User talk:Hesperian/Archive 11

The following text is preserved as an archive of discussions at User talk:Hesperian. Please do not modify it. Subsequent comments should be made on User talk:Hesperian. No further edits should be made to this page.

Hebrew

I did that one page, but I see that there is more (the book has a Hebrew back-title page - this works because Hebrew books open the other way!) and I will take care of that and the Hebrew interspersed within the book. --Eliyak T·C 13:11, 29 June 2012 (UTC)

Stats

Thanks Hesperian for that heads up. On that basis we have 40 very active editors rather than the 7 recorded by stats.wikimedia and I salute you as the most active! Is there any more convenient listing than Special:ActiveUsers?

I'm not sure it disproves my basic point that en:ws is under-volunteered compared with others. e.g. French have 32 v. active although they have only 1/3 the articles. It'd be ok if the articles were mostly tidy but my impression is that that is far from the case. And so far the "reform month" is conspicuous by its lack of activity. Chris55 (talk) 09:08, 6 July 2012 (UTC)

Gosh, I didn't even know about Special:ActiveUsers. Hesperian 10:53, 6 July 2012 (UTC)

Thanks Again

I think I remember thanking you last year, but I hadn't realized till recently just how many versions pages you had created for Mrs. Coates' poems. Still coming across more as I go through my to-do list... Thanks again for making my job easier! Londonjackbooks (talk) 21:08, 8 July 2012 (UTC)

:-) Hesperian 02:15, 11 July 2012 (UTC)

Disambiguation page, how similar is similar

Just created the disambig pages Sea Gull and Venice, and it is easy for those that disambig, though the question becomes how similar is similar for pages that might fit onto such a page. I have listed an example on each page of sort of like works, and would value your thought bubbles. Do we? Do we not? If we do, then how far? eg. Merchant of Venice?

We can probably add some words to Help:Disambiguation to clarify. Also as we have the search intitle box, that may be sufficient to direct people to for the purpose. — billinghurst sDrewth 11:35, 10 July 2012 (UTC)

Personally I woouldn't include either of those so-called "similar" pages. The advice at Wikipedia:Wikipedia:Disambiguation#Partial title matches sums it up for me, and I think it just as applicable here. The other way to look at it is, I wouldn't include an entry in a disambiguation page unless, in the absense of ambiguity, the disambiguation title would be the actual work title or a redirect to it. We would redirect from "Seagull" to a work entitled "A Seagull". We would redirect from "Seagull" to a work entitled "The Seagull". I don't think we would redirect from "Seagull" to a work entitled "Sea-Gulls at Fresh Pond". Hesperian 03:17, 11 July 2012 (UTC)

Also note that the disambiguation template embeds links for 'Search for titles containing or beginning with: "Sea Gull."' Hesperian 03:18, 11 July 2012 (UTC)

On the other hand, I would favour expanding this disambiguation page to cover orthographic variants of "Sea Gull": i.e. Mackellar's Seagull and Checkhov's The Seagull. Hesperian 03:21, 11 July 2012 (UTC)

Done +/-. And I am thinking that the disambiguation pages may be better off at the simplest term, sans A/The/..., as we start to add encyclopaedic refs, they will fall easier into the list, and if it is simpler, then the search function works more effectively.<shrug> — billinghurst sDrewth 16:13, 11 July 2012 (UTC)

I agree with having disambiguation pages sans A/The/.... How far does the ellipsis stretch? "My Seagull"? "Our Seagull"? I think I'd stop at A/The/An. Hesperian 01:00, 12 July 2012 (UTC)

Supplementary. Burial at Sea ←→ Sea-burial ? — billinghurst sDrewth 16:32, 11 July 2012 (UTC)

I wouldn't, personally; but we're now getting very close to genuine edge cases that are down to a matter of personal preference. Another case to be agreed upon is Bill Bloggs ←→ Bloggs, Bill and then what about William Bloggs, Bloggs, Bill B., etc? Hesperian 01:00, 12 July 2012 (UTC)

I didn't, and that is why it is always worth getting a second opinion. Note, that I would do the latter examples where we have encyclopaedic works, and due to the way that enWP has people listed, and how we followed for author naming, though I would fill up bits with redirects. The WHY is that we have not progressed Portal: pages for people, and we need a means to list multiple pages about the same person where they are not an author (ref. one of our historical arguments, somewhere, sometime). — billinghurst sDrewth 04:33, 12 July 2012 (UTC)

War of the Worlds

Thank you Hesperian. I still question the closing, but as long as it is still here, then everything is cool ;) - Theornamentalist (talk) 01:45, 11 July 2012 (UTC)

---

Hello. I request renaming my following accounts:

• محمد الجداوي → Avocato
• GedawyBot → AvocatoBot
• Confirmation link: [1]
• Reason: Privacy reasons

Please, unblock my bot. Thanks in advance.--M.Gedawy 07:45, 15 July 2012 (UTC)

August PotM works?

Hi, I've suggested for August that we return to the Sciences. WeeJeeVee has suggested a forest ecology work on the Oak. I don't think it will be long enough to cover the whole month. Do you have any botanical works that you've been wanting done? Either general or a single taxon would be fine. Beeswaxcandle (talk) 08:18, 18 July 2012 (UTC)

Beeswax candles actually existed (still do in colonial Williamsburg, Va.) and were used in colonial America. They're very soft and colored from yellow to brownish tints. How about a work on something that connects to this whether history or science? --William Maury Morris II (talk) 13:59, 18 July 2012 (UTC)

Hi, sorry for ignoring you. I've been busy and on holidays. No, I can't think of anything in particular. Hesperian 01:36, 25 July 2012 (UTC)

Vacation does that to people. How about, http://archive.org/details/cu31924003267170

Chapter I. Bible references to wax; — In mythology — Pliny's mention of wax — Bee-keeping in ancient times — Writing tablets — Uses of wax in the arts — Wax figures — Modelling in wax — Encaustic painting — Sorcery and witchcraft — Use by early Egyptians — Plato's reference to poisoning and sorcery — Divination and discovery of crime — Wax used for giving light — Pagan Rome and processional candles — Festivals of Saturn, Bacchus, and Ceres — Use by early Christians — Early use of candles in England — Artificial flowers and fruits — Votive offerings in Roman Catholic Church — Uses of wax in medicine; in embalming — Wax seals first used by Romans — Shakespeare's allusion to wax — The Worshipful Company of Wax Chandlers; its objects; its Normansell Cup and Beadle's Staff-head. —--William Maury Morris II (talk) 02:41, 25 July 2012 (UTC)

Blank edit pages

At Page:The Aborigines of Victoria and Riverina.djvu/11 if you edit the page, are you getting a black null image? I am getting it for all edit thumbnail of the work. <mumble grumble> — billinghurst sDrewth 13:41, 18 July 2012 (UTC)

Yes, black as a darkhole in space + "Error generating thumbnail

Error creating thumbnail: terminate called after throwing an instance of 'DJVU::GException' pnmtojpeg: EOF / read error reading magic number " -- --William Maury Morris II (talk) 14:03, 18 July 2012 (UTC)

Of course - the thumbnail averages 5100px by some just as crazy amount. Forced 1200px on the edit page and all is fine. -- George Orwell III (talk) 22:50, 18 July 2012 (UTC)

I have been seeing this all over the place lately; hence Wikisource:Scriptorium#javascript workaround for edit page image failure Hesperian 00:02, 19 July 2012 (UTC)

---

Thanks to all. It was too late for clear thinking. Does anyone know if it has made it into bugzilla? WMM where you seeing the error message, as I wasn't even seeing that. — billinghurst sDrewth 04:20, 19 July 2012 (UTC)

I still see them when I click on image to view the full image and nothing else. It shows on every image when I click on "image" to view the full image. But the image beside text shows now. The following will show you the specifics. I highlighted, copied, and pasted onto notepad and here are those specifics——

• Click on image to see full image and no text. Page|Discussion|Image

• A url is shown. http://upload.wikimedia.org/wikipedia/commons/thumb/e/e7/The_Aborigines_of_Victoria_and_Riverina.djvu/page11-5100px-The_Aborigines_of_Victoria_and_Riverina.djvu.jpg

• An error message is shown. Error creating thumbnail: terminate called after throwing an instance of 'DJVU::GException'

pnmtojpeg: EOF / read error reading magic number

Respectfully, (WMM) --William Maury Morris II (talk) 14:25, 23 July 2012 (UTC)

---

---

adminship

Hesperian, as the only bureaucrat I've corresponded with, can I ask what happens about my nomination for adminship? The guidelines say it lasts a week and then a bureaucrat acts, but nothing's happened. Chris55 (talk) 09:25, 21 July 2012 (UTC)

At least a week. We don't rush by the minute, you'll be right, it isn't going anywhere. This isn't enWP. — billinghurst sDrewth 11:47, 21 July 2012 (UTC)

You're right, it's less than 2 weeks - seemed longer. Chris55 (talk) 14:49, 21 July 2012 (UTC)

Hi, sorry, I've been on holidays. I will have a look in a few hours. Hesperian 09:19, 23 July 2012 (UTC)

Source files with substitutes

Hi, hope this finds you & your's well,

After monitoring your work on generating lists of Index: files and their status, I was wondering if its possible to identify another variant with the same type of list output.

I've been going over the Unknown, Issue with Source File and Needs an OCR text layer categories over the past week or so and have managed to better sort out the issues with the various files listed in those 3 cats - properly fixing a bunch of source files and [re]uploading them to Commons along the way.

