User:Spangineer/Sandbox/archive

Principia
Hallo, thanks a lot for working on The Mathematical Principles of Natural Philosophy (1846). Best regards, --D.H (talk) 17:27, 26 March 2010 (UTC)
 * My pleasure. It'd be great to get it entirely validated.  Let me know if any of my edits seem questionable to you. —Spangineerwp  (háblame)  17:31, 26 March 2010 (UTC)
 * I can't get the paragraph with the image in it to connect with the paragraph in the previous page. Any suggestions?  Do I create a section in the page? ResScholar (talk) 12:16, 1 April 2010 (UTC)

Spanish Academy Dictionary
Is there are an appropriate place, here or on the Spanish Wikisource, to link a reference to the "Spanish Academy Dictionary"? The reference is at Page:Don Quixote (Cervantes, Ormsby) Volume 1.djvu/19. Hesperian 13:24, 10 April 2010 (UTC)
 * Very interesting. I'm almost positive this is a reference to the "Diccionario de autoridades," the precursor to the RAE's Dictionary of the Spanish Language; I've added the link. —Spangineerwp  (háblame)  05:46, 11 April 2010 (UTC)
 * Thanks! I had a look over there, couldn't find it, but left with the impression that it was there to be found. Hesperian 05:50, 11 April 2010 (UTC)

Gif->Djvu
I'm going to send my email by EMailUser. I use DjvuSolo, it's free and really simple to use. My mail account has a limit of 2oMby, so split accordingly your files. --Alex brollo (talk) 23:26, 4 May 2010 (UTC)
 * Done, tell me please if you got the djvu file (it's pretty large, since I assigned the max possible resolution to images), I'll be happy to know its final name too to "follow the works"! --Alex brollo (talk) 08:34, 7 May 2010 (UTC)
 * Sorry about that&mdash;I got it, and it's at Index:Unconstitutionality of the Fugitive Act.djvu. —Spangineerwp (háblame)  11:53, 7 May 2010 (UTC)
 * Thanks! I'll follow the works! --Alex brollo (talk) 12:12, 7 May 2010 (UTC)

MediaWiki:Revdelete-reason-dropdown
Are you going to explain all about it? Maybe just to admins? Yours cluelessly. — billinghurst  sDrewth  03:25, 19 May 2010 (UTC)
 * Looks cool, based on w:Wikipedia:Revision deletion. I'll start a discussion on WS:AN. —Spangineerwp  (háblame)  04:04, 19 May 2010 (UTC)
 * Many thanks. This in the last upgrade (whichever date) escaped my notice. We probably need to have a more thorough watch. — billinghurst  sDrewth  17:44, 19 May 2010 (UTC)

Lectures on Modern History
Have you noticed that some of the pages for this text are subpages of Lectures on Modern History and some aren't? This means that most next and previous links don't lead anywhere, even though the text is complete. I'm not sure what you and the others transcribing the text want to do, hence my note. —innotata 18:29, 20 May 2010 (UTC)
 * It's only complete in the sense that there are placeholders for all the chapters. There isn't text on most of the pages yet, so I haven't prioritized fixing the structure.  Probably worth doing though, now that a few works are in place. —Spangineerwp  (háblame)  18:35, 20 May 2010 (UTC)
 * Should be all fixed now. —Spangineerwp (háblame)  19:58, 20 May 2010 (UTC)

Comment
Sorry about all that, at New texts. I am disappointed that my changes were not well-received, but I thank you for your polite explanation. Cheers, -- Cirt (talk) 01:47, 19 June 2010 (UTC)
 * Thanks for understanding. Let's see if this alternative that I've put in place is satisfactory to others.  In any case, I like the addition of the link to the month archive; that will make things easier for sure. —Spangineerwp  (háblame)  02:13, 19 June 2010 (UTC)
 * Okay, thank you very much! -- Cirt (talk) 02:14, 19 June 2010 (UTC)

Quotation marks at the start of each line?
Hello. Please can I draw your attention to a query I made at the Scriptorium.--Longfellow (talk) 17:16, 11 July 2010 (UTC)

Nullification Controversy in South Carolina Clean Up Request
This is done as you requested. --Mattwj2002 (talk) 05:15, 25 July 2010 (UTC)

nop
Hi Spangineer,

Thanks for the info on the nop template -- makes sense, and very helpful!

