From Wikisource
Jump to: navigation, 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 webclient.
Have you seen our help pages and FAQs?

Categories gone missing?[edit]

I'm not quite sure if I'm not just missing something obvious here, but the categories on The Wounded Cricketer don't seem to be showing — with the exception of the date category that's added by the header. Could someone point out my error? Thanks! — Sam Wilson ( TalkContribs ) … 23:41, 31 May 2015 (UTC)

I've put the categories into the header as well, and it seems to work now. Beeswaxcandle (talk) 00:04, 1 June 2015 (UTC)
Thanks! I didn't know about that header parameter. Can it be added to the auto-filled {{header}} text? Although, it does seem a bit odd that it'd preclude the use of manual categories. — Sam Wilson ( TalkContribs ) … 00:11, 1 June 2015 (UTC)
It shouldn't preclude the manual addition — billinghurst sDrewth 13:24, 2 June 2015 (UTC)
The pages-tag wasn't closed. That always messes things up.--Thurs (talk) 09:47, 13 June 2015 (UTC)
Aha! Thank you very much :) I was mightily confused (but had given up, in favour of other things hehe). — Sam Wilson ( TalkContribs ) … 11:02, 13 June 2015 (UTC)

Help me - Where is my work?[edit]

My page has been redirected by wiki, then by wikisource and now by wikibook; someone called billinghurst has moved it again and frankly, I don't have a clue where its at now.

{{helpme}} Its not me that creating all the links - I was quite happy for my page to remain where it was at. Can you tell me where the link to my work is now at? Dougie3211 (talk) 14:30, 8 June 2015 (UTC)

Your page is at User:William McDougall/A social and economic history of Balnamore, 1637 to 1886. Hesperian 15:14, 8 June 2015 (UTC)

How to create png file from djvu page ?[edit]

I'm proofreading the book - Index:Immigration and the Commissioners of Emigration of the state of New York.djvu When the table of contents was set several pages were set as missing image - high-res raw scan is available ... I can go through the download, crop, etc to complete the proof of these image pages. What I would like to get is the high-res raw scan of six more pages, specifically Page:Immigration and the Commissioners of Emigration of the state of New York.djvu/259 through 264 and 269 because these are landscape pages with tables of numbers. With high-res png files that I can download I will be able to get usable ocr results. Thanks Robin2014 (talk) 17:15, 8 June 2015 (UTC)

If you edit the page file, there should be a tab at the top called "Image", which if you click will give you the high quality scan. —Beleg Tâl (talk) 18:12, 8 June 2015 (UTC)
The high-res page scans are uploaded by HesperianBot i.e. me, and they are much higher resolution than the ones provided by the "Image" tab. Uploading of right-res page scans happens semi-automatically for pages tagged with {{missing image}} or {{raw image}}. I discourage tagging tables that way, but we can make an exception here if Beleg Tâl's proposed solution doesn't work for you: tag the pages with {{missing image}} and wait for HesperianBot to upload high-res page scans. It might take several days. Hesperian 01:42, 9 June 2015 (UTC)
I followed Beleg Tâl's advice for 1 page - rotate page, view page as png, save png to my pc - then used ABBYY FineReader 9.0 to get ocr output which I put at The results are an improvement over the initial ocr BUT there are 31 rows by 15 columns = 465 cells to proofread. I guess marking the pages as 'raw image' for the bot to process will probably not provide OCR. I set 5 pages to raw image and will wait for the bot to take the next step. I guess the bot will give a larger high-res than the manual steps - the image pages have resolution as high as 3,944 x 2,488. Using firefox as the browser, I did the manual steps for the 5 pages and saved png on my computer just to compare with what the bot does. When I saved to my pc the highest res was 1024 x 645 which is the laptop screen res. When I would repeat the manual steps for a page the png resolution would drop to 921 x 581. Thanks Robin2014 (talk) 15:42, 9 June 2015 (UTC)

Annotation help[edit]

I want to link the term LAW OF LOMBARDY on Page:Confessions of a Thug.djvu/7 with Can anyone please help? Hrishikes (talk) 14:44, 10 June 2015 (UTC)