Though I am at a loss to provide a specific example at the moment, I know we have several Indexes based on flawed or missing pages in their source files that have been "worked around" in some way not to reflect their true PR status indicating that such flaws or ommissions exist. Typically, this was accomplished in the past by substituting stand-alone images of said flawed or missing pages uploaded separately and then merging them in as needed in the final transclusion to the mainspace to make it appear as the source file is indeed without issue.

I am hoping you can think of away to identify and list such Indexes so that I can go back and properly merge/amend the substitute stand-alone files into the source file once and for all. Of course, the fixed files will require bulk moves in Page namespace and corrections to the pages command in the mainspace at the same time (I hope all the BOT jockeys lobbying to get the flag bit around here will pick up that workload slack but my single BOT request to do just that has gone unanswered so who knows if this entire project of mine is really worth it or possible.)

TIA for any comment, consideration, etc., you might have on this. -- George Orwell III (talk) 03:35, 1 August 2012 (UTC)

Hi George. Looks interesting. I'll explore it but it might take me until the weekend to find the time. Hesperian 04:00, 1 August 2012 (UTC)

Anytime is fine by me. There are plenty of other things to fix in the interim anyway. Prost. -- George Orwell III (talk) 04:02, 1 August 2012 (UTC)

Okay, this is what I've done. On the assumption that an index named Index:Blah blah blah.djvu ought to contain links into the page namespace only of the form Page:Blah blah blah.djvu/XXX, I grabbed a list of the outgoing links for each index.

• There are a great many indices that contain only "good" links. These are indices that correctly index a single multi-page file.
• There are many indices that contain only "bad" links; these are indices that cover multiple files, for example Index:"Polly". I assume these are not of much interest to you.
• There are a very few indices that contain both "good" and "bad" links. These, I thought, would be what you are looking for, but a brief perusal suggests to me that they are all false hits: each is problematic in some way, but not in the way we were searching for.

Oh well, here they are:

1. Index:Bancroftcopy-1.jpg
2. Index:Blisscopy-1.jpg
3. Index:Book of Mormon (1840).djvu
4. Index:Everettcopy-1.jpg
5. Index:Haycopy-1.jpg
6. Index:Nicolaycopy.jpg
7. Index:Popular Science Monthly Volume 32.djvu
8. Index:Popular Science Monthly Volume 74.djvu
9. Index:Sandbox.djvu
10. Index:Sandbox2.djvu
11. Index:Sandbox3.djvu
12. Index:The International Journal of Psycho-Analysis III 1922 2.djvu
13. Index:The International Journal of Psycho-Analysis III 1922 3.djvu
14. Index:The International Journal of Psycho-Analysis III 1922 4.djvu
15. Index:The International Journal of Psycho-Analysis II 1921 2.djvu
16. Index:The International Journal of Psycho-Analysis II 1921 3-4.djvu

I have another idea for something I want to try; I'll let you know how I go. Hesperian 08:07, 4 August 2012 (UTC)

Well that is sad (maybe I am crazy). The only one that matches what I was thinking of is PSMv32 @ the missing plate between p454 & p455. I'll wait to see if you can come up with something before I post a general request for these on WS:S. -- George Orwell III (talk) 21:45, 4 August 2012 (UTC)

If I were going to implement the kind of fix you are talking about, I would upload the extra images, create pages for them, and patch them into the mainspace. Would I also update the index? Maybe, maybe not. So possibly these fixes are out there but you can't tell by looking at index links. Next step is to look for page-namespace pages that aren't linked to by any index. Hesperian 23:42, 4 August 2012 (UTC)

Well I would undertake the work if and when such source Files are identified. Work arounds & patching defeats integrity. I feel one of the few characteristics that differentiates WS from other online entities is our integrity, attention to detail and overall high standards. I have to laugh at those knowingly transcribing a book that has missing/corrupt pages because that work is ultimately junk in the eyes of the new visitor no matter how much patching and 3rd party sourcing takes place to finish proofreading the work. I can't and don't want to control the behavior of editors who do not see things the way I do, but I can control the source files being uploaded and worked on. I was hoping to fix the files that fall into this slim category without getting into the debate over integrity or editing practices was all. I know one man's junk is another's gold as far as the subject matter goes - this is not what I'm talking about. But the more I dig the more it becomes apparent folks have uploaded lots of "junk" (seriously flawed or lacking .DjVus & .pdfs) for the sake of the title or author rather than for completeness or workability; abandoning the project rather than correcting the work soon after these issues are discovered. -- George Orwell III (talk) 01:08, 5 August 2012 (UTC)

Sorry, I think you've misunderstood me, which in hindsight is understandable because I was very unclear. I agree wholeheartedly with your comments on integrity. You are undertaking valuable work here, which is one reason why I'm keen to help you.

When I said "If I were going to implement the kind of fix you are talking about," I should have said "If I were going to implement the kind of 'fix' that you are want to fix properly, then...". That is, I was trying to put myself into the mind of the perpetrator, so as to better identify the signature of the crime. (But read "perpetrator" and "crime" as tongue-in-cheek, for I also agree with your attitude of simply quietly fixing the problem, rather than reforming those who see it differently.)

Hesperian 01:36, 5 August 2012 (UTC)

Hi George,

I submit for your consideration User:Hesperian/Sandbox, which contains a list of pages that are (putatively) not linked to by any index. Therein you will find oddities such as Page:Democracy and Education page 339.jpg (which is transcluded into Page:Democracy and Education.djvu/358), Page:Cyclopaedia of English Literature 1844 Volume 1 page 548.djvu, and Page:Origin of Species 1859 facsimile.djvu/79, as well as any number of false positives due to people using "empty" in pagelist calls. It looks like a fine mess to me, and I don't envy you your desire to wade right in.

(Note that the large number of reports from United States Statutes at Large Volume 122.djvu are presumably false positives caused by the fact that Index:United States Statutes at Large Volume 122.djvu has more than 5000 links and so hits the bot limit).

Hesperian 12:16, 20 August 2012 (UTC)

Thanks - I've imported the list to an User: subpage and will start whittling down the entries this weekend (hopefully). -- George Orwell III (talk) 13:36, 21 August 2012 (UTC)

Title 3 CFR

Hi again,

I see you've found my mess - please hold off creating any pages starting with Title 3 CFR, the DjVus themselves need to be edited for about 20 pages for each so any existing pages will wind up being offset by that number. Sorry - I'll get to them in the coming week.

Actually the files in question are from 1976 to present - anything before that is already done & OK to edit. -- George Orwell III (talk) 10:30, 5 August 2012 (UTC)

No probs. Hesperian 11:23, 5 August 2012 (UTC)

Oh dear, I forgot about this and posted a whole lot of blank pages. I hope I haven't caused too much inconvenience. Hesperian 14:08, 1 September 2012 (UTC)

meh.. its more my fault for not addressing these sooner so don't sweat it. Do what you have to do - at this point if I have to fix 16 Indexes instead of 6 won't make much difference in the grand scheme of "files to fix" anyway; that list just keeps growing :( -- George Orwell III (talk) 14:15, 1 September 2012 (UTC)

Usurpation reminder

Hi. This is just a gentle reminder about my usurpation request. The target user does not seem to have responded to the notice. Thanks a lot. --FarzanehSarafraz (talk) 03:28, 9 August 2012 (UTC)

Account delete(?) request

Hesperian, I would like to delete my SonjaNBohm alternate account, and was wondering what the best route to do so was. GOIII suggested that the Speedy delete route that I was pursuing was probably not the best method. Thanks, Londonjackbooks (talk) 22:03, 18 August 2012 (UTC)

Hi there. It isn't possible to delete an account. What we can do is rename the account to something that obscures your identity, and then, if you wish, delete the user and talk pages. You may then wish to scramble and forget the password. Unfortunately, you can't entirely put this cat back in the bag: there will always be old revisions of talk pages that contain your signature.

If you decide to proceed with this, all I need is for you to choose a user name and affirm that you want to go ahead with it.

Hesperian 00:15, 19 August 2012 (UTC)

No renaming is necessary, nor obscuring of my identity. If it can't be deleted, I'll just put the Talk page back as it was, and tidy up the User page. I knew page edits/history/signatures would still be around & had no issues about it/them. Thanks for the explanation. Londonjackbooks (talk) 03:36, 19 August 2012 (UTC)

I am still happy to delete the user and talk pages if you wish. Hesperian 03:52, 19 August 2012 (UTC)

Oh—if that can be done, then yes please. Thanks again, Londonjackbooks (talk) 03:55, 19 August 2012 (UTC)

I would suggest that you keep your name and embrace pride for your descendants of this and the next generations. Your name and photo is on Internet anyhow and before you came to WS (I think). I saw an image of a petite and pretty young lady embraced by a strong and young man--on a genealogy forum -- I think. I can find it again. Your works on Florence Earle Coates are superb and with that alone you have locked yourself and your name in good history so take pride in your hard-earned accomplishments instead of just being another anonymous alias whom nobody ever knows to give credit where credit is due in the future. Kindest regards, William Maury Morris II (talk) 01:26, 19 August 2012 (UTC)

No intent to hide my identity—just simplifying/consolidating, and Londonjackbooks is where the meat is—alias aside. Not looking for any credit, just happy to get Mrs. Coates' name out there. My family knows I'm proud of them, and there are other venues available for expressing those sentiments! Thanks, Londonjackbooks (talk) 03:36, 19 August 2012 (UTC)

I think I understand and it is true that you have done exceptionally well with Mrs. Coats (beautiful work!) on Internet, WP, WS, just as I have worked with Commander Matthew Fontaine Maury. I suppose we can rest with what we've done now--we have made them known once again. They are no longer "footnotes on the pages of history" just as my family's deeds of yore was once thrown at my face 20 years ago. Godspeed, William Maury Morris II (talk) 04:02, 19 August 2012 (UTC)

Flood setting?

