Wikisource:Scriptorium/Archives/2010-04

= Announcements =

PDF page cropping template
I am trying to fix OCR on a document (scientific publication) that is on Commons as a PDF, and that has inline images. Instead of extracting and uploading them into separate files, I stole a template from en.wikipedia and adapted it so it can show a single, cropped PDF page. Template is here, demo is here for the page and here for the join-up (look near the end of the page). Note that this increases the size of the "thumbnail" considerably; however, it needs to load each page thumbnail only once, no matter how many images from that page thumb are used (2 in my example). Also, reducing the page thumb size will reduce transfer size significantly. --Magnus Manske (talk) 23:15, 11 March 2010 (UTC)
 * Very clever... but this seems to me not optimal. Would you agree that cropping out the figure and including it in the usual way would be an improvement? Can we make the template categorise into a subcategory of Category:Problematic so that we can track pages that neede improving in this way? Hesperian 00:06, 12 March 2010 (UTC)
 * It would improve the page load times on slow connections, and give an opportunity to adjust colors/contrast/etc., but it would be more work to set up. Also, it would mean (in this case) uploading 15 images that were cropped from the PDF. Is this proliferation really necessary? I think not. Ideally, MediaWiki would do the crop server-side and only deliver the cropped part... I will try to reduce the resolution in my example to reduce the required transfer size. If you think the template should go into "problematic", I'm OK with that; but, I would like to continue to use it at least on this example, for my convenience ;-) --Magnus Manske (talk) 13:32, 12 March 2010 (UTC)
 * Can you make it work for DjVu files, especially where I need to rotate images? (/me prostrates himself with a big pleeeeeease.) — billinghurst  sDrewth  02:36, 13 March 2010 (UTC)
 * OK, rotation seems to work (Firefox, Chrome, Safari; IE, Opera untested). Use Rotation=-90 / 90 / 180. --Magnus Manske (talk) 14:00, 13 March 2010 (UTC)

New source for Australian works - Trove
The National Library of Australia have just put their works online at Trove. Among other things they have Australian newspapers from 1803 - 1954 and government records from The National Archive of Australia and also State and Territory records. Kathleen.wright5 13:35, 28 April, 2010 (UTC)

= Proposals =

Leave background colors alone
I think Wikisource should follow Wikipedia in formatting as much as possible, by this I mean there should be consistency from page to page. So, for example, on Wikipedia you will not find one page with a pink background color and another with a green background color. Why should Wikisource be any different? Case in point: THIS. First, the reason why the DjVu scans look brown is because the paper is acidic. Acidity in paper is not a good thing. Paper should be white. Wikisource pages should be white. Grab a modern book and you will find a statement like "this book is printed on acid-free paper", and the book is printed on acid-free paper to save the paper from turning that shit-brown color. (Sorry, but I do not see how the color of highly acidic paper is aesthically pleasing.) I therefore propose that manipulating background color should be strongly discouraged (if not completely disallowed). Or at the very least, the Ph should be lowered a few shades.Ingram (talk) 05:22, 1 February 2010 (UTC)
 * Faux pas. Sorry.Ingram (talk) 00:06, 7 February 2010 (UTC)
 * I gave the responsible HTML tag an id: "PMSLayoutTop". If you add the line
 * to your custom CSS, it will be white again, regardless of what the template defaults to. Paradoctor (talk) 08:38, 1 February 2010 (UTC)
 * to your custom CSS, it will be white again, regardless of what the template defaults to. Paradoctor (talk) 08:38, 1 February 2010 (UTC)


 * I would agree on the basic premise. It is no longer a book, and if they want the old book look, they can print out the djvu. That said, it is nothing that I am going to get too fussed about if that is the way that a project wishes to manage it. I am not going to encourage a wider proliferation of the look. billinghurst  sDrewth  10:36, 1 February 2010 (UTC)

Periodicals are not works, but collections of works
Why do we need unnecessary redirect pages? See HERE. Obviously, anyone searching for "The Antecedents of the Study of Character and Temperament" on Wikisource will type in "The Antecedents of the Study of Character and Temperament" and not "Popular Science Monthly/Volume 86/The Antecedents of the Study of Character and Temperament". So for a Wikisource search to go directly to "Popular Science Monthly/Volume 86/The Antecedents of the Study of Character and Temperament", the page "The Antecedents of the Study of Character and Temperament" will need to be set up with a redirect to "Popular Science Monthly/Volume 86/The Antecedents of the Study of Character and Temperament". What a mess! Why not simplify the matter and forego the "Popular Science Monthly/Volume 86/…" altogether and simply use the title of the work to begin with? The "Popular Science Monthly/Volume 86/…" nonsense, besides subverting the search function on Wikisource, also makes a mess of the category page, see here. Works that otherwise would not need a DEFAULTSORT, now need it (and obviously won't get it). Also, the "title" space of the header template is designed for the name of the work, and not the name of periodical in which it appears. Using the "title" space for the periodical name in turn obviates the "author" space, and so the work title and author are cramped into the "section" space. Why do we have to make this harder than it needs to be? And what if the work was published in more that one volume as with this work? Eventually a The Study of Sociology page will have to be set up anyway, so why is " Popular Science Monthly/Volume 1/May 1872/The Study of Sociology I " any better than " The Study of Sociology/Chapter I "? My modest proposal: let us treat works as works, (and thereby avoid rewriting templates, doubling page spaces with unnecessary redirects, and making a mess of the category pages).Ingram (talk) 05:22, 1 February 2010 (UTC)


 * There are previous discussions that have taken place in this forum that address this and I invite you to read some of that as the full background. Here are some of my thoughts.


 * People are working on the the periodicals, so they need to have a form of presentation, and one shouldn't make a decision to disregard the items as a collection and from their historical perspective
 * Aspects of the periodicals work to being better displayed in that format, especially with regard to the simple navigation of prev/next
 * To the Category page, there is quite an easy means to manage that. Tag the redirect with the category, and not the namespace article and it will display and sort fine
 * articles that continued over multiple editions can still be transcluded together, and that should not stop the the presentation by periodical, in fact, is the strength of the existing system
 * which templates have been rewritten? Yes, some adaptions to some, however, I think that they have remained remarkably stable, and they are used just as much on subpages of complete works in a similar manner. billinghurst  sDrewth  10:33, 1 February 2010 (UTC)


 * (1.) I work with periodicals, so forcing a policy change on naming conventions will effect me directly. Also, I see no "historical perspective" imperative. For example: Herbert Spencer's The Study of Sociology. Yes, this was published in PSM, but it was also—at the same time—published in the Contemporary Review, and then published over and over again in book form. Why does the PSM printing override all the other printings of the work? Or how is this:

Popular Science Monthly/Volume 1/May 1872/The Study of Sociology I  Popular Science Monthly/Volume 1/June 1872/The Study of Sociology II   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology III   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology IV   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology V   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology VII   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology VII   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology VIII   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology IX   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology X   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology XI   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology XII   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology XIII   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology XIV   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology XV   Popular Science Monthly/Volume ?/Month 187?/The Study of Sociology XVI


 * better than this?:

The Study of Sociology


 * For that work I would see that we could easily look to house two versions of the work. One the historical perspective of the development of the work, and your later version which may be from a separate source. Alternatively, the main concatenated article could be achieved by transclusions of the parts to one article, and a history of where the parts are derived. Exceptions will certainly exist.  Know the principle, then be rule guided, not rule bound.


 * (2.) The prev/next makes sense when reading a book, obviously you would want to go from one chapter to the next. But with periodicals? I think this is the "historical perspective" imperative again. If a user is looking for "On Denoting" by Bertrand Russell, I doubt they could care less what periodical it first appeared, let alone what articles preceded/followed. Why the imperative of having two red links in the header? Are you going to add the articles in question? I'm not. My God, we can't even complete the Encyclopædia Britannica and now we are taking upon ourselves every article in every periodical in the public domain? For historical perspective? Sorry, but this will never happen. We are lucky to have "On Denoting".


 * (3.) Look at the Sociology category page. What a mess! So to clean this up, we need to start creating #REDIRECT pages? Who exactly is going to do that? And why? If we started with the work titles to begin with, it would be totally unnecessary. (Notice The Study of Sociology is listed under T—we are still struggling with DEFAULTSORT).
 * I am not excusing any messes, I proposed a solution. It is one that I have used successfully with alternate names for authors


 * (4.) See (1.) and (2.).


 * (5.) I was referring to the header template. Tell me, if the "title" space now belongs to the name of the periodical in question, what goes in the "author" space? The editor of the periodical? Or leave it blank, e.g. here, and add Herbert Spencer to the "section" space? And if the work has the misfortune of having a subpage, where does the text that belongs in the "section" space go? The "notes" space? My point was, following your system, the header template would have to be rewritten for it to make any sense to a new user.
 * Really good points, probably indicates that have a version that can be used for periodicals, and we probably should. For DNB we have modified the header to reflect each contributor as the author see Jane, Joseph (DNB00). It is still header, though with a variation of the presentation.  Which is different from what I have done with obits from The Times where I have modified how I use it.billinghurst  sDrewth  05:26, 2 February 2010 (UTC)


 * I think for Wikisource to be successful we need to encourage participation, and to encourage participation we need to keep Wikisource simple to use; simple for contributors and non-contributing users looking for texts. In this I think we have a fundamental philosophical disagreement.Ingram (talk) 02:08, 2 February 2010 (UTC)

Adding an work without the subpage structure is fine, it's not a requirement of a contribution, others will organise it if and when it becomes problematic. Providing the source for a transcription is important, if not full details then some clue to discovering it, arranging these work's variants by subpages becomes logical. Categorising redirects is a good idea, especially if they become Versions. If the reader has been given a choice of versions, they wont be disorientated by the full citation appearing as the title, in I found this convenient when I use wikisource elsewhere. That Spencer's work was also serialised makes for an interesting example, as mentioned above leaving it in place and creating a separate version would be trivial. This longer work would still be subpaged in some way so the simplicity of the draft structure at The Study of Sociology would include the year of publication. The header doesn't accomodate some types of texts automatically, but once a solution or work-around is found it can copy/pasted. If a contributor wants to add a sourced transcription of an article, poem, short story, etc. they can do so under that title, if someone wants to arrange it within a deeper navigational structure it is hardly objectionable - it doesn't make work for anyone. Cygnis insignis (talk) 04:58, 2 February 2010 (UTC)

I couldn't possibly disagree more with the thrust of Ingram's comments: periodicals are works, and our subpaging convention rocks. I am sorry that I don't have time right now to do any more than register counter-assertions. I'll make more constructive comments later. Hesperian 08:34, 2 February 2010 (UTC)
 * Bah, I should be doing something else, but I can't resist the pull of Wikisource. :-( I have a book on my bookshelf entitled The Journal of Peter Good. That's a "work", right? And if I refer to it as "The Journal of Peter Good (Bulletin of the British Museum (Natural History) Historical Series Volume 9)", then it's still a work, right? A specimen of the botany of New Holland was released in four parts over two years; but it is a coherent work by a single author. It's a work, right? I think if Ingram examines his position carefully, he will find that "periodical" is a red herring. His issue is not with works that appear over a period of time. His issue is with works that are aggregates of other works. So stop me if I'm wrong, but it seems that Ingram's position is, or ought to be, that an aggregate of works cannot itself be a work. Either the aggregate is a work and the components merely parts of a work (as in chapters of a book), or the components are works and the aggregate is not so. I don't buy that. If you'll forgive me for stepping outside our scope in the search for compelling examples, I cannot see why I cannot consider to be "works" both the Pink Floyd song Comfortably Numb and the Pink Floyd album The Wall. To my mind both stand alone as works of art, despite one being a component of the other. Here's a hairy example. I have in my house a book entitled Stories and Poems, itself a volume in a series, that is essentially a collection of juvenile fiction teasers. It starts with a single chapter from Kenneth Grahame's The Wind in the Willows, which concludes with the teaser "Will Badger, Mole, and Water Rat ever be able to save the reckless toad from himself? Read The Wind in the Willows, the book from which this story came, and find out for yourself....". Turn the page and you're reading a chapter of Farley Mowat's Owls in the Family; and so on. If, as Ingram appears to think, no work can be an aggregate of works, I would ask which are the "works" in this case. Hesperian 09:15, 2 February 2010 (UTC)


 * Oppose any changes. I like the ... what do you call it? nesting? I think having Popular Science/Volume/etc allows for an organizational structure that cannot be achieved by having it all in the main article space. The redirect also allows for someone searching for the page to find it. I don't see how the current system doesn't work as a compromise to achieve both desires. Ottava Rima (talk) 16:14, 2 February 2010 (UTC)


 * Ingram, thanks for finding this dead end redirect. This remained from early attempts, and I regularly search and remove such links. If you find any others, please notify me and I will correct the related links. Also, I just read the various posts relating to the . Since I spend so much time on the project, I will gradually respond to everyone's comments, since all issues raised here were considered.— Ineuw (talk) 20:24, 2 February 2010 (UTC)
 * That is not a "dead end redirect". It is a perfectly functional and useful redirect, and one that should be kept. I think, given the current structure, that Ingram also would agree. I have again declined to speedy it. Hesperian 23:26, 2 February 2010 (UTC)


 * Hesperian, I disagree. There are numerous identical article titles and it's not possible to put a plain name. Please see the post below. The case with TPSM Project is just as bad, if not worse. — Ineuw (talk) 00:33, 3 February 2010 (UTC)
 * Yes, I know. The page should be located at Popular Science Monthly/Volume 86/The Antecedents of the Study of Character and Temperament. Anyone who searches for "The Antecedents of the Study of Character and Temperament" should be taken there. Hence the redirect is useful and should not be deleted. Hesperian 02:20, 3 February 2010 (UTC)


 * Good morning Hesperian. A refreshed mind is much better at realizing that redirects are of no consequence. I was fixated on eliminating my initial aborted attempts. I also realize that there will be numerous new redirects created by others who wish to access a limited number of articles and don't wish to wade through thousands titles. Have a nice day. :-) — Ineuw (talk) 13:51, 3 February 2010 (UTC)
 * Hello. Jolly good; I'm glad that's sorted out. Good night. ;-) Hesperian 14:53, 3 February 2010 (UTC)

My radical suggestion in all of this is: let's treat works as works. Let's treat Spencer's "The Study of Sociology" as a work in its own right. Let's use the search function as intended, let's use the category pages as intended, let's use the Header Template as intended. Allow me one final thought, though, if you will entertain it. My position is—just to be clear—when adding a work published in a journal or magazine treat it as any other work when naming it. Simple. The consensus above is no, no, no, we need the full path title. Right? y'all luv the forward slash (Br. stroke). Fine. Here is a list of popular periodicals:


 * 1) The Contemporary Review
 * 2) Nature
 * 3) Fortnightly Review
 * 4) The Nineteenth Century
 * 5) The English Historical Review
 * 6) The Critical Review
 * 7) The American Journal of Psychology

Now, tell me the official Wikisource naming convention in each of the above cases. This is the burden you want to place on me and the other users of Wikisource, so I place it on you. It's not as easy as you think. Do you think thousands of Wikisource users will produce an identical list of name paths ../../ ? Assuming we can agree on the title of periodical in the first place (remember Popular Science Monthly used to be Popular Science). After PeriodicalName/volume ?/.. imagine the chaos. Look at the redirects for PSM right now. And you want to duplicate that for every periodical? I know you have PSM set up tight and pretty, but have mercy on the rest of us. Even in the PSM project you have things like:

Popular Science Monthly/Volume 86/The Antecedents of the Study of Character and Temperament

vs.