Would it not be better to link it to The Law of Lombardy, and leave it at that until the work linked is added to Wikisource? —Beleg Tâl (talk) 16:04, 10 June 2015 (UTC)
And in the "Wikilinks as annotations" section on the Wikisource:Wikilinks page, it states that "Links to non-Wikimedia pages are not acceptable"—unless my interpretation is incorrect. Londonjackbooks (talk) 16:06, 10 June 2015 (UTC)
There are templates here like {{ext scan link}}, {{IA}}, {{IA small link}}, even a category called Category:External link templates. The link I want is in line with these items. Hrishikes (talk) 16:32, 10 June 2015 (UTC)
It's added to Wikisource now. —Beleg Tâl (talk) 16:35, 10 June 2015 (UTC)
Thanks. Hrishikes (talk) 16:41, 10 June 2015 (UTC)
@Hrishikes: The templates you cited above are usually used for Author page purposes, etc., and not for annotations within texts themselves. Londonjackbooks (talk) 17:23, 10 June 2015 (UTC)

Should the word Main: be removed from the title of this book ?[edit]

Should the word Main be removed from ? I guess that would be renaming ?

As I am working to put the chapters into subpages I noticed that my first transclusion inadvertently set title = Main:Irish Emigration and The Tenure of Land in Ireland, and this has carried forward to chapter 1 and chapter 2. Do these titles need to be renamed to remove the word Main: ? Robin2014 (talk) 14:55, 13 June 2015 (UTC)

Done. You could Move them as well, In More tab, upper right of page.— Mpaa (talk) 17:53, 13 June 2015 (UTC)
Thanks ! Now I'll learn about the more tab and move. Robin2014 (talk) 15:27, 14 June 2015 (UTC)

Transcluding across pages using hashes[edit]

Using hashes to distinguish sections, is it possible to transclude across pages correctly, à la the <section> tags? The hashes are a much nicer way of doing it, but while they work over a single page (like here), they seem to encroach onto the next section if the transclusion runs over a page (like here). Either there is a bug or I have made a silly mistake, it would seem. Ostrichyearning (talk) 23:16, 14 June 2015 (UTC)

Yes check.svg Done But I think only separate articles/chapters should be transcluded sepatately by using section tags, and not sections/divisions of an article/chapter. The Visitors and The Preliminary Tests on p72 do not require section tags. Hrishikes (talk) 02:03, 15 June 2015 (UTC)

Index:UK Traffic Signs Manual - Chapter 8 - Part 1 (Traffic Safety Measures and Signs for Road). Designs 2009.pdf [edit]

This should have been named as in the title for the document and I've only just noticed this :(

Can someone that's able to do a mass page move to the new name assist?

The correct name should have been Index: UK Traffic Signs Manual - Chapter 8 - Part 2- Traffic Safety Measures and Signs for Road Works and Temporary Situations) - Designs - 2009.pdf and I intend to request a rename at Commons, once I know the pages will get migrated here. ShakespeareFan00 (talk) 15:21, 15 June 2015 (UTC)

Wikisource webpages keep on loading non stop[edit]

Wikisource keeps on loading a page long after it's fully loaded. In Windows 7, this is true with Firefox, Comodo Dragon (Chrome) and Opera, but not with IE 11. Every other website loads and stops normally, including Wikipedia and Wikimedia Commons. This ongoing loading is also in Linux Xubuntu 14.04 Firefox, but not Chromium (Chrome). I suspect that it's some modification of a javascript that causes it. — Ineuw talk 21:44, 18 June 2015 (UTC)

I think it happens when is slow, as that's where various javascripts are served from (e.g.). Annoying. Not sure what can be done. It's not just you though. — Sam Wilson ( TalkContribs ) … 03:20, 19 June 2015 (UTC)
Thanks, I figured that I wasn't the only one, but it was worth to find out. Let this be a warning to A Smiley.jpgIneuw talk 04:04, 19 June 2015 (UTC)
Yep, labs is down, due to a disk failure:
Some things should be okay again now. — Sam Wilson ( TalkContribs ) … 05:39, 19 June 2015 (UTC)

Need help creating an .svg image file based on complex text[edit]

Is there anyone out there who can make a .svg image file out of the text below?


The raw HTML used to generate the above is...

