Wikisource:Scriptorium/Help

From Wikisource
Jump to navigation Jump to search
Scriptorium Scriptorium (Help) Archives, Last archive
The Scriptorium is Wikisource's community discussion page. This subpage is especially designated for requests for help from more experienced Wikisourcers. Feel free to ask questions or leave comments. You may join any current discussion or a new one. Project members can often be found in the #wikisource IRC channel (a web client is available).

This page is automatically archived by Wikisource-bot

Have you seen our help pages and FAQs?


My customized editing toolbar is not appearing[edit]

Is this a known issue? Thanks, Londonjackbooks (talk) 19:34, 24 August 2018 (UTC)

Is anyone else having this issue? @Ineuw, @Beeswaxcandle:? Londonjackbooks (talk) 19:05, 26 August 2018 (UTC)

Old toolbar is showing customised buttons for me in monobook skin, with Firefox browser — billinghurst sDrewth 21:56, 26 August 2018 (UTC)
@Londonjackbooks:I no longer remember your setup. So I uploaded two images of my Preferences\Edit File:Ineuw preferences toolbar settings 1.jpg and its result when I open a page for editing File:Matching toolbar 2.jpg. This would be the advanced toolbar! — Ineuw talk 01:16, 27 August 2018 (UTC)

Thanks all. Brief wikibreak for a few, then I will explore further. Londonjackbooks (talk) 17:09, 27 August 2018 (UTC)

Basically, this is the same issue as I had a couple years ago &c. (see this conversation). I have updated the old photos with current depictions of what I should/should not be seeing. Utilizing the customized <br /> button is the deal breaker for me for proofreading poetry. Thanks anyone, Londonjackbooks (talk) 20:06, 15 September 2018 (UTC)
@Londonjackbooks:I remember the issue and the solution. Your special keys are defined and exist in the User:Londonjackbooks/common.js. - In when looking at a page edit mode there is the Charinsert bar where your special editing keys are under "User:" at the bottom of the dropdown list. In the same common.js module, it is set to be shown above your editing window and the Wikisource editing bar. — Ineuw talk 20:21, 15 September 2018 (UTC)
@Ineuw: Okay, good deal. I found the break option under "User:", thanks. That should be sufficient :) Londonjackbooks (talk) 20:28, 15 September 2018 (UTC)
@Londonjackbooks: Went through again your common.js and it was also defined and inserted in the advanced toolbar as you wanted. When was the last time that the toolbar was OK? Please check your "Preferences/Editing" page if the "Enable enhanced editing toolbar" is the only toolbar checked. — Ineuw talk 20:27, 15 September 2018 (UTC)
@Ineuw: Last time the toolbar was present ("OK") was maybe a few days prior to 24 August. Londonjackbooks (talk) 20:30, 15 September 2018 (UTC)
@Londonjackbooks: I am not proficient to check if there was a change in the Advanced toolbar's implementation but your old code is intact. Our Mediawiki software was updated on 2018-09-11. Perhaps billinghurst would know more. — Ineuw talk 20:46, 15 September 2018 (UTC)
Calling it a night here. The User option should be sufficient. Don't want to make things too complicated :) Have a good one! Londonjackbooks (talk) 20:50, 15 September 2018 (UTC)

Ampersand in italics display error[edit]

There is an ampersand here in italics, which does not display properly. I am unable to get this to display and can't find a solution. Celuici (talk) 09:00, 15 September 2018 (UTC)