With so many pages going through RC, would it be worthwhile setting the flood feature for your account? — billinghurst sDrewth 13:37, 19 August 2012 (UTC)

Can do. Hesperian 13:39, 19 August 2012 (UTC)

... and run one of those 5K API scripts you mentioned the other day while in that state - it would be nice to know if its possible for future reference. -- George Orwell III (talk) 14:05, 19 August 2012 (UTC)

Heh heh. Running now. Goodnight. Hesperian 14:11, 19 August 2012 (UTC)

It's working; I'm seeing numbers greater than 500 in the outputs.... Hesperian 14:23, 19 August 2012 (UTC)

Good to know. Also; using the ff-bit supersedes a recent wmf upgrade's changes to session timeout settings for inactivity. Under the bit I am once again able to upload very large files without the system falsely registering inactivity on my end and dropping the upload before it finishes. -- George Orwell III (talk) 00:26, 25 August 2012 (UTC)

Without text

Hi. I tried your "without text" that you posted and it used to work. Now it does not work any longer. Is it the same for you? I also noticed that also another function (Eyestrainer) in my commons.js which was relying on addOnloadHook does not work in Page: any longer (though it works elsewhere). Might the 2 things be related? Thanks--Mpaa (talk) 23:32, 23 August 2012 (UTC)

Far out, it is broken for me too! How annoying. I will try to diagnose it over the weekend. Hesperian 00:53, 24 August 2012 (UTC)

As far as the "Eyestrainer" mentiond above is concerned (a function created by Inductiveload to gray out background when editing), I noticed that it gives problems only in Page:ns and Main:ns of pages that transclude text. All other cases I tried are OK. So it looks related to areas related to the Proofread extension. Can it be that the two things stopped at the same time for the same root cause? Moreover, I have problems with the layout choice link as well. I use Firefox 14.0.1.--Mpaa (talk) 22:45, 25 August 2012 (UTC)

Fixed. I'm not sure what happened. It kinda looked as though firefox has subtly changed the way it builds a DOM out of html code... but on the balance of probabilities it is far more likely that Mediawiki has changed the page code in a manner too subtle for me to detect. Anyhow this new version should be a little more robust I think. Hesperian 03:36, 26 August 2012 (UTC)

Thanks. Don't know if the problems above were related but now everything is back to normal.--Mpaa (talk) 09:22, 26 August 2012 (UTC)

That would be because I edited your js file. ;-) Hesperian 11:32, 26 August 2012 (UTC)

MediaWiki:Gadget-Without text.js

I missed your update in the code to "Without text", and didn't see the fix. As more than one will be using it, and you haven't got it poked away as a separate file, I have created MediaWiki:Gadget-Without text.js. Not sure whether you want to move it to your user space, and leave the redirect. At this point in time, I am not sure that I want to have it as an open gadget as it has some power with its auto save, and may be problematic, though maybe we could add [rights=autopatrol] to restrict it to those. — billinghurst sDrewth 12:12, 1 September 2012 (UTC)

Eternity (Blake) :-) Hesperian 12:21, 1 September 2012 (UTC)

Ah, got it. Leave it there. — billinghurst sDrewth 15:50, 1 September 2012 (UTC)

Line breaking with {{gap}}

... moved to Template talk:Gap

Disambig and Versions

I cannot remember what we do when we have to do both disambiguation, and versionify. At Perdita I have done both on the same page, which is probably not what you have been doing. Happy for you to beat that page. — billinghurst sDrewth 12:08, 28 September 2012 (UTC)

As I had an interest in the page, I beat Hesperian to it :) Londonjackbooks (talk) 12:54, 28 September 2012 (UTC)

Yep, that's exactly what I would have done. Hesperian 13:28, 28 September 2012 (UTC)

Thanks (duh!). Making such decisions after a very long week tiring wasn't on the agenda. — billinghurst sDrewth 23:47, 28 September 2012 (UTC)

Federalist papers

When you deleted all the pages of the Federalist papers with titles such as this: The Federalist Papers/No. 67, you also broke all the Wikisource links from Wikipedia. Links from Wikipedia to Wikisource for the Federalist papers now just lead to a page where the reader is told that the page has been deleted. :-( V85 (talk) 21:06, 30 September 2012 (UTC)

:-( is right. The redirects were needed to reflect the "traditional" numbering being cited nowadays vs. the original publication's numbering. The Dawson edition was the easiest to work with for this while still remaining true to the content as published (i.e. the numbering is more of an academic thing). The "plan" was to eventually provide other editions (like Ford's) to somehow compare and contrast this nuance through some sort of disambig.

Is it possible to restore these through a script or something? Otherwise, I'll get on putting these back in the coming days. Thanks for any attention in advance -- George Orwell III (talk) 22:01, 30 September 2012 (UTC)

My bad; I will restore. Hesperian 00:35, 1 October 2012 (UTC)

Done. Hesperian 01:02, 1 October 2012 (UTC)

Western Australia climate

Hi,

You recently changed the status of Index:The_climate_of_Western_Australia,_from_meteorological_observations_made_during_the_years_1876-1899.djvu to needing an OCR for a text layer which isn't exactly the case. The original from IA created in 2007 added a text layer but opted out of running OCR on pages it couldn't "understand" (the tables) which was the norm until somepoint in 2011 where the folks at IA switched or tweaked their standard AABBY settings to better handle such nuances.

Still, the problem with this Australian work (and many others currently statused for lacking OCR) is that the default settings both at IA & at Any2DjVu can only handle around 4 to 5 columns of a simple table and about 3 columns of text at best before it all starts to fall apart and becomes recognized as one run on sentence. What we need is for someone to address these manually and set the AABBY reader software to expect columns or tables well beyond the defaults prior to running the OCR. The best that I can do is re-OCR these via Any2DjVu and those results are far less than the current best that I've seen from IA lately.

Now I don't mind redoing this particular file - you can judge if my last upload is any better - or any others you may find in the future. I'm not quite sure why the changes from the original in 2008 however. All the others (Gazettes & the like) with the OCR status I've tried to fix already with little to no difference. There are just too many columns per [newspaper] page for OCR to make heads or tails out of the resulting text. I'm at the point where I don't know what to do with them moving forward. -- George Orwell III (talk) 21:04, 3 October 2012 (UTC)

I like the way you take every index status change as a personal challenge. :-) Before I changed the status, I had a look through the uncreated pages and couldn't find a single page with OCR, so I assumed the text layer was completely empty. Possibly all the pages with OCR text have already been created. I've reverted myself; I think that was the only sensible thing to do here.

Well not so fast - I see you edited the original IA digitization (2007) several times in October 2008 & I'm not so sure you didn't make things worse in the process. Was thre any particular reason to stray from the University of California / Microsoft version? I only ask because I'm thinking re-doing the original still might be better than what we have now is all.

Background is at User talk:Moondyne/Archive 1#Index. I was only just starting to learn the joys of DjVu back then. It is entirely possible that in the course of "improving" the scans, I replaced good IA OCR with crappy Any2DjVu OCR. If so then I believe I know how to re-embed the original OCR.... Hesperian 01:32, 4 October 2012 (UTC)

Oh, and feel free to deal with the file however you see fit, including reverting to the original. I'm comfortable with the idea that I was doing stuff four years ago that looks like a foolish waste of time now; it means I've been learning.... Hesperian 01:39, 4 October 2012 (UTC)

Hmmm... too much going on there to make it worthwhile table-wise it seems. The IA PDF itself has the best hairlines out of all the ones on Commons including my last and that is what the OCR routines seem to look for to distinguish between table cells &/or text columns from one to another. Shame; it looks like the PDF has about as much of a text layer as the DjVu does - would be better to re-upload and re-convert the PDF on the sneak over on IA for the best possible text-layer under today's terms & not the 2007 regime. Unless this becomes a priority for someone anytime soon, I'll leave it on my local list for possible test runs and play around with it in Acrobat first when I get the time. -- George Orwell III (talk) 02:00, 4 October 2012 (UTC)

By the way, we should apply for a WMF grant to buy you an AABBY reader.

No - somebody should add the existing ability to OCR on the fly to Commons properly where it makes the most sense to have ('use external application to edit this file' appears on on every summary page for just such a reason). Don't get me started on utilizing XML over plain text dumps either....

Oh, and while I've got you, do you have the ability to OCR single pages? I have run across a djvu file in which the otherwise-adequate text layer is completely missing two pages.

