2017 (UTC)

RevisionSlider

Birgit Müller (WMDE) 14:39, 16 May 2017 (UTC)

15:19, 16 May 2017 (UTC)

21:08, 16 May 2017 (UTC)

Building a "small language" corpora

Is Wikisource the right place to compile a small language corpus, perhaps on a test Wikisource? The source of the works is a university that has published literature compositions of an indigenous language as well as student's literary works including poems and short novels. Unfortunately there is a pressing need to archive these documents by means of digitization. -Masssly (talk) 10:06, 14 May 2017 (UTC)

Multilingual Wikisource (a.k.a. Old Wikisource) is the place for works in languages that aren't big enough to have their own wikisources. They already have a small corpus of Cherokee texts and Inuktitut textsBeleg Tâl (talk) 16:57, 14 May 2017 (UTC)
Good to know Beleg Tâl, thanks. By small I actually meant between 3K and 4k (and growing) pages of texts. I suppose that is sufficient to have their own Wikisource. I could use some help setting it up in the incubator -Masssly (talk) 20:38, 14 May 2017 (UTC)
@Masssly: multi language Wikisource is our incubator for Wikisource, it can be set up now and easily and you should ask for help at their mul:Wikisource:Scriptorium. From there they can spool it into its own independent source at the appropriate time, they will also be able to push it through the language committee more easily. — billinghurst sDrewth 00:38, 15 May 2017 (UTC)
Make sure to check the licenses; projects like these don't tend to worry about copyrights of the students or the works they've translated as well as Wikimedia would prefer.--Prosfilaes (talk) 03:48, 15 May 2017 (UTC)
@Prosfilaes: I didn't envision this could be a problem. But thinking of it now it seems to me that the only option available to not get into trouble is to convince the school to release their material under BY-SA. My question is: and if they did, could I upload on their behalf? -Masssly (talk) 07:45, 18 May 2017 (UTC)
If the work has scans to be uploaded to Commons, then I would think that it should have supporting documentation through the OTRS process would be beneficial (see c:Commons:OTRS). Following this should cut out a lot of potential grief, and set a steady base to work upon. — billinghurst sDrewth 10:54, 18 May 2017 (UTC)
@Masssly: I could help you starting on the Multilingual/Old/Incubator Wikisource. Could you tell us what language you are talking about? (and if possible, the corresponding ISO code). Cdlt, VIGNERON (talk) 16:54, 16 May 2017 (UTC)
@VIGNERON: Language is Dagbani and the ISO is dag.Masssly (talk) 07:34, 18 May 2017 (UTC)

For the records oldwikisource:Main Page/Dagbanli (and I've search on the websites - like Internet Archives - but didn't find any text in this language). Cdlt, VIGNERON (talk) 16:06, 2 August 2017 (UTC)

Wikimania 2017 "Wikisource Year in Review"

hey all, i have been accepted for a talk about wikisource at wikimania. please get your suggestions, gripes, talking points in. i would be happy to present your brief.- cheers. Slowking4SvG's revenge 01:32, 20 May 2017 (UTC)

Proofread namespace xwiki alignment for WS sisters

As a note, the Vienna hackathon (currently running) will be pushing ahead with the long-discussed plan to harmonise the namespaces for Index: and Page:. There is no specific timetable for the implementation / migration, though the process to migrate is now in play. Whenever it happens it will mean a small disruption, and this will be announced to us well ahead of the change. There is no requirement for the community to do anything except note the forthcoming change, (and if you do hardcode namespaces into templates, etc. then you will need to update such). Technical detail is in the phabricator ticket and further background at Mediawikibillinghurst sDrewth 16:25, 21 May 2017 (UTC)

22:02, 22 May 2017 (UTC)

21:05, 23 May 2017 (UTC)

File:Dante (Oliphant).djvu

The File:Dante (Oliphant).djvu needs to have the Google notice REPLACED with a blank page. The person who scanned the volume did not include the front cover. --EncycloPetey (talk) 01:05, 23 May 2017 (UTC)