I was actually just about to message you with a question. I am very interested in the process of preparing a document here -- seems it's become a much more refined (and more inscrutable) process since I last looked. I'm wondering if you can help me learn how to create an Index: page, and how to properly prepare a DjVu file.

As you can see in my user space, I'm trying to create a Wikisource version of a document significant to my home city; this one is a scanned PDF of low resolution, which of course introduces some additional challenges.

I tried to follow the instructions on Help:Djvu files, but didn't get terribly far. I also used a Mac program, VolOCRaptor, to create a PDF with (very poor) OCR text.

Can you give me any pointers on creating a high-quality DjVu file, and/or how to use that to create an Index: page?

Also, a much smaller question -- I have not been able to figure out (in the Full Disclosure pages) how you get the header (book title and page number) onto the top of the page.

Finally..is there a convention about where people discuss the process of wikifying a longer work like this? Is there Index talk: namespace the place to use?

Thanks so much for your help with Full Disclosure, and for any answers you can provide! -Pete (talk) 06:09, 26 July 2010 (UTC)


 * Very helpful, thank you! I've already tried the "Running header" part, thanks for solving that mystery. I should have mentioned, I did try the Internet Archive method too, but ran into some trouble (upload hung interminably). I could try it again, but I think I'll try your method first. More later! -Pete (talk) 15:56, 26 July 2010 (UTC)

Missing text in the Federalist
I glanced at this diff, and noticed there is missing text; do you happen to know why that might be? Cygnis insignis (talk) 00:27, 27 July 2010 (UTC)
 * I did a fair amount of research attempting to identify the "best" (closest to the original) copy of the Federalist, and the result was that the version that was on Wikisource (and what I used in the match and split) is slightly different from the version being proofed against. The DjVu version I uploaded is an original Dawson edition from 1863, upon which are based most modern editions of the Federalist (including the one that was on Wikisource).  Dawson's edition rejected the changes that Gideon's edition introduced.  Besides the obvious formatting differences, there are a few differences between Dawson and modern editions, most notably in the numbering of the essays and occasional differences in text (like the one you point out). But for the most part the differences between Dawson and modern-day editions are slight. For what it's worth, in the case of your example, the text of my print version (Bantam, 2003) matches the text in the page image. —Spangineerwp  (háblame)  02:08, 27 July 2010 (UTC)
 * Great that we have the best edition, and to see its transcript improving. I had supposed that there was a match and split, but couldn't see where it happened. I usually correct ocr, so often nearly perfect, and find detecting editorial changes takes longer. I'm thinking we should gives users a heads up about what's going on with the text layer in these situations, no idea how to do that though - keep moaning about it? ;-) Cygnis insignis (talk) 02:29, 27 July 2010 (UTC)
 * In general I tend to agree with you; pulling from the OCR is often better than using an existing edition. In this case I felt like it the two options were a bit more even, though I probably didn't realize the number of editorial differences when I performed the match and split.  The good news is that it was relatively easy to apply mass conversions: an OCR isn't always going to give the same result for a ligature, for example, and there are lots of those.
 * As for alerting people, I put a number of notes on the index talk, some of which relate directly to the different versions. But I don't think the index talk is a widely frequented namespace =). —Spangineerwp  (háblame)  02:41, 27 July 2010 (UTC)
 * Tends to indicate that we probably need a bold statement (emblazoned banner) that YELLS to people to read the talk page for that work. I know that for specific (tricky/clever) formatting that I put notes on the talk pages too. — billinghurst  sDrewth  08:13, 27 July 2010 (UTC)
 * Hoping that it will be useful, I have created and added index talk — billinghurst  sDrewth  11:55, 27 July 2010 (UTC)
 * Looks great. —Spangineerwp (háblame)  02:33, 28 July 2010 (UTC)