Popular Science Monthly/Volume 1/May 1872/The Recent Eclipse of the Sun

Get your own house in order. ''What? I'm just say'n'' … Let me end with a reasonable proposal:

Any agreement? ~ :

I concede a dysfunctional search function and messy category pages—Header Template for another day.Ingram (talk) 03:01, 4 February 2010 (UTC)


 * Oppose, for reasons given above. Ingram, if you simply restate your position in the form of a formal proposal, without first engaging with people's questions, comments and objections, then the failure of your proposal is more or less inevitable. Hesperian 03:14, 4 February 2010 (UTC)
 * Your comments didn't make any sense to me, sorry, you lost me at Pink Floyd. Sorry, really. No offense. "Will Badger, Mole, and Water Rat ever be able to save the reckless toad from himself?" Yes…?Ingram (talk) 03:53, 4 February 2010 (UTC)
 * I don't mind trying again. You're seem to be arguing that titles like The miscellaneous botanical works of Robert Brown/Volume 1/Character and description of Kingia represent a failure to treat Character and description of Kingia as a work. I disagree; and conversely, I think your proposal represents a failure to treat The miscellaneous botanical works of Robert Brown as a work. All you have to complain about is an unwieldy title. But under your proposal the aggregate work would be ripped to pieces and denied identity as a coherent work. This is a much worse outcome. Hesperian 04:08, 4 February 2010 (UTC)
 * OK, I think I understand. The paragraph above—your immediate comment—illustrates a difference in opinion in editorial choices in presenting a work. You are correct as far as editorial choices I personally would make (although, for me, it would depend on the work, so in "../../Character and description of Kingia" I might have made the same choice you did, [but probably not, so you point stands]). I think editorial choices belong to the original contributor of the work; they are the ones who puts the most into making it [[Image:100%.svg]]. In my mind, "../../Character and description of Kingia" is User:Hesperian’s work, and if I wanted to do anything substantial to it I would first contact you on User talk:Hesperian and we would go from there. I simply want the same respect in my contributions. Please read my proposal again. I conceded almost everything in order to give myself a way out of this discussion. The only concession I ask for is time. Can we wait until a project is maturing before we run roughshod over other people's work? Consider this page. If I add "The Naturalism of Hume", I will add it as The Naturalism of Hume . If, sometime in the future, another user starts adding to Vol. 14 and wants to make it like PSM Vol. 1, who am I to say anything? But until that time, and since I am the only person working on Mind (journal)/Volume 14 (New Series), can we let things stand as they are? I don't think the name space The Naturalism of Hume is in anyway detrimental to Wikisource. At this point, take a deep breath, and try to empathize with what I am faced with. If I wanted to name this in the manner you are insisting upon…How? I won't make a long list of all the different combinations, but consider the "New Series" thing. Should it be "(New Series)" or "../New Series/.."? And if you give definite answer to that question, is that your personal preference? (Here, we are back to the "editorial choices" thing again). I calculated over 256 various combination in just the above example. Is it too much to ask for, for me to wait until another editor comes along (who would actually care about Mind (journal)) before any firm decisions on naming are started (leading to something like the various PSM redirects that litter Wikisource)? I have been criticized for not being delicate, but I have been reasonable through all of this. I have concerns, I have expressed my concerns, I think the source of my concerns will effect more than my projects. I have expressed my concerns in good faith for the betterment of Wikisource. I have given a reasonable proposal to allay my concerns. We are at an impasse, so I am done here.Ingram (talk) 01:09, 5 February 2010 (UTC)
 * An impasse? On the contrary. Now that I understand you, I agree with you, and always have. So long as your efforts don't interfere with the efforts of others, you should be given a free hand as much as possible. Hesperian 01:20, 5 February 2010 (UTC)
 * I have to laugh at myself, I am like a bear protecting its cubs. If I offended anyone, I am sorry, I was just being overprotective.Ingram (talk) 05:07, 5 February 2010 (UTC)

Periodicals and reoccurring titles
Hi, I have been working a bit on The Dial/Volume 15 which contains The Dial/Volume 15/Number 169 up to number 180. In each number there is a section called The Dial/Volume 15/Number 169/Briefs on News Books, Briefer Mention, and Literary Notes and News, as well a a couple of other reoccurring sections. I am sure this is true for most other periodicals as well; and I can think of no better way to organise them within the current system. Ingram may I suggest that your next post be a bit more delicate. And try to remember what Fitzgerald said about the exclamation marks :) P. S. Burton (talk) 00:04, 3 February 2010 (UTC)
 * That was me being delicate.Ingram (talk) 03:01, 4 February 2010 (UTC)

Customised header for periodicals
As part of the discussion, there is comment about the author/contributor on articles, rather than the whole periodical. I agree on the basic premise that for a periodical the editor is less relevant to be named, and it is somewhat misleading for the author of an article to be placed above the section. The options that I see to address this are There are strengths and weaknesses to each. billinghurst  sDrewth  00:29, 3 February 2010 (UTC)
 * manually modify a section to have an author wikilink below the section name by use of a &lt;br />
 * forget it
 * create a new header design format specifically for periodicals based on header parameters
 * add a new parameter to header that triggers a modified layout
 * add a new parameter eg. that is adds extra bit but relies on other header tags being omitted
 * Could we create a super header (super in the sense of "above")? I imagine a shorter header just above the regular header with the name of the periodical in the center with the prev/next aligned left/right respectively. The text would look like this:


 * This is an evolutionarily approach to works on Wikisource. First, a work from a periodical is added normally with the normal header. Then if/when the periodical is realized (Hegel, get out of my head…) as a work by and for itself (ahh…), at that point the super-header could be added with relative ease (someone would have already done the heavy lifting with the header, which would only need minor tinkering).Ingram (talk) 23:11, 5 February 2010 (UTC)


 * We could use this for all collected works, and if I had this option I would be more inclined to treat collected works as works in their own right. E.g.:


 * We would also avoid confusing the inexperienced Wikisource user who would quickly figure out that the text below matches the title in the green box thingy.Ingram (talk) 02:04, 6 February 2010 (UTC)

= Other discussions =

Helping new editors with common templates
To aid new editors (like myself) in the use of common templates, I've been bold and created the Help:Templates page and linked to it from the Help:Contents page. I would be very grateful for any assistance I can get in adding content to the page, as I'm likely to learn a great deal in the process!!! Many thanks! -- Clifflandis (talk) 18:40, 11 February 2010 (UTC)
 * It is helpful to everyone to get this stuff updated, cheers! I added a few of my most frequently used templates to Help:Editing Wikisource, make use of that if you like. Help:Side by side image view for proofreading has some recent stuff about using them in Page:namespace and more, there is also stuff in the Wikisource:namespace like Tools and scripts. Many of the templates have some documentation, which I see is linked, a nice example is right's doc. Cygnis insignis (talk) 19:31, 11 February 2010 (UTC)

= Questions =

Public Domain Day in the UK
Hi all. WMUK have put out a press release today about Public Domain Day in the UK. It would be great to upload some new content here as a result of works coming out of copyright - however things are somewhat complicated due to international copyright law. Does anyone know any specific examples of works that are now coming out of copyright and can be put here? Examples of authors whose work comes out of copyright in the UK include Yeats and Freud. Thanks. Mike Peel (talk) 17:32, 31 December 2009 (UTC)
 * We believe that New Poems (1928) by Yeats will be out of copyright this evening in the UK, and is not in copyright in the US as its copyright was not renewed. The only issue may be if the poems have been published in other compilations. As such, we plan to upload these early tomorrow morning. If anyone knows of any problems with this, please let me know asap... Thanks. Mike Peel (talk) 21:58, 31 December 2009 (UTC)
 * 1928 or 1938? You added the latter date on the page. You wouldn't happen to have a scan of the work?  If you do, we can do some serious work with it if we convert it to a .djvu file. billinghurst (talk) 06:19, 1 January 2010 (UTC)
 * Sorry, that should have been 1938. Unfortunately only 450 copies of the book were produced in 1938, which makes getting a copy to scan a tad difficult... Mike Peel (talk) 09:47, 1 January 2010 (UTC)
 * Full text is now up. I've also done a book at User:AndrewRT/Books/Yeats' New Poems but the tag isn't rendering correctly. Any ideas? AndrewRT (talk) 11:25, 1 January 2010 (UTC)

"The only issue may be if the poems have been published in other compilations." Not an issue in US. Copyright circulars at copyright.gov will explain in more detail if you are interested or need more information but this part of a response from the Copyright Office addresses your issue concisely. Refrigerator Heaven (talk) 10:20, 3 March 2010 (UTC) However, works in the public domain can be freely used by anyone because they are no longer under copyright protection. This means that if you add something to a public domain work, you can file a copyright claim in only what you added to the PD work and the claim can only be in what was added, never in the PD work itself.

Thank you. cfk

For those who want the stats
ThomasV has created some more templates (thanks!) with stats that replicates the EN data at oldwikisource:Wikisource:ProofreadPage Statistics. I have added it to Administrator's noticeboard/snapshot in case anyone wishes to transclude it, etc. billinghurst  sDrewth  14:39, 23 January 2010 (UTC)


 * What is with the Greek Wikisource stats? It says Greek Wikisource only has 6 texts, while on their website they claim 2,757 texts.  ResScholar (talk) 04:22, 2 February 2010 (UTC)


 * Apparently it means they have 6 scans in the Page: namespace. ResScholar (talk) 04:36, 2 February 2010 (UTC)

Need some guidance
Hi all,

I'm curious if there's a clear statement of how Wikisource defines "previously published" (in terms of what texts qualify for inclusion). I ask because I'm in the early stages of exploring an idea with some professors: using Wikisource to publish cases that result from rigorous research, and are used in graduate level classes in public policy and related fields. Although the process used to produce them is rigorous, I'm not sure whether or not they are "published" in a traditional sense (as in, made available to the general public for purchase, etc. etc.)

There is potentially a great deal of research that is currently unavailable to the public; we're exploring the possibility of publishing it on a Wikimedia project. Can anybody help me understand whether Wikisource is the right home for this sort of work? -Pete Forsyth (WMF) (talk) 20:56, 23 January 2010 (UTC)


 * I don't know what exactly constitutes previous publication, but seeing as this material is used in teaching, Wikibooks might want your material in case it is not appropriate here. Paradoctor (talk) 21:40, 23 January 2010 (UTC)


 * What's In and Out and What Wikisource includes covers the topic in the principle. Published doesn't need to be retail sale, especially if you consider WikiProject Academic Papers.  I would think that if a paper has been through a traditional rigorous peer review process; is willing to be put into the public domain under an acceptable licence, then I would personally encourage us to further develop the idea.  From my POV, the publishing aspect is about being able to prevent a plethora of self-written works that haven't met a community standard (of the time) from being included. Not yet a yes, but personally something that I would like to see further explored. billinghurst  sDrewth  23:30, 23 January 2010 (UTC)


 * Thank you -- those are both helpful answers. As I'm not sure yet what exactly we're dealing with on our end, I wasn't hoping for anything definite -- but both these answers help me understand the general lay of the land. Much appreciated, and I'll stay tuned for more feedback. -Pete Forsyth (WMF) (talk) 10:19, 24 January 2010 (UTC)


 * On the other hand - I suppose "case studies" may fit rather better into Wikiversity, which in effect collects teaching materials. Charles Matthews (talk) 13:52, 24 January 2010 (UTC)


 * Yes, that was my initial thought -- and there may be uses for both. The point here, I think, is that they contain substantial research that may be useful outside of the intended educational use. I'll see if I can get a copy of one of them CC-By-SA'd (very likely) and upload so we have something specific to discuss. Might take a week or so. -Pete Forsyth (WMF) (talk) 18:18, 24 January 2010 (UTC)
 * Two things. 1) We could look to a cross-wiki project, where the info sort of sits seamlessly in appropriate sites. That would be new and interesting for us. 2) We can always create Author: namespace pages that links to the WV/WS pages again as required. billinghurst  sDrewth  06:54, 25 January 2010 (UTC)


 * That would mean the hosting importing wiki(s) would have to enable scary transclusion, right? Paradoctor (talk) 07:23, 25 January 2010 (UTC)


 * For a beginning/sample, I'm planning to ask the professor to simply email one of the cases to OTRS. Then we can all discuss, and figure out the best way to handle them. Sound like a good plan? Is there a set of instructions written up somewhere for that? If not, I'll just send her the commons instructions, and modify them to suit a text document. -Pete Forsyth (WMF) (talk) 23:24, 25 January 2010 (UTC)

(outdent) What format is being looked at to supply the information? We encourage image files with text layers, though that is usually when we are proofreading / transcribing, which isn't going to be happening on this occasion. Our preferred means is .djvu file, and these are uploaded to Commons, which obviously makes them universally available. If they create a PDF file with a text layer, see instructions about conversion at the djvu link. Alternatively, mention where we can download a PDF file, and a number of us can assist. billinghurst  sDrewth  07:08, 26 January 2010 (UTC)


 * Any reason why it can't just be an RTF, or even wiki code included in the body of an email message? -Pete Forsyth (WMF) (talk) 01:25, 27 January 2010 (UTC)


 * No, none. Though beyond a page history, there is no ability to demonstrate who did what when, the original form of presentation, etc. Id that is not important to them, then easy peasy. If it is wiki code, we can slap it straight in. If RTF,  we will need to convert to something. If it it coming that way, we can pretty well deal with anything, and get it to a format that we can manage, from postscript to MS Word. We have the skills. billinghurst  sDrewth  05:11, 27 January 2010 (UTC)


 * I left a note on Peter Forsyth's talk page that links to something at Wikiversity as an example on what is published over there. (I felt a note there would get more attention than a note in an old thread). Ottava Rima (talk) 00:04, 2 February 2010 (UTC)

Sister links need some work
Would someone with a bit of graphic skills, please have a look at Wikipedia and Commonscat and see if they can standardise them. Maybe see what we can standardise crosswiki. If you have a look at The Musical Times you will see that they look butt ugly. :-( Thx billinghurst  sDrewth  15:04, 26 January 2010 (UTC)


 * These may be fine in the external link section of a wikipedia article, but they are ungainly here. I think the style given by the author template is a lot neater, if one prefers ad-style templates over wording with a link. This came up before, the example was Yagan. Cygnis insignis (talk) 15:19, 26 January 2010 (UTC)
 * Fair points. I am not so certain that I want something as insignificant as those in the author template, as they fade into the morass of a page. I would be happy with something small with an icon and no specific text beyond a generic [ICON link]. Maybe a line and a half in height (2/3 height of toolbar button?), and 4-6em in length (2x width of toolbar button). Standard, and neatly stackable. Usually it is only one per article, and very occasionally two. If it is a minor link, then I usually use a link. billinghurst  sDrewth  16:14, 26 January 2010 (UTC)
 * I started working on some standardization of that awhile back; see template talk:sister for an example. -Steve Sanbeg (talk) 18:37, 26 January 2010 (UTC)


 * I think merging this into a single template is the way to go here. I've put it on my to do-list. Paradoctor (talk) 18:48, 26 January 2010 (UTC)


 * This is meta data associated with the content, which is included in existing templates. The links in the author template appear at the right of the notes section, any[one] using these [having used this] once or twice knows exactly where to find. Adding a link is generated by the extant argument "wikipedia=" and so on. This is simple, intuitive and elegant. To achieve the effect described above I would have to stand on the other side of the room, I will concede that a banner-style amidst the actual content remains 'loud and proud' from this vantage point. Cygnis insignis (talk) 04:07, 27 January 2010 (UTC) Modified blather to to be slightly easier to read. 04:05, 28 January 2010 (UTC)


 * I'm not sure I understand what you mean by "achieve the effect", but formatting of the link set can have several options, with context-sensitive default behavior, so compatibility with local layout conventions should not be a problem. Paradoctor (talk) 23:59, 27 January 2010 (UTC)


 * We already have a method a method for linking sister sites, it is very widely deployed as it appears in the author template. It was suggested above that this would not stand out, it was "insignificant" and "they fade into the morass of a page". I don't how this could be, it appears at the same place in the existing template, not amidst the content that is the primary purpose of this site. It is not necessary to create a new template, with a new box and formatting, to be placed among the actual content; this creates noise and a jumble of formatted metadata all trying to stand out lest they be overlooked. The author template is not broken, the same functionality could be added to the header template; if a reader uses the sister links note at one page they know exactly where to look for at any other page. If it doesn't appear, and any user thinks to include a sister-link, they can add it to an part of the existing template rather than discovering and adding one of the various other templates. Cygnis insignis (talk) 02:30, 28 January 2010 (UTC)


 * Your statement finally passed my parser. My turn to clarify: All your concerns are addressed by "compatibility with local layout conventions should not be a problem". I'm not going to force anything on anyone (not that I could), merely refactor and add (optional) functionality. I already had to deal with similar concerns on Commons, so I know how to work with that.
 * "simple, intuitive and elegant": We agree about the first. It was not intuitive for me, even though I knew that there was something to find. Icons have not become ubiqitious for no reason. But as I said, that is nothing that stirs me. And hey, if you see the kind of chrome I can offer, you even might like it. ;) Paradoctor (talk) 03:25, 28 January 2010 (UTC)
 * I forgot to add that I have no objection to a small icon appearing next to "quotes" etc., like the 'inline, sentence style' template at wikipedia. 'Icon and short text' would assist the recognition, 'branding', without shouting about it, or 'advertising', with a separate banner. I will reiterate that it should appear with other metadata in the header; this is more convenient than the default placement at Commons, some article on the same topic would appear in the sidebar at that site. Cygnis insignis (talk) 03:51, 28 January 2010 (UTC)