Hesperian 00:25, 4 October 2012 (UTC)

Sure -- if I understand you right that is. If the text-layer appears on every other page in the source file then the pages in question were most likely inserted after the original OCR or, if these are consecutive pages, the boundaries between pages are jumbled up somehow and the text dump gets skipped for those pages because DjVuLibre gets "confused" where one ends and the next one starts (& vise versa). -- George Orwell III (talk) 01:20, 4 October 2012 (UTC)

The pages I have in mind are Page:U.S. Department of the Interior Annual Report 1871.djvu/16 and Page:U.S. Department of the Interior Annual Report 1871.djvu/24. Hesperian 11:45, 4 October 2012 (UTC)

Done & replied to your last on my talk page -- George Orwell III (talk) 20:18, 4 October 2012 (UTC)

Thanks! Hesperian 23:58, 4 October 2012 (UTC)

{{-}}

I set out to look into Robert Meni Lyon's "A glance at the manners..." and sort of got distracted. BTW, what is the point of this? [2] Moondyne (talk) 14:21, 12 October 2012 (UTC)

{{-}} creates a line space. There's a better explanation here—if that's what you were asking about. Londonjackbooks (talk) 03:45, 13 October 2012 (UTC)

The question was really why do we want a line space there? Its not a big deal but I know H. is deliberate in most things and so I wondered if it was something I'm not seeing. I thought it looked better without the break. Moondyne (talk) 04:45, 13 October 2012 (UTC)

It doesn't just create line space; it clears floats. Immediately above I have a float right, a float left and a center. If your screen is too narrow then the floats will wrap, and if that happens I want the government notice to be pushed down under the floats, rather than the float floating into it.

The template is particularly useful when someone posts an image in a discussion on your talk page, and a new, unrelated discussion starts underneath it. You don't really want that new discussion wrapping around the unrelated image. You want the discussion to push down to the next clear line under the image. That's what the template does. See for example User talk:Hesperian/Archive 9#Index:Irving Berlin Michigan.

There have been some big changes in how things render following a recent mediawiki update, necessitating a lot of template rejigging. I think it likely that line space wasn't there when I first saved the page.

Hesperian 06:18, 13 October 2012 (UTC)

Here I removed the {{-}} and as far as I can see Govt Notices behave regardless of how narrow my screen is. The only visible effect of the template in the above is to add an unwanted <br>. Moondyne (talk) 06:37, 13 October 2012 (UTC)

Couldn't the problem be resolved by combining the functions of float left, float right, and center? Isn't the {{running header}} template designed to do all this with far less clutter? --EncycloPetey (talk) 16:29, 11 November 2012 (UTC)

{{rh}} seems to work well. Moondyne (talk) 00:54, 12 November 2012 (UTC)

Index:Tracks of McKinlay and party across Australia.djvu

Not sure whether you are interested in this work that I have found and not really well mentioned in the modern corpus and not available elsewhere that I can find. It should be more widely available as it is about the search for Burke and Wills. Anyway, it is one that I will start upon soon, and just thought that it was worth sharing its availability. — billinghurst sDrewth 11:34, 18 November 2012 (UTC)

Very cool! I expect Moondyne will like it too. Hesperian 00:25, 19 November 2012 (UTC)

Yep, that does look interesting. Will have a look soonly. Moondyne (talk) 15:23, 19 November 2012 (UTC)

Question on pulling a specific report

Hey there,

I was planning on doing some housekeeping on our MediaWiki:Common.css file in the coming year and had a question or two. The file is roughly 40Kb now which is still twice the size of the most common .css file normally pulled from our servers. Over the years, much has been added while little has been removed - a good portion of it seems outdated, under-used or currently redundant.

I was wondering if I compiled a list of suspect classes & ids, would it be possible to ascertain what pages/templates are actually using these through some report so I can replace/ amend any usage that is wasteful or better served by consolidation. TIA. -- George Orwell III (talk) 22:50, 9 December 2012 (UTC)

Hi George, let me have a think about that and get back to you. Hesperian 00:18, 10 December 2012 (UTC)

Probably have to be done from a dump. I once asked the question of MZM and he said that we should try putting such requests on m:Tech. MZM is css clueful so may be able to assist us there if we are specific. — billinghurst sDrewth 09:10, 10 December 2012 (UTC)

Well there are some obvious things that can go (we updated the main page a couple of weeks ago and didn't use any of the classes or ids defined there for example) but I'm leary of just ripping things out that only I don't recognize for obvious reasons. I'd start a running list if I knew what was the best layout for ease of use/application by the techie folks for starters. -- George Orwell III (talk) 09:22, 10 December 2012 (UTC)

Dear bot

I think that you have forgotten to take off the hat. — billinghurst sDrewth 23:22, 9 December 2012 (UTC)

Oh! Hesperian 23:54, 9 December 2012 (UTC)

Not at all to be taken seriously: missing image tags

Are out to get me? I just caught this morning up with populating the last round of {{missing image}} pages in Index:Things Seen In Holland (1912).djvu, and you've already marked some more! (Yes, I knew I hadn't got them all, yet.)

Is this turning into some kind of competition? MODCHK (talk) 07:51, 28 December 2012 (UTC)

Yay, someone is actually processing them, I had no idea! I have a bold plan to deal with those missing images eventually, but at the moment I'm just cruising through tagging them and nailing blank pages. Hesperian 08:44, 28 December 2012 (UTC)

Well I grabbed the entire JP2 zipfile for this book off Internet Archive, and at last count I still have 36 images to process. (And regrettably I don't consider myself a great shakes with graphics at the best of times!) More fool me for starting on a book which is part of a series where every damn title is followed by the catchphrase "...With 50 Illustrations!" Cheers, MODCHK (talk) 09:27, 28 December 2012 (UTC)

I have a disk here with 550 of those zipfiles, totalling 133Gb. From them I have extracted 3473 page scans of {{missing image}} pages, totalling 29.7 Gb. Now I just have to convince the community to let me systematically upload all those images so that the {{missing image}} template can link directly to them... and then you won't ever have to download a JP2 zip again. Hesperian 09:55, 28 December 2012 (UTC)

Well I'm impressed, at least. If I may be so bold as to extrapolate from my own minor experience, the effortedium of separating JP2s-of-text from JP2s-of-wanted-plate-images is half the battle. I can only surmise you must have automated the process...but still! However, what is really bogging me down on this particular run is attempting to obscure the Univ. of California "tattoo" on every page. I get quite happy when I run across the rare image which happens to have missed the mutilated edge of the page.

Not that I personally look forward to the thought of more image manipulation, but as a concept I think directly linking {{missing image}}s to any kind of source image is a complete winner. Why anybody would want to oppose such a good idea completely defeats me. MODCHK (talk) 11:01, 28 December 2012 (UTC)

Yes, it is tedious, and I don't have a JP2 viewer, so for years I would convert every page image to PNG, and then delete the text ones; i.e. 99% of them. But, as you say, I've now automated that: the internals of the DjVu file format associate each page with a page "name", and archive.org have used the same internal page name as the page scan files. So if /242 is tagged {{missing image}}, then I simply pull the page name of the 242nd page out of the DjVu file, and then extract that file with that name from the zip. I don't have to unzip the whole thing, I don't have to manually search for the image, and I don't convert pages needlessly.

Those watermarks are bloody annoying. I am wondering: many of the archive.org files have both jp2.zip and jp2_raw.zip files. Perhaps the raw ones don't have the watermarks yet.

Thanks for the feedback. I'm likely to revive this proposal in the next month or so, so watch this space.... Hesperian 11:48, 28 December 2012 (UTC)

{{missing image}}

Slightly new topic: I saw your changes to the above template and on the whole entirely concur. However (and feel free to do the evil laugh if I've misunderstood something and/or are statin' th' bleeding' ovvius!), I can envisage a coupe of failure modes:

1. Doesn't this assume the Internet Archive project name will be kept all through the process, because I certainly did not with IA:thingsseeninholl00rochrich.djvu ==> Commons:Things Seen In Holland (1912).djvu...
2. Yet another opportunity to stuff up assumptions arises when pages are transcluded (once again, I changed capitalisation to match that used in the internal text and the case of WS:Things Seen In Holland (1912).djvu ==> Things Seen in Holland.

Now I'm not suggesting every project is as pig-headedly mucked around as this, but it does give the template an impossible task to guess what the potential PNG file file might be for any given {{tl:missing image}}-containing page... (In short, I doubt PAGENAME quite provides the right magic; the rest of the logic looks O.K. ― that is if my opinion can be trusted!)

Finally returning to the case of a tagged page being transcluded: is there any way {{tl:missing image}} can present a pointer back to the appropriate Page: namespace page; because (e.g. see Things Seen in Holland/Chapter I#59) occasionally when the page-number-in-frame-border gadget misfires (as it often does not for me) there is no hint how to navigate back to to correct page to fix the problem. If this is "too hard" and/or my attempted explanation is too unclear please don't bother; but just a thought if it is "simple" to add... MODCHK (talk) 09:52, 29 December 2012 (UTC)

Regarding 1 and 2, my system appears to be immune to these pitfalls, but I'll take this opportunity to talk you through it in case you can see problems that I have missed:

1. Start at a page in Category:Texts with missing images; e.g. Page:Things Seen In Holland (1912).djvu/65.
2. The corresponding DjVu file must be named File:Things Seen In Holland (1912).djvu.
3. Locate the DjVu file on Commons and download it.
4. List the DjVu file index. Note that the 65th page is internally named "thingsseeninholl00rochrich_0067.djvu".
5. Read the File: page on Commons. Search for an archive.org link. Found it: "http://archive.org/details/thingsseeninholl00rochrich". I've found that archive.org links always follow one of three patterns, and it is never hard to identify and pull out the archive.org name string i.e. "thingsseeninholl00rochrich".
6. Go to archive.org and follow the links to the jp2.zip file. Download it.
7. We don't need to extract the entire zip; the file we want is in there under the name "thingsseeninholl00rochrich_jp2/thingsseeninholl00rochrich_0067.jp2". The name always has this pattern and is always consistent with the internal name of the DjVu page; this can be relied upon.
8. Convert to PNG, because MediaWiki doesn't support JP2 files.
9. Upload under the title File:Things Seen In Holland (1912).djvu-65.png.