@Celuici: If you mean the ampersand in "See Various Prospects, &c, p. 113-123", so I can see it well in both Firefox and Chrome. What exactly does it display to you? --Jan Kameníček (talk) 11:01, 15 September 2018 (UTC)
It displays as follows: (See Various Prospects, &amp;c, p. 113-123.). I've tested this in both Firefox and Chrome. Celuici (talk) 11:07, 15 September 2018 (UTC)
@Celuici: Hm, that is weird. May be you can try to type "&amp;" or "&#38;" instead, which should both render "&". --Jan Kameníček (talk) 11:25, 15 September 2018 (UTC)
BTW: Do you see the ampersand well at the Page:Philosophical Transactions - Volume 054.pdf/105, or do you experience there the same problem? --Jan Kameníček (talk) 11:29, 15 September 2018 (UTC)
Sorry, only now I noticed that you actually used the code in your previous contribution so I suppose you had also tried it there and it did not work. Hopefully somebody else will help, but I am still curious about the page I asked above, because I also used it several times thinking everybody can see it properly. --Jan Kameníček (talk) 11:56, 15 September 2018 (UTC)
@Celuici: Sounds like a browser or a cache issue, it displays fine for me. Noting that where we transclude a whole page that we wouldn't add sections. Section transclusions should only be used for partial pages. — billinghurst sDrewth 14:09, 15 September 2018 (UTC)
I've finally discovered that this problem stems from one of the tools I've installed. Blanking my common.js page has fixed the problem. Celuici (talk) 22:38, 15 September 2018 (UTC)
@Celuici: Made a minor adjustment. Can you please check if it's still OK? — Ineuw talk 06:27, 18 September 2018 (UTC)
@Ineuw: It's now displaying fine for me (having reverted by commons.js back to how it was). Celuici (talk) 06:50, 18 September 2018 (UTC)
@Celuici: Thanks for letting me know. I am using the "typoscan.js" which picks up on a variety of typo errors and remembered that the &c. must not have spaces between the three characters. That is what triggered the error. The reason I asked you to check is because the typoscan script is constantly active in my environment and don't see if it looks different for others. — Ineuw talk 06:58, 18 September 2018 (UTC)

Sidelinks to pages[edit]

The sidelinks to individual pages at Philosophical Transactions/Volume 54/An Account of the Effects of Lightening at South Weald, in Essex are not displayed properly for some reason. By various experiments I found out that only two things make them to display well:

a) removing the first paragraph (title) from the page Page:Philosophical Transactions - Volume 054.pdf/256 (as I tried here). The whole title has to be removed with all the text, removing just the templates like {{hi}} does not help.

b) dividing the text in the following page Page:Philosophical Transactions - Volume 054.pdf/257 into at least two paragraphs (as I tried here).

I have no clue what is really happening, why the title works only if the following page has more paragraphs and what is the connection between these unrelated things. Other similarly designed pages like Philosophical Transactions/Volume 54/Observations... work well.

What could be the cause of this weird behaviour? --Jan Kameníček (talk) 22:45, 15 September 2018 (UTC)

Without being able to see the problem, I can't offer any suggestions. It looks fine for me right now. --EncycloPetey (talk) 00:43, 18 September 2018 (UTC)
Strange, now it looks OK to me as well, although a couple of days ago I saw it broken in both Firefox and Chrome, trying it on two different and completely independent computers. Nevertheless, I thank you for looking at it. --Jan Kameníček (talk) 11:51, 19 September 2018 (UTC)

Large print—what have I done?[edit]

Yesterday, I did something? and now all text on WS pages is large and things are jumping around - sorry, best description I have. I have rebooted the computer, shut all WS tabs, other pages aren’t affected. Not sure how to attach a screenshot, hoping it’s something basic. Please tell me how to get it back to normal, Cheers Zoeannl (talk) 00:05, 17 September 2018 (UTC)

Use Ctrl+0 to zoom out? —Beleg Tâl (talk) 00:28, 17 September 2018 (UTC)
Alternatively ctrl-key and your mouse scroll wheel. — billinghurst sDrewth 04:29, 17 September 2018 (UTC)

Creating pages of a book[edit]

I am asking to have the bot create the pages for this book Doing this manually is insane. — Ineuw talk 19:11, 17 September 2018 (UTC)