 * It's not what Billinghurst asked for but, for Cygnis insignis, would something like Interwikidiscreet be what you want? I made it by cannabalising bits of the author template and existing interwiki templates.  I've also added some functionality so it can be used for more than just wikipedia, commons and wikiquote.  It could be placed anywhere, the header or footer. - AdamBMorgan (talk) 23:41, 28 January 2010 (UTC)
 * For example:


 * Actually, no, I typed too soon. It doesn't work in headers but it works elsewhere. I will attempt to debug the template.  However, would this be helpful if it did work properly? - AdamBMorgan (talk) 23:53, 28 January 2010 (UTC)
 * That should be ok now. Also, I've found out that the Wikipedia template is a table while Commonscat is a div.  As the latter is more advanced html, I've converted Wikipedia to the same format.  I think the font size and box width might be a little off but that can be easily modified with the tags in the first div tag of each template.  Wikiquote already has a div-based version called wikiquote2, so I'm not sure whether to change that one.  I haven't looked at the others but I can get around to that in a day or so (the converting isn't difficult, I'm just going to be busy). - AdamBMorgan (talk) 00:53, 29 January 2010 (UTC)
 * ✅ All of the interwiki templates should now use much the same code and have identical widths. Note they are set at 90% text size, which seems to be shrunk even further if they are included in the notes section of a header template. - AdamBMorgan (talk) 13:24, 1 February 2010 (UTC)
 * Thanks Adam. I just made the icons smaller too, to lessen the depth. Can you check that it looks okay as per The Musical Times and if it looks okay, can you propagate it? Thx. billinghurst  sDrewth  22:51, 1 February 2010 (UTC)
 * That should all be done now too (see Sister projects to compare them all). The Musical Times looks fine. The text seems a little too small but I can't seem to correct that and it might be more suitable for the header anyway. - AdamBMorgan (talk) 20:36, 3 February 2010 (UTC)
 * I used the discreet version recently, there was a wiktionary and wikipedia entry, some stuff at quote, and two corresponding links at commons. Adding right-aligned boxes would have created a lot of empty space for a user, before giving them the targeted left-aligned content. Another option would be to put these links with the sisters in the sidebar, the language links produced by link and so on. If the creator doesn't mind I will play around with Interwikidiscreet, or swipe the code, or offer some suggestions on how it could meet most needs for links leaving en.wikisource. Cygnis insignis (talk) 21:17, 3 February 2010 (UTC)
 * Be my guest. - AdamBMorgan (talk) 00:32, 4 February 2010 (UTC)

Help:Introduction
With a bit of a heavy heart I refer the community to recent edits to that page and to the discussion here. Input is invited regarding both the specific matter at hand as well as the overall atmosphere in the discussion. Dovi (talk) 21:51, 28 January 2010 (UTC)


 * As for the policy issues raised, I don't really see any point in adding our own illustrations, unless it's a case where another edition of the book, possibly in a different language, had the illustrations. Annotations are useful, if problematic; I'd like to see how the other namespace used in the Hebrew WS works on this, but I know no Hebrew. I think strongly that translation is an important part of what we do here. I don't think that moving Ho, mia kor' or Rat on a tray to another Wikiproject is going to help anything; it's certainly not magically going to create someone with the knowledge and interest to check them. Could Rat on a tray use some serious work? Sure; but even without it, it's a valuable addition to what we have. Through a glass darkly, but at least we have the glass. Trying to move translation elsewhere or drop it altogether would detract from Wikisource being a repository of all literature.--Prosfilaes (talk) 03:01, 29 January 2010 (UTC)


 * Indeed. Category:Wikisource translations has far more material than I even realized until I looked now in the wake of this discussion. Some of it is quite nice; other parts need a great deal of work. But that is the wiki way: Many people contribute over the course of time to eventually improve things and complete them. In the meantime they can be rated for completeness and accuracy, or a note can be posted that they have not yet been reviewed.


 * Regarding the "Annotations" namespace in Hebrew, it contains some of the most popularly viewed (and high-quality) material at Hebrew Wikisource, and is partially responsible for our high page-view ranking. There is also much material there that needs a great deal of work. One user did indeed object that annotations don't belong in Wikisource since they are creative, but the fact that there is a separate namespace seems to mollify even people who hold that view, since it clearly indicates that a different, though related, kind of Wikisource work is going on. An "annotations" namespace also helps us serve the public by overcoming the limitations inherent in aged, public domain versions of classical literature from generations ago, and include updated sourced scholarship. At the same time it allows for simultaneous cooperative work on related source texts and their Wikisource annotations, and the transclusion of source-texts into the annotated edition.


 * In general, the wiki way at its best is to be inclusive and encourage a wide variety of useful contributions. By the logic that "99% of people are not interested" (from the discussion at the talk page), then we also should have only Wikipedia and not Wikisource, and we should have wikis in only a handful of languages.


 * Sorry that I'll have to absent myself from the continuation of this discussion over the weekend. Dovi (talk) 04:41, 29 January 2010 (UTC)


 * The quote is "99% of the work that is done at Wikisource is not about creation of original content, ..." diff


 * by the logic that 99% of wikipedians were not interested, Wikisource was created as a separate entity from Wikipedia. It is a good thing to gather the users who share the same project on the same wiki. ThomasV (talk) 20:10, 29 January 2010 (UTC)


 * I think tossing around "99%" isn't very helpful; we have a lot more Wikipedians interested in The Open Window (Saki) than 2009 New Mexico State Aggies football team; it's about separation by form, and policy, not mainly social groups. Wikisource stores the literatures of the world; if I'm looking for the works of Arkady Averchenko in English, I don't think I should have to search Wikibooks for some of them. Nor is Wikibooks set up to deliver bilingual texts and small independent texts like we are. I think that should overweight arguments about "original content". I think that Wikisource is more likely to form a group of translators or to receive casual translation fixes than Wikibooks is.--Prosfilaes (talk) 03:03, 30 January 2010 (UTC)


 * "It is a good thing to gather the users who share the same project on the same wiki." Likewise, it is good to be inclusive of the similar yet not identical ways that the same project can be accomplished. Let users who share the same project complement each other, rather than trying to fragment them or to mandate a single method for contributing texts. We are building a library, and that is our shared project; a library needs and includes more than what is available through OCR alone. The OCR tools are of course extraordinary, and we are all grateful for them. Dovi (talk) 10:50, 2 February 2010 (UTC)

ref name
Is it possible to use "ref name" in Wikisource to introduce several references to one footnote?

I tried to use it at Page:Gesenius' Hebrew Grammar (1910 Kautzsch-Cowley edition).djvu/50 and it didn't work. Am i doing something wrong or is it really disabled at Wikisource?

Thanks in advance. --Amir E. Aharoni (talk) 22:08, 1 February 2010 (UTC)
 * Yes, it is enabled and it does work, eg. Page:The life of Matthew Flinders.djvu/366, and I have no idea why it isn't working for you. billinghurst  sDrewth  23:05, 1 February 2010 (UTC)


 * I can get a test to work on the previous page. billinghurst  sDrewth  23:09, 1 February 2010 (UTC)
 * I have it working. I copied the tag from the first, and pasted it over the second, and added the forward slash and it works fine.  There is a diff, so all I can think is that there is some character there that is different but not visible to the eye. Weird! billinghurst  sDrewth  23:24, 1 February 2010 (UTC)
 * Thanks! I am baffled, too... --Amir E. Aharoni (talk) 23:44, 1 February 2010 (UTC)
 * This is similar to this one. I can confirm that your page contains numerous non-printables, apparently unmatched combining diacritics. You should always be able work around this by deleting and retyping OCR input. Paradoctor (talk) 00:15, 2 February 2010 (UTC)
 * Thanks! Is there some easy way to see that they are there? I can't retype whole pages... :) --Amir E. Aharoni (talk) 00:21, 2 February 2010 (UTC)
 * You mean, you don't want to. ;) Install Notepad++ and the hex editor plugin, that will allow you to see everything in the text. If you don't want to istall anything, you could try to retype all occurences of special characters after deleting them and one "normal" character to each side. This wouldn't catch single occurrences of invisible cruft, of course, but this is probably not a problem here. In the long run, you'll probably want an editor capable of displaying non-printing Unicode characters. Another thing you could try is to mail a copy of the text to yourself, but that is probably no better than retyping the specials. Paradoctor (talk) 01:03, 2 February 2010 (UTC)
 * Unprintables and invisibles are never really a concern to me, it is only the visible cruft that I dislike. It is unusual for invisible cruft to confuse that often. Apart from that, if you do know the characters that are regularly appearing in your works, we can readily & easily do a [search and remove] script that can be added via Pathoschild's regex gadget to your left toolbar by editing your monobook.js.  Something like a line  rm cruft  that you would click once per page and voila, cruft removal. You would need to identify the cruft symbols to be removed so we could search for them. billinghurst  sDrewth  01:33, 2 February 2010 (UTC)
 * That requires determining when non-printables are ok and when not &rarr; to do list. Paradoctor (talk) 03:13, 2 February 2010 (UTC)
 * Thanks for all the comments.
 * Since we're discussing it, do you happen to understand what exactly is this invisible character here? It's between the closing bracket and the less-than sign: ')<r. In a hex editor i see the numbers ef bb bf. What kind of an unprintable is it exactly? And is there some way to force Firefox or at least Notepad++ to show them? "Show all characters" in NP++ doesn't help... --Amir E. Aharoni (talk) 06:39, 2 February 2010 (UTC)
 * Voilà! ﻿ I simply copied the offending character into Wikipedia. --Eliyak T · C 07:08, 2 February 2010 (UTC)
 * It is Unicode Character 'ZERO WIDTH NO-BREAK SPACE' (U+FEFF), and my favourite site for these things is fileformat.info billinghurst  sDrewth  08:59, 2 February 2010 (UTC)

Regarding Leave background colors alone
The essence of Ingram's first point is that he finds the background colour offensive. There was no consensus taken on a color scheme and in the wiki world, results are the fastest way to evoke a response desired by this writer. No proofreaders voiced their opinion. So, consider yourself the first. :-)

There is no way I wish to force users to waste their time making code changes to suit their reading pleasure and comparisons to Wikipedia are invalid. Colours and other minor aesthetics are being slowly introduced there as well, and 800 pixels wide lines are very unpleasant to read, although, not in the same class of unpleasantness, as Ingram's reference to the current page color. :-)

All I really need is a lineation, a contrast, between the approximate page size and the overall screen. To this end, I placed a THIS of the PSMLayout template for another possibility. Also, please look at THIS PAGE for standardized font sizes of the main title and sub-titles, which I am using, and were based on the analysis of the scan and preferences by the majority of previous proofreaders.

As time permits, I will provide additional info about academic interest in certain topics of the PSM, and how this relates to my efforts in this project.— Ineuw (talk) 19:12, 3 February 2010 (UTC)


 * Very interesting, thanks, Ineuw. Could you show how the annotations might look like, if there are any? Is it possible to find some way to have them different from the author's text if they were written by another person? --Zyephyrus (talk) 23:08, 3 February 2010 (UTC)


 * "800 pixels wide lines are very unpleasant to read": Really? I'm starting to find anything less than 1000 pixels annoyingly short. ;) Paradoctor (talk) 10:02, 4 February 2010 (UTC)

In reference to Periodicals are not works, but collections of works


My Dear Ingram,

Just received your telegram and hastened to return to this side of the ocean (and century). I discerned your anguish, especially after counting the use of the word "Why" 8 times!

First things first. I telegraphed Alex (Bell), and he told me to tell you that, the reason one needs "unnecessary" redirect pages is guided by the same principle as assigning a name to a 10 digit phone number in a cell phone directory. Sadly, gone are the days of Ernestine, or the 5 digit rotary dial phone. He also mentioned a rumor about something inexplicably called Google which lists all web pages and prominently distinguishes the article title from the surrounding text.

Regarding the categories, I was just as dismayed after seeing the results and marked it down as another matter to be dealt with. The question is whether the [[Category: link can be modified to display the title in alphabetic order, while masking the path. Then, new issues arise, since there are numerous recurring titles in the project and these would require some visual identity of note.

Also, in this project, the publishers of PSM broke up The Study of Sociology and published the articles periodically, and each article deals with a different subject requiring its own sub-categories, apart from the general category of sociology.

The header template can be manipulated to display what and how we want. So, please eliminate another "why" from your list and present us with a show and tell.

We have the same goal in guiding visiting readers, (and not editors), to quickly access articles by title, or category. To this end, I created the first version of a proposed ARTICLE TITLES' INDEX which hopefully, and with community input, will evolve in some acceptable form to be part of a gateway to the project. Additional pages relating to the project and possibly to your post, are AUTHORS AND CONTRIBUTORS and CATEGORIES

The intention is to produce a quality professional display within the Wiksource framework and if the contents cannot be promoted to readers then it's a useless undertaking.

To summarize, there are also other news, of the gossipy kind. While sitting on this newfangled invention called the "loo", I've just read that a personal typing machine is being developed, Nikola T. is on his way to complete his masters, Thomas E. is still sucking the air out of a "glass bulb???", Albert is only a gleam in Mr. & Mrs. Einstein's eyes. Mrs. Vicki Brown is grossly embarrassed by Bertie's scandalous public behaviour, while her teenage grandson, little Willy, had ten more whiskers grown into his mustache.— Ineuw (talk) 21:25, 4 February 2010 (UTC)


 * The unfortunate thing here is that I didn't read this until after offering my apology above. You want to make this personal, but my concerns were not intended as a personal attack against you, and I offer you a special apology if you took them as such. I just don't think Wikisource pages should have tan lines.Ingram (talk) 06:01, 5 February 2010 (UTC)

Absolutely not taken personally. :-) It was just my humorous take on the project, since I spend so much time on it that one gets affected by the speech style of the times. It's like living in the 19th century and comparing to the 21st century concept, I find a lot of humour in the arcane. I am glad someone commented, and I hope that this continues to be so.— Ineuw (talk) 16:52, 5 February 2010 (UTC)


 * An addendum:

By the way, can I just go ahead and change the PSM background? — Ineuw (talk)


 * Sorry, I did you an injustice by assuming malice. The background thing: I think I have seen other editors experiment with it, so I was concerned with this getting out of hand (in my mind I multiply actions by the number of users on Wikisource, so thousands of WS users correctly adding—say textinfo—would be great; thousands of editors experimenting with background color, watch out). If (#F9EACD;) makes you happy, please keep it, I won't say anything else on the matter. Just don't change it to (#FF69B4;). Tan lines are sexy enough.Ingram (talk) 02:27, 6 February 2010 (UTC)

Playing with the colours
No offence taken at all, and sorry for the weird sense of humour, but one must take time out and inject some levity into it all. The object of colouring was to make the project stand out and be easily identifiable to non-editing readers; medical students being introduced to the history of 19th century medicine, particularly physiology. The acid paper colour was chosen on an impulse, the desire for an instant solution and move on, and familiarity because I collect old books. But, most importantly, it was a successful attempt at initiating discussion and exploring other opinions.— Ineuw (talk) 15:52, 6 February 2010 (UTC)

Ingram, border color will be set to #F9EACD and let's move on. — Ineuw (talk) 18:54, 6 February 2010 (UTC) :-))

US Dept. of the Interior
Are budget reports such as the one found here appropriate for Wikisource? If so, is the a Wikiproject,category, etc? Thanks, The New Mikemoral (talk) 02:31, 5 February 2010 (UTC)
 * Personally I dislike the things, however, the one thing that we have at the community is the ability to rise above our own dislikes and look to our principled approach. If you think and can justify that fits within What Wikisource includes with a quick glance at WS:IO and would be prepared to politely argue a case against that criteria, then it presumably fits until the community decides otherwise. If there is a perspective from the above documents that feel make it borderline, then that is the time for discussion. As an explanation of my dislike, it is more reflective of things that aren't otherwise readily available; desirous of being made into a collection; or have an enduring historical perspective. billinghurst  sDrewth  03:50, 5 February 2010 (UTC)

