Wikisource:Scriptorium/Archives/2018-10

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

17:35, 1 October 2018 (UTC)

Encrypted PDF of PD book

The text of this book: [5] is out of copyright (Author:George Bramwell Evens, died 1943) but is only available as an encrypted PDF to "borrow". Does anyone have suggestions for uploading it? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:35, 4 October 2018 (UTC)

Two points:
  • The original work is PD in the UK by the 70 pma rule, but it is PD in the US, as it was first published in 1932?
  • The copy you link to is a 2002 edition, so it's hardly surprised that access is restricted, if it contains copyrighted modern material.
BethNaught (talk) 20:58, 4 October 2018 (UTC)
not renewed here [6]; [7]; [8]; [9]; [10]; [11] and no hits at https://cocatalog.loc.gov/ (after 1978 renewal) = i would say PD-US no renewal - do not see a 1932 scan at Internet Archive; i see there is a copy of 1946 edition at Drew University in New Jersey, and Michigan State University, i can drive down and scan a copy [12] - name your price. Slowking4SvG's revenge 21:44, 4 October 2018 (UTC)
The 2002 edition contains, AFAICT (and I'll check against my paper copy once I can access it), no new material. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:15, 5 October 2018 (UTC)
I see no evidence that it was ever published in the US, so the URAA would have made it publication+95 in the US, or in copyright in the US until 2028.--Prosfilaes (talk) 03:04, 5 October 2018 (UTC)
we can have that discussion on commons. Slowking4SvG's revenge 16:23, 5 October 2018 (UTC)

Licence check: anonymous 1929 Australian article

Please can someone advise what licence should apply to this anonymous 1929 article, published in Australia: Examiner (Launceston, Tasmania)/1929/"A Romany in the Fields"? If there's a URAA issue, should it be moved to the Canadian site? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:51, 5 October 2018 (UTC)

{{PD-anon-1996|1929}}billinghurst sDrewth 14:26, 5 October 2018 (UTC)
@Billinghurst: Thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:10, 5 October 2018 (UTC)

Text OCR cleaned up, anyone want to Proofread? ShakespeareFan00 (talk) 21:46, 5 October 2018 (UTC)

Problematic PDF: The Migration of Birds - Thomas A Coward - 1912

There is a problem with File:The Migration of Birds - Thomas A Coward - 1912.pdf; please see c:Commons:Village pump#Problem with PDF, and advise if you can. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 10:35, 5 October 2018 (UTC)

Some of the PDFs are overly compressed for display in Mediawiki. I think that either @Mukkakukaku, @Hrishikes: has fixed some of these previously, I cannot remember whom. We had one in the past couple of months that should be in the archives. We have a section further up to park broken files, for whatever reason. — billinghurst sDrewth 14:32, 5 October 2018 (UTC)
@Pigsonthewing: -- Done. OCR is not there, however. If you insist on OCR layer, then I'll do some more experiment. Hrishikes (talk) 15:35, 5 October 2018 (UTC)
@Hrishikes: Working well now, thank you, What did you do to fix it? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 18:17, 5 October 2018 (UTC)
The file was in pdf 1.5 format with compression. I resaved it in pdf 1.4 format without compression. Hrishikes (talk) 02:48, 6 October 2018 (UTC)
I guess it was a similar problem as https://phabricator.wikimedia.org/T203402. --Jan Kameníček (talk) 09:52, 7 October 2018 (UTC)

Mangled page numbers. It seems someone needs to rethink the module, so it is in fact COMPATIBLE with Proofread page, as currently once you are inside a section generated using {{aligned table}} the automatically generated page numbers aren't displayed correctly (Long term issue). ShakespeareFan00 (talk) 10:01, 7 October 2018 (UTC)

As has been discussed for an extended period, within templates put the table row markers at the beginning, rather than at the end. I have never understood why people close with a row open statement, especially at the end of a table, an extra row marker is like "why?" — billinghurst sDrewth 11:47, 7 October 2018 (UTC)
As an extra comment, the template itself says that this is problematic for page numbering, and it is your choice to continue to use the template. — billinghurst sDrewth 11:52, 7 October 2018 (UTC)
Yes, I know.. Sometimes it would be nice to have long term solutions, (It was me that documented the incompatibility originally). ShakespeareFan00 (talk) 12:38, 7 October 2018 (UTC)
The fix you suggest about row openings would need to be made in Module:Aligned table, All other table handling in the work is based on that template. I'm using it rather than direct table syntax because of concerns about transclude limits for table rows.ShakespeareFan00 (talk) 12:44, 7 October 2018 (UTC)
Yes, which is why I haven't fixed it. I can fix templates, LUA is beyond my capacity, or maybe that is my patience-level. — billinghurst sDrewth 13:07, 7 October 2018 (UTC)
Same here. I'll consider if a different approach might work. The work will need to be split into sections anyway.. ShakespeareFan00 (talk) 13:33, 7 October 2018 (UTC)

23:38, 8 October 2018 (UTC)

Google Books PDF

What's the best way to upload the PDF available here - do we have a tool for that, like ia-upload? Note that it includes a Google Books cover sheet. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 13:48, 9 October 2018 (UTC)

@Pigsonthewing: -- It can be done with url2Commons tool. Hover the cursor over the "Ebook - Free" notice, then right click the pdf option and copy the link address. Use this as the url in the first box of url2Commons. Use the main Google Books address as the source url in the second box. OAuth authorization will be required. OAuth often shows failure in case of this tool. It is false failure. Keep the OAuth screen as it is and go to the tab having the tool window to complete the transfer. If you want to remove the frontsheet, you will need to download, edit and re-upload. Hrishikes (talk) 15:06, 9 October 2018 (UTC)
@Hrishikes: Thank you. The simulation failed, complaining about an invalid URL. I trimmed the "?" and everything after it, and then the simulation worked, but the upload failed with " ERROR: null". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:52, 9 October 2018 (UTC)
@Pigsonthewing: -- c:File:A Discourse on the Emigration of British Birds.pdf -- Hrishikes (talk) 16:19, 9 October 2018 (UTC)

BHL IDs

The w:Biodiversity Heritage Library is a rich source of out-of-copyright texts, and a good ally for Wikimedia projects. We store BHL author IDs in Wikidata, as P:4081. Can we add these IDs to {{Authority control}}? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:58, 8 October 2018 (UTC)

Looks like a change to Module:Authority control, which seems rather straight-forward. --Mukkakukaku (talk) 23:56, 8 October 2018 (UTC)
@Pigsonthewing: -- Done. It needed change in the module, not the template. Hrishikes (talk) 15:22, 9 October 2018 (UTC)
Looks like it's working; thank you. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:15, 10 October 2018 (UTC)

Notable printers

The plate at Page:The birds of Tierra del Fuego - Richard Crawshay.djvu/180 (like others in the same work) was printed by West Newman & Co. I have created a Wikidata item for that company, d:Q57166684. What's the best way to link them? I've used {{Reasonator}}, for now, but am open to counter suggestions. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 14:25, 9 October 2018 (UTC)

We have created portals for some publishers, no problem for doing that for printers, though that has usually been for complete works. If it is just the images, then maybe Commons alone is sufficient. No need to replicate what is done better elsewhere. — billinghurst sDrewth 22:05, 10 October 2018 (UTC)

Two-page table

I should be grateful if someone could verify the table on Page:The Migration of Birds - Thomas A Coward - 1912.pdf/114, which also incudes data from Page:The Migration of Birds - Thomas A Coward - 1912.pdf/115, and advise on formatting. I have created a single image, above, to aid this.