Text layer is there, what is the problem? What is different from normal? — billinghurst sDrewth 21:24, 17 September 2018 (UTC)
Creating the pages one by one is incredibly time wasting. I know that there is a bot that can create the text layer pages. I don't see why I cannot take advantage of an existing tool.— Ineuw talk 22:23, 17 September 2018 (UTC)
I don't understand. You'll still have to proofread/validate them one by one. The only thing a bot would be able to do is save the OCR layer which seems like a waste of time. --Mukkakukaku (talk) 23:27, 17 September 2018 (UTC)
Ditto. The book is not so big, around 300 pages only. The ocr layer is good enough. The ocr is not going to disappear suddenly. I am not able to see the problem at all. We can talk about solutions if you can be more explicit about what the problem actually is. Hrishikes (talk) 01:50, 18 September 2018 (UTC)
The application of text layers means someone getting onto their bot account, downloading the image file, setting up scripts, then when it is all finished, a clean-up, so is not an effortless task. I will do it where there is the benefit of having text layers for searching/looking for text, as has been done for biographical works, eg. Index:Catholic Encyclopedia, volume 1.djvu. I don't do it for any of the general works that I otherwise transcribe as I don't find that it saves time over simply creating the page and proofreading it, especially when applying headers. — billinghurst sDrewth 03:48, 18 September 2018 (UTC)
It is not my intention to have someone assemble a scripts and monitor the running of a bot. I assumed that such a bot exists because all the pages of the Popular Science Monthly were created by the User:Mjbot before I began to proofread them.
I now realize that our editing/proofreading approach is different. I imagine that most editors work on a book by creating the page, and completing the proofreading process in one go. I tried this as well, but found it difficult to work like that. Will explore another method of page creation using AutoHotkey. So please forget this request. — Ineuw talk 04:20, 18 September 2018 (UTC)
Mjbot's operations were of a different time and place in the development of Proofread Page, and the ready ability to grab a text layer were more fraught. Back then it definitely was beneficial, these days not so. Plus Matt loved botting and disliked proofreading. — billinghurst sDrewth 05:40, 18 September 2018 (UTC)
I figured that it was Matt who did it. I just thought that it was a regular practice for all works. - These Mexico related books I am preparing is mostly for User:Gumr51 who loves proofreading, subjects about Mexico, but not the formatting and standardizing, and this works for me. His way of proofreading is altogether different. He downloads the text layer and using two monitors, he edits in MS Word, converts the result to ANSI text and then pastes it.— Ineuw talk 06:50, 18 September 2018 (UTC)
@Ineuw: That sounds like a reasonable reason to run the bot through. I will try to get to it, though have limited time with access to my PC from where I get access, so I cannot guarantee anything, or anything, anytime soon. Suggest stick something at WS:BR giving the Index: link, the File: link, and ensure that the Index: has the required header and footer detail prior, and you are asking from someone to run djvutext.py through the work. — billinghurst sDrewth 07:26, 18 September 2018 (UTC)
wikisourcetext.py does not require to download the file. It fetches text from the page itself.— Mpaa (talk) 19:30, 18 September 2018 (UTC)
@Billinghurst: Please don't waste your time. I have a good chance to implement an automated method with AutoHotkey, with help from their forum members. AutoHotkey understands precise display screen coordinates and emulates mouse clicks well. I plan to give an explanation and a demo.— Ineuw talk 07:32, 18 September 2018 (UTC)

┌─────────────────────────────────┘
In some books, eg. in Index:Ossendowski - From President to Prison.djvu (pages 314-360) the OCR layer displayed by MediaWiki is shifted (while internally in DjVu it is aligned correctly). It would be nice to have a bot that can extract the text layer directly from DjVu and upload it in such cases. Ankry (talk) 05:13, 18 September 2018 (UTC)

For mass creation of ocred pages, one can use the OCR4wikisource script. The software will: 1. Download the book from Wikimedia Commons, 2. Split the file into individual pdf/djvu pages, 3. Convert the individual pdf/djvu pages into jpg files, 4. Upload the jpg files to Google Drive one by one for doing OCR, 5. Download the OCRed text and 6. Upload them to respective Wikisource pages. But still I feel that such an approach is not required in English Wikisource, except when, as Billinghurst said, searchable text, even in not-proofread pages, is desirable. Hrishikes (talk) 12:04, 18 September 2018 (UTC)
@Hrishikes: I think, the procedure metioned by you will be too heavy and unnecessary overload when there is a good OCR in djvu, but it is incorrectly aligned to pages by Mediawiki/Proofreadpage software. Using djvutext.py seems to be the best solution in such cases. However, as there was only 45 pages affected in the above example, I have already aligned the text manually. Ankry (talk) 14:10, 19 September 2018 (UTC)

To SIC or not to SIC[edit]