Why don't zero-width spaces behave properly?
At Page:Judson Burmese Grammar.djvu/21 I have inserted zero-width spaces (U+200B) in between Burmese characters at appropriate places so that the long phrase translated "God said, let there be light" will divide across a line break as appropriate. But it doesn't work! The phrase is kept together as a block on a single line rather than splitting. (This should be true even if you can't see the Burmese characters; even if you just see squares or question marks, they should still be splitting across the line break, but aren't.) Anyone know why not? Angr 16:51, 6 February 2010 (UTC)
 * Not I, though a question, how do they behave when transcluded into the main ns? This will test whether it is an artefact or inherent. billinghurst  sDrewth  12:34, 7 February 2010 (UTC)


 * I do. Because they are not in the text! ^_^ I inserted a few HTML entities, and they work just as they should. I do not know their proper places, of course, but you should be able to follow my lead. What did you do to insert them? Paradoctor (talk) 13:20, 7 February 2010 (UTC)
 * I either use the Myanmar character picker and click the ZWSP button, or I add them in Microsoft Word with 200B + Alt-X. They're definitely there in Word, because they give line breaks there. I can't imagine why they get lost when I copy and paste the text from Word to the edit box, though I have encountered other weirdnesses when doing that. (For example, if you type the raised dot · (U+00B7) in Word and then copy and paste it into a text box at a Wikimedia project, it gets converted to the bullet • (U+2022).) Anyway, I hadn't thought of forcing them by means of their HTML entities, but that seems to work, so I'll do that from now on. Thanks, Paradoctor! Angr 14:48, 7 February 2010 (UTC)


 * Word? Maybe the free alternative gives you proper respect. ^_^ Seriously, that is either a function of Word that converts text for export to the clipboard, or your browser does not fetch the proper version from the clipboard. Anyway, the Myanmar character picker works for me in Firefox v3.6. Happy editing! Paradoctor (talk) 15:27, 7 February 2010 (UTC)
 * Word just seems evil when it comes to paste, it just collects lint! Can I recommend either FileFormat.info or the Firefox plugin International Sideboard. billinghurst  sDrewth  16:28, 7 February 2010 (UTC)
 * Well, Word is easiest for me, probably simply because I know it best. I'm familiar with most of its bugs and know how to get around them. But your answers here were helpful, and helped me come up with the solution to another problem I'd been having with Burmese text, one that isn't Word's fault and one that (to judge from pages I've seen at the Burmese Wikipedia) happens to everyone: Whenever both MYANMAR SIGN ASAT (U+103A) and MYANMAR SIGN DOT BELOW (U+1037) are supposed to occur on the same character, they have to occur in that order (first 103A and then 1037), otherwise they don't display correctly. But when you type them in the correct order in the text box, they get saved in the opposite order, and then don't display right. But if you use the HTML entity &amp;#x1037; (or &amp;#4151;) for DOT BELOW, then they don't switch and do display right. Angr 16:58, 7 February 2010 (UTC)

Wrong email subject in deletion notification
Yesterday i requested the deletion of two pages that i created by mistake. Hesperian deleted them and about each of them i received an email with the subject:
 * "Wikisource page TITLE has been created by Hesperian"

I would expect that it would say "deleted" instead of "Created". --Amir E. Aharoni (talk) 08:55, 8 February 2010 (UTC)
 * This is a known bug, see this Commons discussion. Paradoctor (talk) 09:59, 8 February 2010 (UTC)

Nada Bindu Upanishad
Hello, I am endeavouring to make an interwiki at the abovementioned Wikipedia article for the corresponding primary source at Wikisource, नाद-बिन्दु_उपनिषद् but for some reason it is not working. I would appreciate assistance in troubleshooting. Blessings and godspeed B9hummingbirdhoverin'chittychat 10:21, 8 February 2010 (UTC)


 * Done. You have to use the ludicrous interwiki code oldwikisource. Hesperian 11:40, 8 February 2010 (UTC)

Strange
Since updating my monobook.js this morning, on pressing show preview after an edit, the edit box appeared completely blanked. When I pressed save it not surprisingly saved a blank edit. I tried blanking monobook.js and it seems to have stopped doing this, but because the behaviour is intermittent I'm unsure if its fixed. Maybe its got nothing to do with monobook.js, but it does seem a coincidence. In the course of typing this message its happened to me twice. Just wondered if anyone else has seen this. Moondyne (talk) 08:29, 9 February 2010 (UTC)
 * It seems to have sorted itself out now. Moondyne (talk) 12:44, 9 February 2010 (UTC)
 * Were you using Beta? The most recent code update was very buggy, and had to be rolled back. Hesperian 23:22, 9 February 2010 (UTC)
 * Beta what? Please explain.  Its just done it again.  Moondyne (talk) 00:23, 10 February 2010 (UTC)
 * Up the very top of the page, next to the link to your user page, it either says "Try Beta" or "Leave Beta". If the latter, then click it. Hesperian 00:30, 10 February 2010 (UTC)
 * (FWIW, there's nothing in your monobook.js that could be causing this. Hesperian 00:32, 10 February 2010 (UTC))
 * Mine says Try Beta. Perhaps its a pc-specific issue, seeing it was working OK at home last night.  I'll switch over to IE for a while to see.  Am currently on FF3.6.  Moondyne (talk) 00:40, 10 February 2010 (UTC)
 * Me too; and no trouble here. Hesperian 00:47, 10 February 2010 (UTC)
 * Still broken. I'm stumped.  Moondyne (talk) 03:02, 10 February 2010 (UTC)
 * You might try opening Tools | Error Console. Probably too many errors in there to be useful for diagnosis, but if you clear it, then keep an eye on it as you browse, maybe it will spit out an enlightening error at the moment of failure. Hesperian 03:06, 10 February 2010 (UTC)
 * Seems to be fixed now. I'd also had FF lock up a few times so I reinstalled that and both problems seem to have now gone away.  I have no idea what the specific cause was.  Moondyne (talk) 14:38, 11 February 2010 (UTC)

Formatting for a book's index
Howdy! I'm fiddling with The Craftsmanship of Writing's index, and I've been using the Running Header template to space each line so that it passes basic usability standards. But I know that's not the original intention of the Running Header template. So my question is: Is there a template for books' indexes? If not, should I try to whip one up based on the Running Header template? Thanks! Clifflandis (talk) 17:53, 9 February 2010 (UTC)


 * Hi. THIS style may be help but there may be better ideas.— Ineuw (talk) 17:58, 9 February 2010 (UTC)


 * This one uses Dotted TOC page listing to make nice with all those dots. Paradoctor (talk) 22:19, 9 February 2010 (UTC)

Quite an interesting layout there, the way the page numbers are right-aligned, and intrude to the left only as far as the entry, at which point they wrap and the entry doesn't. I've made you a template and deployed it several times at the page linked above. If you want the dots too, maybe you can ask Paradoctor to tweak the template for you. Hesperian 01:07, 10 February 2010 (UTC)
 * Nothing maybe, just poke me. ;) Paradoctor (talk) 07:07, 10 February 2010 (UTC)
 * With the index
 * we'll need to see how it transcludes as it has a 100% width
 * as many indexes are going to be similar, it would be nice to see us look to a generic code for indices, this could either be as templates or as a class, consideration may need to be given to whether they are a 1 or 2 column index, and then even whether we wish to maintain that or not.
 * with respect to tables and class, I have put a few components into common.css, though still need to document, and need Jack Merridew to his special sanity checks. I also have a few shortcut templates for code t/ar, t/c2 and t/c3
 * personally, I don't see the need to necessarily replicate dot leaders as sometimes they just make the page too busy, and become a lot less relevant in an electronic format.
 * billinghurst  sDrewth  12:10, 10 February 2010 (UTC)


 * &rarr; Nevertheless, there are instances where certain collections use this 'wrapping-dots'-type of TOC for just about everything. Well, I know I could find use for it if it worked as it should George Orwell III (talk) 08:15, 11 February 2010 (UTC)


 * I'll see what I can come up with. Paradoctor (talk) 10:22, 11 February 2010 (UTC)

Thanks everyone! This answered my ultimate question: whether there are templates just for books' indices to choose from. I agree with billinghurst that it would be useful to create generic index templates, as most indices are similar. I created the Category:Index templates to collect templates for formatting indices.Clifflandis (talk) 21:22, 10 February 2010 (UTC)


 * Isn't this already covered by Category:TOC templates? Paradoctor (talk) 21:54, 10 February 2010 (UTC)
 * To be honest, I'm not really sure. It probably is.  Since there's not a lot of documentation for templates, I'm not really sure which templates are used for what, how they work, or where they go.  So being relatively new I'm somewhat lost trying to figure things out.  But so far folks have been great at letting me know gently where I screw up, and fixing stuff behind me. -- Clifflandis (talk) 22:14, 10 February 2010 (UTC)
 * That's the wiki way. You're doing fine in my book. ;) As regards the two categories, I note that neither category explicitly states its purpose, so this looks like a good time to talk about how to use them. Paradoctor (talk) 23:01, 10 February 2010 (UTC)

TOC... is for various Table of Contents, either within a work, or within the working space. Primarily that is to provide special linking. It would appear to belong in Category:Specific article templates in its current form, and creating another Index category is probably less successful. Oh and Category:Specific navigation templates

FWIW Wikisourcians are crap at documenting. No independent thought here! billinghurst  sDrewth  12:35, 11 February 2010 (UTC)

Browser survey
Which browser and version do you use (most of the time)?
 * F/Fox 3.6. Moondyne (talk) 03:05, 10 February 2010 (UTC)
 * FF 3.6 — Ineuw (talk) 03:23, 10 February 2010 (UTC)
 * Firefox v3.6 Paradoctor (talk) 07:07, 10 February 2010 (UTC)
 * Most recent FF and Chrome. Dovi (talk) 07:47, 10 February 2010 (UTC)
 * Firefox 3.6. Charles Matthews (talk) 12:25, 10 February 2010 (UTC)
 * Firefox 3.6 --Angelprincess72 (talk) 17:27, 10 February 2010 (UTC)
 * Firefox 3.6. Otherwise, Chrome. Blurpeace  05:24, 11 February 2010 (UTC)
 * Chrome 4.0—Zhaladshar (Talk) 13:42, 11 February 2010 (UTC)
 * Firefox (most current) as User:Jeepday (love that spell check). JeepdaySock (talk) 16:26, 11 February 2010 (UTC)
 * Firefox (various versions from 3.0.5 up) -- Clifflandis (talk) 17:14, 11 February 2010 (UTC)
 * I like DosLynx, but it is a bit of a memory hog. NetTamer is better on older boxes. Hesperian 02:19, 12 February 2010 (UTC)
 * IE v6.0 mostly. Occasionally FF 3.5 or Safari 4. Depends which computer I'm on. Beeswaxcandle (talk) 07:02, 12 February 2010 (UTC)
 * Firefox 3.5.7 Mike Peel (talk) 09:08, 12 February 2010 (UTC)
 * Firefox (various versions depending on which computer I work with). --Zyephyrus (talk) 14:14, 12 February 2010 (UTC)
 * FF 3.0.17 Sherurcij Collaboration of the Week: Author:Thomas Carlyle. 18:36, 12 February 2010 (UTC)
 * FF 2.0.0.14 on Linux. As I grow older, I've grown lazy about updating my software. :-/ Llywrch (talk) 23:35, 12 February 2010 (UTC)
 * FF 3.6 --kathleen wright5 (talk) 12:01, 13 February 2010 (UTC)
 * FF 3.57--Birgitte SB  16:56, 13 February 2010 (UTC)

Public Domain Question
Based on this: http://en.wikipedia.org/wiki/List_of_countries%27_copyright_length_based_on_publication_and_creation_dates I assume that Yakov Perleman, a Russian who died in 1942, would have the copyrights on his works expire in 1992, meaning all the work published during his life would now be in the public domain. Can someone else confirm this? Basically, I am mostly wondering if it would be permissible to post one of his books/translation to Wikisource, and I want to make sure this is safe. Thanks. Minute Lake (talk) 07:34, 12 February 2010 (UTC)
 * So, according to that page, Russia was Life + 50 years for deaths before 1953, so as you say his works went out of copyright in Russia in 1992. Works that were PD in the home country before 1996 are supposed to be PD in the US too. So, looks fine to me. Mike Peel (talk) 09:07, 12 February 2010 (UTC)
 * As we are talking about a translation, that adds a level of complexity. The translation usually would be considered to have a copyright of its own, so we would need to know
 * when the translation was published;
 * who did the translation; and
 * the translator's date death
 * to give more authoritative commentary on whether we could house the work. billinghurst  sDrewth  11:36, 12 February 2010 (UTC)
 * Well, I wanted to do a translation of one of his works myself, so I think the translation will be fine if the source text is fine. Thanks for your help! Minute Lake (talk) 23:04, 12 February 2010 (UTC)
 * Excellent, then I will just point you towards Translations — billinghurst  sDrewth  05:07, 13 February 2010 (UTC)

(Qn. for community) Dealing with book indices
Some have drawn my attention to book indices, and asked me to consider how we utilise them.

The issues are billinghurst  sDrewth  04:48, 13 February 2010 (UTC)
 * Page numbers: Linking for the index to the page in the Page: namespace is reasonably trivial in most books, though start to complicate when page numbering is not with a consistent offset. Of course, any Page ns only linking is not exactly that beneficial.
 * Page numbers: Linking for the index in the main ns, is not so straightforward as one needs to know the relevant chapter. So less easy to botify (which is limited anyway), and as a regex has a level of complexity, and the code is not exactly pretty when viewed.
 * Wikilinks: For the text component, Do we link it? If so, where to, to author namespace, back to the book, to offsite? Do text links, if done, just belong in the body of a work?
 * /me throws in Page:The Craftsmanship of Writing.djvu/283 for an example. billinghurst  sDrewth 


 * I think we need a Wikisource Glossary, but I will attempt to describe my approach. Linking from an index requires - Title/subpage#[anchor]|[text] - the 'anchor' and the 'text' are the page number given in the scan and adjusted at our Index. If I find an explicit internal reference in a work, I create an anchor to link it; this is not required with a work's index because the link to the Page:namespace creates an anchor in mainspace - the actual page number. Sure keeps things simple, any offset and other palaver in fixed in place and the rest looks after itself. Creating these links within a work, in mainspace, is marvellous and well worth the labour and bits of code. When these 'original navigational pages' are transcluded into their Index's page they provide a guide to the existing and paginated links. A long way of describing something that becomes a simple way of enhancing a work in mainspace, I see many advantages to readers and contributors and little possibility of confusion or misdirection. I don't see how cross namepace links are useful for these pages when they are:
 * Page: -> Page:
 * Index: -> Page: (tranclusion)
 * Main: -> Page: (makes me dizzy too)

What is the advantage of adding namespace switching links in the Page:namespace? Cygnis insignis (talk) 08:00, 13 February 2010 (UTC)


 * I have yet to make my first substantial contribution requiring such considerations, so this is a question from ignorance: What is the disadvantage? In general, I tend to think that more links are a good thing. Paradoctor (talk) 09:22, 13 February 2010 (UTC)


 * Cyg, I think you misinterpreted. I am not talking the Index: namespace, I am talking the book index in the Page: namespace (as per the link and this). Remember those quaint things at the end of the book if you were lucky?!?  billinghurst  sDrewth  09:55, 13 February 2010 (UTC)    So we have
 * Arnold, Matthew, 	54, 253, 254
 * Arnold, Matthew should link where, if anywhere. In the main ns, and in the Page ns, same links or different?
 * 54, 253, 254. In the Page namespace, these link to the relevant Page: page, and is easy. At the moment in the main ns, they are not linked, though probably should be, there is a whole lot of them, and it is not quick or simple task.
 * I endeavoured to clarify the concepts surrounding this finer issue, which is seeking a solution to a problem I don't recognise without a suggestion of advantage. This is what I did as a response: I opened two pages in main space, the Index subpage and the Title page with the contents. With these I was able to give the example of what I build by copy and pasting and put into page space:
 * Which produces:
 * Arnold, Matthew, 54, 253, 254
 * I see no reason to link anything other than the page number in the original, why does the template link to the Page:namespace at all? In response to the notion of 'why not', I have already suggested confusion and misdirection being undesirable properties, I am extremely wary of anything that makes contributing to this site even more arcane. Cygnis insignis (talk) 11:09, 13 February 2010 (UTC)
 * I see no reason to link anything other than the page number in the original, why does the template link to the Page:namespace at all? In response to the notion of 'why not', I have already suggested confusion and misdirection being undesirable properties, I am extremely wary of anything that makes contributing to this site even more arcane. Cygnis insignis (talk) 11:09, 13 February 2010 (UTC)

I agree 100% 99% with Cygnis insignis. The main namespace is what our readers read. All our efforts should be directed at getting the main namespace right, so as to optimise the readers' experience. The Page: namespace is our editorial space, where we write, proof and validate works. No-one needs to be there unless they are contributing. Keeping the two namespaces as separate as possible is a fundamental design principle. Links that take the unsuspecting reader from the readers' view into our editorial namespace are Evil. To address Billinghurst's example, I put it to you all that any reader who went to that index in search of references to Adam's Diary, would expect that "51" to take them to The Craftsmanship of Writing/The Power Of Self-Criticism, and would be utterly bamboozled when they instead find themselves at Page:The Craftsmanship of Writing.djvu/65. Hesperian 11:36, 13 February 2010 (UTC)


 * DYK ... that 99% of comments quoting "99%" are inaccurate? Cygnis insignis (talk) 12:16, 13 February 2010 (UTC)
 * DYK... that 86% of all quoted statistics are made up on the spot? Hesperian 12:40, 13 February 2010 (UTC)

DYK (ec is very frustrating)


 * I am not advocating anything, I am here for the discussion, heck I brought it here in an open manner and I said that page to page was not beneficial in my opening dot point. There is only one page with the page numbers entered, and that was done in discussion with someone in IRC, so no whipping the messenger with bookmarks. Sounds like you have Natanui fixation. ;-)


 * Page numbers
 * Page -> Page is just that, no user in the main ns space sees it, I just commented that it was easiest of the lot, rueful comment. I was not advocating them, they did allow me to demonstrate something, and they do allow one to x-ref that the indexing is correct, easily (so was useful for maintenance reasons). I also said that they didn't transclude through to main ns.