Is this the best way to show a table which runs horizontally over two pages? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 15:45, 9 October 2018 (UTC)

For maps and tables that spread, it does sound best to handle them on one page, and comment on the second. It is one of the adaptations that makes sense to me. — billinghurst sDrewth 22:02, 10 October 2018 (UTC)

Match and Split bot

As reported by both @Jasonanaggie and @Beleg Tâl on Wikisource:Bot requests, the Match and Split functionality is not currently working. Going to @Phe-bot's page (https://tools.wmflabs.org/phetools/match_and_split.php) says match_and_split robot is not running. Please try again later. @Phe has been pinged at least twice about this. -Einstein95 (talk) 20:45, 9 October 2018 (UTC)

He has been active on another wiki recently, I have asked there if he was no longer supporting the tool whether it is something that we can migrate. — billinghurst sDrewth 22:00, 10 October 2018 (UTC)

RFC: Automating "Wikipedia" link in Header if WD main topic is activated

I have been tromping through transcluding and WD'ing Dictionary of Indian Biography which has been proofread, though predominantly, not transcluded. Quite a number have Wikipedia articles, and it is pretty tiresome to transclude, then add WD, and identify whether they have a main subject link, then have to go back to the biographic article again. Whereas where I have added "main subject" to d:Q57008414, it would be my preference if the database pulled and automagically added the Wikipedia link, rather than the extra edit.

I am trying to identify any downsides to such an approach, and apart from wrong additions (which can equally happen here. About the only one that I can identify is if someone added more than one main subject, where we would be forced to choose one (if rank preferences where used), or maybe choose none, though mark as problematic and needing resolution. Otherwise, I am unable to identify major stumblings.

@Samwilson, @Mike Peel: from your WD experience, I am guessing that this is a relatively easy data pull. [Mike this happens through {{plain sister}} which is embedded within {{header}}, and in the main ns is an indirect pull as it is a many to one relationship, unlike {{author}} which plain sister does as a straight pull of the interwiki data).

So I am seeking the community opinion on

  1. their thoughts on automating the linking;
  2. any hurdles for implementation; and
  3. the technical aspects for implementation.

Thanks. — billinghurst sDrewth 11:37, 7 October 2018 (UTC)

(comment) I think you mean this property? --Mukkakukaku (talk) 17:51, 7 October 2018 (UTC)

comment

My first comment is that there already many links in place for Wikipedia, so as we have done for other migrated data, where the the parameter is implemented within the existing header it overrides any WD data pull. This approach allows projects to work out what they wish to do with their data. This allows us to identify where we have overrides in place (current situation for images and dates of life). — billinghurst sDrewth 11:43, 7 October 2018 (UTC)

  •  Comment I can't see any issues where the data item will have a single "main subject" for biographical articles, but aren't there situations where we would pull information that isn't suitable, say for non-biographical non-dictionary data items? Some books will have a "main subject", but the WP article of primary interest is actually the WP article about the book, and not the article about the book's subject. --EncycloPetey (talk) 19:12, 7 October 2018 (UTC)
    Fully agree about biographical/people. Maybe that is part of our decision-making process. If it is "edition" d:Q3331189 it should one path, if it is an article, it should follow another path. Let us try mapping these. — billinghurst sDrewth 03:00, 8 October 2018 (UTC)
    This does not account for editions of articles though, nor articles which themselves have wikipedia articles. In my opinion, we should either a) have the article's wp link override the subject's wp link, or b) have two wp links (like how we have two commons links for gallery and category), or c) not use plain sister but perhaps have a special template for such cases. —Beleg Tâl (talk) 11:27, 8 October 2018 (UTC)
    Not just "editions" but also versions pages and translations pages for works, which will have any of several possible values for "instance of" (novel, poem, short story, etc.) --EncycloPetey (talk) 22:04, 9 October 2018 (UTC)
    Further to this, there are some works with multiple "main subjects", and this will need to be accounted for. —Beleg Tâl (talk) 00:15, 8 October 2018 (UTC)
    I am guessing that multiple main subjects is due to there being no single useful subject. To me, if one is given priority (higher ranking) then we show the preferred, if two are equal, maybe we ignore them., or maybe we flag them for review, and again not displayed. — billinghurst sDrewth 03:00, 8 October 2018 (UTC)
    An alternative: create a wikidata item for the group of multiple subjects and link to that from the article. —Beleg Tâl (talk) 11:27, 8 October 2018 (UTC)
    I think that this alternative happens from case 3 of flag as problematic, with an fix eventuating. — billinghurst sDrewth 14:23, 8 October 2018 (UTC)
  • The code for this is demo'd at User:Mike Peel/main topic - for Dictionary of Indian Biography/Aliverdi Khan, {{User:Mike Peel/main topic|qid=Q57008414}} will show User:Mike Peel/main topic, and if used without a QID then it will follow the page's sitelink. It should be straightforward to migrate this to Lua and to embed it into the appropriate templates directly (it's basically a couple of Lua module calls and an if statement - just written in parser functions rather than lua right now). Thanks. Mike Peel (talk) 07:00, 10 October 2018 (UTC)
    But, as noted above, this creates more than a few problems that have yet to be solved. --EncycloPetey (talk) 14:00, 11 October 2018 (UTC)

I was just attempting to validate this single page article, but have encountered an issue where the source image text is cropped early. I found a link referencing back to the original source page (https://timesmachine.nytimes.com/timesmachine/1907/02/04/106705459.pdf) where the complete text can be found.

I have added in the missing few lines of the article, would somebody be able to recreate the source image for this page using the above link so that it can be completed?

Thanks Sp1nd01 (talk) 14:30, 9 October 2018 (UTC)

Done -Einstein95 (talk) 22:11, 9 October 2018 (UTC)
Thank you! Sp1nd01 (talk) 07:50, 11 October 2018 (UTC)

Narrow no-break space for contractions?

This is something I've thought about for a long time and would like to hear what people think. In a lot of old books contractions like 'll for will and 's for is are preceded by a narrow space that never breaks for a new line. Until now I've been deleting the space (or following whatever's the trend on projects that are already well advanced, usually an ordinary space), but felt I should really be using u+202f. Anyway, on this page from Oliver Twist there's a good example of why it's important: "Why, a beak 's a madgst'rate; and when you walk by a beak's order, …" where there's a contrast between the possessive "beak's" and the contraction "beak 's". That page I proofed entering the unicode character directly, and the following page I used the entity   … Does anyone have advice on which would be better to use? Or should I just use   which would be less confusing for validators? — Mudbringer (talk) 01:58, 12 October 2018 (UTC)