That's pretty much it. The script ain't pretty, but it seems to work. What do you think?

Regarding {{missing image}} providing a pointer, that would be great but I don't think it is possible.

Hesperian 10:51, 29 December 2012 (UTC)

Assumptions can really hang (well: me, in this case!)

Thank you very much for explaining the process in such painstaking detail. I am now quite satisfied my concerns are both moot; as I realise I had assumed a different order of events. In fact the more I think about it; the less likely my originally envisaged process order seems to be. I plead (I hope temporary) insanity.

Your process seems pretty solid.

With regards the pointer/back link idea; I really didn't have high hopes for it, but was rather hoping that was due to my ignorance. Sometimes it is a pain to find out I am at least on the right track.

I thank you again for your kind patience, MODCHK (talk) 11:15, 29 December 2012 (UTC)

Page:Things Seen In Holland (1912).djvu/1

I know, I know. I am being a complete PITA, because every time you create a nice example case in Things Seen in Holland, I seem to come along and spoil the party... Yes, I like what you are doing with (or should that be to?) {{missing image}}/{{raw image}}.

Please don't let my fiddling discourage you. You are in fact motivating me to concentrate on at least this task! MODCHK (talk) 06:58, 31 December 2012 (UTC)

Gosh and wow! :-)

Now permit me to reciprocate: when I'm fiddling with someone else's work, any change I make should be regarded as submitted for their consideration. If you prefer to do things differently, then please feel free to do so. Hesperian 07:22, 31 December 2012 (UTC)

I could have worded that better. What I meant to say was that I hoped we were at least cooperating with one another, even though I was concentrating on one little book; whilst you were addressing the larger mass issue. (Does that make it sound too "grand plan"? A term, I might mention of peculiar abuse in my family?)

I "think" I have now populated all the image pages in Things Seen in Holland, so should go back to good honest proofing for a while. I really don't think I am wired right for graphics tasks! MODCHK (talk) 08:05, 31 December 2012 (UTC)

restoring images is hard work... which I guess is ultimately why there are so many missing image tags here. It's nice to collaborate every now and then, isn't it? The wikis are supposedly collaborative spaces, but Wikisource is relatively isolating. We're all working on our own little projects, and it isn't often we cross paths. Hesperian 08:08, 31 December 2012 (UTC)

Restoring images does take time but it's not too difficult with the right programs. A lot of that time, work, and annoyance (although necessary) is placing them on Commons and filling out all required details including naming, source, description et cetera. I don't mind the image work except for what I have just stated. It's good that we all have our projects lest we have fewer editors here. Hesperian, the books you noted as "missing image" were books that nobody is working on as I posted a little earlier. Naturally all of the images are missing but so is the text in an abandoned book. Such books, in my opinion, should be the proofreads of the month -- or something along that idea. The present proofreads of the month are already lined up now to be done for a year, or perhaps that is something else but I don't think so. Those books not done should be for validation of the month before any new ones are introduced but this is not what happens. The problem I have seen with that is some editors just drop in and do very few pages and often near the end of the month so the person can get an "award". It is very clear that the "awards" are a motivating factor. Perhaps there can be created a different award for x# images done, or x# pages done for proofread of the month? I also believe some people bring in a new book and do nothing with it, or a little with it, and hope others finish it. These are my observations and my 6 pence and 12 shillings. Happy New Year to all, —Maury (talk) 09:29, 31 December 2012 (UTC)

Hi Maury. I love how structured, ordered progress can emerge out of what looks like utter chaos. All these contributors uploading works they won't proofread, proofreading random pages according to whim, tagging pages, adding metadata, writing scripts, issuing awards. You can't control them, and you wouldn't want to. You can't tell them what's important and what's not. You can't tell them what their priorities should be, or what ought to motivate them. And yet we — a motley collection of I in constant flux — produce beautiful things every day. Hesperian 12:22, 31 December 2012 (UTC)

Request

Hesperian, I respectfully request that this account be terminated to the fullest extent. Thank you and Happy New Year. Brother Officer (talk) 20:15, 29 December 2012 (UTC)

It is impossible to terminate an account. I can do any or all of the following: rename your account, indefinitely block you, and/or delete your user page. How shall I proceed please? Hesperian 01:31, 30 December 2012 (UTC)

I didn’t know an account couldn’t be terminated. I suppose there must be a good reason for that although I don’t know what it would be. Anyhow, none of the above is my choice. I had planned on creating a new account. This account has a lot of edits plus transcribed books so I will keep the same account name. I thank you and I wish a Happy New Year for you. Brother Officer (talk) 21:06, 31 December 2012 (UTC)

Perth Gazette question

Hi,

Back in 2008, you uploaded about 2 dozen of these four-page periodicals and had them statused as needing text-layers. Well I tried every online service (and all the tricks I know) to generate a layer for these and I keep coming up short.

The problem basically grinds down to the fact these were created at 100dpi and none of the latest OCR routines recognize settings that low (even Internet Archive won't run AabbyyFine on it unless its over 150 dpi and thats for PDFs not DjVus). I don't want to propose these for deletion without checking with you first to see if these can be replaced by PDFs straight from the AU gov't source or something (it would be even better to load them straight to IA for [re]precessing).

Otherwise, they are just taking up space if we continue to keep them statused that way. -- George Orwell III (talk) 11:21, 30 December 2012 (UTC)

I've just been fooling around with the source website and have figured out how to get much higher res images. But it is a pain. Now I need to figure out whether I can be bothered doing it. Can you give me a couple of weeks please? Hesperian 14:04, 30 December 2012 (UTC)

Sure... but I will hold you to that. Again, I don't know what that site is all about exactly (crashes my pooter while trying to load) but the real point here is to try and save whatever the range is as a single PDF file and let IA deal with the majority of processing it. We can split up the resulting output as needed. Don't know if you can see this from where you are or not but it clearly shows IA updates their routines far more than Any2DjVu.org does (every other month vs yearly if at all). -- George Orwell III (talk) 15:37, 30 December 2012 (UTC)

This remains something that I might get to eventually, but it has already been a lot longer than 'a couple of weeks', so I'll have no hard feelings if you run out of patience and nuke them. Hesperian 01:39, 11 February 2013 (UTC)

Is there a reason they can't they just be re-statused? Moondyne (talk) 04:50, 11 February 2013 (UTC)

Look, there is no "rush" at the moment to get these re-worked to a DPI that most freeware might have a chance of OCR'ing into a workable layer but the pile of "problem" files is steadily shrinking which makes that batch stand out more and more over time is all.

Sure, we can status them back as needing plain old proofreading (everybody else has taken that liberty to some degree, why not us?), but that won't change the fact there is little chance anybody will pick one up and make a dent in proofreading it from scratch.

So if we want to go down the road of fooling ourselves by shuffling them in and out of one status or state in the coming year, you can just count me out now. I'd rather get them up to par (one day but not today) and try to aid towards reaching that workable goal if I can, but at some point this "trash" will just stink too much to ignore any further and they will be gone from at least en.WS. at that point... tick-tock. -- George Orwell III (talk) 05:25, 11 February 2013 (UTC)

Forgive me if I misunderstood. I thought you were saying we should replace the djvus with pdfs (without text layer). I was wondering why we needed to bother. I will butt out now. Moondyne (talk) 06:49, 11 February 2013 (UTC)

Not at all - the point was that maybe 4 or 5 years ago it was a good idea to get, split and set a series like that up before it was really ready for consumption but thats not-so-much the case in today's terms & with the benefit of hindsight. I mentioned PDF only because it can handle odd page sizes such as a newspaper's and retain some ability to still be converted text-rich to DjVu at the same time. Many services are partially set up for that already (Google's electronic quilt of old newspapers is still Adobe/.ps friendly for example). -- George Orwell III (talk) 07:54, 11 February 2013 (UTC)

Pardon me, but...

... since when do we mark an entire Index: as faulty for just one more bad scan-image thumbnail ? -- George Orwell III (talk) 14:16, 1 January 2013 (UTC)

Apparently we don't. Thanks for the clue. Reverted. Hesperian 14:30, 1 January 2013 (UTC)

You can if, say, you intend to transclude the bulk of a proofread work to the mainspace as long as you let the community (or me) know about it specifically. But that wasn't the case here; this is just another grab-ass upload done to justify this or placate that - it will be months (if not years) before somebody who is really interested in the subject matter steps-in and actually works the file all the way to the end... & normally that is OK by me. But there are plenty of other things I could be fixing for folks who actually need resolution sooner rather than later; dealing with a dead end like that for nobody in particular is not the optimal way to try to schedule all that.

The other thing I need to mention is that statusing of Files needing an OCR Text Layer. Last week you were pretty much dead-on but at other times - not so much. I won't bore you with the details of how or why if you don't know them by now already but suffice to say it is near standard practice nowadays to visit a "misbehaving" file's main-page over on Commons, clicking 'Purge' one or twice from tab along the top while there, and then coming back to en.WS to see if the issue has corrected itself in about 10 or 20 seconds after your purging cycles through. An entire Index's sudden inability to pull the text-layer upon first-creation of one or more Page:'s in that Index: is just such a scenario that I'm alluding to. Thanks. -- George Orwell III (talk) 15:07, 1 January 2013 (UTC)

Okay, thanks for that. And my apologies for when it was "not so much", I don't want to waste your time. Hesperian 15:13, 1 January 2013 (UTC)

Do Pardon, but... ?