<div style="display: inline-block; color: #000000; font-family: 'Linux Libertine',Georgia,Times,serif; font-size: 2rem; line-height: 0.75; font-weight: 500; text-align: justify; overflow: hidden;"></div>

... and the text itself (under the Linux Libertine font) in hex is:


Appreciate any effort or guidance on how to make this seemingly simple task a reality. TIA. -- George Orwell III (talk) 07:08, 21 June 2015 (UTC)

Is this of any use? Beeswaxcandle (talk) 07:45, 21 June 2015 (UTC)
In principle this ought to be fairly close to what you asked for. May need some border trim (which is where my knowledge runs out: width/height/viewBox probably need changing at the very least?):
<?xml version="1.0" standalone="no"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 20000303 Stylable//EN"
<svg xmlns="" width="11.7in" height="8.3in" viewBox="0 0 20157 13858">
<g style="stroke-width:.025in; stroke:black; fill:none">
<!-- Text -->
<text x="0" y="200" fill="#000000" font-family="Linux Libertine" font-style="normal" font-weight="normal" font-size="500" text-anchor="start" >
AuFCL (talk) 08:11, 21 June 2015 (UTC)
Thanks guys. I didn't have much luck with the online-converter (no Linux Libertine font-family online?) but managed to "alter" AuFCL's idea some....
<?xml version="1.0" standalone="no"?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 20000303 Stylable//EN"
<svg xmlns="" width="auto" height="auto">
<desc>Text based Wikisource logo substitute</desc>
<g style="display:inline-block; font-family:'Linux Libertine'; font-size:2rem; line-height:0.75; font-weight:500; fill:#000000; text-align:justify; overflow:hidden;">
<text x="0" y="32" dominant-baseline="reset-size" text-anchor="start">&#xE02F;&#xE059;&#xE05B;&#xE059;&#xE063;&#xE05F;&#xE065;&#xE062;&#xE053;&#xE055;</text>
I'm lost when it comes to this (viewport, stroke, etc.) but the above renders locally at least. I'll let this simmer for a day or two to see if somebody comes up something "better". -- George Orwell III (talk) 10:48, 21 June 2015 (UTC)
Have you tried the SVG editor at WMF labs? sDrewth 12:38, 21 June 2015 (UTC)
(As a complete aside I have finally figured out what attribute stroke et al achieves—and have further come to the conclusion it really is not what you wanted in any case. SVG1.1 overview specifications may be found here.)

I took the liberty of submitting your "improved" version to the W3C markup validation service and fiddling with the thing per suggestions made there until it finally passed muster. Here is the result (I completely misled you first time round with the chosen DTD and encoding—apologies! Also note things like dominant-baseline="reset-size" force SVG1.1 as SVG1.0 simply doesn't incorporate that feature…) -- AuFCL (talk) 08:32, 22 June 2015 (UTC)

<?xml version="1.0" encoding="UTF-8" standalone="no" ?>
<!DOCTYPE svg PUBLIC "-//W3C//DTD SVG 1.1//EN" "">
<svg version="1.1" xmlns="" xmlns:xlink="" width="10.0em" height="2.25em" preserveAspectRatio="none meet" xml:space="preserve">
<desc>Text based Wikisource logo substitute</desc>
<a xlink:href="">
<text style="display:inline-block; font-family:'Linux Libertine'; font-size:1.5rem; line-height:1.3334; font-weight:500; text-align:justify; overflow:hidden;" x="16" y="24" dominant-baseline="middle" text-anchor="start">&#xE02F;&#xE059;&#xE05B;&#xE059;&#xE063;&#xE05F;&#xE065;&#xE062;&#xE053;&#xE055;</text>
Made some additions/revisions to prior incarnation after doing some research. Damn if I still can't my head around the coordinate scheme being used; would be nice to shrink the line-height without clipping the text for example. I did manage to embed a link under the text that points to our main page at least. (Note: syntaxhighlight extension appears to be semi-broken; enclose, style settings have no effect). -- George Orwell III (talk) 01:58, 25 June 2015 (UTC)
@George Orwell III: Firefox 38.0.5 will not display anything when presented with this latest; but will accept (i.e. display and act upon) the linkable image if <svg> attributes height and width are relaxed to (both) values "auto". Does a viewBox set to all zeros even make sense (and if not would it be best to remove the specification?) Hope this helps somewhat. AuFCL (talk) 08:07, 25 June 2015 (UTC)
Additional note: I missed this before but expunging the viewBox clause at the same time as leaving height and width unaltered also results in a working image+link in FF. So presumably there is an internal conflict between the three attributes (at least inside the Gecko parser?) AuFCL (talk) 08:52, 25 June 2015 (UTC)
GOIII's 20:23 version of File:Wikisource-logotext.svg is much better. Now how about tweaking y="32" to y="16" to alleviate that "chopped-off-at-the-knees" effect? (Or does it work perfectly in I.E. already?) AuFCL (talk) 10:21, 25 June 2015 (UTC)
I've tweaked the uploaded version some more and my latest is now above but I cannot for the life of me understand where the "extra" offsets are coming from (beginning to suspect that Libertine font isn't all that svg friendly?). The straight hex text under plain old html (as applied by the flat-list side-bar gadget) is far easier to manipulate but its just not as "crisp" as the .svg variant would seem to be. I'll keep pounding away I guess. -- George Orwell III (talk) 12:13, 25 June 2015 (UTC)
Not a main-stream issue but appending attribute title="Visit the main page" to the <text> (or indeed to the <a> tag if you prefer—inner tag seems to take precedence in FireFox at least) tag emulates the flat-list gadget hover popup. The "crispness" argument seems to defy logic—I cannot usefully advise. AuFCL (talk) 12:49, 25 June 2015 (UTC)