I think that any of those options are acceptable so long as it's consistent within a work. I personally would probably use nonbreaking space if I were to use a space at all. I like the idea of narrow nobreaking space if you're willing to put in the effort for it. —Beleg Tâl (talk) 23:22, 12 October 2018 (UTC)
For dialectical speech, I tend to use a full space. I tend to treat such instances of elision differently from contractions. I've come across cases where a half-space is used in the source, but also cases where a full space is inserted. There are also situations where "connecting" the two parts with a non-breaking space would imply a connection not implied in the source text. For example, consider the final paragraph on this page, especially the phrase fellers 'll which is divided between two lines in the final paragraph. Using a non-breaking space when the source text allows for a line break in such a place would not be faithful to the style in which the original was printed. --EncycloPetey (talk) 01:43, 13 October 2018 (UTC)
Thank you both for the comments. The example from Red Badge of Courage is very interesting. In that work I found 'd 'll 'm 'n 're 's 've with elided initial vowel, which all appear both connected with the previous word and with an intervening space, often for the same combination of forms, such as we 're and we're. The only example of one of those appearing following a line break is the case of feller 'll that you pointed out. In Oliver Twist there's also variation between inserting spaces before these forms and joining them to the previous word, but I can find no cases in the three volumes where they appear after linebreak. …… Does anyone directly insert no-break spaces, or is it better to use the html entity? — Mudbringer (talk) 15:29, 13 October 2018 (UTC)
I would look at later editions of the same text for clues to transcription, the fashion for thin spacing to indicate a semantic distinction from a regular space did n't last. I have seen them slipped in and out to aid with the justification of the text block, as if the typesetter was in two minds about the whole business. The trend for justified text (i.e. flush left and right margins), which lasted much longer, was always going to confound the practice; requiring variable width 'full spaces' to be read as distinct from the thinnest space between the type. — CYGNIS INSIGNIS 00:04, 14 October 2018 (UTC)

We were lacking a Portal for the Renaissance period from our list at Portal:Era, so I began one.

Anyone with an interest in texts from this period (c. 1420-1630) please feel welcome to improve the meagre start that I've made. --EncycloPetey (talk) 00:13, 14 October 2018 (UTC)

Orphée aux Enfers

Could someone please strip the Google notice from File:Orphée aux Enfers (Chicago 1868).djvu (on Commons) in preparation for hosting the work on Wikisource? --EncycloPetey (talk) 01:54, 14 October 2018 (UTC)

Done --Mukkakukaku (talk) 18:34, 14 October 2018 (UTC)

Descriptions from Wikidata

In {{author}}, can we pull |description= from the English-language description in Wikidata, if there is no locally-entered value? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:14, 10 October 2018 (UTC)

It is my understanding (from an earlier time) that we cannot pull the description from the description field. We decided to not pull the occupation alone and continue to add our own description. <shrug> — billinghurst sDrewth 21:59, 10 October 2018 (UTC)
I am sure that's not (or is no longer) the case; no doubt User:Mike Peel can advise. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 12:29, 11 October 2018 (UTC)
As Mike is busy, @RexxS: for help, please. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:55, 14 October 2018 (UTC)
Try (for George Orwell, Q3335):
  • {{#invoke:WikidataIB |getDescription |qid=Q3335 |wikidata}} -> British writer and journalist
Documentation is at Module:WikidataIB #Function getDescription. You could use it in a template something like this:
  • {{#invoke:WikidataIB |getDescription |qid={{{qid|}}} |{{{desc|wikidata}}} }}
That takes optional parameters |qid= and |desc=. If qid is omitted, then it uses the current page; while desc is a local description, which overrides the wikidata. You can supply |desc=none if you want to suppress the description. HTH --RexxS (talk) 17:51, 14 October 2018 (UTC)
I have poked it into template:author/sandbox with an example visible in special:diff/8873313. I haven't done an example where we have no description, and pull from WD. — billinghurst sDrewth 20:29, 14 October 2018 (UTC)
Question, do we wish to track where we have used the WD description? — billinghurst sDrewth 01:53, 15 October 2018 (UTC)
@Pigsonthewing: It is only technically possible, and a fine idea. One of the benefits of having WD fill in blank fields is that the usual names of authors, as given in citations or a user's search, could be spilled out as synonyms of the author page's title here. — CYGNIS INSIGNIS 11:33, 14 October 2018 (UTC)
Additional comment: What would appear in the description field here that is not data or facts, better served at the respective sister sites? Errant content forking across wikimedia is one of the things WD can resolve, and there is no mechanism to address it here if a description or fact is given without references other than bluff. I would prefer that author pages function as a library index card, merely links to sources with all relevant and labelled data providing disambiguating context for the reader. — CYGNIS INSIGNIS 04:22, 15 October 2018 (UTC)
The descriptions at Wikidata are sometimes too brief, sometimes overly verbose. We do often want information in the description such as pseudonyms, pen names, other forms of their names, as well in some cases tha names of close colleagues, family members they might be confused with, or information specific to their status as author rather than whatever else they might be known for. I've seen all of these things and more placed in our descriptions, but they are not generally included in the description field at Wikidata. Neither can we wikilink or bold portions of text pulled from the Wikidata description. --EncycloPetey (talk) 04:27, 15 October 2018 (UTC)
I was not very clear on how I think the pages should be configured, it is very different to the individual creation and maintenance of the information by users here. A key point in my upcoming proposals is that labelled data is the solution to untidy workarounds and verbosity. That information is available in other statements at the the WD item with a reference, each site and reader would be able too choose a preference for what is displayed by default and an opportunity to gather or access further information. This allows any whim to be fulfilled by being able to create a query across wikimedia: are there incomplete books here or at commons, who are the coauthors, who are the notable collaborators, what was their birth name …? I cannot think of an example of an author page I created or modified here that required a unique and unreferenced description, only those that required me to manually copy paste data from other sites, — CYGNIS INSIGNIS 05:52, 15 October 2018 (UTC)
And you will still be able to do all of those things. Using Wikidata descriptions would be a mere fallback, for where none is provided locally. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 11:00, 15 October 2018 (UTC)
 Comment There will be some situations where we want more information than is in the description, and there are some authors for whom the description maintained on Wikidata does not meet our needs, but for the majority of situations, I don't see why we wouldn't want to do so. --EncycloPetey (talk) 15:35, 14 October 2018 (UTC)
Yes, I did specifically say "...if there is no locally-entered value". But where we currently have none, surely something from Wikidata is an improvement. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:54, 14 October 2018 (UTC)

Cool image

Kudos to our friends on the Polish Wikisource for creating this image! Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:37, 15 October 2018 (UTC)

22:40, 15 October 2018 (UTC)

Natural History of the Nightingale

Natural History of the Nightingale is ready for a second set of eyes, if anyone has time to kindly check it over. There is a gallery of images of the original publication on the talk page.