Hesperian, are we now replacing raw images where scanned images are to be placed? Example,

(Improve this image)

If not what is the purpose of the raw image? In looking over a book one could easily see a raw image and think it isn't but instead a scanned image. Apparently we cannot resize a raw image either, or rotate one. —Maury (talk) 13:55, 2 January 2013 (UTC)

It's a stop-gap measure. The raw image should be replaced by a proper image... eventually. But meanwhile we have several thousand missing images tagged with {{missing image}}, many of which have been like that for years. We could be showing raw images to our readers and instead we are displaying error boxes. If we can push half those {{missing image}}s over to {{raw image}}s, we'll be displaying to our readers 2000 images that we otherwise wouldn't have, and that has got to be a good thing. For example, have a look at Fasting for the cure of disease. Yesterday it was full of error boxes. Today it has pictures. The pictures aren't as good as they should be, but there's no doubt in my mind that the reader's experience has been vastly improved by displaying them. I made them display simply by replacing {{missing image}} with {{raw image}} on those pages where the page scan comprised a suitably oriented image. Hesperian 14:07, 2 January 2013 (UTC)

As for the issue of readers possibly thinking the image is a finished product, the idea is for {{raw image}} to display the image but also an unobtrusive indication that the image needs to be improved. See Page:A song of the English (1909).djvu/167 for an example (look closely). See how it works? If we really don't have an image, and the scan can't be used, we tag the image as missing using {{missing image}}. But if our raw page scan is adequate for illustration, we tag it with {{raw image}}, and the raw image gets displayed along with a gentle nudge to improve it. Hesperian 14:14, 2 January 2013 (UTC)

Thank you. I have been following your work and conversations and had already looked at the raw images. I just wanted a confirmation that they would not be permanent. When you stated, "It's a stop-gap measure", that was my full confirmation that I was understanding what you are doing. The idea is _absolutely brilliant_! I have not been impressed with an idea so much for a long time. It's incredible in replacing thousands of images of some sort. If it was your idea, my hat, if I wore one, would be off to you! Respectfully, —Maury (talk) 16:47, 2 January 2013 (UTC)

You're very kind. Actually it was George's idea. I just wanted to put high quality scans at readers' fingertips. The idea of using the better images as stop-gaps and thereby eliminating many error boxes — that was George. Hesperian 01:13, 4 January 2013 (UTC)

Any "idea" is forever useless until put into action. —Maury (talk) 04:26, 4 January 2013 (UTC)

Raw images after update

Hi. Should I delete the local WS file after update with reworked image? What if someone wants to improve the image even further after my work? Or maybe are you plannng for a bot to clean up every now and then?--Mpaa (talk) 20:16, 18 January 2013 (UTC)

I was thinking that the raw image should be deleted. You should feel free to delete, but I will also be running cleanup every now and then. Hesperian 01:18, 19 January 2013 (UTC)

Loeb Classical Library and the Greek text

Hi, what's your intention with this work (Index:Loeb Classical Library, L001 (1912).djvu)? All the even pages have the Greek text with the English on the odd pages. I'm not keen to deal with this much Greek through {{Greek missing}} and I wonder if the Greek text belongs on the Greek WS. Beeswaxcandle (talk) 04:37, 19 January 2013 (UTC)

If it's a multi-language work then it shouldn't be here at all; it should be on the original language-agnostic wikisource. Hesperian 07:10, 19 January 2013 (UTC)

We could probably export it and import it over there. — billinghurst sDrewth 03:09, 24 January 2013 (UTC)

It's not a "multi-language" work. It's a Greek text and an English translation text; the book format merely allows parallel comparison of the two texts. There should be no ideological problem with having the English text here, and some people will definitely come here looking for an English translation. --EncycloPetey (talk) 23:49, 24 January 2013 (UTC)

Not much different than what Zyephyrus is doing with Eleven Poems. See this page in edit mode. I'm not sure how it works, but it seems similar to the Greek/English setup of Loeb. I copied certain aspects of the Eleven Poems for Byron's Francesca of Rimini. Londonjackbooks (talk) 00:06, 25 January 2013 (UTC)

Encyclopetey, I am wondering on what grounds "a Greek text and an English translation text [formatted to allow] parallel comparison" is not a multi-language work. Hesperian 00:14, 25 January 2013 (UTC)

[Hesperian, for some reason, your response above doesn't show up in Recent changes, but it does appear in the history of this page. Londonjackbooks (talk) 00:23, 25 January 2013 (UTC)]

Sorry, that's because I'm mid-flood, and have the flood flag on. You would need to turn on "see bot edits". I won't turn the flood flag off and on again every time I want to post an message, so occasionally an message will come up as a bot edit. Hesperian 01:41, 25 January 2013 (UTC)

Aah... That makes sense. Thanks for explaining, Londonjackbooks (talk) 02:21, 25 January 2013 (UTC)

@Hesperian: Because there are two monolingual works in the volume. Neither work is multi-lingual; each work is written in a single language. It is merely the binding (not the works) that includes more than one language. --EncycloPetey (talk) 20:30, 25 January 2013 (UTC)

I think I understand your point of view, but I am unable to adopt it: I still regard this as a fairly straightforward case of a multi-language work that belongs on the multi-language Wikisource. However, this friendly conversation is the full extent of my interest in contesting the matter. ;-) @Beeswaxcandle: I agree, I won't be tagging any more of these pages with {{Greek missing}}. Hesperian 21:45, 25 January 2013 (UTC)

Categories and errors

I think that I saw recently that you were sneaky and had something that created an error message. My naïve search of mediawiki found nothing, and I am a bit time (and very tolerance) poor today. Cannot remember where you did, so do have a pointer to where you did some and where it may be documented in the MW blancmange. Thanks if you can. — billinghurst sDrewth 03:06, 24 January 2013 (UTC)

I thought error messages were commonplace. You are probably thinking of {{raw image}}, which whines if you don't tell it the page name. Hesperian 09:02, 24 January 2013 (UTC)

Comparative/Annotated/WS Translations (CATs)

Not sure whether you have been perusing WS:S and seen the conversation Wikisource:Scriptorium#Versions_guideline, though it may be a tad of TL;DR. Anyway, one of the discussion points was about the CATs (it was a point of slight confusion/misinterpretation), and one in which you reverted an attempt to return to a non-agreed position. As this is one of your particular points of interest, I am wondering where/when there may be some time in your schedule to look to see if we can have that discussion, and your thoughts to where, and to temporarily pull you from your comfy couch where you're patting cat. I would like to discuss the three components together(ish) as they are all where we are not just reproducing a work, but adding something that isn't there, s making it a little more than annotations. alone. — billinghurst sDrewth 13:20, 31 January 2013 (UTC)

Oh, I was unaware there was an ongoing discussion on annotations. I thought that whatsisname had simply appeared out of nowhere to revert a page simply because s/he was unhappy that no-one seemed to care that there was no consensus on it.

I have looked into the versions guideline discussion a couple of times, but saw nothing to demand my input or hold my attention. I did not notice the discussion point on comparisons / annotations / translations. If you wish, I can pop in there; or feel free to start a new, focussed discussion, and I will willingly pat that CAT instead. Let me know.

Hesperian 13:26, 31 January 2013 (UTC)

Noice. CATs was just muddying the water, and I believe that the appearance of the green flash was just coincidence. I will leave it for the moment, and put to bed the existing issue. Then breath, then stir the pot. — billinghurst sDrewth 13:41, 31 January 2013 (UTC)

I've been noticing the "Users looking for help" (in Recent changes) tag has not gone away for some time, so I looked to see what the topic was some time ago. It seems that the User's question would be at least tangential to the Annotations 'question'. Londonjackbooks (talk) 14:07, 31 January 2013 (UTC)

js problem

G'day mate. I hope you're finding some relief from the heat. Would you mind looking at my User:Moondyne/common.js and tell me why the Running header wont work? I'm out of my depth. Moondyne (talk) 13:48, 10 February 2013 (UTC)

Hiya, good to hear from you. All my js code stopped working a while ago, no idea why. I ended up nuking it all and haven't got it sorted out again yet. So I'm not in a position to fiddle with your code. But it looks like you've ended up with something I wrote a while back, so I'll have a shot at diagnosing it.

Do you see the specialFormats variable? It looks to me like you haven't defined that. The purpose of the specialFormats variable is to specify what you want the header to look like for particular works that you are working on. The code you are looking at checks whether you are currently working on a work that has a specialFormat, and if so, applies that specialFormat, and if not, applies a generic header instead. Without any specialFormat variable defined, it is crashing.

Long story short, you need to copy the contents of User:Hesperian/works.js into your js file, adapt it to your own works, and then you should be good to go.

If you can't make it work then let me know and I'll sort my js out and have a look... but I expect I wouldn't have time to do that until Saturday.