 * Main -> Main, yes, it can be done, it never was doubted, it was the difficulty factor that I was addressing for an extended series of numbers. In this work it may be feasible to do manually, for 20pp like this I think not manual, though it is of even more value than the example that I chose.


 * Text links Main ns: index pages
 * Link or not link? If link, to where? To the page of the link seems superfluous if the page numbers are linked. To author/existing works pages may be confusing, though beneficial; to external pages has a detrimental feel.

billinghurst  sDrewth  12:50, 13 February 2010 (UTC)


 * Ugh, sorry, did I go overboard with the rhetoric? I wasn't waterboarding bookmarking you; just giving my opinion on a matter I feel fairly strongly about. Difficulties in mainspace linking can be overcome. I did all these by wrapping every page number in . Granted User:Hesperian/mbb took me over an hour to get right, but it was fun rather than tedious, and it made linking all those page number a cinch. Hesperian 13:04, 13 February 2010 (UTC)


 * Hesperian, the reason why I brought the matter here, at this time, was that I was about to do that to the work for the numbering. However, I decided that I wanted to organise something that could be reproducible for other works, AND then maybe in a more simplistic form that can be used who are not regex-ge+ks. I doubt that this is a specialty service that we are going to want to offer. I pounced on Pathoschild in IRC, and discussed the matter with him, I even played the universal "take pity on the weak and uncoded" card. As I thought it through, all relevant linking seemed relevant to address in this forum seeing there were a variety of different approaches. Maybe that is a context that I should have expressed earlier, however it seemed to be about the background and process, and it was just going to muddy the water, ie. low KISS rating. Getting the community to address so that we are all singing from the same index page in the hymnbook. billinghurst  sDrewth  13:47, 13 February 2010 (UTC)


 * I'm not sure there is a simple solution. One option is to create a linking template at The Craftsmanship of Writing/Page link, so that users can type and automagically get the link to the right chapter. But some geek still has to write the template. Another option is create a redirect from The Craftsmanship of Writing/Page 51 to The Craftsmanship of Writing/The Power Of Self-Criticism... and then more redirects, from every other page. It would be pretty easy to write a bot that looks at the page ranges that each namespace section covers, and creates the necessary redirects; and it would trivialise page linking. But some geek still has to write the bot; and it would pollute our nice clean universe. I fear I'm still missing your point, and hijacking the conversation. In response to your original post, I acknowledge that it can be horribly tedious/complicated to link page numbers to the mainspace location where the page appears, but I reckon it should still be done. Hesperian 14:12, 13 February 2010 (UTC)


 * Regarding the page numbers, my preference is along the lines of linking to The Craftsmanship of Writing/The Power Of Self-Criticism. Regarding the alpha list I would link within Wikisource primarily and to Wikipedia secondarily mirroring how the item is linked within the text.  I don't think we should give a different link in the text than we do in the index.  Although I think we might sometimes refrain from linking in the text while offering a link in the index. --Birgitte SB  17:21, 13 February 2010 (UTC)
 * I question when it is appropriate to synthesise new purposes for existing content. Wouldn't that make the navigational function of a document's internal directory, a source's explicit reference to a subject, a little fuzzy ... Cygnis insignis (talk) 20:06, 13 February 2010 (UTC)
 * I don't think people would be at all fuzzy about the page numbers being an internal navigation as they always have and the alpha stuff which was never the internal navigation not being an internal navigation. You generally question whether linking in texts is appropriate any where, any time, any how.  So I have trouble regarding your objection as addressable as I have already realized I need to agree to disagree with you on the issue.--Birgitte SB  21:41, 13 February 2010 (UTC)

I have been using the "fools rush in" approach to "enhancing" the Navigation for the Dictionary of National Biography, 1885-1900. Our problem is that we are a small project team with a huge project: 63 volumes, 29,000+ articles, 25,000+ pages. We are working on articles in order of importance rather than front-to-back. But what should we tell a reader to do if we have not yet transcribed the article the reader is looking for? Well, I unilaterally decided to pierce the namespace boundary by providing a "source cescription page" for each volume. This page tells the reader (briefly) what to expect and how to navigate that volume in page space. This violates the pure distinction between mainspace (for readers) and page and project spaces (for editors,) but it was the least bad solution I could come up with. For smaller works, the time to completion is small enough that we can (implicitly) ask the user to just wait until we are done, but for this project, Wikisource represents the only coherent, organized way to access the scans. I wanted to give a user who had already navigated to the correct volume in mainspace, some help, rather than an (implicit) "tough luck: you are on your own." So, my postion is: Yes, page space is generally not for readers, but sometimes we need to make exceptions. -Arch dude (talk) 00:14, 14 February 2010 (UTC)


 * I attempted to modify Hesperian's template so it would take an input page such as User:Eliyak/test/2 (notice the "section" tags) and use this as a readable TOC. Unfortunately, at some point Mediawiki seems to have changed so that template looping (through redirects) no longer works. This also disables my beloved Mishnah Tractate (it used to generate pages like this one)! Does anyone know how to get User:Eliyak/test/1 working? I understand the desire to avoid infinite loops, but surely the transclusion limits should suffice? --Eliyak T · C 16:37, 14 February 2010 (UTC)


 * Gosh, that's ambitious!
 * It seems template recursion is dead or disabled. There's a simple test case at User:Hesperian/Recursion. It should print
 * Recursing... recursed.
 * What it actually prints is
 * Oddly enough it does work at one location: User:Hesperian/Recursion itself. Hesperian 23:43, 14 February 2010 (UTC)
 * Hesperian 23:43, 14 February 2010 (UTC)
 * ... oh, I see. Recursion never worked. You found a loophole by recursing via redirects, but it has been closed. As for how to make the above work, you can fake recursion by deciding upon the maximum depth you'll need to go, then posting the same template at User:Eliyak/test/1, User:Eliyak/test/2, User:Eliyak/test/3, User:Eliyak/test/4, and so on down to whatever depth you need. Then make User:Eliyak/test/n invoke User:Eliyak/test/n+1 instead of itself.
 * ... oh, I see. Recursion never worked. You found a loophole by recursing via redirects, but it has been closed. As for how to make the above work, you can fake recursion by deciding upon the maximum depth you'll need to go, then posting the same template at User:Eliyak/test/1, User:Eliyak/test/2, User:Eliyak/test/3, User:Eliyak/test/4, and so on down to whatever depth you need. Then make User:Eliyak/test/n invoke User:Eliyak/test/n+1 instead of itself.

Boo to closed loopholes. Actually, now that I think about it, one page should be all that is necessary, but the question is how many chapters to allow for. I suppose a few dozen should do it. --Eliyak T · C 03:34, 15 February 2010 (UTC)


 * A few dozen? Ahem. Hesperian 03:59, 15 February 2010 (UTC)
 * Or should I say Ahem. Hesperian 04:00, 15 February 2010 (UTC)
 * This is fun; you should try it. :-D Ahem. Hesperian 04:02, 15 February 2010 (UTC)
 * I think this is my last Ahem. Hesperian 04:03, 15 February 2010 (UTC)

I've created TOCIndex, which creates a section-ed off TOC page, and pg, which uses that TOC page to create page links. They go up to 200 chapters. (You don't need to take that as a challenge, Hesperian!) They were a fun challenge to make, and I hope someone finds them useful. --Eliyak T · C 08:12, 22 February 2010 (UTC)


 * Thanks, though it doesn't make sense to me at this hour of the day. I merged the two parts of documentation so that they can be seen side by side. — billinghurst  sDrewth  14:41, 22 February 2010 (UTC)

Congratulations to the WS:PotMers
The Proofread of the Month has been completed in just 12 days. Congratulations to all involved! In celebration, I have picked another work to be undertaken. With Mike Peel having done WMF proud recently in the Britain loves Wikipedia - British photograph bonanza, I picked a work that is dear to his heart and out of this world. Again congratulations to the enthusiastic participants! billinghurst  sDrewth  04:58, 13 February 2010 (UTC)

Nasa Documents
Would it be appropriate to upload NASA factsheets/guides to wikisource? For example Solar Dynamics Observatory Fact Sheet?Smallman12q (talk) 14:27, 13 February 2010 (UTC)
 * The base principles are published or historical documents. After that we generally don't prejudge works and we would ask people to justify that they meet the criteria. So I will would point you towards What Wikisource includes and also to WS:IO.  If you think that you can justify it after looking at those documents, then they are probably cool, and less likely to get much argument from us. If you think that it is borderline, then bring that conversation here about what you see as the potential issues.  We try to be principle-based, and flexible on rules if there is a justification for such.
 * Now my personal opinion. Here are some of my *relevant* internal rules of thumb for web documents
 * is it public domain?
 * is it an authoritative document? (referenced / peer-reviewed / undeniable authority)
 * is it a dynamic document? (if yes, it probably isn't right for WS yet, eg. FAQs ...)
 * is it fixed in a point and time? (then it meets my first criteria of being published, even if it updates at a later fixed point and time, eg. legislation; drill down further)
 * hope that helps for how I approach it. billinghurst  sDrewth  05:42, 14 February 2010 (UTC)
 * Personally, I would say that NASA factsheets/guides tend to be fairly authoritative (at least when it comes to communicating the info to the public), and in the public domain (work of a US governmental body), so I wouldn't envisage a problem with ones such as the linked-to fact sheet. Mike Peel (talk) 10:51, 14 February 2010 (UTC)
 * Yup, pretty much everything from NASA is public domain as long as you remove their logo (which they do defend) Sherurcij Collaboration of the Week: Author:Thomas Carlyle. 13:08, 14 February 2010 (UTC)

Looking for portraits
I'm looking for original portraits of the Irish Edgeworth family: namely Francis Ysidro, Henry Essex and Maria.
 * User Billingshurst had an entry.
 * My eaddress is tilleard@madasafish.com

If you mean portraits as images, I found these two. We also have literary portraits. A third: File:HenryEssexEdgeworth.jpg
 * File:Edgeworth.jpeg
 * File:Maria Edgeworth.jpg


 * It is 455th on my to do list to get portraits or images, either from Wikipedia over to Commons, or from books here or archive.org into place. Check Wikipedia at there still many images there that need to be moved to Commons so that we can use them here. Plus check frontispiece of books by or about the authors in works at archive.org. — billinghurst  sDrewth  04:53, 16 February 2010 (UTC)

Call for proposals for Wikimedia UK initiatives
Hi all. The Wikimedia UK board has been putting together a budget for the next year (You can see this, and help with its development, here) and we have some money left over. We are looking for proposals for projects/iniatives with budget requirements in the range of £100-£3000 (GBP). These projects can be either online or offline, but they should be primarily focused on the UK and they must further the objects of Wikimedia UK (broadly, to collate/develop/spread freely licensed material).

The deadline for proposals is the end of this month (i.e. 0:00 UTC on 1 March 2010). You can find more details of the requirements, and how to submit proposals, on our blog. Thanks. Mike Peel (talk) 23:02, 17 February 2010 (UTC)

Preferring image files and a better system for imported Gutenberg texts and have images
Not a whinge, more wanting community involvement on suggestions with how we manage random users pasting Gutenberg texts into place. Today, all 480kb+ of Battery and the Boiler arrived on our doorstep. It has all the hallmarks of a GB text. Having a look at archive.org a djvu file is here, which makes this a candidate for a Match and Split.

We are pretty much saying that one of our features/strengths is that we provide linked text and images. So we need a means to identify files that originated from GB or equivalent and we know that the image file exists.

I would like to see us now start to better reflect in all of the relevant pages that we have a hearty preference for images to go with the text.

Note: we have migrate to djvu which we use once the file and Index: page are in situ, however not set up for absent file. I am wondering whether we should create a separate template, or modify this template to identify that files have corresponding image files and we should be looking to source them for Commons. — billinghurst  sDrewth  12:52, 18 February 2010 (UTC) (that may not make any sense at all with this space brain).
 * Could identifying plaintext sources be done reliably by script? Finding corresponding page scans from different sources like the GB-Archive.org example you provided would have to be done manually it seems, but if long text comparison isn't too complicated it seems like matching texts and tagging the source, and if they're hosted by GB, the pagescans, could be bot work. Also more generally on the subject of texts of unknown provenance, we have a Template:No source, which, bizarrely, dumps texts into the Category:Doubtful fidelity, shared by Template:Fidelity (should that be changed?). Would it make sense to have a bot mass-apply it to texts in the first level of the main namespace which aren't transcluding pages or don't possess a Template:Edition? Things like poems which are part of a collection but still considered independent works don't really need it, I guess, but they don't not need it either, so it doesn't seem that problematic. There's probably something I'm not taking into account, though. Lastly, regarding the proposed template, it seems to me a new one should be made, since they're separate steps in the process and require different actions. Prosody (talk) 05:05, 21 February 2010 (UTC)
 * Ack, categories are case sensitive. It should read Category:Doubtful Fidelity. And I should be previewing comments before posting them. Prosody (talk) 05:06, 21 February 2010 (UTC)

About illustrations
Would it be possible to offer the reader a choice between having the text with illustrations or the text without illustrations? I mean in the Main space. The idea is that they may keep the illustrations when they like these particular images but that they aren't obliged to have them mixed with the text when these images happen to spoil the text if hurting their taste.--Zyephyrus (talk) 08:10, 19 February 2010 (UTC)


 * Yes. Give me an example, and I'll show you how. Paradoctor (talk) 09:37, 19 February 2010 (UTC)


 * Could we look to make this a global option, and stick it into special:mypreferences ? That said, if we did anything like that, I would want a big statement in the NOTES that specifically stated that YOU HAVE IMAGES TURNED OFF. — billinghurst  sDrewth  12:43, 19 February 2010 (UTC)


 * This could cause big problems in places where images are meant to be read as part of the text; e.g. ornate initials such as at Page:The songs of a sentimental bloke (1917).djvu/21. Not displaying images is browser functionality, or at least browser extension functionality. I don't think we should be offering it server-side. Hesperian 12:56, 19 February 2010 (UTC)


 * (edit conflict) We could make it a dynamic toggle, too. The important question is: How can the images be identified? If merely switching CSS class "image" is sufficient, then it's almost trivial. Just add  to the custom CSS for your skin, and you're done. Toggling and an additional warning requires a bit of JavaScript. I'm pretty sure that's too blunt a tool, though, and the result is ugly. Paradoctor (talk) 13:08, 19 February 2010 (UTC)