It's quite complex, being originally spread over two issues; and a lengthy footnote, that includes a subordinate footnote. Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 20:01, 10 October 2018 (UTC)

Is there a reason this wasn't worked as a transcription project using those images in an index? --Mukkakukaku (talk) 03:05, 15 October 2018 (UTC)
There is potential for improvement, I would be proof reading it now if the images were in index, but if a knowledgeable user with access to hathi trust were to bring it over … Is there a reason why that would not be the simple solution? — CYGNIS INSIGNIS 06:26, 15 October 2018 (UTC)
A perhaps not-so-elegant solution(?): Create a single PDF file using the available images, upload to IA, then Commons &c. Unless IA is *still* not generating DjVu files? Londonjackbooks (talk) 06:57, 15 October 2018 (UTC)
Another solution is to check if the file is already hosted IA, but I can't do that and type this message. I'm also limping along on an antique that is allergic to pdf, as am I, limping and allergic. It's a quandary … CYGNIS INSIGNIS 07:25, 15 October 2018 (UTC) P.S. A very enjoyable text, Andy, nicely sourced, transcribed and linked. Any note within a note will resolvable in the Page: namespace, but I'm wondering if another was missed; a dagger † often refers to the second footnote of a page, following the use of an asterisk * — CYGNIS INSIGNIS 07:44, 15 October 2018 (UTC)
Couldn't find it at IA searching text. It would be doubtful anyway (wouldn't it?) for both sections to be pieced together at IA unless someone had taken the pains to do so. I am limping, but my computer is not; and neither of us are (is?) allergic. I can't get to it this minute, but maybe later today, unless/until someone else gets to working out a solution. Londonjackbooks (talk) 07:54, 15 October 2018 (UTC)
Is the watermarking problematic? I can't do anything about that for most images. Londonjackbooks (talk) 08:35, 15 October 2018 (UTC)
Went ahead and uploaded a file to IA. Don't see a 'regular' djvu file derived (forgive my questionable terminology). I have failed in the past to to the pdf to djvu conversion via Commons (or wherever)... If we can get it to Commons, I can set it up here, but confess I have yet to learn how to do match and split (not without an offer to help). Soup sandwich I am :) Londonjackbooks (talk) 17:11, 15 October 2018 (UTC)
DjVu file is now at Commons. Will create an Index here. Londonjackbooks (talk) 05:50, 16 October 2018 (UTC)
What is the point of this, when the work is already proofread and published? I simply asked for someone to "check it over". Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:12, 16 October 2018 (UTC)
I don't regard it as obligatory, although I presonally have a strong preference for localised images. The few times I proofread something the old way did not encourage me to do it again (most users cut/paste gutenberberg texts), but I have checked thousands of pages since. This includes a couple of small improvements to this text, that I frankly would not have bothered to do if were not for the scan. Secondly, verifiability, crucial to the work we do here, confirming that the text matches makes it easier for London, for example, to find all the things I miss. — CYGNIS INSIGNIS 17:13, 16 October 2018 (UTC)
Some even {{ls}}eem to be po{{ls}}{{ls}}e{{ls}}{{ls}}ed of a different {{ls}}ong from the re{{ls}}t, and contend with each other with great ardor.
Proof reading is a bit of a challenge, is the preference that the long esses appear in main? — CYGNIS INSIGNIS 12:18, 16 October 2018 (UTC)
  • Generally preferred not to have long s in mainspace, but it's up to you and the other proofreaders of the work. You can post your discussion and decision here. —Beleg Tâl (talk) 12:33, 16 October 2018 (UTC)
    • Er, as I transcribed and proofread the entire work, isn't it up to me? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 16:12, 16 October 2018 (UTC)
      • That is the practice, but the guidelines say that deviation from whatever is deemed 'standard'—by consensus or otherwise—is liable to be challenged by others. I've only done one work with long esses, just one mind, the argument against their display became even more persuasive. — CYGNIS INSIGNIS 17:21, 16 October 2018 (UTC) P. S. the reason I ask is that I will need to replace the template with the character, which is trivial when compared with your investment in applying them. — CYGNIS INSIGNIS 17:26, 16 October 2018 (UTC)
        • "I will need to replace..." You will? Why? Andy Mabbett (Pigsonthewing); Talk to Andy; Andy's edits 19:22, 16 October 2018 (UTC)
          • The shorter version of the protracted chapter in WS history is that the template does not display in mainspace [!] unless you install a script to show them, although there was an idea to put another option in the sidebar (maybe this has been implemented). In my view the existence of Template:ls is unhelpful, a user is either using it or not; my recommendation is to always check any template documentation. — CYGNIS INSIGNIS 21:03, 16 October 2018 (UTC)

I changed the fake ToC on the main page to use {{AuxTOC}} and changed the links to point to arabic numeral numbering rather than roman, eg. "Chapter 4" over "Chapter IV". Only issue now is, the old pages are at "Chapter IV" but there are already existing, non-scan-backed pages named with the roman numerals so I am unable to move the page over it (due to lacking permissions). Can someone either move the pages or do a mass edit of the non-scan-backed pages (1-24) to basically contain the content from the recently made pages (I-XXIV) and redirect the roman numeral pages to the arabic numeral ones? -Einstein95 (talk) 03:05, 16 October 2018 (UTC)

Done.— Mpaa (talk) 20:20, 17 October 2018 (UTC)

Patrolling

Recently I've been noticing IP edits that have been "patrolled" by someone but which included changes to the text so that the Wikisource copy no longer matched the source text.

Wikisource copies must match the source text in matters of spelling. So if someone changes the spellings in a document, and the text no longer matches the source, it is not OK to let that edit stand and mark it as "patrolled". The change should be undone, and a courtesy notice given to the IP (or any editor) regarding Wikisource and fidelity of the text. --EncycloPetey (talk) 02:27, 19 October 2018 (UTC)

I will typically go down the list of unpatrolled recent changes, marking everything I've seen as patrolled and keeping problem pages open in tabs to be dealt with later as a batch, sometimes after a break. I believe I always get around to it if noone beats me to the punch, the case that prompted this was something I came back to look at just now, to find you addressed it, and then found this. I've tried other ways of doing patrolling, like not marking the edit as patrolled until after it's corrected, but I wasn't able to keep up with the volume of edits any other way. If this is more harm than help, I'll stop patrolling, although I would like to be assured that other people will take care of it. Prosody (talk) 05:12, 19 October 2018 (UTC)

Pagelist checking...

Category:Index - File to check

Any chance of this being emptied by the end of the year? I'd done a few more, but the remainder are ones I don't necessarily feel happy about doing the page-listing for various reasons. ShakespeareFan00 (talk) 09:19, 19 October 2018 (UTC)

Help to regain my sanity

I hereby offer anything within my means, to the successful elimination of the additional empty line created by each click of the [Show preview] button. You may not realize, but each click of this button adds an empty line to the end of the "textarea" text and this drives me "crazy!!!". Five clicks of the [Show preview] button adds five empty lines at the bottom of the text edit area. This problem exists in all Wikisources which use the Proofread module. and I tested this (as a renegade of sorts), at the French Wikisource. So, I know that this issue is not exclusive to English Wikisource, but originates in the Main Proofread module of Wikimedia. Volunteer programmers there overlook this problem as not being important, and simply refuse to correct it because they are programmers and not proofreaders.

So, Please support this request to help me proofread and avoid mental health issues. . . . Thanks for your support. — Ineuw talk 05:01, 21 October 2018 (UTC)

ticket 188844 (created by Ineuw) presumably refers? 114.73.248.245 05:12, 21 October 2018 (UTC)

23:11, 22 October 2018 (UTC)

Bilingual book

One of the books I am considering to upload and proofread is Modern Czech Poetry, ed. Paul Selver, 1920. The book contains a collection of poems of Czech writers in original Czech language on one page and the English translation of the opposite page, see https://archive.org/details/modernczechpoetr00selviala/page/12 . The editor probably wanted to present readers both versions and so it seems to me that we should also present here the English translation together with the original poem. What do you think, is it a good idea or should only the English translation be added to the English Wikisource, as original Czech version belongs to the Czech Wikisource?