Cheers, Hesperian 00:39, 11 February 2013 (UTC)

Direct uploads

Hesperian, why doesn't your bot upload scans tagged with {{raw image}} (not {{missing image}}) directly to Commons? Advantage: one has to simply upload a new version of the file instead of having to build up the entire page with description, license, etc.--Erasmo Barresi (talk) 11:05, 11 February 2013 (UTC)

Hi Erasmo. I see your point of view. There are several reasons for uploading to Wikisource:

1. This is a Wikisource maintenance project, approved by the Wikisource community. I don't think it would be welcomed at Commons.
2. Even if it were welcome at Commons, I would prefer it to be here, under the control of this community.
3. When I mess up, I like to know I'm amongst friends.
4. In order to upload to Commons, I would have to make firm and clear assertions as to copyright status. I take copyright very seriously, and I am not prepared to run a bot that automatically makes copyright claims on my behalf. I am more comfortable to be gesturing vaguely in the direction of the djvu file and saying that the copyright status is presumably the same. The person who uploads the final version can be responsible for asserting copyright.
5. I am uploading all images in PNG, because this is a lossless format. PNG compresses diagrams, logos and some drawing well, but is terrible at compressing photos. Therefore many of these images ought be uploaded in their final form as JPG rather than PNG, which will prevent uploading over the top.

Hesperian 01:40, 12 February 2013 (UTC)

Taking you up...

... on your Pent. Pprs. IA-to-Commons offer.

This, as well as the following, week do not look promising enough for me to play with this on my own - better to get it off the radar any way possible and that means your offer (even if it is a week or more from now).

List is   From--tab--To

Pentagon-Papers-Part-IV-B-4.djvu	Pentagon-Papers-Part IV. B. 4.djvu
Pentagon-Papers-Part-IV-B-5.djvu	Pentagon-Papers-Part IV. B. 5.djvu
Pentagon-Papers-Part-IV-C-1.djvu	Pentagon-Papers-Part IV. C. 1.djvu
Pentagon-Papers-Part-IV-C-2a.djvu	Pentagon-Papers-Part IV. C. 2. a.djvu
Pentagon-Papers-Part-IV-C-2b.djvu	Pentagon-Papers-Part IV. C. 2. b.djvu
Pentagon-Papers-Part-IV-C-2c.djvu	Pentagon-Papers-Part IV. C. 2. c.djvu
Pentagon-Papers-Part-IV-C-3.djvu	Pentagon-Papers-Part IV. C. 3.djvu
Pentagon-Papers-Part-IV-C-4.djvu	Pentagon-Papers-Part IV. C. 4.djvu
Pentagon-Papers-Part-IV-C-5.djvu	Pentagon-Papers-Part IV. C. 5.djvu
Pentagon-Papers-Part-IV-C-6-a.djvu	Pentagon-Papers-Part IV. C. 6. a.djvu
Pentagon-Papers-Part-IV-C-6-b.djvu	Pentagon-Papers-Part IV. C. 6. b.djvu
Pentagon-Papers-Part-IV-C-6-c.djvu	Pentagon-Papers-Part IV. C. 6. c.djvu
Pentagon-Papers-Part-IV-C-7-a.djvu	Pentagon-Papers-Part IV. C. 7. a.djvu
Pentagon-Papers-Part-IV-C-7-b.djvu	Pentagon-Papers-Part IV. C. 7. b.djvu
Pentagon-Papers-Part-IV-C-8.djvu	Pentagon-Papers-Part IV. C. 8.djvu
Pentagon-Papers-Part-IV-C-9a.djvu	Pentagon-Papers-Part IV. C. 9. a.djvu
Pentagon-Papers-Part-IV-C-9b.djvu	Pentagon-Papers-Part IV. C. 9. b.djvu
Pentagon-Papers-Part-IV-C-10.djvu	Pentagon-Papers-Part IV. C.10.djvu
Pentagon-Papers-Part-V-A-Vol-IA.djvu	Pentagon-Papers-Part V. A. Vol. I. A.djvu
Pentagon-Papers-Part-V-A-Vol-IB.djvu	Pentagon-Papers-Part V. A. Vol. I. B.djvu
Pentagon-Papers-Part-V-A-Vol-IC.djvu	Pentagon-Papers-Part V. A. Vol. I. C.djvu
Pentagon-Papers-Part-V-A-Vol-IID.djvu	Pentagon-Papers-Part V. A. Vol. II. D.djvu

>> I hate Monday's :( << -- George Orwell III (talk) 07:49, 12 February 2013 (UTC)

Okey dokey, I'll tackle it this weekend. (They're not a particular favourite of mine, either.) Hesperian 00:24, 13 February 2013 (UTC)

What's that? I said I'd do it last weekend? You've forgotten to account for time zone differences. It's still last weekend here. Uploading now.... Hesperian 03:19, 23 February 2013 (UTC)

Done. Hesperian 03:52, 23 February 2013 (UTC)

Many thanks. With those now re-statused, we only have 61 files needing text-layer attention. -- George Orwell III (talk) 07:09, 24 February 2013 (UTC)

A ponder about music

Hi, we've now got three categories for missing music. Category:Pages containing sheet music, Category:Pages requiring musical examples and Category:Texts with missing musical scores. The first is populated by a template. I didn't know about it and set up the second when I started work on the Grove Dictionary of Music and you didn't know about the other two (I assume) when you set up the third in December. What's the best way forward from here? Beeswaxcandle (talk) 05:19, 13 February 2013 (UTC)

Pick one, and get us to align them accordingly. Surely they are all in templates so not a tricky beast. — billinghurst sDrewth 10:21, 15 February 2013 (UTC)

flood on?

? — billinghurst sDrewth 10:19, 15 February 2013 (UTC)

Ugh. I really am terrible at managing that so far. I leave it on when it should be off, then leave it off when it should be on. Sorry. Hesperian 13:16, 15 February 2013 (UTC)

A Song of the English

Hi, Hesperian. Question: All the text that appears in this Index (published 1909) appears also within this Mainspace page (part of larger work published 1896). Granted, the former has many images, and someone may come along one day who would like to finish the work for that reason alone; is it desirable then to keep a [redundant] work around for that reason? Thanks, Londonjackbooks (talk) 21:31, 19 February 2013 (UTC)

I should certainly be in favour of keeping both versions in this case. Hesperian 00:38, 20 February 2013 (UTC)

Ok, thanks. Londonjackbooks (talk) 02:14, 20 February 2013 (UTC)

Piggybacking

I have uploaded Index:Departmental Ditties and Ballads and Barrack-Room Ballads, Kipling, 1899.djvu to eventually supercede the two unindexed Kipling works we have hosted here, Departmental Ditties and other verses and Barrack-Room Ballads. The Indexed work covers all poetry listed on both pages excepting the poems "Recessional" and "The Vampire" from the former, and "Bobs" from the latter, and it also contains even more poetry in its "Ballads" section. I would progressively usurp all related (hosted) poetry pages (e.g.) to be incorporated into the Indexed work. Only one issue I have come across: Some of the poems hosted here are linked to this source, but that particular Index has not yet been proofread. Can I "usurp" those poem pages as well (e.g.), or should I create versions pages to include those unproofread versions? Thanks again, Londonjackbooks (talk) 13:55, 20 February 2013 (UTC)

I'm no authority on this but personally I would turn the mainspace pages into versions pages, and allow people to read whichever version they want. Even if the versions are identical. Partly because I like the way that versions pages function also as a publications history. Also because unless we are going to ban anthologies, it is inevitable that we will be hosting the same poem multiple times, so we may as well get used to it. Hesperian 00:43, 21 February 2013 (UTC)

Will do. I think there's only a few. Thanks, Londonjackbooks (talk) 01:55, 21 February 2013 (UTC)

How about if a poem is not linked or even related to a hosted Index, like this one, for example... I would want to supercede (usurp) it by linking (transcluding) it to the Index I am working on, but what to do with the corresponding Talk page material?

I like to think that if a work has no index then you should feel free to usurp. However I upset someone once by doing that — the work had no page scans but nonetheless had excellent provenance — and I ended up reversing my actions and apologising. Therefore I would advise to have a chat to Birgitte first. Hesperian 12:15, 21 February 2013 (UTC)

Agreed. Londonjackbooks (talk) 12:23, 21 February 2013 (UTC)

I did some digging, and the issue I referred to above was me overwriting authoritative Stallworthy editions of Wilfred Owens' poetry with first edition versions backed by scans. The first edition versions are known to contain errors that were corrected by Stallworthy, so the existence of scans did not justify me in simply over-writing them. Hesperian 02:02, 22 February 2013 (UTC)

"textus receptus"

Came across the above phrase the other day, where it is described as being "... the last collected edition to have the benefit of the poet's supervision..." (from the Preface of The Poems of Matthew Arnold, Longman's Annotated English Poets, 1965) Would that be something we should be concerned about here at WS? I was thinking about your comment with regard to anthologies, and it made me think about whether there was a way to find out which works (collection, etc.) an author 'authorized' in their lifetime. It would seem to go to author's intent. The volumes I am working on by Byron were obviously published after his death, but they are annotated so that you basically know everything about every line of his poetry you never wanted to know about (alternate versions, etc.)... Can we assume that anthologies and collections published in an author's lifetime have/had the approval of the author? I don't know how that works or worked (pre 1923). Londonjackbooks (talk) 17:27, 21 February 2013 (UTC)

I read a lot of classics, and I always read the editor's introduction because I enjoy the scholarship behind the editing. Sometimes it's the first edition that is considered 'authoritative' because subsequent editions were fiddled with by over-zealous editors, sometimes it's the 'textus receptus' (thanks for the terminology) because an unknown author could not prevent the first edition from being edited, but had more influence over later editions. Sometimes it isn't until after a author's death that the original manuscripts and/or galley proofs fall into the hands of scholars, allowing them to produce an authoritative 'author's cut'. This is why I like the idea of hosting all notable editions, rather than blessing a single edition with the label 'authoritative'. Hesperian 01:53, 22 February 2013 (UTC)

Thanks all around, Londonjackbooks (talk) 03:36, 22 February 2013 (UTC)

presentation copies

Another couple questions for you (off-WS-topic, if that's alright): I purchased a copy of William Blake by Chesterton. All pages are the same as this version, but mine has green covers. The title page is 'stamped' with "PRESENTATION COPY" over the title (not embossed or debossed, but cut through the paper—don't know what that's called). Would that likely be a publisher's presentation copy? There are no other markings but for a (assumed) bookseller's inscription in pencil of "Presentation Copy ?1920ish". (My questions—one implied—are italicized) Thanks, Londonjackbooks (talk) 20:25, 25 February 2013 (UTC)

(1) "perforated"? Londonjackbooks (talk) 21:17, 25 February 2013 (UTC)

I have no idea! This is very interesting. In stamp collecting, perforations are the punch holes along the edge of the stamp. If letters or a logo are punched into face of the stamp they are called perfins, for perforated insignia.