Problem of /* Proofread */ not registering.[edit]

A number of pages I proofread, marked and so saved, show up as not proofread. When the page is reopened to check the radio button - it's on proofread, but the summary line is empty. Is this part of the crash? Or perhaps mice chewed through the javascript or php? — Ineuw talk 18:34, 21 June 2015 (UTC)

@Ineuw: Have you verified as much by checking the Page Status section when viewing the revision Diff of a Page: in question? If the diff shows the proofreading status was changed, then I suspect the problem indeed lies with something off-site (or with server caching maybe?). -- George Orwell III (talk) 19:05, 21 June 2015 (UTC)
I'll do that - but following was noticed - yesterday, I ran the volume status checker, to get the proofread page count and it had about 20-30 sequential pages marked as not proofread, but the pages showed the color status properly, only the summary was missing. Today, it's different in that the Summary is missing and the color indicator remains red. But, since it comes from wmflabs, I know that there was a crash and it is more likely the cause of the problem. — Ineuw talk 19:25, 21 June 2015 (UTC)
It seems edits from around June 11th & 12th are the ones specifically behaving that way in Checker. Locally, all else indicates the correct status just fine. It must be something wmfLabs related; & I am of no help when it comes those pseudo off-site entities. -- George Orwell III (talk) 22:20, 21 June 2015 (UTC)

A simple table row problem, which I can't resolve.[edit]

Hang my head in shame but try as I might, can't figure out why the text in this single row table don't line up on the top. — Ineuw talk 21:47, 21 June 2015 (UTC)

Fixed. Only the table cells (TD, TH) within a row (TR) "accept" vertical-align:top .css stylings; not rows themselves. -- George Orwell III (talk) 22:15, 21 June 2015 (UTC)

Remove pages from .djvu file[edit]

Help needed to remove two pages (scans of extraneous materials) from the file Chesterton's Alarms and Discussions. The pages in question are 73 and 74. Thank you very much! Cheers, Captain Nemo (talk) 03:54, 22 June 2015 (UTC).

Yes check.svg Done -- Hrishikes (talk) 08:45, 22 June 2015 (UTC)

Thank you! Captain Nemo (talk) 10:16, 22 June 2015 (UTC).

Editing a book to fix typos (93 by Victor Hugo)[edit]

I recently found this site and have started to read 93 by Victor Hugo ( that I downloaded from it. So far I am 25% in and have found about 15 instances of typos. Some are missing quotation marks. Some are clearly where a scan thought the letting "h" was "li" or some such thing. And one case is a double word. Is there a way for me to fix these or submit them to someone? I briefly scanned through the help page and couldn't make heads or tails of if it was alright for me to do these repairs (or confirm if they are part of the original text and should be left). I am happy to do the edits myself if that is ok but I am new and didn't want to act incorrectly.

Update: Ok I went to edit the document and found that the original text is displayed next to the book text and I fixed a few that do not match. However, there are some where there is a missing quote mark (First word on page 87 that I am not sure what to do about. Clearly the quote mark is missing but it is missing in the original text too. Regards, Adam unsigned comment by AdamWuertz (talk) .

@AdamWuertz: For my money the changes you have made so far are perfectly correct (and please keep going if you see any more corrections!)