If both versions could be added, what would be the best way of their transclusion to the main space, so that they stayed next to each other? --Jan Kameníček (talk) 08:07, 19 October 2018 (UTC)

--Jan Kameníček (talk) 08:07, 19 October 2018 (UTC)

@Jan.Kamenicek: Multilingual works belong at mul:. —Justin (koavf)TCM 08:16, 19 October 2018 (UTC)
Thanks for the answer. I have looked at it and it seems to be a strange site to me. The main page is just a disambiguation page referring to various language wikisources including en.ws and many others. After a long time I found a list of languages included at mul and it seems it is intended for some minor languages, but not for English or Czech. I am trying to browse the site but I am completely lost there, unable to find any local rules or whatever. If the work really belongs there, I am afraid it would be completely lost there with a minimal chance to be found by readers (as the main page sends readers somewhere else) and so it seems a loss of time adding it there. So if the Czech text should not be here at en.ws, I will add just the English versions of poems. --Jan Kameníček (talk) 08:40, 19 October 2018 (UTC)
@Jan.Kamenicek: OldWikisource (the multilingual Wikisource) serves several purposes: one is to be the landing page for Wikisource in general, just like how https://wikibooks.org/ introduces Wikibooks and https://wikivoyage.org/ does for Wikivoyage. Another is to hold material for languages with very small literature corpuses (e.g. some dead languages or Papamiento which is generally only spoken and not written). A third is to act like incubator: where a language subdomain can "graduate" to its own site. Finally, it hosts multilingual works, such as s:mul:Index:Festival Romanistica.pdf or s:mul:Hail Mary or s:mul:Index:Boletín RAE VI (1919).djvu or s:mul:Bukvar staroslovenskoga jezika glagolskimi pismeni za čitanje crkvenih knjig. —Justin (koavf)TCM 02:07, 20 October 2018 (UTC)
@Koavf: I see, thanks for the explanation very much, now I understand it better. The biggest problem I see with this site is that it was necessary to explain it at all, as the reader does not get this information on the main page and in fact I did not find it even after quite a long time of searching. I was not able to find any page explaining the system of the site, its rules, anything. As a reader I am directed to various single-language sites and do not get the information that multi-lingual works can be found there and how/where I can find them. So I got the impression that adding there some work is like throwing it into a black hole :-( --Jan Kameníček (talk) 07:20, 20 October 2018 (UTC)
Putting this work at the Multilingual Wikisource would mean that users looking for Czech poetry in English on the English Wikisource would not find it. We need to store works in English on this Wikisource so people looking for works in English can find them. We could have a link from the relevant author pages here to the Multilingual Wikisource, but I don't see what that gains us.--Prosfilaes (talk) 22:29, 23 October 2018 (UTC)
@Jan.Kamenicek: If the work is a collection of single-language works, rather than a single multi-language work, this is frequently handled by splitting the content between enWS and the relevant language WS. For examples, have a look at my laWS user page. — If the text is completely parallel, it may also be acceptable to only transcribe the English content and leave the rest to other editors; see Index:Aida Libretto English.djvu and Index:National anthem act Canada.pdf. — You may also find the templates documented at Template:Iwpage/doc useful. —Beleg Tâl (talk) 12:18, 19 October 2018 (UTC)
@Beleg Tâl: I see, thanks. I just thought that as the editor of the book wanted to present the English readers the English translation of the poems along with the original text, we could keep the original text too. But if you feel it is better to add only the English text as in your examples, it is fine to me as well. --Jan Kameníček (talk) 12:28, 19 October 2018 (UTC)
@Jan.Kamenicek: I think it is (always) better to have both the English text (on enWS) and the Czech text (on czWS). You will see that the examples I gave on my laWS user page (especially The seven great hymns of the mediaeval church where the original and translation are presented in parralel) are like this: the English and Latin texts are both present and accounted for. The last two examples were provided because you said "I will add just the English versions of poems". If you are not comfortable working in multiple wikisources, it is completely acceptable for you to only do the English parts and to leave the Czech part for someone else at a future time. For example, I did this with Aida, because I don't speak Italian and was not comfortable trying to work on itWS. —Beleg Tâl (talk) 12:54, 19 October 2018 (UTC)
@Jan.Kamenicek: You can have a look at this work. Hrishikes (talk) 12:58, 19 October 2018 (UTC)
@Hrishikes: I don't know if this is the best method for this particular collection. I thought we generally discouraged formatting these works in parallel like this, except where it is clearly inappropriate to separate the original from the translation. —Beleg Tâl (talk) 13:16, 19 October 2018 (UTC)
@Beleg Tâl: As for cs.wikisource: the poems are much older than this bilingual publication, so it would be much better to add them to cs.ws from the original Czech sources (in fact some of them are already there). Besides that, there is no clear consensus on cs.wiki as for using proofreading extension and the index and page namespaces. Inf fact conservative local admins oppose it very much and discourage contributors from using it, and I gave up fighting with them. Another problem is technical: cs wikisource uses a lot of templates which are not compatible with en.wikisource environment. Once I tried to transclude something from there to en.ws and failed for this reason.
So I personally do like the solution suggested by @Hrishikes:. --Jan Kameníček (talk) 13:32, 19 October 2018 (UTC)
  • I see no reason to exclude part of the work, and would think that can be duplicated at the Czech wiki source. The attempts to avoid this situation just begets problems, and putting it where it won't be seen is unhelpful. Notwithstanding some questionable interpretation of a rule inferred from the historical split of the wikisource library, the complete text can be welcomed at this site; here is a featured text Le Corbeau that displays both languages (having just restored that version from "experiments" and an undiscussed revert). It is preferable that users concern themselves with solutions to matters that have defined and undesirable consequences, Yet again I ask, what are those consequences in this situation? — CYGNIS INSIGNIS 08:12, 20 October 2018 (UTC)

Hi all, on other Wikisources (at least the French, the Latin, the Breton and the Multilingual - that I know of), we put each parts of a multilingual book in the corresponding Wikisource (with all the templates and tools to make it easy and smooth). Is it not the rules also on the English Wikisource? (if not, it will create problem for bilingual books containing English, like on Le Corbeau where I thought I was applying the usual formatting to only discover afterwards that I was not by Cygnis insignis). Cdlt, VIGNERON (talk) 12:43, 20 October 2018 (UTC)

Large-scale copying of ancient inscriptions into Wikipedia articles

There is a discussion going on regarding large-scale copying of the text of ancient inscriptions, and whether this kind of primary source copying is more appropriate for inclusion in Wikipedia or Wikisource. Your feedback would be welcome at w:WT:CP#Large-scale copying of ancient inscriptions into articles. Mathglot (talk) 05:46, 23 October 2018 (UTC)

A wandering student in the Far East vol.1

I have encountered an issue in the display of CHAPTER II. SHANGHAI TO ICHANG. [19]

On Page:A wandering student in the Far East vol.1 - Zetland.djvu/81 there is a Footnote containing a table which spans over onto the following page.

I am unable to make this display correctly. Would someone be able to take a look and correct my formatting error please.

Thanks.