 * Your contribs prompt me to mention what I've been doing lately, inspired by lots of things and users and hardly an original idea. I get the text file from the source, make some search and replace fixes in a text editor, then add a section. Replacing  with   400+ times with a  single click assured me I was on to something. As I go along I find more 'consistent errors', the ignorant machine is working in our favour because it keeps reading 'the' as 'tlie' or 'é' as 'd' and so on. It does requiring pushing the text to page, but tools like match and split could make do each section as the file improves. Cygnis insignis (talk) 02:51, 27 July 2010 (UTC)
 * P.s. the biggest time-saver is the paragraph breaks are not removed, Cygnis insignis (talk) 02:52, 27 July 2010 (UTC)
 * Interesting. I had thought of doing something like this a while back, but never got around to it. The approach I was considering was to pull the text out of the djvu file using djvused, make the fixes, then embed the corrected text back into the djvu file. Hesperian 05:08, 27 July 2010 (UTC)
 * I did this last night. (1) djvused to generate a dsed file; (2) simple python script to pull out all the words, throw away numbers and common English words, and sort the rest by frequency of occurrence in the document; (3) manually identify the frequently occurring OCR errors; (4) another simple python script to fix them all in the dsed file; (5) djvused to embed the corrected text. Works nicely. Hesperian 03:25, 28 July 2010 (UTC)
 * Cool, I used replace all for the certain fixes, and manually checked the rest. Does your way keep the paragraph breaks, or are they still collapsed in the Page:ns? One reason I keep the text file open, instead of pushing it out at once, is that I find fixes as I go along - even a dozen occurrences in later sections saves some hunting and typing. Cygnis insignis (talk) 03:44, 28 July 2010 (UTC)
 * Text appears exactly the same except for OCR fixes. I just did File:History of West Australia.djvu. The top five OCR errors were "iu" for "in" (450 times), "tile" for "the" (351 times), "aud" for "and" (254 times), "Iu" for "In" (63 times), and "wRh" for "with" (51 times). In total there were just over 60 OCR errors that occurred at least ten times each, totalling 2231 errors fixed. It took me about 45 minutes to go through the whole process, from downloading the djvu to uploading the new version. Hesperian 04:53, 28 July 2010 (UTC)


 * Hmmm... are you saying you take the text file that is offered at say archive.org, run a bunch of regex, create a page with that text (say in the user namespace) and match and split that into the page namespace? Sounds pretty good... might have to try that next time I feel the need to match and split. —Spangineerwp  (háblame)  02:33, 28 July 2010 (UTC)
 * No match and split necessary with H's way, it changes the text layer. The corrected text layer would be offered when someone first edits the page - as it is currently. Cygnis insignis (talk) 03:44, 28 July 2010 (UTC)
 * But H's way requires dsed files and python scripts... so... I'm hoping that there's a way that doesn't fly over my head at 30,000 feet. —Spangineerwp (háblame)  03:29, 29 July 2010 (UTC)
 * For some unfathomable reason I am eternally confusing you with Pathoschild. You would think I would have you two sorted out after all these years, but no, sorry. :-( Now I remember, he's the tech-savvy one.... ;-) Hesperian 03:51, 29 July 2010 (UTC)
 * Heh. My mother and grandparents think I'm tech savvy, but that's about it =). —Spangineerwp  (háblame)  03:56, 29 July 2010 (UTC)
 * My crude method uses a text file, in an external editor, but that is not useful to anyone else unless I match and split it (from user space). Fixing the spacing around quotes and keeping the paragraph breaks were my biggest successes, but any fix I find along the way can be applied to the rest of the text with a few keystrokes. I used to move each page of text back and forth from Page:ns, before I started using scripts in the page ns, switching to a text editor is faster than it sounds. Text editors also have keystrokes to select text, so moving a whole text to Page is not that difficult or tiem consuming; using search and replace scripts, and page markers opens up all sorts of possibilities too.
 * My current notion is to merge the two methods by making the whole text layer available to anyone with a bright idea for a fix, preferably through a wikipage which can show a history of these. This is already a possibility, by dumping the text and using the match/split when a part is moved to main space; a clunky method and quite a different approach. Cygnis insignis (talk) 11:44, 29 July 2010 (UTC)