 * I agree that sometimes the illustrations may be part of a work; what I thought of first was when an artist has illustrated a text after the author's death; the illustration may be liked or not, and some readers may prefer to read the text only. I didn't mean all the pictures, I meant the choice of seeing or not seeing some pictures. --Zyephyrus (talk) 20:21, 19 February 2010 (UTC)

Wikisource:Possible copyright violations
I did some archiving at Possible copyright violations I left three that look like candidates for WikiLivres before they are deleted. As I am working from the US I hesitate to post them on WikiLivres myself. Would some one like to move them, and next time I sign on I will delete and archive here. Jeepday (talk) 14:05, 20 February 2010 (UTC)
 * Usually Jusjih is the champion there. I have dropped a note on their talk page. — billinghurst  sDrewth  21:44, 20 February 2010 (UTC)
 * Exported the following seven pages to Canadian Wikilivres and deleted them here:


 * 1) The Shape of Things to Come
 * 2) Talk:The Shape of Things to Come
 * 3) H.G. Wells on the Polish Corridor
 * 4) The Earliest Latin Commentaries on the Epistles of St. Paul
 * 5) Talk:The Earliest Latin Commentaries on the Epistles of St. Paul
 * 6) The Earliest Latin Commentaries on the Epistles of St. Paul/Pelagius
 * 7) A Nice Cup of Tea
 * I have been very busy recently. Thanks for reminding as my email tells me if anyone posts new message to my talk page here. By the way, I would also like to remind the prolonged backlog of Possible_copyright_violations/Special_discussion_for_pages_tagged_as_PD-manifesto. If no valid comments to send them to Canadian Wikilivres with Manifesto, I may want to delete them without exporting or further warning. We cannot let them hanging around forever.--Jusjih (talk) 02:26, 21 February 2010 (UTC)

More Aggressive stance?
Possible copyright violations says To protect the legal interests of the Wikimedia Foundation, these will be deleted unless there are strong reasons to keep them within at least two weeks We have not been doing this, with several  months passing before a decision is made. Is it time to be a little more aggressive in the deletion of suspected copyright violations or should we change the header at Possible copyright violations. JeepdaySock (talk) 17:39, 22 February 2010 (UTC)

Listing of Categories
Is it possible to create pages, or extract a list, of the Topics by subject category tree, which when copied and pasted into a text file, retains the tree structure indentation? I opened the tree branches and pasted it into a text file, but it looses the indentation of the subcategories. — Ineuw (talk) 18:06, 20 February 2010 (UTC)
 * Some years ago I used PrintScreen and had them as images on fr.ws like this ; but perhaps there are other possibilities now? --Zyephyrus (talk) 09:24, 22 February 2010 (UTC)

Avadhuta Gita
I need some assistance with identifying and specifying a license for the abovementioned item when someone has time. The Sanskrit as Devanagari is in the public domain. I generated the Roman transliteration with diacritics. The English translation and commentary is based on a version that was fully released into the public domain in 2006 with no reservations. I would like to give attribution but am not sure how to do so. I will be building and altering slowly all of the English translation and commentary as I progress with the verse by verse grammatical analysis and parsing. Indeed a number of the verses with commentary are already my work. Thanx B9hummingbirdhoverin'chittychat 03:03, 21 February 2010 (UTC)

Concerning religous texts.
The Vedas(particularly the Atharva Veda) and Talmud haven't been completely add yet. Someone should use the link to finish the Atharva Veda. —Preceding unsigned comment added by 71.171.130.187 (talk) 01:29, 21 February 2010


 * To broaden this, Wikipedia has a thousand articles which are strongly recommended for every Wikipedia to produce. There could be a good WikiProject in something analogous--identifying and getting significant works which we don't yet have. Prosody (talk) 07:10, 21 February 2010 (UTC)

Featured Texts and Match & Split
The Time Machine was a featured text, though it is a text without a images to support the text. We now have Index:The Time Machine.djvu and we could Match and Split. Doing this, however, will mean that we would need to return to undertake another proofread. I am not about to undertake such a task without bringing that to the community for their decision. — billinghurst  sDrewth  13:55, 21 February 2010 (UTC)
 * I say we do it. We've already featured it, so I think we should bring it up to our current standards.  This wouldn't be any kind of a formal proofread effort, would it?  (Like, this wouldn't be a PotM?)—Zhaladshar (Talk) 14:20, 21 February 2010 (UTC)
 * Different editions complicate things, PG gives "1898" without indicating whether they mean their version is the first edition. The djvu is an American 1920s edition, the intro is a bit vague about the additional text mentioned at the Time Machine. The first lines show significant differences,


 * Submitting the djvu edition to FT would be bringing both up to current standards of versions and page scans. Getting the early FT linked to first editions would also be desirable, passing it by FTC again honours the original effort. Cygnis insignis (talk) 16:16, 21 February 2010 (UTC)
 * It was late, and I was tired, so it may have been more simplistic than I had intended. I would actually in this scenario suggest we attempt to hunt up and match the specific edition, however, that aside, I was more thinking of the general approach.— billinghurst  sDrewth  14:29, 22 February 2010 (UTC)

Need djvu help
I am trying to upload a United States Army Field manual from a PDF. Following the instructions at Help:DjVu files and Help:Beginner's guide to Index: files, I created a DjVu with Any2DjVu and uploaded it to commons. I created an index at Index:United States Army Field Manual 3-13 Information Operations.djvu.

What's next? Help:DjVu files says something about using User:JVbot to preload pages with the OCR text from Any2DjVu, but the bot appears to be inactive. Do I need to extract the text with DjVuLibre and then do it manually? Any help is appreciated.

Also, it appears that Help:Beginner's guide to Index: files is incomplete. The final steps says something about "below," but then the page ends. Is there another page somewhere that I am missing?

Thanks, Cerebellum (talk) 00:44, 26 February 2010 (UTC)


 * I added the so now the pages are available. I also purged the file at commons and there is no textlayer present so you may have to OCR each page manually as you go along proofreading the work, unless there is a active bot that can do it for you. --Xxagile (talk) 01:01, 26 February 2010 (UTC)
 * So that just means pushing the little OCR button on the toolbar, right? I don't need to use any external software?  --Cerebellum (talk) 01:29, 26 February 2010 (UTC)
 * Yes, just push the button. It might take a minute to load but it should be readable.--Xxagile (talk) 01:36, 26 February 2010 (UTC)
 * Works like a charm. Thanks for your help. Cerebellum (talk) 01:56, 26 February 2010 (UTC)
 * Okay, I'm having some troubles, see this pages for an example:Page:United States Army Field Manual 3-13 Information Operations.djvu/56. Are the results usually that garbled?  Cerebellum (talk) 04:28, 26 February 2010 (UTC)
 * Pretty much. Computers are not really very good at reading text.  More advanced (and expensive) software can do it better but some garbling is likely to take place.  The software is trying to (a) find, and then (b) recognise each individual letter.  It often falls down at both stages (for example, finding "in" as one letter and matching it to a preloaded pattern for "m").
 * Incidentally, the software is going to fail quite badly at the table you are about to reach. It may recognise the text but it will not format it correctly.  I did Page:United States Army Field Manual 3-13 Information Operations.djvu/65 to check things and just the sidebar caused some problems there.  The method I used was to save the garbled OCR text, remove the side bar, lay out basic formatting (such as headings), save again, correct and retype text as necessary, save again, enter the sidebar by hand and save one last time. - AdamBMorgan (talk) 13:29, 26 February 2010 (UTC)
 * Thanks, I was wondering what I was going to do about those inset sidebars. Wikicode to the rescue!  The tables will be pretty rough, but I have a little bit of experience with formatting tables so it won't be too bad.  Thanks again!  --Cerebellum (talk) 15:20, 26 February 2010 (UTC)

OK, I was able to avoid the OCR issue by using Adobe Reader to rip the text from the original work, but now I'm wondering if there's any way to use a bot to add the text to each page, or if I need to do it manually. Help:Match and Split says that the Page: pages already need to have a text layer for Match and Split, so I don't think that will work. Any ideas? --Cerebellum (talk) 20:55, 4 March 2010 (UTC)


 * Why aren't we converting the PDF file to a DjVu file and uploading it with its text layer already in place? To do that you can either upload to Archive.org and wait a few days until they convert it, or use http://any2djvu.djvuzone.org/.  For me, the former is more reliable.
 * Hey, sorry it took me so long to respond! The answer is that I converted the PDF to a DjVu with any2djvu, but for some reason the DjVu has no text layer.  Should I convert again and re-upload?  That might mess up the already-proofread pages, though. --Cerebellum (talk) 20:51, 6 March 2010 (UTC)
 * Hmm, then I would suggest uploading to archive.org. I just find it less problematic. — billinghurst  sDrewth  00:14, 7 March 2010 (UTC)
 * Sounds good. Thanks!  --Cerebellum (talk) 00:16, 8 March 2010 (UTC)

Tab order in Page: namespace
I would expect that the tab would jump from the editing field to the Summary field. It works this way in the main namespace, but in the Page: namespace the edit summary field is very far from the main editing field in the tab order.

Is it a known bug? --Amir E. Aharoni (talk) 14:19, 27 February 2010 (UTC)
 * This behaviour of tab is unexpected in the Page:namespace, here are the swings and roundabouts: hitting  in the edit box moves you to the drop-down palettes of special characters, to move to the edit summary use  . If you have the header and footer sections open, via the button [+] or your preferences [recommended], the tab keystrokes move you to those first - the summary field is reached by using   twice. Pretty sure this is universal, not browser/platform specific. Cygnis insignis (talk) 19:52, 27 February 2010 (UTC)
 * Thanks.
 * And is it possible to do anything about it? --Amir E. Aharoni (talk) 22:25, 27 February 2010 (UTC)
 * It will be a construct of the template put in place, hence would be occurring in all WS wikis. I would suggest that this should be a request to Wikisource talk:Proofread Page. — billinghurst  sDrewth  23:16, 27 February 2010 (UTC)

A table on two pages
Hi,

This is a question for the wiki table wizards.

Look at


 * Page:Gesenius' Hebrew Grammar (1910 Kautzsch-Cowley edition).djvu/532
 * Page:Gesenius' Hebrew Grammar (1910 Kautzsch-Cowley edition).djvu/533

In the original book this table is spread on two pages. It is already formatted well enough on separate pages here, but i would like to put it on a single page so the complete table would be seen without having to maintain two copies of the same information.

Is there an elegant way to do it? --Amir E. Aharoni (talk) 11:11, 28 February 2010 (UTC)


 * It can be done, but it is a bit yucky:
 * On the second page, wrap each individual line in section tags:

|-
 * thy
 * ,, in
 * thy.
 * On each line of the first page, import the rest of the line from the second page, but only when transcluding into the main space:
 * On each line of the first page, import the rest of the line from the second page, but only when transcluding into the main space:
 * On each line of the first page, import the rest of the line from the second page, but only when transcluding into the main space:

|-
 * Then, when you transclude into the mainspace, transclude only the first page. It will pull in the bits of the second page as required, and you will end up with a merged table.
 * Hesperian 11:28, 28 February 2010 (UTC)
 * I am not as competent or as patient as Hesperian, I would have just tried to transclude them side by side for those two pages. So inhale the pages before and after normally, and then looked to transclude the two pages of interest into a (2x1) table. That may take a little more fiddling on the table setup, however, makes the proofing easier. — billinghurst  sDrewth  13:11, 28 February 2010 (UTC)
 * That said, one cannot use "prettytable" and would need to be constantly aware, so Hesperian's methodology would seem to be easier with that beast. — billinghurst  sDrewth  13:22, 28 February 2010 (UTC)
 * Transclusion is a wonderful tool that provides two important benefits: (1)reduced initial effort, and (2) reduced effort to correct errors. But in this case, the effort to use transclusion is higher than the effort to simply cut and paste, so cut and paste is appropriate. Do, however, place big internal notes in both copies to indicate that corrections must be duplicated manually. There is a third possibility: If someone with advanced template-fu is willing to do so, a small set of templates can be created to solve the problem: one for table head, one for table body part and one for table end. Use these with the appropriate noinclude and include-only or sections. A fourth possibility is to do the tabes using HTML markup instead of wiki markup, and then use transclusion. -Arch dude (talk) 13:34, 28 February 2010 (UTC)
 * Then, when you transclude into the mainspace, transclude only the first page. It will pull in the bits of the second page as required, and you will end up with a merged table.
 * Hesperian 11:28, 28 February 2010 (UTC)
 * I am not as competent or as patient as Hesperian, I would have just tried to transclude them side by side for those two pages. So inhale the pages before and after normally, and then looked to transclude the two pages of interest into a (2x1) table. That may take a little more fiddling on the table setup, however, makes the proofing easier. — billinghurst  sDrewth  13:11, 28 February 2010 (UTC)
 * That said, one cannot use "prettytable" and would need to be constantly aware, so Hesperian's methodology would seem to be easier with that beast. — billinghurst  sDrewth  13:22, 28 February 2010 (UTC)
 * Transclusion is a wonderful tool that provides two important benefits: (1)reduced initial effort, and (2) reduced effort to correct errors. But in this case, the effort to use transclusion is higher than the effort to simply cut and paste, so cut and paste is appropriate. Do, however, place big internal notes in both copies to indicate that corrections must be duplicated manually. There is a third possibility: If someone with advanced template-fu is willing to do so, a small set of templates can be created to solve the problem: one for table head, one for table body part and one for table end. Use these with the appropriate noinclude and include-only or sections. A fourth possibility is to do the tabes using HTML markup instead of wiki markup, and then use transclusion. -Arch dude (talk) 13:34, 28 February 2010 (UTC)

Deprecated template blank line - notification
After a series of discussions and consultations in a variety of places (talk pages and irc; sometimes vigorous) and quite a few trials, I have been bold and deprecated blank line. For the background and the internal workings please see the templates mentioned, as I am trying to not get into too technical jargon here.

Blank line was set up as a means to circumvent the issue of the collapsing of whitespace, especially when adjoining pages were transcluded to the main namespace from the Page namespace where a paragraph terminated at the end of a page (ie., a new paragraph on a page). By trial we have determined that the addition of nop on a new line after the end of the paragraph (at the end of a page) serves the same purpose, is less intensive and looks neater.

Presuming that this is acceptable to the community, the way forward is 1) amend the instructions at H:SIDE, 2) change our practice, 3) look to whether it is necessary to systematically convert pages using the methodology to be converted or left as they are and let changes happen organically; and 4) if we undertake the previous point, then we can make blank line then be a redirect to nop or similar. — billinghurst  sDrewth  02:50, 2 March 2010 (UTC)

Help with Swedish pdf -> djvu
I would like to have some help with converting File:Så fungerar Wikipedia.pdf to a .djvu-file. I need no text-layer in the file, since the text can be found on a user-page on Swedish Wikipedia. (The text is written by the chairman of the Swedish WMF-chapter.)

Best regards! -- Lavallen (talk) 19:25, 2 March 2010 (UTC)


 * Convert it at http://any2djvu.djvuzone.org/ and don't OCR. If you already have the text, then just selectively copy and paste once you have it setup in the Page: namespace. The text layer is most useful when there is no alternate text, or it is a big work and we would look to use Match and Split. — billinghurst  sDrewth  00:01, 3 March 2010 (UTC)

Format/layout of scanned PD magazine stories & images/illustrations questions
If I've just failed to find the info in Help, I'd appreciate links and advice on the local vocabulary that will help me do better searches. I have spend a good deal of time looking before asking here.

Is keeping things like kerned center-justified double columns, smart quotes and ALL CAP first lines of stories considered important? I haven't even figured out how or if I can do this in Text Only. I do see that double columns could be important to integrate illustrations as in the original presentation if they are also verifiably public domain and it's possible to integrate them in original position and scale from WikiMedia Commons. More to learn than I'm likely to be willing to learn any time this year and for the story I'm doing now the two illustrations can be made to match original positions closely without using double columns. Frankly verifying PD for illustrations is often harder than stories for pre-1964 non-renawals anyway and the scan quality from old pulps and digests can be very disappointing. But these are intended as general questions. My feeling is that for the most part this sort of thing has nothing to do with the author, many PD stories were reprinted in collections or anthologies with different layouts and the truly important thing is to preserve the work and get the actual text of the story (or even book) correct. Is there a solid guideline on this?