Sp1nd01 (talk) 13:45, 24 October 2018 (UTC)

The <ref> tag was in the header and so wasn't transcluding, should be ok now —Beleg Tâl (talk) 20:15, 24 October 2018 (UTC)
Thanks for your help! it's displaying correctly now. Sp1nd01 (talk) 21:02, 24 October 2018 (UTC)

This work has already been fully transcribed years ago, but there are still many pages that need to be created for individual poems (see red links in the contents). Isn't there an automated way to do this? Thanks ~ DanielTom (talk) 20:16, 19 October 2018 (UTC)

The automated way to do it is transclude the entire work to that page. The ToC can then link to the page further down, which is easily done by wrapping the page numbers with the wikicode #170|. If someone then wants to make 170 redirects or versions, that full transclusion need not wait and will still return a search result. — CYGNIS INSIGNIS 08:27, 20 October 2018 (UTC)
I have done the single-page ones. I did a few a couple of the multi-pages ones but I stopped as I was unsure how to fix <poem> across pages (you can follow my traces in Page ns to see my fix). If someone more familiar with poems could take a look, I appreciate.— Mpaa (talk) 14:46, 20 October 2018 (UTC)
BTW, djvu page 218 is unreadable. If someone finds a replacement, happy to fix it.— Mpaa (talk) 21:12, 20 October 2018 (UTC)
@Mpaa: Sample here of how multi-page <poem> may be achieved. At a basic level the poem extension is almost incompatible with the ProofreadPage extension (the two fight over the implied </div> between the page body and page footer. Neither really wins! The compromise is to put up with the result of transclusion being effectively two <div>s nested inside a third being… sort of acceptable. 114.73.248.245 04:49, 21 October 2018 (UTC)
i have stopped using poem, and rather use block center, with lots of br line breaks. i find it more stable, if a little more work. Slowking4SvG's revenge 14:40, 25 October 2018 (UTC)
@Mpaa: I don't know if this is deliberate, but several pages have the character "⋅" in the translator and notes fields of the page header. For a list of affected pages, see Category:Works with non-existent author pages (which flags them as Author:⋅ is not an existing page). —Beleg Tâl (talk) 20:27, 22 October 2018 (UTC)
@Beleg Tâl:, thanks, should be fixed now.— Mpaa (talk) 20:34, 22 October 2018 (UTC)

Large-scale copying of ancient inscriptions into Wikipedia articles

There is a discussion going on regarding large-scale copying of the text of ancient inscriptions, and whether this kind of primary source copying is more appropriate for inclusion in Wikipedia or Wikisource. Your feedback would be welcome at w:WT:CP#Large-scale copying of ancient inscriptions into articles. Mathglot (talk) 05:46, 23 October 2018 (UTC)

Difficult to read Latin text

Can anybody read the text at the top of Page:Bohemia's claim for freedom.djvu/11, please? A better legible version is also at https://archive.org/details/bohemiasclaimfor00proc/page/6 . It seems to me that it starts with the name of Johan Hus, but I am not able to read much more. Or should I add it there as a picture? Thanks! --Jan Kameníček (talk) 12:15, 27 October 2018 (UTC)

I would put it as a picture, since it's only an image of a mauscript, as evidenced by the way it's cut off at the right margin. If it were transcribed it would need to preserve the abbreviations which can be difficult in Unicode. The full expanded text is "Johannes Hus Magister in Artibus Sacrae theologiae Baccalaureus [studii universitatis] Pragensis Rector et praedicator verbi Jesu Christi in capella S[anctorum]". —Beleg Tâl (talk) 13:08, 27 October 2018 (UTC)
OK, you are probably right with the picture. Thanks for both the transcription and the advice. --Jan Kameníček (talk) 13:41, 27 October 2018 (UTC)

Big untranscluded works

I am just looking at some of the untranscluded works that we have, and these are some of the biggies that need addressing

Top 5
Migration from text only to image-based

Probably a couple of thousand pages here. — billinghurst sDrewth 04:59, 11 October 2018 (UTC)

The US Statues are a nightmare because of the weird templates and they use in mainspace. Is there a list of untranscluded works somewhere? I've never found one. --Mukkakukaku (talk) 04:53, 12 October 2018 (UTC)
Also, I'll take a stab at A General History... so we don't all go stepping on each other. :) --Mukkakukaku (talk) 04:54, 12 October 2018 (UTC)
We have category:Transclusion check required which is proofread works with something needing to be done. There is also the active list generated at toollabs:phetools, though that a listing of untranscluded pages, irrespective of the status of the work. — billinghurst sDrewth 05:07, 12 October 2018 (UTC)
i see Confederate Military History is missing 2 volumes, and needs some index love for others. i could scan those volumes at LOC, if there is milhist interest. Slowking4SvG's revenge 03:53, 17 October 2018 (UTC)
the 2 missing volumes are at hathi https://babel.hathitrust.org/cgi/pt?id=hvd.hx2n9w;view=1up;seq=1 ; https://babel.hathitrust.org/cgi/pt?id=hvd.hx2n9z;view=1up;seq=5 if someone from a US institution could upload to internet archive. Slowking4SvG's revenge 15:33, 25 October 2018 (UTC)
I'll take care of the A Child's History as well since A General History is completed. --Mukkakukaku (talk) 03:12, 26 October 2018 (UTC)
That one's done too. I transcluded it separately from the existing works since the text was actually different, so they weren't from the same edition. Not only that, but the work was actually two volumes, only one of which was The Child's History of England. Either way, it's been transcluded and cleaned up some. --Mukkakukaku (talk) 06:30, 28 October 2018 (UTC)

Rotated table overlaps text

At the bottom of Page:United States Statutes at Large Volume 1.djvu/772, there appears a table whose contents have been rotated left 90 degrees relative to the text of the page. At least as it is displayed in my browser, however, the rotation has caused the table to cover up the last several lines of text in the paragraph immediately before the table. Assuming that this is not caused by some idiosyncratic behavior of the computer I am using, does anyone know how to prevent the rotated table from covering up the preceding text? Tarmstro99 20:11, 20 October 2018 (UTC)

Rotation doesn't resize the container, so you have to add a gap above (and below) the table for the long ends to rotate into. If the container is square everything will fit nicely. —Beleg Tâl (talk) 20:16, 20 October 2018 (UTC)
It’s odd (to me) that the same problem does not appear at Page:United States Statutes at Large Volume 1.djvu/778, which also contains a rotated table but doesn’t (in my browser, at least) appear to overlap the text above and below. Tarmstro99 15:40, 28 October 2018 (UTC)
That table is only different because 1) it is more square and 2) it has a gap before and after for the long sides to rotate into. If you make it less square (by removing some rows), or if you remove the paragraphs before and after that contain only a single non-breaking space, you will see that it rotates into the text above and below just like the other one. —Beleg Tâl (talk) 15:52, 28 October 2018 (UTC)
The observation about the second table being “more square” gave me the idea to add a width parameter to the table on Page:United States Statutes at Large Volume 1.djvu/772, which seems to have done the trick (for now, at least; we’ll have to see how it looks when transcluded. :-) Tarmstro99 15:56, 28 October 2018 (UTC)

Pseudonyms

I would like to ask the community whether it could be acceptable in some cases to name an author page with the pseudonym of the author instead of his civil name. I believe it could be possible if the author is known almost only by the pseudonym. I have recently founded Author:Petr Bezruč, whose civil name Vladimír Vašek is practically unknown and unused both by readers and literary critics. The same case is Author:Otakar Březina (who probably changed his name because his civil surname sounds obscene to contemporary Czech readers). A short time ago this practice was still accepted by Help:Author pages, but today it was removed by Billinghurst.