With regards the missing opening quote that is a somewhat hard case: you might consider any of these alternatives:

  • leaving the page unchanged (on the basis of it matches the book page image—even though the typesetting technically is at fault),
  • marking the missing quote mark "silently" (i.e. note to proofreaders only—a bit late in this instance!) by adding {{sic}} at the point where the quote ought to have appeared, or
  • marking the missing quote mark by substituting the opening word "The" with {{SIC|The|"The}}.

As you see there are several ways of approaching this particular issue (each of which has both pros and cons) and I think it likely somebody else will raise further possibilities I've not thought of as well. AuFCL (talk) 09:24, 22 June 2015 (UTC)

EB11 Volume 27[edit]

Anyone know what happened here? I was trying to look for the article on the popes Urban but it looks like someone took a torch to the scan index. Or did it just not exist? — LlywelynII 13:20, 23 June 2015 (UTC)

The short story is that something is broken in how djvu files are handled somewhere by Mediawiki and/or ProofreadPage extension. There are more details in the archive but the bottom line is that no solution has been found yet (I am not even sure someone is still looking for it <sigh>) There should also be a bug filed.— Mpaa (talk) 21:18, 23 June 2015 (UTC)
See also Wikisource:Scriptorium#DjVu_file_okay_at_Commons_but_looks_faulty_from_hereMpaa (talk) 21:22, 23 June 2015 (UTC)

Paragraph breaks within poems[edit]

Is it possible to insert a paragraph break within a <poem>? The usual double-space results in a <br><br> instead of a </p><p>, which makes the line break much larger than usual. Ending one poem and starting another just collapses into one poem. The only other solution I can think of is to forego the poem extension and use line breaks on every line, but this will be unwieldy and ugly so I want to see if there is another option. Example: Page:Book of common prayer (TEC, 1979).pdf/46Beleg Tâl (talk) 19:39, 23 June 2015 (UTC)

This is part of the reason that I've almost stopped using Poem tags completely and use explicit line breaks. I've had a go at that page to show what I mean. Feel free to revert. Beeswaxcandle (talk) 20:30, 23 June 2015 (UTC)
Yes, that works well. Unfortunately it results in complecated markup, and will take much longer to proofread :( —Beleg Tâl (talk) 20:53, 23 June 2015 (UTC)

Moving over redirect[edit]

What's the proper way to move a page over redirect? The desired move is Author:M. G. Robinet to Author:G. Robinet (M is for Monsieur, of course!). Thank you, Captain Nemo (talk) 00:27, 24 June 2015 (UTC).

Yes check.svg Done . The easiest way is to delete the target to make way for move. Beeswaxcandle (talk) 01:56, 24 June 2015 (UTC)

Dividing two authors in the header template[edit]

How can the two authors be divided in the header template on this page? Macfadden's Fasting, Hydropathy and Exercise Jpez (talk) 13:43, 25 June 2015 (UTC)

You can use the "override_author" field for this, as I have done. —Beleg Tâl (talk) 15:01, 25 June 2015 (UTC)

Yes check.svg Done Thanks! Jpez (talk) 15:29, 25 June 2015 (UTC)

Two different references pointing to the same footnote[edit]

On this page Page:Arcana Coelestia - Volume I.djvu/36 there are two astericks in the first paragraph pointing to the same reference. How do you implement this into the transcription?

You use the "name" attribute. I have done it in the text. —Beleg Tâl (talk) 15:00, 25 June 2015 (UTC)

Yes check.svg Done Thanks! Jpez (talk) 15:29, 25 June 2015 (UTC)

Pages missing in djvu, how to insert?[edit]

I know this has been asked in the past, but the responses I found are of the "there, I did it for you" variety, and I want to learn how to fix this problem in the future. I am working on Index:Memoirs_of_a_Trait_in_the_Character_of_George_III.djvu and discovered that two pages are missing from the scan, specifically work pages 229 and 230, which should be in the index between 286 and 287. I have found another scan which includes those pages here.

I don't even know what tools to use. How does one:

  1. Extract two pages from a PDF file
  2. Insert those pages in a djvu file
  3. Correct the WS index after updating the file at commons?