@Mpaa: where are we with getting the fix implemented at Commons? — billinghurst sDrewth 10:41, 23 May 2017 (UTC)
Still far away unfortunately. At least pywikibot has a now a method that can handle it (see https://phabricator.wikimedia.org/rPWBC461f747f272d8d8c99beadcf5c51f4fa0fa423e0). it is actually called "whiten_page(()", the commit message was not updated. It is not clear to me how to identify the files to work on (a template? A category?)— Mpaa (talk) 20:09, 23 May 2017 (UTC)
At least I have a proto-script that can do it just with one command.— Mpaa (talk) 20:30, 23 May 2017 (UTC)
@Mpaa: Billinghurst started a list at User:Wikisource-bot/Lead google page, which I've been adding to as I find them. However, the ones that I've added require removal of the page rather than replacement. Beeswaxcandle (talk) 07:23, 24 May 2017 (UTC)
IMO, the preferred way would be to place files in a Maintenance category (e.g one for page blanking and one for page removal). In that way, it would be possible to use built-in pywikibot features. Another option would be to put a template on them.
A list in a page needs to be parsed and entries need to be deleted once processed successfully. It is not impossible but the other way is supported better by the library.--Mpaa (talk) 18:56, 24 May 2017 (UTC)
Thanks, Mpaa! --EncycloPetey (talk) 13:35, 24 May 2017 (UTC)
Welcome.--Mpaa (talk) 18:56, 24 May 2017 (UTC)
This section was archived on a request by: --EncycloPetey (talk) 13:35, 24 May 2017 (UTC)

Hi all,

I'd like to have another crack at adding support to {{plain sister}} for including links to sister projects where they're not from the directly linked item, but rather are from the parent 'work' level. I had a look at this a while ago, but it turned out that there are some editions that are editions of multiple works. Now I've fixed up for this scenario, and all available site links will be listed. (Actually, I'm having trouble finding an active example of this, but you get the idea.)

This is the change I am suggesting. The output will do things like this: Script error: The function "interprojetPart" does not exist. Where our The Nether World currently has to have a manual override to display a sitelink to Wikipedia; with the above change, it will follow the Wikidata connections.

What do you think? :-)

Sam Wilson 20:08, 24 May 2017 (UTC)

Oh, also: have a look at Module:Plain sister/sandbox testcases and see if you can suggest any other tests I might add (i.e. page names, Wikidata IDs, etc.). Thanks! (Tomorrow is hack day at WikiCite, where I am now, and I'll be working on this with some other Wikisourcerers.) Sam Wilson 20:12, 24 May 2017 (UTC)

Template:media

The {{media}} template generates different text about available sound files depending on whether it is used in the Author namespace or the Main namespace. And this works well under most circumsatnces, except when the Main namespace page is a versions or translations page. Witness Vita Nuova (Dante), which ought to explain that:

The icon identifies that the work includes a spoken word version.

But because this is the main namespace, it states:

one or more chapters are available in a spoken word format.

Could someone familiar with this template suggest a means for making this work properly for versions and translations pages? --EncycloPetey (talk) 01:22, 27 May 2017 (UTC)

It would be a bit of faffing around to make that modification for a one-off use. I would suggest using the "icon" parameter, and then add your preferred text next to it. — billinghurst sDrewth 07:08, 27 May 2017 (UTC)
Not a one-off. I'll have the same problem for several plays by Aeschylus when I set that up in a few weeks, and this has the potential to be a regular issue. We just don't have many people yet adding audio files or setting up versions pages. Fixing the problem either requires adjusting the template, or following what the Italians have done in creating a "Work:" namespace for versions/translations to keep those pages clearly separate from the editions (which remain in their Main namespace). Neither solution would be simple, but adjusting the template would be simpler than creating another namespace. --EncycloPetey (talk) 15:15, 27 May 2017 (UTC)
Would it perhaps be best to integrate {{media}} into the header template, since that's where the distinction is made between editions and versions? —Beleg Tâl (talk) 16:11, 27 May 2017 (UTC)
If it is going to be needed then there is no issue making a change. Please describe the circumstances that will be different from now. Will it only be main ns, or will it be others? What text is required to be utilised in this situation. Probably best to have the amendment conversation on the template's talk pgae. Re part of header ... IMO I don't think that we would need to add it in the header configuration, it isn't used that much so no point in bulking up templates and modules for low use templates. — billinghurst sDrewth 01:00, 28 May 2017 (UTC)
We could add some 'custom text' parameters to the media template (e.g. 'before', 'icon', and 'after' or something). I don't think it should be part of {{header}} because it's not applicable to most works. Sam Wilson 07:30, 28 May 2017 (UTC)

12:18, 30 May 2017 (UTC)

How do you fix 1911 Encyclopædia Britannica/Hetaerism ?