So I see a word misspelled in the original. Only, it's *really* obvious that 'spiine' is 'spine'. Would you {{SIC}} the dogs tags on it? Shenme (talk) 04:15, 28 September 2018 (UTC)

Er, you can use the template, that's why it exists. In this case, if I saw "spiine" in the mains namespace, I would assume it was the proofreader's typo if it wasn't annotated and go to the page namespace to fix it. I personally would use the annotation in that scenario. --Mukkakukaku (talk) 20:33, 28 September 2018 (UTC)
Okay, thanks. But I've got another obvious misspelling, only it isn’t, only it will be seen as a misspelling by everyone born since 1900. This page hurts my sense of right/wrong, yet when I look around I find a short mention in Wikipedia:
Before about 1850, the spelling "Frith" was more common.
The book being edited was published 1850.
Is there a method for adding an explanatory note to a word/phrase, whereby the modern editor can explain to the modern reader what the heck is going on? Is there an {{AntiSIC}}? Shenme (talk) 22:36, 28 September 2018 (UTC)
There are too many archaic spellings (such as compleat x complete, shew x show...) and some texts could be flooded with SIC templates. Besides that, somebody's sense of right and wrong can be hurt also by archaic grammar (e. g. "broke" being past participle of "break"). I think that readers of old texts should be prepared to meeting obsolete spellings and grammar. If we just want to avoid other contributors correcting such seeming mistakes, we can notify them using the lowercase template {{sic}}. --Jan Kameníček (talk) 22:51, 28 September 2018 (UTC)
Agreed. We aren't here to expand people's vocabularies, only to preserve writings in their original form. Explanatory notes are already discouraged by the community as noted at WS:ANN. I would use {{SIC}} for printer errors but use {{sic}} otherwise, if at all. —Beleg Tâl (talk) 23:05, 28 September 2018 (UTC)
Ugh I didn't even realize {{sic}} was a separate template. Which ... doesn't even do anything? :(
But yes, {{SIC}} for printer's errors/typos. I also use it for egregious mistakes on behalf of the auth, such as when they're talking about some groups ABC and DEF and indicate one where they obviously meant the other. --Mukkakukaku (talk) 16:07, 29 September 2018 (UTC)
In fact {{sic}} works in a similar way as <!-- -->. --Jan Kameníček (talk) 16:42, 29 September 2018 (UTC)

Partial works[edit]

Hi. Over the last few years I've digitised a whole lot of books with Supreme Court decisions and I'm interested in uploading the scans to Wikisource. In my communications with the Supreme Court they are not ready to state that anything outside the actual decisions is also out of copyright. However, I see nothing on the help pages if it's allowed or not to upload only certain parts of works or not due to copyright reasons. If it's allowed, what's the best method to indicate a page has been retracted for that reason so it can be uploaded later when it's indeed in the public domain? Thanks in advance for your responses. -Svavar Kjarrval (talk) 19:05, 29 September 2018 (UTC)

We generally do not host the scans of works here at Wikisource. Scans of works are housed at Commons. So it would be better to broach this issue in discussion at Commons. --EncycloPetey (talk) 19:16, 29 September 2018 (UTC)
@EncycloPetey, You are technically correct, what I meant to say was: to include it here after uploading it to Commons. The reason why it's better to discuss it here (first) is that the work would eventually (if allowed) be referenced here, as there wouldn't be much point in uploading it to Commons otherwise, and the community here is more likely to have a more detailed experience in uploading works for that purpose. -Svavar Kjarrval (talk) 10:30, 30 September 2018 (UTC)
Two points:
  1. Per our inclusion policy, "Random or selected sections of a larger work, are generally not acceptable." However if the portion can be considered a work in its own right (either as one work out of a collection, or because that passage has been published separately as a work in its own right), then there should be no problem hosting it. I assume that a Supreme Court decision can be considered a work in its own right so there should be no problem there.
  2. I would like to provide you an example of a work I added in which significant portions of the publication were redacted for copyright reasons: The Tale of Kitty-in-Boots, with associated scan File:Kitty-in-boots.pdf. This may be a useful example in setting up your own project.
I hope this is helpful. —Beleg Tâl (talk) 02:34, 30 September 2018 (UTC)
@Beleg Tâl, Thanks for your example. I'll look into it and see [if/how well] it can be applied in my case. -Svavar Kjarrval (talk) 10:30, 30 September 2018 (UTC)