(And feel free to point me at the documentation, which must exist somewhere around here.) Thanks in advance. -Xpctr8 (talk) 13:56, 29 June 2015 (UTC)

PDF pages can be extracted as images using the PDF24 Creator application. There are many such ways. After extraction of the pages as images, assemble them in a separate folder. Convert that folder to djvu using Djvu Toy or such other application. Then open the original djvu file using the edit option of Djvu Toy. Insert the new djvu file at appropriate location. Hope this helps. But mind, in the cyber world, there are many roads going to the same destination. This is just one of the roads. Hrishikes (talk) 14:35, 29 June 2015 (UTC)

A question about CSS classes[edit]

Do we have a single file for CSS "classes," if not, where are the CSS class declarations like "fsInherit"? — Ineuw talk 18:09, 29 June 2015 (UTC)

Technically we never had a "single file" for .css classes but to answer your question, the two main site specific .css files most likely for revisions, etc. are:
I can list the rest if one of those doesn't hold what you're looking for. -- George Orwell III (talk) 22:01, 29 June 2015 (UTC)
Thanks GO3, found it in MediaWiki:Gadget-enws-tweaks.css. — Ineuw talk 23:01, 29 June 2015 (UTC)

Language help[edit]

From time to time, I do the proofreading of Index:A History of Hindu Chemistry Vol 1.djvu. This work frequently has quotations and footnotes in Sanskrit, Greek, Latin, French and German. I have done the proofreading as best I could, but would appreciate if someone knowledgeable in languages could do some checking. Hrishikes (talk) 11:14, 1 July 2015 (UTC)

Hi Hrishikes, if you add the template {{greek missing}} to any Greek text you may find I'll add it for you. I can't help with the other languages though. Jpez (talk) 16:26, 1 July 2015 (UTC)
@Jpez: I had just gone ahead and put the text there, so right now I can't remember where the Greek words are; however, in another of my projects, you can help with the footnotes on this and this page and other pages. Thanks in advance. Hrishikes (talk) 16:45, 1 July 2015 (UTC)
I also keep an eye on Category:Pages with missing Greek characters, so anything tagged with {{greek missing}} will be gotten to eventually. I can help with Latin and French too if/whenever there is a list of problematic passages. —Beleg Tâl (talk) 16:57, 1 July 2015 (UTC)
So that explains why when I make my weekly visit to this category it's empty these days. Thanks for doing this. Beeswaxcandle (talk) 06:07, 2 July 2015 (UTC)
Speaking of which, would it not be prudent to make Category:Problematic language templates for common Latin-alphabet languages, for precisely this reason? —Beleg Tâl (talk) 16:59, 1 July 2015 (UTC)
We've avoided doing this, because it's assumed that the diacritics of the European languages that use the Latin alphabet are easily accessed through the Latin option on drop down menu on the CharInsert bar. But if it's felt to be something that would be useful, then {{diacritics missing}} could be created. I'm suggesting just one template, rather than proliferating for all the Latin and Germanic languages. Alternatively, could just use the existing {{symbol missing}}. Beeswaxcandle (talk) 06:07, 2 July 2015 (UTC)
For starters:

With thanks, Hrishikes (talk) 17:13, 1 July 2015 (UTC)

I've gone through most of them. They are all fine (better than some English-language proofreading I've seen) so there is no need for anything beyond the regular validation process. —Beleg Tâl (talk) 21:11, 2 July 2015 (UTC)
@Beleg Tâl: Many thanks. I was not really confident that they were OK. Though I try to be meticulous. Hrishikes (talk) 01:26, 3 July 2015 (UTC)

Reworking material from elsewhere[edit]

I teach history at the university level, and want to ensure that I am giving students texts that are accurate, cited properly, and legal. The Internet Sourcebook is a nightmare in these respects, and cannot be corrected since it is no longer maintained. Wikisource strikes me as one of the best available solutions for fixing this situation due to its ability to provide a clear and verifiable relationship to an original document, and to break things down into appropriate sections rather than giving students a link to one massive file representing the entire book. But the process for importing material from elsewhere is unclear to me. I've posted a few books on Project Gutenberg, creating files in their PG RST format, which can be easily converted into MediaWiki markup using Pandoc. It would be ideal if you had software that could take such a file and automatically break it up to match a set of imported photos, using the in-text page numbers.

I would also appreciate the ability to show in the text where I have corrected the original, and on looking through the archives I see many other limitations, such as the lack of a standard way to insert sidenotes: these are much easier to encode in TEI, and it seems that there has been some discussion of using it, but I cannot find anything more recent than 2013. It would be brilliant if you would consider supporting something like TEI Simple, since this would allow all sorts of accurate materials from elsewhere to be added and corrected, such as from the Text Creation Partnership.AndrewNJ (talk) 21:26, 4 July 2015 (UTC)