On page 1911 Encyclopædia Britannica/Hetaerism, "HETAEHISM" should be "HETAERISM" and "¥racpa" should be "ἑταίρα", but when I click on the edit button, there's no content there, just a template which I don't understand... AnonMoos (talk) 14:29, 30 May 2017 (UTC)

@AnonMoos: Click on the page number to the left of the heading ([415]). That will lead you to the index where you can make necessary changes. Londonjackbooks (talk) 15:25, 30 May 2017 (UTC)
Maybe it's because I'm using an older browser version, but I don't see any "415" link -- just links to Hesychius of Miletus and Heterokaryota. AnonMoos (talk) 15:48, 30 May 2017 (UTC)
@AnonMoos: Go here: Page:EB1911 - Volume 13.djvu/431. Hrishikes (talk) 16:04, 30 May 2017 (UTC)
if you check "display options" on left hand menu, you may have "Page links hidden" enabled, if so, you would need to toggle to "Page links displayed" - they keep improving display. Slowking4SvG's revenge 16:22, 30 May 2017 (UTC)
Hrishikes -- thanks, got it... AnonMoos (talk) 16:58, 30 May 2017 (UTC)
And to close the loop, it is my understanding that it is the default for all new users and IP editors to have the page links displayed (dynamic layout). — billinghurst sDrewth 22:31, 30 May 2017 (UTC)

Community Collaboration

Poll with 2 questions:

(1) Is it time to switch to a new Community collaboration? We've had Thoreau since January, and have made great progress. Only two items are still missing: The Maine Woods, which will be PotM in July, and his Journal, which has not been attracting much attention from editors.
(2) What about Edward VII (of the UK) as our next collaboration? Edward VII (d. 1910) is the most recent monarch of the UK likely to have much material in public domain. George V, his successor, died in 1936. There are already offsite links to several books listed on his Author page, but we have almost nothing here on Wikisource. --EncycloPetey (talk) 21:22, 15 May 2017 (UTC)
  •  Support yes; yes --EncycloPetey (talk) 21:22, 15 May 2017 (UTC)
  • I support Ed7 as next collaboration. I thought we were going to keep Thoreau till the end of the year though, though maybe I misunderstood somewhere. —Beleg Tâl (talk) 21:58, 15 May 2017 (UTC)
    There is no set date for changeover on these. Sometimes they flip after a few weeks, sometimes they linger for months. I had thought initially, we'd keep Thoreau until July, but (as I noted above) most of the work likely to be done for Thoreau has now been accomplished or is scheduled, so we might as well rotate to a new collaboration. --EncycloPetey (talk) 22:34, 15 May 2017 (UTC)
    Yeah, I think we can move on since we've got most of the important stuff already taken care of. —Beleg Tâl (talk) 01:41, 16 May 2017 (UTC)
  • Happy for a change, though I would like to see us look to have us consider something like the Famous Women Series (see partial list). Women writing about women. Some of these I had been thinking that we could do as PotM, though think that the community collaboration would be a better means. — billinghurst sDrewth 12:11, 16 May 2017 (UTC)
    I'd like to see future collaborations concerning minority groups, women, or non-English-speakers. The Famous Women Series would fit nicely, and I know I've raised the issue before of 19th-century African-American biographies, which I also think would make a good choice. I've also jotted down in my own notes the names of several women for whom we have shockingly little coverage and who deserve better. Once we get started on whatever we choose next, I hope to resurrect the nominations page for the community collab, which has lain dormant for a very long time. We didn't get any new nominations during our work on Thoreau, so part of the reason for suggesting Ed VII is that there are already external links to books, and there aren't likely to be too many more. I expect it will go quickly because of that. It's also very different from anything we've done recently, so I hope it will draw in new editors with different interests. If you can start a list for the Famous Women Series with external links to scans in the nominations page, I think it would make a very strong nom. --EncycloPetey (talk) 14:28, 16 May 2017 (UTC)

Done Based on support for the change and marginal favor of Edward VII, I've switched the Comm. Collab. over for June. We'll run this for a while and see what happens. In the meantime, I'll archive and clean up the Collaboration Proposals page, and then advertize its existence so that we can establish a set of proposals from which to draw. I see that billinghurst has already initiated his proposal, so we'll have at least one new option ready. --EncycloPetey (talk) 03:43, 1 June 2017 (UTC)