Heinrich Oskar Sommer (1861 - when?)[edit]

I'm trying to upload The recuyell of the historyes of Troye, the first book published in English. The problem is, the version commonly available is edited by Heinrich Oskar Sommer (Wikidata) in England, and while I could upload it to Wikisource, I was hoping that someone could establish a death date before 1948 so I could upload it to Commons.--Prosfilaes (talk) 04:43, 8 October 2018 (UTC)

Died in 1929: http://www.mcbax.de/info/s/heinrich_sommer.html -- Hrishikes (talk) 05:03, 8 October 2018 (UTC)
Thanks.--Prosfilaes (talk) 06:31, 8 October 2018 (UTC)
Did a quick search, and added some research to author talk page, and migrated the data to WD. — billinghurst sDrewth 14:46, 8 October 2018 (UTC)

Catholic Encyclopedia, volume 17[edit]

I would like to proofread a couple of pages of Index:Catholic Encyclopedia, volume 17.djvu, but I have noticed there the warning that the "source file must be fixed before proofreading". The problem probably is that subpage 15 and subpage 17 are identical, while subpage 16 is missing. May I ask for help with fixing it, please? --Jan Kameníček (talk) 14:45, 9 October 2018 (UTC)

@Jan.Kamenicek: -- Pages 16 and 17 are extra. They can be tagged as "without text". Hrishikes (talk) 15:35, 9 October 2018 (UTC)
@Hrishikes: Page 16 is completely missing so there is nothing to tag, it will stay as a red link forever. The only reason why I am writing here is that it is explicitely written at the index page that editors should not start proofreading before the issue is fixed. Or can the warning be ignored/removed? --Jan Kameníček (talk) 17:38, 9 October 2018 (UTC)
@Jan.Kamenicek: I think you can go ahead. All pages have been created, so if fixing the file will require a page move, the Proofread status will not make any difference at that point.— Mpaa (talk) 20:15, 9 October 2018 (UTC)
@Jan.Kamenicek: You can mark them as "without text" and give explanation in the header. For page 16: Duplicate of page 14 (yes, it is not missing, it is a blank). For page 17: Duplicate of page 15. In the pagelist, the query marks can be replaced with Dup. If there is no other problem, the work status can be changed to to be proofread after taking care of those two pages. Hrishikes (talk) 01:16, 10 October 2018 (UTC)
Thanks for the advice. --Jan Kameníček (talk) 11:39, 10 October 2018 (UTC)

Endnotes[edit]

Esme Shepherd (talk) 19:40, 10 October 2018 (UTC) I have been looking at the instructions for endnotes. I will be posting a document soon that has 44 endnotes spread over 13 pages in the original. If I include them as footnotes, they will all appear in order when transcluded, so that seems to be the best way forward. What I don't understand is how they are to be proofread. For one thing, as far as I can see those 13 pages will be blank, as the notes will all have been put into < ref >< /ref >s in the 51 main text pages, but the text will not therefore appear side by side and some of them are quite complex, being in foreign languages.

@Esme Shepherd: Are you saying the work itself has endnotes, and has references interspersed pointing to these? Endnotes as endnotes with pointers are a bit trickier to represent as neatly in both Page: and main namespaces, though we have some means to do that. I would want to see the specific work to comment more thoroughly.

I have created a template {{authority reference}} which can be use directly, or if there is some complexity, it can be adapted for use like I did with A Compendium of Irish Biography and {{IrishBio ref}}. This is not basic stuff to create, and happy to assist. Don't let it hold up the proofreading, as we can retrofit templates, so if you put in normal references markings like <ref name="ref1" />, <ref name="ref2" />, etc. then we can a bot through to resolve. — billinghurst sDrewth 21:35, 10 October 2018 (UTC)

In the Mediæval Hymns example, which puts the endnotes at the end as it were and therefore proof-readable, I can see the use of < ref name="xxx" > and < ref follow="xxx" > but do not understand the meaning of ​ Is this a code for a blank?