Featured work
Thank you for your proofreading help! Now that Houston: Where Seventeen Railroads Meet the Sea is totally proofread, what is the process for making this a featured work of Wikisource? WhisperToMe (talk) 01:09, 16 August 2010 (UTC)
 * Just follow the "nomination" instructions at Featured text candidates. It's pretty simple--add a template to the talk page of the work, and start a new section at the bottom of the FTC page saying why you think the work should be featured. —Spangineerwp (háblame)  16:06, 17 August 2010 (UTC)
 * Thank you! Lemme get to work on that... WhisperToMe (talk) 19:58, 9 September 2010 (UTC)

In the Roar of the Sea
Hi, I noticed in the "Recent changes" that you have been fixing the naming problems for In the Roar of the Sea. Thank you so much for doing that! It is very appreciated by me. Hopefully I have learned from this and won’t screw it up the next time. Best regards, Another editor (talk) 18:39, 23 August 2010 (UTC)
 * No problem. I actually made a similar mistake fairly recently: Index:Whatsocialclasse00sumnrich.djvu.  I haven't had the energy to go back and fix the file name, so I figured I'd help you fix yours while it was still easy to do so =). —Spangineerwp  (háblame)  18:41, 23 August 2010 (UTC)
 * Well, again I thank you all the more. Does it all have to be done by hand? {It seems like the sort of job that there should be a script for.) If there is a way I could help you with your page, let me know. I am willing to do repetitious work, as long as I know I am doing it correctly. (But I don’t know how to do such tasks as suppress redirects. ) Another editor (talk) 19:18, 23 August 2010 (UTC)
 * You are right, it is best done with a bot. Unfortunately I'm not that tech-savvy, so I don't worry about it; if someone with a bot comes along and wants to fix it, they can go ahead. Thanks for your offer, but your time is better spent proofreading, especially since the ugly file name isn't really hurting anything in the meantime.
 * I believe that to suppress redirects you have to be an administrator. When I click the "move" button, there's a checkbox that says "Leave a redirect behind".  If I uncheck that box, the redirect is suppressed.  For non-administrators, the best thing to do is to add sdelete or a dated soft redirect. —Spangineerwp  (háblame)  19:47, 23 August 2010 (UTC)
 * ok, I assume that the template is added to the redirect? (I don’t know what a "dated soft redirect" is.} Another editor (talk) 20:01, 23 August 2010 (UTC)
 * See Dated soft redirect. Unfortunately it's not automatically added when you move the page.  When you move a page, a redirect is created, and the page confirming your action includes the exact template that you should use (with all the parameters filled in).  Thus if you want to replace the redirect with a soft redirect, it's as easy as copying and pasting. The purpose of the dated soft redirect is to give people a few months to change their incoming links.  It's most useful when moving works in the main namespace, since other websites (most notably wikipedia) often link to them, and just deleting the redirect would result in a lot of broken links.  For the page namespace, and especially for new works, no one is linking to those pages, so redirect suppression is fine. —Spangineerwp  (háblame)  20:10, 23 August 2010 (UTC)