In my opinion this makes the author page less comprehensible to readers, because a note about pseudonym written in much smaller font somewhere in the description or note is not the same as the name of the page. I also believe we should follow the same practice as big the libraries do: most links in the authority control call him O. Březina too (including the Library of Congress or Czech National Library ), mentioning his civil name as an alternative. --Jan Kameníček (talk) 21:54, 21 October 2018 (UTC)

It is acceptable to use either the civil name or the pseudonym (or the religious name or whatever) at the discretion of the page creator. There should be a redirect in place from the other names the author goes by. The notes field of the author header should also specify names the author goes by. —Beleg Tâl (talk) 22:21, 21 October 2018 (UTC)
Agreed. For some authors, the pseudonym is the name under which the individual is indexed in VIAF, LoC, BNF, GND, etc. and for whom the civil name is almost wholly unknown. --EncycloPetey (talk) 22:30, 21 October 2018 (UTC)
There are plenty of examples here, where names other than civil names are used: Author:Sri Aurobindo (civil name: Aurobindo Ghosh), Author:Ramakrishna Paramahamsa (civil name: Gadadhar Chattopadhyay), Author:Swami Vivekananda (civil name: Narendranath Dutt), Author:Ishwar Chandra Vidyasagar (civil name: Ishwar Chandra Bandyopadhyay), Author:Paramahansa Yogananda (civil name: Mukundalal Ghosh). There are plenty others. This is a time-worn scheme, it is not prudent to discontinue this. We will always find cases where civil names of authors are virtually unknown, except to scholars in the field. Hrishikes (talk) 06:51, 28 October 2018 (UTC)

{{sp}} vs. {{gesperrt}}

Time for another template showdown, with two templates with only a difference of 0.05em of spacing. Special mention to {{letter-spacing}} which has no defaults but is marked as for merging with {{sp}}.

Results:

  1. {{sp}}: 8940 pages
  2. {{letter-spacing}}: 1325 pages"
  3. {{gesperrt}}: 1144 pages"

This will be a little more work than a simple redirect unless the 0.05em doesn't matter much. -Einstein95 (talk) 09:45, 28 October 2018 (UTC)

I think we should merge the three of them into one.— Mpaa (talk) 21:15, 29 October 2018 (UTC)

Removal of {{textinfo}}?

When one of the old non-scan-backed works is replaced with a transclusion, should we also be removing the old (no longer accurate) {{textinfo}} on the talk page?

For example, consider Call of the Wild (London). The talk page's text info claims that this is the text from Project Gutenberg, but it's now transcluded from a scan of the original work, so it's not relevant anymore. (And all contributing users can be found in the history anyway.) --Mukkakukaku (talk) 00:50, 29 October 2018 (UTC)

Agree. If the text is from an original scan, the textinfo cannot give the reader information that it was taken from somewhere else. --Jan Kameníček (talk) 06:50, 29 October 2018 (UTC)

20:09, 29 October 2018 (UTC)

The Community Wishlist Survey

11:05, 30 October 2018 (UTC)

November's PotM

Anyone interested, can you please take a look at some items parked at the PotM talk page for November 2018, as well as BWC's remarks in the Option C section for that same month? November is almost here, and it would be great to get some feedback from more people. Thanks, Londonjackbooks (talk) 21:08, 29 October 2018 (UTC)

Decision time: We need a follow-on work for PotM. There are several options on the table. Thanks, Londonjackbooks (talk) 16:05, 3 November 2018 (UTC)

Decided. Can someone please create a disambig page for "The War and the Future" before I forget? There are two instances here. I am signing off or I would. Thanks, Londonjackbooks (talk) 22:00, 4 November 2018 (UTC)

Support ends for the 2006 wikitext editor

This toolbar is being removed from MediaWiki.

The 2006 wikitext editor will be officially removed next week, on the normal deployment train (i.e., Wednesday, 24 October 2018 for the Wikisources). This has been discussed since at least 2011, was planned for three different dates in 2017, and is finally happening.

If you are using this toolbar (and most of you aren't), then you will be given no toolbar at all (the 2003 wikitext editor). This default was chosen so that your editing windows will open even faster, and to avoid cluttering the window with the larger toolbars (a particularly important consideration for Wikisource's PagePreviews). Of course, if you decide that you would prefer the 2010 or 2017 wikitext editors (or a gadget like WikEd), then you are free to change your preferences at any time.

Although it is not a very popular script overall, I know that some editors prefer this particular tool. If you are one of its fans, then you might want to know that some long-time editors are talking about re-implementing its best features as a volunteer-supported user script. I believe that any announcements about that project will be made at mw:Contributors/Projects/Removal of the 2006 wikitext editor. Whatamidoing (WMF) (talk) 17:48, 15 October 2018 (UTC)

Please see mw:Contributors/Projects/Removal of the 2006 wikitext editor#Alternatives (and perhaps w:fr:Discussion utilisateur:Arkanosis#Page, where he talks about changes that you might need to make to some older scripts). Whatamidoing (WMF) (talk) 17:20, 29 October 2018 (UTC)
see also m:Wikimedia_Forum#mw.toolbar_back_or_global_gadget_as_a_replacement apparently, this toolbar change is contentious. Slowking4SvG's revenge 05:04, 9 November 2018 (UTC)

e-book reader down loads

is there interest in providing a "e-book" download option for readers? i see there was previous discussion about selling those,[24] but could we make it easier for readers to download for off line reading? Slowking4SvG's revenge 13:19, 23 October 2018 (UTC)

We can barely come to a consensus for how to support mobile users; I don't think we have the raw manpower/wherewithall to support a variety of additional formats. :( --Mukkakukaku (talk) 02:25, 25 October 2018 (UTC)
if it is an external tool, do not need a consensus. can change interface via phrabricator. it is unclear how hard it is to build a tool. maybe a summer of code could do it. Slowking4SvG's revenge 14:54, 9 November 2018 (UTC)
Most of our templates are not compatible, and any changes to templates would require consensus. We had similar issue with that "book download" functionality. --Mukkakukaku (talk) 00:18, 10 November 2018 (UTC)

Bilingual book

One of the books I am considering to upload and proofread is Modern Czech Poetry, ed. Paul Selver, 1920. The book contains a collection of poems of Czech writers in original Czech language on one page and the English translation of the opposite page, see https://archive.org/details/modernczechpoetr00selviala/page/12 . The editor probably wanted to present readers both versions and so it seems to me that we should also present here the English translation together with the original poem. What do you think, is it a good idea or should only the English translation be added to the English Wikisource, as original Czech version belongs to the Czech Wikisource?