#8203 is unicode for a zero width space, see http://www.codetable.net/decimal/8203. Hard to comment further without exactly seeing what you are seeing. I would normally remove these in the works I am transcribing as they are often OCR artefacts. — billinghurst sDrewth 21:17, 10 October 2018 (UTC)
I added that instruction. If you put nothing inside <ref name="xxx">, then it doesn't work. If you put something inside <ref name="xxx">, then it does work, but the thing you put inside it is now in the footnote also. That is why I put &#8203; — it allows the footnote to work but also does not show anything in the footnote. (As an aside, I have recently been using {{em|0}} instead of &#8203; for this purpose, and the effect is the same.) —Beleg Tâl (talk) 01:51, 11 October 2018 (UTC)

Esme Shepherd (talk) 11:28, 11 October 2018 (UTC) Thank you for looking at this. I wasn't going to post this just yet but, as there are other similar cases, I have now done so. It is in Felicia Hemans - Subcategory:Hemans Published Volumes - Modern Greece. The first note reference is on page 3 of the poem and the note appears on page 55. Note also that note 3 overlaps from page 55 onto page 56, something that occurs a number of times. I have all the text ready to add, so I will start doing so soon. Hope you can find a good solution.

Esme Shepherd (talk) 15:30, 11 October 2018 (UTC) I see Beleg Tâl's method now. If the note number in the text is indicated by <ref name="p3.14">blank </ref> and my note on page 55 is enclosed by <ref follow="p3.14">....... </ref>, the two will be linked and will appear as note 1. As to the blank, I have myself used gap|0em in the past. Does this work in practice and is it an acceptable solution? Note 3 would then be name="p6.12" and have ref follow="p6.12" on both pages 55 and 56.

To that I can only say: try it, and if it doesn't work you can use one of the alternate methods we have discussed. —Beleg Tâl (talk) 23:42, 11 October 2018 (UTC)
If we are talking continuing footnotes and a nomenclature, I usually do the page number of the first page of the note, e.g. p60, then as many follow=p60 on all subsequent instances of the continuing footnotes. I don't go poking any other code, that is just making things hard for oneself, best to keep it simple so others can follow. — billinghurst sDrewth 04:44, 12 October 2018 (UTC)
Esme Shepherd (talk) 08:10, 12 October 2018 (UTC)Agreed about simplicity: the reason for the p3.14 is that 14 is the line number and there will be both a p12.16 and a p12.17. However, maybe when there is only one note to a page I should stick to p3, as you suggest. Page 21 has three notes to it.

Confused frog inquiring about Lilypond[edit]

I thought that at one point I had found the version of Lilypond in use here at Wikisource, but now I can’t confirm that. I thought I’d seen 2.12.xx. How can one find the current version numbers for tools used here?

I note that I found this note for the Score extension 0.3, that allows use of Lilypond to create music scores:

This extension was tested with LilyPond 2.12.3 through 2.18.2.

The concern is that, if indeed we have only 2.12.xx here, then much is documented for the current version 2.18.2 (2014) that is in fact unavailable to us. Lots.

There have been considerable changes since 2.12.xx (2.12.3 December 2009):

New features in 2.14 since 2.12 (2.14.2 July 2011)
New features in 2.16 since 2.14 (2.16.0 August 2012)
New features in 2.18 since 2.16 (2.18.2 March 2014)

Obviously knowing the version installed here will help people select the correct documentation. As it is, all our links here to lilypond.org get the most recent stable version docs. Shenme (talk) 17:41, 12 October 2018 (UTC)

@Shenme: Special:Version: LilyPond 2.18.2. —Justin (koavf)TCM 17:48, 12 October 2018 (UTC)
Music to my ears! Thank you. Shenme (talk) 18:35, 12 October 2018 (UTC)

Need some help[edit]

Hi all, I am currently working on Indian parliamentary committees on the english wikipedia - Politicoindian.

I saw that the Constitution of India has been uploaded here and was wondering if I could upload other Indian parliament relevant documents here as well.

I have PDFs of these documents. I am trying to understand how I can upload the following documents to wikisource and what would be the process for it.

Cheers - --Politicoindian (talk) 19:29, 16 October 2018 (UTC)

Edit - All these documents are in the public domain in India and covered under the GODL-India licence. --Politicoindian (talk) 19:33, 16 October 2018 (UTC)