User talk:98.154.149.97
Hi, thanks for adding the block warning, I was searching a template or something like that... Phe (talk) 19:53, 23 August 2010 (UTC)
 * My pleasure. The templates are at Category:User warning templates.  I added "subst:" before the template name (i.e.,  ) to save the contents of the template rather than the template itself. —Spangineerwp  (háblame)  20:01, 23 August 2010 (UTC)

Talkback
possible solution — billinghurst  sDrewth  08:08, 27 August 2010 (UTC)

Question about restoring the scanned page
Hi, I proofed http://en.wikisource.org/wiki/Page:In_the_Roar_of_the_Sea.djvu/159 but instead of saving my proofed page, gibberish was saved in its place. Is there any way of deleting the gibberish page in order to restore the scanned page so I can re edited it? Otherwise, I don’t know what to do. Thanks, Another editor (talk) 19:09, 27 August 2010 (UTC)
 * It looks fine to me. I'm seeing text that matches the page image. —Spangineerwp  (háblame)  19:17, 27 August 2010 (UTC)
 * pardon, I fixed it cygnis insignis 19:20, 27 August 2010 (UTC)


 * Yes, most kind. Thank you. Another editor (talk) 19:22, 27 August 2010 (UTC)
 * Okay, good, I'm not crazy =). —Spangineerwp (háblame)  20:25, 27 August 2010 (UTC)

infra section links
I forgot to mention that after our chat about nav within a section, I made use of section headings in this chapter. This paid dividends immediately, check the link for "glass globe" (i.e. 'crystal ball') and notice that a search would have failed to find the point where the "glass egg" is mentioned. cygnis insignis 14:45, 23 September 2010 (UTC)
 * Excellent. I have a few works in my queue that have similar text at the top of a chapter; the anchors will be useful.  Have you considered using Template:Namespace link to avoid the bad links in the page namespace? —Spangineerwp  (háblame)  17:37, 23 September 2010 (UTC)
 * Yes. Why add it? What is bad about the linking in the workspace, which serves the main page. That namespace is used a few times to give the result, the actual point, why does it need its own navigation? The validating user might like the option, using the main page is much less confusing. Is the case that must labour to make the workspace run in parallel to ns0, is it imagined that anyone would not be using, reading, maybe tweaking from main. Excuse me, but I fairly well exhausted by how much of a big deal people can make about getting text into main-space - it ought to be fairly bloody simple. cygnis insignis 18:27, 23 September 2010 (UTC)
 * Fair enough; I don't have any issue with not using it. Carry on. —Spangineerwp  (háblame)  19:43, 23 September 2010 (UTC)
 * As I ended previously, excuse the candour. Without supporting documentation on why, not how, this stuff makes the site more complex and arcane. It takes a user about year to discover that need about seven templates, the odd trick, and some good examples to do >99% of the job here. Everyone has a cause célèbre, mine is getting content focused contributors the simple and rationalised guidelines that allows them to carry on :( cygnis insignis 20:33, 23 September 2010 (UTC)
 * My response was based on a misunderstanding, what I said relates to Template:DJVU page link, which does the reverse of the one above. Sorry about that. cygnis insignis 20:43, 26 October 2010 (UTC)
 * Makes sense. Really neither template is particularly necessary.  Personally, I don't mind the extra code, because I feel like it often saves me time navigating the work later.  But I can tell that it also needlessly complicates matters for new folks. —Spangineer  (háblame)  21:18, 26 October 2010 (UTC)

help pages
If you can see anything else to do, please, do it. I keep trying to simplify the related pages, but I lose heart after unravelling and merging the stuff - there are a lot pages with too much detail, and that shift from the purpose. It needs a lot more work, and I'm not sure I am always leaving it improved. On other matters, be sure that I am acting with a great deal of consideration. cygnis insignis 15:06, 20 October 2010 (UTC)