Are original pagination, page numbers, footers and headers important? Again this is something that will vary by source for works that have been reprinted though I can see where page numbers for reference might be important even with works of fiction.

Are images for proof-reading necessary? I didn't save both images and text for current short story, don't want to put the magazine through the scanner more than necessary and it's not difficult to locate a copy of this story in reprint at the moment. In future I can remember to scan one page or pair of pages at a time and save as both image and OCR text output but I have limited image choices. Honestly, I don't want to mess around with all the .djue stuff that seems to involve intermediate software I also don't have and would need to find and learn how to use some freeware programs for. I'd be happier just posting .jpg images in Commons for proofing reference. 150 dpi or maybe 72 dpi should be adequate for that, 300 dpi definetely is. OTOH, I imagine there is some reason for the apparent .djve preference.

Most of what I think would be worth the trouble of scanning should have something for proofing reference since one of the top reasons for me scanning something would be that it is rare but still of contemporary interest, people should be able to see that what they are reading is correct and I'd be pleased if other people would do the hardcore proofing.

Advice on how to do this most efficiently would be welcome. And I might occasionally take requests from people willing to proof from a basically uncorrected scan. 1950s and 1960s SF magazines are what I have most of, also some mystery magazines and books from these genres and other genres or "mainstream" though tackling whole books sounds daunting. Some books, including short story collections or anthologies, might go back decades earlier.

Is the normal procedure to create sandboxes in the User Page and move or copy from there to somewhere else? Refrigerator Heaven (talk) 14:39, 3 March 2010 (UTC)
 * Hi! Wikisource is mainly about the textual content: the illustrations support it. Thus, pagination and columns are generally not mandated, an image positions are not a critical part of the work: the text integrity is. Having said that, having the illustrations on the page somewhere is highly encouraged, as they are often integral to the text. Generally speaking, the text will be in a single column, but it is up to the contributor's judgment (and is usually respected by others if it is a sensible choice). For example, Remarks_in_Recognition_of_the_30th_Anniversary_of_CORA is arranged as one column, but the original is newspaper-style in many narrow columns.
 * As for the JPG/DJVU thing, DJVU is preferred because it allows side-by-side proofreading and it allows automatic transfer of OCR data from the file. However, illustrations may well be badly degraded by the DJVU compression, so you may also want to upload the JPGs to Commons as well, so that full-quality pictures can be extracted. People with bots (me for example) can help you batch-upload large numbers of files if this will be a problem. If the text is just text however, the DJVU alone generally suffices. I can also create DJVU files out of sets of JPGs for you, or tell you how to do it yourself quickly and easily with a script.
 * Headers, footers, page numbers etc are added to the page-by-page proofread text, but not to the overall text. This is handled automatically by the side-by-side proofreading tool. For example, this page has the running header, but when the text is transferred to the main page the header is not taken, and the text is combined with the preceding and succeeding pages seamlessly.
 * I hope this helps, any more questions, ask, or come into the IRC channel and ask the friendly denizens for real-time help. Inductiveload (talk) 01:22, 4 March 2010 (UTC)

Possible Changes to Collaboration of the Week
A few people and I were discussing on IRC about changing the Collaboration of the Week (or CotW) to the Collaboration of the Month (or CotM). In addition, we are would like to change the focus from picking a single author each interval to authors, subjects or ideas (that was Billinghurst's idea). In addition, we need more ideas to bring new life into the project. So what do you guys think of these and any other ideas? --Mattwj2002 (talk) 00:29, 4 March 2010 (UTC)
 * I think it's a great idea to move away from the author per week format to a more flexible one. I think it would be good to occasionally have a focus on a specific maintenance action (e.g. fixing damaged or incomplete texts, tying up loose ends in a certain area etc.) as well.
 * One more thing about the past handling of the CotW is that when the Machiavelli one had no real contributions, it was left like that, gathering dust on the front page. Instead, I think it would be good to make executive decision about terminating an unpopular subject prematurely (or extending a very popular one past the week/month mark), giving the CotW more time to gather useful edits from interested people. Inductiveload (talk) 00:37, 4 March 2010 (UTC)

Interesting annotation solution
User:TomS TDotO has proposed an interesting method for Wikisource annotations at Talk:Evolution and Ethics (it can even annotate within footnotes). I would be interested to know what others think of this solution. --Eliyak T · C 02:13, 5 March 2010 (UTC)


 * To me it is interesting to speculate as to what a fully integrated notes system would look like. For example, the header notes should be presented in some more sophisticated way (not just a single text line). Errata handling, which this example shows, should be more explicit in the header: where errata are being handled by sidenotes, the reader should be told that, where the text uses inline editorial [sic] this should be flagged up, and if there is another text being referenced as published errata, there should potentially be a link. The footnotes really need to be in a "footnote footer" of some kind, also accessible easily from the header: this example shows that the footnotes can start in the middle of the page, practically speaking. So I think there are some design issues to address, beyond what you could call "text placing" for annotation. Charles Matthews (talk) 10:03, 5 March 2010 (UTC)


 * An editor has returned Evolution and Ethics to its "clean" state, and created a second version, Evolution and Ethics (wikisource annotations) with the annotations added. I'm not going to change this, unless and until there is some kind of consensus. However, it seems to me that this is not the best solution to the question of annotation in Wikisource. Having essentially two versions of the same text is going to raise problems of synchronizing the text between the two versions. If someone discovers, for example, a typo, then one must make the identical correction to the two versions. This seems to me to be, at best, placing an additional burden on an editor, to make certain that the identical change is made to the two versions, and also to check to see whether all previous editors have taken that care; and - well, maybe "at worst" is overstating it - the more likely situation is that the two versions will slowly drift apart.

On Annotations it is said that "editors are encouraged to experiment and try new methods", and in that spirit, I have experimented on my own pages with an alternative. I keep in mind the two major policies, (1) that a clean text be available for those who desire it and (2) NPOV is maintained. (NPOV, I trust, will be always adhered to and needs no further discussion.) What I am, for the moment, suggesting is that:

1. A subpage of the clean text be created by appending to the title "/Annotation". 2. The subpage consist of some brief introductory and concluding material, with the substance of the subpage being the annotations to the main page clean text. 3. Each annotation in the subpage is a separately titled section, distinctively named according to the particular text in the main page clean text which is being annotated. 4. In the main page clean text there be a link to the section in the subpage where the annotation is made.

For example, if in the text is found in ThisIsTheMainText, the annotations are in ThisIsTheMainText/Annotation. In the main page, there is the text "this is an obvious tpyo", it will be replaced by a link like this: this is an obvious tpyo. In the subpage, there will be a section with the title "===tpyo===" and it will say something like this "this is an obvious tpyo [sic] should be this is an obvious typo".

This means that the main page text remains as clean as possible. The text as it appears is only distinguished from the original by the indication, by color change, that there is a a link. If, for example, one prints the text, it is identical to the clean text.

I trust that I do not have to say that I welcome - I indeed encourage - comments, criticisms, changes, and such. I'm a newcomer to Wikisource, only editing here for a couple of weeks, and recognize that I have a lot to learn. For those who want to see what I have in mind, I have set up a brief example on my personal pages: User:TomS TDotO/Evolution and Ethics and its subpage User:TomS TDotO/Evolution and Ethics/Annotation. TomS TDotO (talk) 13:07, 8 March 2010 (UTC)


 * Personally, I prefer and will always prefer SIC for tysoptypos [sic] and will always do that inline. To your general premise, to me it depends on the number and nature of the annotations. To me the rules can stretch, the principle is solid. The text is king. — billinghurst  sDrewth  15:05, 8 March 2010 (UTC)


 * First of all, I thank you. I wondered what the template was for [sic]. That is very handy. Secondly, I realize that I have (once again) expressed myself poorly. What I was thinking of was when an editor discovers a mistake in the transcription from the original to the "clean text" as presented in Wikisource, and wants to make a correction. It has nothing at all to do with annotations. If the Wikisource "clean text" differs from what Huxley wrote, then no one will question a silent editorial correction of the "clean text" to what Huxley wrote. My point was that, under the system where there are two copies of Huxley's text, one a "clean text" and one an "annotated text", then the editor has to be aware of the existence of two versions and make the identical change, simultaneously, to two texts. Even in the case where there is strict discipline, the problem of keeping copies synchronized is a problem. In a wiki, I think that it is impossible. As to the "number and nature of the annotations", the immediate case that started me off on this has about a dozen minor annotations. In brief, the standards for citations in Huxley's day were far looser than the standards of today. TomS TDotO (talk) 15:53, 8 March 2010 (UTC)

No rest?
What happened to the rest of 1 John? Only Chapters 1 and 2 were on-line. Please post complete.


 * Thanks, please provide some more details. There a few versions here, look in your history menu for the page name or url and we can see what the problem is? Cygnis insignis (talk) 16:38, 6 March 2010 (UTC)


 * The complete book is available at Bible (King James)/1 John and Bible (American Standard)/1 John. —Spangineerwp (háblame) 16:08, 8 March 2010 (UTC)

Too sensitive or too demanding?
is adding many statements for Author:Rajiva Wijesinha‎ and was thinking of asking them to lift the standard of their work to include scans and pages and from that transclusions. I would envisage that I would be given the bird, however, as part of our lifting our works to being with scans I am wondering whether it is a reasonable request? That we are being used as a press release station may be the case, however, we can add some rigour to our expectations of how the work is added. — billinghurst  sDrewth  12:08, 8 March 2010 (UTC)
 * Seems like a reasonable request to me; scans would be a great addition. Hopefully his response will be better than you expect. —Spangineerwp (háblame) 16:05, 8 March 2010 (UTC)
 * Just Curious - have you asked "them" about their efforts directly or are you infering or injecting your own perceptions here about the project being a "press station"? George Orwell III (talk) 16:57, 8 March 2010 (UTC)
 * I don't really have a view one way or the other regarding this, but I'm just wondering what the usual situation is with documents that never have a hardcopy 'original'. The above documents, from the couple I looked at, are originally published online.  With asking for scans to be included, would we just be asking for screenshots of the website from which they're from?  &mdash; Sam Wilson ( Talk &bull; Contribs ) &hellip; 08:56, 9 March 2010 (UTC)
 * Nearly every one I checked had a link to the same content already being hosted online elsewhere too (which is why I questioned the rationale given in the first place). I think making a case prior to OTRS approval for requiring a very modest percentage of the overall posted content be done using "originals" (for starters) would make much more sense than imposing/suggesting changes be made at some point after the fact. George Orwell III (talk) 10:59, 9 March 2010 (UTC)
 * But what are 'originals'? In this case, the originals are probably the online HTML copies.  (I don't mean to say that I think that these particular examples should be on Wikisource or not.)  The presence of other online copies of works doesn't per se mean that we don't also host those works here—look at Project Gutenberg for a classic example!  What I mean is: yes, Wikisource may be being used as a dumping ground for this material, but—and I'm sure someone will be able to answer this; it is in no way rhetorical—on what grounds do we judge this?  &mdash; Sam Wilson ( Talk &bull; Contribs ) &hellip; 11:55, 10 March 2010 (UTC)

Cold Fusion resources
Most are under copyright and would not be easily copied to WikiSource. A few, however, are public domain. This one is the 1989 US Govt Dept of Energy report: http://www.ncas.org/erab/index.html --It says right there on the page that "Distribution is unlimited". That means it should be easy to copy it here somewhere --except for one nagging difficulty; the Report is scattered across a whole slew of Web pages; there doesn't seem to be a single document that can be easily uploaded. What to do? Thanks in advance! Objectivist (talk) 14:17, 12 March 2010 (UTC)
 * Thought about running Adobe through it and collecting subsidiary page into a PDF then uploading it to archive.org and getting it converted to the other document formats? — billinghurst  sDrewth  02:42, 13 March 2010 (UTC)


 * Well you can pay $15.00 for an electronic copy of Technical Report: DOE/S-0073, November 1, 1989, straight from NTIS. There doesn't seem to be a copy of that freely floating around as far as I can tell. George Orwell III (talk) 14:36, 12 March 2010 (UTC)
 * I had a good old sniff around as well, but I couldn't find any other source for it, even in academic databases. Except for one place you can get it digitised for $85, but since you can get it for $15, that's hardly helpful! Inductiveload (talk) 00:27, 13 March 2010 (UTC)


 * You don't need a scan to create a work here. You can bring in just the text and link that page as the source. If a PDF turns up in future, we have automagical ways of matching the text up to each page. The text is the most important part, the source document is just a nice addition, and can easily be added later. Inductiveload (talk) 01:34, 14 March 2010 (UTC)
 * The advantage of providing a script is that it allows authentication of the work by easily allowing others to proofread and validate the work. — billinghurst  sDrewth  10:35, 14 March 2010 (UTC)
 * Well, yes. I meant that if there is no script around, then it is not a total impediment to putting up the work (unlike the Germans). Of course it is better with the original to hand, but the text alone is better than nothing. I would question the value of PDFing the website, and then converting to DJVU - that would rather create a circular reasoning - the text will naturally be identical to the PDF if it is copied and pasted, but whether it is identical to the original text is another matter. Inductiveload (talk) 17:12, 14 March 2010 (UTC)
 * Maybe I should have said PDF an authoritative text. The issue can be that anyone can come along and edit, then make a statement about this being the correct text. So having an image file, against a text can at least make something an authoritative version, if nothing else. Always nice to brook that argument. — billinghurst  sDrewth  00:57, 15 March 2010 (UTC)
 * No sweat, I agree totally! Just making sure it was clear that scans are not needed. In other news, look what the cat dragged in: File:Cold Fusion Research - ERAB - 1989.pdf. Enjoy! Inductiveload (talk) 03:24, 16 March 2010 (UTC)
 * And page started at Cold Fusion Research. I have also emailed NCAS and asked if they still have the original scanned microfiche (11 years on, may be a long shot). But for now we have a "hard copy" to use as a fallback. Inductiveload (talk) 03:53, 16 March 2010 (UTC)
 * Thanks, folks! By the way, there is another DOE review that was done in 2004.  Since it also is a Government document, it is not copyrighted.  Here is a place to start looking for it, should someone decide to put a copy of it in WikiSource: http://www.lenr-canr.org/Collections/DoeReview.htm

Creative commons

 * Papers published under "creative commons Attribution-Noncommercial-No Derivative Works 2.5 Generic", would these papers automatically accepted by wikisource?Arilang1234 (talk) 23:43, 14 March 2010 (UTC)

No; as a Wikimedia project, we only accept Free Content, and neither noncommercial nor no derivative licenses fits that criteria.--Prosfilaes (talk) 02:16, 15 March 2010 (UTC)
 * Thanks.Arilang1234 (talk) 01:54, 16 March 2010 (UTC)

Rollbacker proposal above
For those who focus on the bottom of this page, please see in the Proposals section above. — billinghurst  sDrewth  11:26, 16 March 2010 (UTC)

Cross wiki post management
Is there a way which would notify here on Wikisource that there is a response to my post on another wiki? — Ineuw (talk) 12:26, 16 March 2010 (UTC)
 * gWatch and w:Wikipedia:Integrated watchlists might be of interest here.
 * Thanks. gWatch does the trick. very neat. :-) — Ineuw (talk) 13:18, 16 March 2010 (UTC)

Wikimania Scholarships
The call for applications for Wikimania Scholarships to attend Wikimania 2010 in Gdansk, Poland (July 9-11) is now open. The Wikimedia Foundation offers Scholarships to pay for selected individuals' round trip travel, accommodations, and registration at the conference. To apply, visit the Wikimania 2010 scholarships information page, click the secure link available there, and fill out the form to apply. For additional information, please visit the Scholarships information and FAQ pages: Yours very truly, Cary Bass Volunteer Coordinator Wikimedia Foundation
 * Scholarships
 * Scholarships FAQ

Source statistics
I'm dragging into activity statistics of variuos source projects, looking for the causes of a trouble of it.source (I saw a nonsense low number of "good" pages). Statistics, in my opinion, are only useful if numbers are meaningful, than can be a reward for contributors.