If both versions could be added, what would be the best way of their transclusion to the main space, so that they stayed next to each other? --Jan Kameníček (talk) 08:07, 19 October 2018 (UTC)

--Jan Kameníček (talk) 08:07, 19 October 2018 (UTC)

@Jan.Kamenicek: Multilingual works belong at mul:. —Justin (koavf)TCM 08:16, 19 October 2018 (UTC)
Thanks for the answer. I have looked at it and it seems to be a strange site to me. The main page is just a disambiguation page referring to various language wikisources including en.ws and many others. After a long time I found a list of languages included at mul and it seems it is intended for some minor languages, but not for English or Czech. I am trying to browse the site but I am completely lost there, unable to find any local rules or whatever. If the work really belongs there, I am afraid it would be completely lost there with a minimal chance to be found by readers (as the main page sends readers somewhere else) and so it seems a loss of time adding it there. So if the Czech text should not be here at en.ws, I will add just the English versions of poems. --Jan Kameníček (talk) 08:40, 19 October 2018 (UTC)
@Jan.Kamenicek: OldWikisource (the multilingual Wikisource) serves several purposes: one is to be the landing page for Wikisource in general, just like how https://wikibooks.org/ introduces Wikibooks and https://wikivoyage.org/ does for Wikivoyage. Another is to hold material for languages with very small literature corpuses (e.g. some dead languages or Papamiento which is generally only spoken and not written). A third is to act like incubator: where a language subdomain can "graduate" to its own site. Finally, it hosts multilingual works, such as s:mul:Index:Festival Romanistica.pdf or s:mul:Hail Mary or s:mul:Index:Boletín RAE VI (1919).djvu or s:mul:Bukvar staroslovenskoga jezika glagolskimi pismeni za čitanje crkvenih knjig. —Justin (koavf)TCM 02:07, 20 October 2018 (UTC)
@Koavf: I see, thanks for the explanation very much, now I understand it better. The biggest problem I see with this site is that it was necessary to explain it at all, as the reader does not get this information on the main page and in fact I did not find it even after quite a long time of searching. I was not able to find any page explaining the system of the site, its rules, anything. As a reader I am directed to various single-language sites and do not get the information that multi-lingual works can be found there and how/where I can find them. So I got the impression that adding there some work is like throwing it into a black hole :-( --Jan Kameníček (talk) 07:20, 20 October 2018 (UTC)
Putting this work at the Multilingual Wikisource would mean that users looking for Czech poetry in English on the English Wikisource would not find it. We need to store works in English on this Wikisource so people looking for works in English can find them. We could have a link from the relevant author pages here to the Multilingual Wikisource, but I don't see what that gains us.--Prosfilaes (talk) 22:29, 23 October 2018 (UTC)
@Jan.Kamenicek: If the work is a collection of single-language works, rather than a single multi-language work, this is frequently handled by splitting the content between enWS and the relevant language WS. For examples, have a look at my laWS user page. — If the text is completely parallel, it may also be acceptable to only transcribe the English content and leave the rest to other editors; see Index:Aida Libretto English.djvu and Index:National anthem act Canada.pdf. — You may also find the templates documented at Template:Iwpage/doc useful. —Beleg Tâl (talk) 12:18, 19 October 2018 (UTC)
@Beleg Tâl: I see, thanks. I just thought that as the editor of the book wanted to present the English readers the English translation of the poems along with the original text, we could keep the original text too. But if you feel it is better to add only the English text as in your examples, it is fine to me as well. --Jan Kameníček (talk) 12:28, 19 October 2018 (UTC)
@Jan.Kamenicek: I think it is (always) better to have both the English text (on enWS) and the Czech text (on czWS). You will see that the examples I gave on my laWS user page (especially The seven great hymns of the mediaeval church where the original and translation are presented in parralel) are like this: the English and Latin texts are both present and accounted for. The last two examples were provided because you said "I will add just the English versions of poems". If you are not comfortable working in multiple wikisources, it is completely acceptable for you to only do the English parts and to leave the Czech part for someone else at a future time. For example, I did this with Aida, because I don't speak Italian and was not comfortable trying to work on itWS. —Beleg Tâl (talk) 12:54, 19 October 2018 (UTC)
@Jan.Kamenicek: You can have a look at this work. Hrishikes (talk) 12:58, 19 October 2018 (UTC)
@Hrishikes: I don't know if this is the best method for this particular collection. I thought we generally discouraged formatting these works in parallel like this, except where it is clearly inappropriate to separate the original from the translation. —Beleg Tâl (talk) 13:16, 19 October 2018 (UTC)
@Beleg Tâl: As for cs.wikisource: the poems are much older than this bilingual publication, so it would be much better to add them to cs.ws from the original Czech sources (in fact some of them are already there). Besides that, there is no clear consensus on cs.wiki as for using proofreading extension and the index and page namespaces. Inf fact conservative local admins oppose it very much and discourage contributors from using it, and I gave up fighting with them. Another problem is technical: cs wikisource uses a lot of templates which are not compatible with en.wikisource environment. Once I tried to transclude something from there to en.ws and failed for this reason.
So I personally do like the solution suggested by @Hrishikes:. --Jan Kameníček (talk) 13:32, 19 October 2018 (UTC)
  • I see no reason to exclude part of the work, and would think that can be duplicated at the Czech wiki source. The attempts to avoid this situation just begets problems, and putting it where it won't be seen is unhelpful. Notwithstanding some questionable interpretation of a rule inferred from the historical split of the wikisource library, the complete text can be welcomed at this site; here is a featured text Le Corbeau that displays both languages (having just restored that version from "experiments" and an undiscussed revert). It is preferable that users concern themselves with solutions to matters that have defined and undesirable consequences, Yet again I ask, what are those consequences in this situation? — CYGNIS INSIGNIS 08:12, 20 October 2018 (UTC)

Hi all, on other Wikisources (at least the French, the Latin, the Breton and the Multilingual - that I know of), we put each parts of a multilingual book in the corresponding Wikisource (with all the templates and tools to make it easy and smooth). Is it not the rules also on the English Wikisource? (if not, it will create problem for bilingual books containing English, like on Le Corbeau where I thought I was applying the usual formatting to only discover afterwards that I was not by Cygnis insignis). Cdlt, VIGNERON (talk) 12:43, 20 October 2018 (UTC)

@VIGNERON: What tools and templates to make it easy? I'm not familiar with how other Wikisources do it, and I think at least part of any difference may be from ignorance and reinventing the wheel, instead of choosing to do different things.--Prosfilaes (talk) 04:56, 24 November 2018 (UTC)
@Prosfilaes: there is a lot of them but mainly the {{iwpage}} (and alike) to do interwiki transclusion and display a text from a Wikisource in an other Wikisource (no duplication, if a mistake is corrected on one side the correction appears in the other side) and the DoubleWiki extension to display side-by-side bilingual texts (it doesn't work well anymore for Le Corbeau (Mallarmé) since now we have french and english on the english side thanks to @Cygnis insignis:). On my side, I work on a lot of different Wikisource (and I'm sysop on 4 of them) so it is always a bit strange to interact with monowikisource people (who indeed reinvent the wheel too many often). Cheers, VIGNERON (talk) 09:16, 24 November 2018 (UTC)
I am likely to endorse any view by prosifilaes, on the basis their view is near identical to mine >95%, but in this case I have lost the thread that leads to a solution (a typo or such?). Is that solution transclusions across wikis? CYGNIS INSIGNIS 11:44, 24 November 2018 (UTC)
Transclusions across wikis is one solution (currently, mainly used for pages), but for texts a display across wiki is more often favoured, it's even easier than a transclusion (usually, not for case like Le Corbeau (Mallarmé) since it's not possible to align two languages with one…). Cheers, VIGNERON (talk) 16:50, 24 November 2018 (UTC)