Page links
''I'm looking at State_Documents_on_Federal_Relations and The Rise and Fall of Society in IE 7 and IE 8 and not seeing page numbers in either browser, even after refreshing. I'm not sure what the difference might be between these works and the ones that are showing up correctly. —Spangineerwp (háblame)  18:42, 21 October 2010 (UTC)''
 * Good on you for addressing that. I found the same thing, but didn't follow it up because it doesn't affect me. I notice both examples use page break, does it occur where there are none? cygnis insignis 19:03, 21 October 2010 (UTC)
 * What Social Classes Owe to Each Other/Introduction has no page breaks nor numbers. Can't find one with both though... —Spangineerwp (háblame)  19:11, 21 October 2010 (UTC)
 * Looks like a purging issue&mdash;if the page has been saved recently, the numbers appear in IE just fine. —Spangineerwp (háblame)  19:18, 21 October 2010 (UTC)

Index:The Cambridge History of American Literature 3.djvu
Hi,

There is also this version which is partly proofed, and maybe more versions on Wikisource already. Mattisse (talk) 23:42, 21 October 2010 (UTC)
 * Thanks for letting me know. I haven't found any other volumes, and archive.org doesn't have any more of that type.  I'm thinking it may be worth moving all the pages for the sake of consistency to the index page I created&mdash;it's the exact same book, just with better scans (white background, text doesn't seem quite as faded, etc.).  Any objections? —Spangineer  (háblame)  12:22, 22 October 2010 (UTC)


 * No objections at all.  Mattisse (talk) 14:03, 22 October 2010 (UTC)

Category deletion request
Hi, I think I successfully depopulated the Category:Audubon and moved the images to Category:Audubon and His Journals. So the Category:Audubon can be deleted on the Commons. Thanks! Mattisse (talk) 21:00, 26 October 2010 (UTC)
 * Done. —Spangineer (háblame)  21:04, 26 October 2010 (UTC)

File:Bigger Firefox Monobook WS.jpg
This jpg one I can’t read at all. Can you on your screen? Mattisse (talk) 20:46, 27 October 2010 (UTC)
 * According to firefox my


 * Window width: 1581px
 * Window height: 871px


 * Viewport width: 1563px
 * Viewport height: 699px


 * That is from the Web Developer addon. The regular window resize seems hard to find on Wikisource pages. Do you know why? Mattisse (talk) 21:22, 27 October 2010 (UTC)


 * OK, I figured it out. At 800 by 600 window size the text is readable. Mattisse (talk) 21:26, 27 October 2010 (UTC)

Thank you
Thank you for pointing out the layer buttons. I finally figured them out and they do indeed make a big difference. All this time, I never knew about "layer" options. The word "layer" to me means something else, like positioning, rather than margin size. (I wonder how many readers know about the choices?) Best wishes, Mattisse (talk) 15:53, 29 October 2010 (UTC)
 * You're welcome. Unfortunately the layout buttons are probably not well known.  Fortunately, for users reading on smaller screens, the standard small margins are probably not a big issue.  But it would be good to figure out a way to make these options more accessible to our readers. Maybe rework Help:Reading?  —Spangineer  (háblame)  17:23, 29 October 2010 (UTC)
 * It is hard to explain as it only appears while reading a "work" version (which I never do, so I didn’t see it). I had trouble even finding the option just now. I have no idea how readers come to Wikisource or what options they are presented with initially. Perhaps some explanation that "if" the layout option appears on the left margin "then" choosing layout options will manipulate the width of the text to read. Perhaps also explaining that once a layer option is chosen, then all "works" will appear in that format (if that is true). Mattisse (talk) 22:10, 29 October 2010 (UTC)

Talkback
I believe that what you desire is what I have been doing. — billinghurst  sDrewth  15:54, 3 November 2010 (UTC)
 * ack I do see that in some I had a combined two edit summaries (didn't turn one off when I swapped scripts), which was decidedly unhelpful and probably has lead to the confusion. Sorry about that. — billinghurst  sDrewth  15:59, 3 November 2010 (UTC)