Taking a good look to http://s23.org/wikistats/wikisources_html.php?sort=ratio_desc, you can see nonsense relationship between total number of pages and "Good" column. I found that this comes from two different issues:
 * 1) different settings of "content pages", I presume that there's differents settings about nsPage: and nsIndex: (are they counted as "contents" or they aren't?)
 * 2) differencies in formatting of the pages. A page is counted as "good" if:
 * 3) it is into a "content" namespace;
 * 4) its code has at least one link to another page of the project, but the counter works only if the link is posted with normal syntax  ...  and NOT if the link is generated by a template; this affects directly the stub ratio and explains a large variety of stub ratios (i.e. Portuguese source: 0.75; English: 0.25; French: 0.11). It's obviuosly wrong to presume that the 89% of French source pages are "stubs"!

The issue and a possible solution for "the lacking link issue" is discussed here: wikt:en:Wiktionary:Page count but the differencies about different settings of "content pages" should be discussed to come to a shared approach IMHO. The main question is: should pages into nsPage: and nsIndex: be counded as "content pages" or not? My temptative answer is no; but I'm interested in your opinion. --Alex brollo (talk) 11:38, 15 February 2010 (UTC)
 * If a text included in a page of the main space and the same text on a page in the page space—two texts that are one text if we consider the truth, don't you think so?—don't you think that this text ought to be accounted for only once and not twice? Is there any risk of counting it twice?--Zyephyrus (talk) 22:09, 2 March 2010 (UTC)
 * You're right. This has been the conclusion of our talk into it.source: to mark ns0 pages only as "content", excluding both nsIndex and nsPage pages, and to add an explicit link code when lacking (since any text page hase lots of links, but they are usually "hidden" because they come from templates). Adding such a code, now our "stub ratio" grew to a good 25% from an "impossible" 10%, a stub ratio very similar with en.source and de.source values. Problem solved! --Alex brollo (talk) 12:16, 26 March 2010 (UTC)

Ai Weiwei conversation
http://www.readwriteweb.com/weiwei/

Ai Weiwei, Jack Dorsey and Richard MacManus had a conversation at the Paley Center for Media in New York moderated by Emily Parker, if I am able to turn the audio into text, would the text be in public domain?Arilang1234 (talk) 01:54, 16 March 2010 (UTC)
 * I wouldn't have thought so. Just being a public conversation doesn't mean that they give away their rights. — billinghurst  sDrewth  12:48, 21 March 2010 (UTC)

Transcluded article with missing source page
I have used transclusion to create article Allen, William (1532-1594) (DNB00), but one of the source pages (this one) is totally illegible (it's a reasonably long article, so there's still plenty of meat there). Not knowing any better I have created a warning box on the source page, so that it gets included into the article. I think I'm asking, is there an already-agreed way of handling such issues? Alternatively, just asking for comments. Jan 1 naD (talk • contrib) 21:57, 17 March 2010 (UTC)
 * Previously I have used under construction or incomplete as a means to manage. — billinghurst  sDrewth  22:09, 18 March 2010 (UTC)


 * I believe lacuna is the most appropriate template for this situation:


 * Perhaps it needs modifying so that the source url can be omitted when on a Page: page, as the source is then obviously implicit. Inductiveload (talk) 15:10, 19 March 2010 (UTC)

The Birth-mark
Comment at Talk:The Birth-mark needs addressed, who has the best suggestion? JeepdaySock (talk) 16:31, 18 March 2010 (UTC)
 * I put one up, though I have a feeling I left a double redirect somewhere. Cygnis insignis (talk) 17:54, 18 March 2010 (UTC)

zap request
Admin required. See here] Thanks Victuallers (talk) 17:35, 19 March 2010 (UTC)
 * Dealt with the luser's sockpuppets just now too. — billinghurst  sDrewth  23:54, 19 March 2010 (UTC)

Naval history book
I am interested in collecting Wikipedia articles on WW II US Navy Ships and organizing into a book. Is Wikisource the tool to accomplish this?
 * Hi! Short answer is that no, it isn't. Wikisource is devoted to published material, rather than user-created material. The good news is that Wikimedia Foundation operates Wikibooks, where you are encouraged to write your own books. Of course, should you find interesting public domain material (such as ships logs, official reports, etc) these can and should be brought to Wikisource, rather than Wikibooks, and cross-wiki linking can be used to tie it all together. Almost all US Navy documents should be in the public domain, as works of the US government, so there is scope for interesting material to be brought in! Good luck with your book. If you need any help, I am also active on Wikibooks, so drop me a message if you need anything at all! Inductiveload (talk) 18:01, 19 March 2010 (UTC)

If you are interested in collecting Wikipedia articles, you may also be interested in Wikipedia:Books. --Cerebellum (talk) 03:04, 22 March 2010 (UTC)

Seeking knowledge through advice
There are two pages in PSM volume 8 marked as Problematic because the OCR image is not visible in edit mode as in THIS PAGE.

Also, THIS PAGE begs for improvement. A .jpg image was inserted VERY temporarily, just to show what the page SHOULDN'T look like.

..... This line is reserved for the usual pleadings and petitions for help. — Ineuw (talk) 15:14, 20 March 2010 (UTC)
 * The thumb of the image has failed, and the image has been cached. It happens occasionally and will fix itself in time. You can see a similar discussion on my talk page, and I cannot get the normal means to get the beast to display. — billinghurst  sDrewth  12:33, 21 March 2010 (UTC)

What to do about words hyphenated across a line break in scanned texts?
Djvu scanned text often words that are hyphenated across a line break. There are two approaches to dealing with this:


 * 1) Use the   macro
 * 2) remove the hyphen

(1) has the advantage that it fully preserves the source text. (2) has the advantage that it is much quicker and is less prone to error. I've so far been taking approach (1), but am finding that this significantly slows down my editing and proofreading of text. Since all that hw does is preserve the intent of the typesetter and looses none of the intent of the author, I've come to the conclusion that I want to start using approach (2).

Any views on this? Martin.Budden (talk) 15:28, 21 March 2010 (UTC)


 * For hyphenation, I use the rationale that it will never be like the original, and why make it hard for non-editing reader?


 * 1) If the word is incomprehensible separately, then I remove the hyphen and join the sections properly.
 * 2) If the words when separated make sense, then they are old fashioned usage of proper English and just remove the space trailing the hyphen. — Ineuw (talk) 22:45, 21 March 2010 (UTC)
 * I now close all hyphens, but try to keep in mind the what Ineuw is referring to: that words that today do not use a hyphen may have used one in the past. So my technique is to keep an eye out for hyphenated words in the text that are not on a line break, and if I see those words hyphenated over the line break, I make sure to keep the hyphen if required to keep the work consistent.  I'm sure I don't catch everything, but it's worth a try and it's far simpler than using hw. —Spangineerwp (háblame) 00:13, 22 March 2010 (UTC)

The fact that we are transcribing printed paper books into an entirely different medium in which we have very little influence over typefaces, font sizes, margins, line widths, etcetera, implies a willingness on our part not to preserve layout exactly. As a guiding principle, I aim to preserve the creative input of the author whilst ignoring the input of the typesetter. In the case of hyphens, this means preserving the "hard" hyphens that the author would have intended to be there regardless of page layout, while feeling free to remove those "soft" hyphens that were inserted by the typesetter whilst trying to make the author's text look nice on the page. Hesperian 00:18, 22 March 2010 (UTC)
 * Exactly. We do not attempt to preserve the left-and-right justification, and more generally we do not attempt to preserve the printed lines at all. Our line breaks are wherever the web browser chooses to place them. But the end-of-line hyphens were placed there by the typesetter: the paper equivalent of a web browser, so we are obliged to remove them as part of our transformation into our new medium. If a reader requires a more faithful reproduction of the published work, then that reader must look at the original. We don't use the same fonts, we don't use the same line widths, and we don't use the same end-of-line hyphens. If in some hypothetical world we were trying to be more true to the printed work, we would need to reproduce the fonts, the line-breaks, and the spacing. There are in fact printed works where these aspects are a fundamental part of the work, and we will need to consider all of them. But the vast majority of Wikisource works are about the author's words, and not about the original typography. -Arch dude (talk) 01:11, 22 March 2010 (UTC)
 * Sufficient general commentary made about line hyphenation, especially Hesperian's comment. Do note that we do wish to maintain hyphenation at the end of pages in the Page:namespace, though even soft-hyphenation disappears too when we transclude it. See hyphenated word start — billinghurst  sDrewth  02:07, 22 March 2010 (UTC)

The New York Times
Hi guys, I just wanted to let you know that have moved a New York Times Collaboration page to the Wikisource namespace. We can use this page to help upload the New York Times in a standard format, format the main namespace pages in a standard format, and general collaboration for the New York Times. If anyone has any particular issues or articles you are looking for please let me know. I can convert the file and upload the DJVU for you. It is probably best to leave a message on my talk page for such request. --Mattwj2002 (talk) 12:00, 22 March 2010 (UTC)
 * Excellent! One important question: how will we will arrange the individual issues in the mainspace? I see three options:
 * One mainspace page per issue. This will be very long (one page is about 50kB of raw text; and issue is about 15 pages)
 * One mainspace page per page of the original. This will produce "quite long", but perfectly manageable pages.
 * One mainspace page per article. While this is attractive for the long articles, it will produce tens of thousands (literally) of tiny pages with only a few lines on them for the shorter articles.
 * I am personally in favour of the second option, at least for the ~1900 period where the paper seems to be dominated by short articles. Perhaps for papers with fewer longer articles, the third option may be better when we come to those. Inductiveload (talk) 02:14, 23 March 2010 (UTC)
 * I think I like the second option, one page = one page. JeepdaySock (talk) 10:45, 23 March 2010 (UTC)
 * One page per article. The article is the item of interest. One page per page is pointless; a "page" is simply an artifact of the medium.--T. Mazzei (talk) 02:59, 25 March 2010 (UTC)
 * One page per article is how we're handling other works that are compilations of short articles (cf. EB1911). —Spangineerwp (háblame) 15:35, 26 March 2010 (UTC)

Odd DEFAULTSORT behavior
Has anyone else noticed that on pages transcluding from the Page: namespace, a stray DEFAULTSORT key is being added to the page?

Look at this test (which transcludes some Pages) which has a DEFAULTSORT key, and compare it with this one which doesn't have the key. Why do I get an error when I add the key? This is going to make a whole lot of pages have an annoying error message at the bottom.—<font style="color: #000000; text-decoration: none; font-weight: bold">Zhaladshar <font style="color: #FF0000; font-size: small; text-decoration: none">(Talk) 15:36, 25 March 2010 (UTC)


 * Yes, noticed in Category:DNB No WP some time ago, assuming we're talking about the same bug. There large numbers of pages (those transcluded in a certain style) end up under "D". I'm told the software reason has been tracked down. See User_talk:Billinghurst/Archives/2010. Charles Matthews (talk) 22:19, 25 March 2010 (UTC)


 * ThomasV has a fix ready in place, which is ready to be rolled out with the next software update. Thomas says that there are a number of issues surrounding this and they are indicated on his user talk page, and over at the Proofread Page talk page at oldwikisource: — billinghurst  sDrewth  00:20, 26 March 2010 (UTC)

Author categories
I just made some changes to Category:Authors-P to give an example for discussion on how this is managed. One of the things I did was to delete the subcategories that contained three or less members, leaving Pu (8), Pl (8), Ph (16), and others with more than thirty members. The sub-cat Pa has 92 members. Some other issues emerge as you poke around, but I want focus an a simple formulation. This is not the largest or smallest category, so I think that makes it an good example. The number of members is of course increasing. The less common first and second letters, such as Pt, are currently found under Authors-P. If the only consideration is the display limit, 200 members is the default I think, there are are four possible approaches: Perhaps there are other considerations on how the categories are managed. For those unfamiliar with the current system, the category is included by using the Author template's parameter. How this relates to other author indexing is, I think, a separate issue. 10:40, 27 February 2010 (UTC)
 * 1) Remove all the two letter sub-cats. [This means the user may need to click 'next' once or more.]
 * 2) Use a subcat to stop that limit being exceeded, by removing the largest sub-group eg. Pa then Pe [one more click to find target]
 * 3) Create a category when the number of a sub-group equals an agreed value - 5 to 200?
 * 4) Always create a subcat for a member of Author-P, leaving it empty.
 * To be honest, I've never really been that big a fan of this sub-categorization. I don't understand how clicking "next" to get to the desired "Ph" last names is any worse than having to click on the "Authors-Ph" category link immediately given on the author page.  Especially when we get large numbers of authors in subcategories (say "Pa") will we need to start sub-sub-categorizing ("Par," "Pas," "Pat," etc.) to get rid of the need to click "next"?  I don't think it actually helps a whole lot, and if display is the issue, that's an aesthetic approach that will merely add a whole lot of extra work to maintain (such as recategorizing large numbers of author pages into sub-sub-categories once the sub-categories we've already put them in has gotten too large).


 * I think it's also a bit counter-intuitive at times, because we have a small number of authors with uncommon combinations of letters (such as Pt or Sz). Someone would first have to check to make sure that their combination of letters isn't a sub-category before browsing.


 * I think a better way of handling this (that will help with scalability) is to put all the authors back in one category again and link to specific letters, like this (I've linked to all poems in Category:Poems that start with "Du"). We still accomplish the same thing, but don't have to worry about any one editor going through thousands of author pages to make sure they are put in the appropriate part of the author category hierarchy.—<font style="color: #000000; text-decoration: none; font-weight: bold">Zhaladshar <font style="color: #FF0000; font-size: small; text-decoration: none">(Talk) 14:55, 27 March 2010 (UTC)

Book reviews/scholarly work
One of the things about adding journals to WS is that many journals contain a few book reviews. This is very important information in some circles (I've had to use them in order to write some papers in college) and helps a great deal with (especially historical) research. How should we go about handling these types of writings?

When we have works written about an author, we link it from the author page in the Works about this author section. But I don't know if this is the best practice for individual works, since the work itself should be a relatively faithful facsimile from the Page: namespace. So where would be a good place to link to them so that people can easily find it? For example, if someone wrote a review of Moby Dick, where should the review go? And an almost identically related issue is if a scholar wrote a critical work on a particular book or author (such as wrote a dissertation on Romeo and Juliet), we would need to find a place to put that work so people interested in Romeo and Juliet can see other works about it?—<font style="color: #000000; text-decoration: none; font-weight: bold">Zhaladshar <font style="color: #FF0000; font-size: small; text-decoration: none">(Talk) 15:06, 27 March 2010 (UTC)
 * If a work has a disambiguation page already, I would add it there. See for example Observations, systematical and geographical, on the herbarium collected by Professor Christian Smith, in the vicinity of the Congo, during the expedition to explore that river, under the command of Captain Tuckey, in the year 1816. If a work doesn't have a disambiguation page, I don't think I would create one for a review; I would probably create a Wikisource: index instead. Hesperian 22:38, 27 March 2010 (UTC)
 * If there aren't too many reviews, the Notes field of the header could be used. - AdamBMorgan (talk) 01:38, 28 March 2010 (UTC)
 * I agree with both Hesperian and Adam on this point. Disambiguation pages of there is more than one work, and Notes field if it is the only work.  I would think that we could even use wikisourcepar2 within the notes if we needed something with a little oomph. These would also be two way links.
 * With regard to where the work would be in the main namepsace, I would encourage it to be as a subpage of the parent work, we have Ineuw's work with PSM where this already has been done, some of the poetry. A redirect from the root level can be added as/if necessary to promote the work. — billinghurst  sDrewth  05:34, 28 March 2010 (UTC)
 * I generally add similar works like a "book review of Moby Dick" underneath the Moby Dick heading on Melville's authorpage. I always like to alert browsers to what exists, since it hardly clutters up the page to say "hey, if you came here because you like Moby Dick and you want to see what else Melville wrote...you'd probably like this antique book review on Moby Dick..." - I would also support adding it in the|notes field of the work itself; in situations where a book has numerous "tangential" works, we could even create a footer template if necessary. I am a fan of Author:Richard Francis Burton, any time somebody has written a poem about him, or a review of one his books, I want to see it linked on his authorpage (which it is, yay!) Sherurcij Collaboration of the Week: Author:Thomas Carlyle. 13:19, 28 March 2010 (UTC)
 * To extend this a bit, I think that the work's "notes" header should link to the review (singular) or to a "reviews" page when there is more than one. The "reviews" page can then link to each individual "review" page. I do not think that a link from the author page of the author of the original is a good idea, but I seem to be in the minority here. I do think that a link from the author of the review is a good idea. -Arch dude (talk) 22:03, 28 March 2010 (UTC)