I've never heard of anything being cut into a book like that. And in book-collecting, a presentation copy is a book that the author signed and then gifted to someone.

Hesperian 00:40, 26 February 2013 (UTC)

Darn. Thought you might know. Maybe I'll try looking deeper into it. I was at least able to find a reference stating that the green cloth copy came before the "cheaper binding of light brown boards"... Whether it's a 1910, I don't know. Good book, though. Lots of good quotations/thoughts gleaned from it. Thanks for the info you could provide :) Londonjackbooks (talk) 01:42, 26 February 2013 (UTC)

nowiki your common.js

Would you mind <nowiki>'ing your common.js file, top and bottom behind commented lines, so it doesn't show in maintenance categories. Thanks if you can. — billinghurst sDrewth 13:57, 3 March 2013 (UTC)

Done. Hesperian 14:23, 3 March 2013 (UTC)

Annotations and derivative works

You commented on the proposal Derivative works on Scriptorium. I have closed this as generally supported and, as I said at the start, we now need to work out details. I have created Request for comment on annotations and derivative works if you would like to comment on this stage as well. I apologise, as I just did on Scriptorium, for the long winded nature of this page but it got contentious last time (as you know; you had to blank the policy page) and I thought this approach might work. - AdamBMorgan (talk) 22:30, 6 March 2013 (UTC)

Question on status Bad Page Scan

Hey,

When you apply {{bad page scan}} in the Page: namespace - is it just for pages with entirely and/or partially blurred, missing or clipped full-page images (regardless of any caption text being involved) as well as entirely and/or partially blurred, missing or clipped pages consisting of nothing but text?

Primarily, the reason I ask is (so far) in approaching a work with many instances of 'bad page scan' induced categorizations, giving the total of both no-images-just-text problematic pages and no-text-just-images problematic pages all tallied-up together in one Cat., looked like and probably is an useful starting point for what to address first for fixing, but in reality, it does not seem to be the optimal characteristic to base that choice on. When it comes to actually securing a suitable full replacement or several replacements for a per-incident substitution solution, further sub-categorization showing just one variant or the other separately would be more helpful (well at least for me it would).

In short - its one thing to work to replace a book with 15 blurry pages of text and maybe 2 missing full-page images and quite another to replace a book with 15 missing full-page images and maybe 2 blurry pages of text. Text has always been easier not only locate but to replicate as needed than suitable images are here, and I was just asking if there was a better way to help facilitate that through including a more targeted sub-categorization scheme while using that template or something. -- George Orwell III (talk) 07:12, 12 March 2013 (UTC)

Hi George. My criterion was "this page cannot be used to establish the veracity of a transcription." That is, at least some page content is missing or so corrupted that it might as well be. With respect to images, I wouldn't tag a page as bad simply because the scan was too poor to be used; it would have to be so poor that you couldn't even use it to check that a better image (if one could be found) was the correct image for that page.

As such, I should think the standard is much the same: it isn't "give me something legible" for text, and "give me a high-quality pic" for images; it is "give me something legible" for both.

Having said all that, if you're going to work on these bad page scans, I'll do whatever the heck you want. Would you like me to move to something along the lines of {{bad image scan}} and {{bad text scan}}? Hesperian 09:22, 12 March 2013 (UTC)

Your rationale and the logic behind it is sound and not lost on me. If it were just a matter of marking these pages to better indicate their place in the overall "completeness" scheme of an Index: pagelist, it would be more than enough. I'm looking more towards the future & what is it going to take to ultimately remove that Problematic page status regardless of whatever may be behind the cause for marking it with that Problematic status.

Unless I'm seriously mistaken, you haven't marked many pages with this template just for blurred, clipped or missing text-only pages - if any at all. So for now, it might indeed be wise to at least move to using {{bad image scan}} for pages with poor or clipped full page images in the scan along with missing full-page images in the scan (although a missing image with a clear caption still in place is frequently due to some loss or mismatch during the DjVu conversion process and its not certain to also be missing from the base PDF version built from the scans -- but without a fast rule-of-thumb to tell 'really missing' from just 'lost in conversion missing' we might as well lump poor or clipped images in along with missing images for sanity's sake if nothing else). I'll keep pondering this for better markers to differentiate the types of faults we see and where exactly they originate from in the meantime. -- George Orwell III (talk) 10:54, 12 March 2013 (UTC)

Reply

Thanks for the appreciation! :D --Wylve (talk) 16:12, 12 March 2013 (UTC)

Module:RawImage

Hello.

Pardon my being finicky, but would you please consider changing the comment line following "Usage:" in this module from:

    {{ Invoke:RawImage | rawimage | pagename }}

to something like:

    {{ #invoke:RawImage | rawimage | pagename }}

?

As there are so few Lua modules on WikiSource at present, idiots like me who copy from this code can completely confuse themselves as to how to later correctly invoke the module. (Yes I did... and was... and am!)

Cheers, MODCHK (talk) 10:05, 18 March 2013 (UTC)

Done. :-) Hesperian 12:28, 18 March 2013 (UTC)

Thank you. You, sir, are a gentleman. MODCHK (talk) 18:30, 18 March 2013 (UTC)

Aww, stop it, I'm blushing. Hesperian 23:46, 18 March 2013 (UTC)

Are these empty lines in {{center}} necessary?

Hello Hesperian.

May I propose modifying this template to remove two internal newlines which I believe to be nonfunctional? The current contents looks (in part) like:

<div class= "tiInherit" style="text-align:center;">
{{{1}}}
</div>

which I would suggest be modified to:

<div class= "tiInherit" style="text-align:center;">{{{1}}}</div>

I have just made the analogous changes to each of {{left}}, {{right}} and {{justify}}; but of course {{center}} is warded about with rather more heavy duty incantations (I refer to your "12:49, 24 June 2010 Hesperian (Talk | contribs) protected "Template:Center"‎ [edit=sysop] (indefinite) [move=sysop] (indefinite) (High traffic page: 20000 uses)".)

If I am incorrect and in fact these empty lines are desirable or necessary I would really like to know your reasoning if you would be so kind; and of course please either re"left", "right" and "justify" or let me know so I may do so myself.

I have been experimenting along with User:Londonjackbooks with an alternate method of formatting poetry, and my code just happens to trip up on those empty lines. (Obviously this is not a sufficient condition for mucking around with heavily used templates unnecessarily, and I may have to rethink my former approach if this situation crops up frequently.)

Regards, MODCHK (talk) 17:22, 22 March 2013 (UTC)

Hiya. See Wikisource:Scriptorium/Archives/2008-12#div needs to go on a separate line. Do what thou wilt, as long as that issue is mastered. Hesperian 01:07, 23 March 2013 (UTC)

Thank you very much. I am responding slowly because I am grinding my teeth. I thought I was addressing the exact same issue (and I was! just in a fashion which mucks up all templates which are formed per those rules...)

I'll undo the damage. Good to know! MODCHK (talk) 01:34, 23 March 2013 (UTC)

O.K. Changes to left, right and justify backed out. Now the question: can you foresee any circumstance where it is even possible to apply those templates across a transclusion boundary (i.e. template "opens" in one page and "closes" in another?) I can certainly understand this happening when the <div>s are hand coded. MODCHK (talk) 01:45, 23 March 2013 (UTC)

Heh, I had no idea I was dropping such a bombshell, or would have been more gentle about it.

MediaWiki's template handling is recursive but not re-entrant. It expands templates strictly top-down, once-only. It parses a page, recognises template calls, and expands them. It doesn't then go back and re-parse the result. Simple case study: suppose {{foo}} produces "{{" and {{bar}} produces "}}; now consider parsing the wikicode "{{foo}}foo{{bar}}". Mediawiki will parse that text exactly once, expand the template calls, and the result will be the literal text "{{foo}}". There is no provision for Mediawiki to feed that output back into the parser and go through the process all over again. Therefore that {{foo}} will not be recognised as a template call.

Once I came to understand this, back in the day, I gave up trying to get templates to span transclusion boundaries. Hesperian 07:30, 23 March 2013 (UTC)

... but Lua is a gamechanger. Suppose we didn't use template transclusion to pull together our pages into mainspace works. Suppose instead that in the mainspace we simply passed a page sequence to a Lua module whose job it was to pull the unparsed wikitext of each page, concatenate it, and feed the result into the parser. Possibly? Maybe.... Hesperian 07:37, 23 March 2013 (UTC)

Not really a bombshell. Well, not in of itself anyway. And as for your example {{foo}}foo{{bar}}. Eek, I see worms; but at least it demonstrates the cross-template-boundary situation is kind-of possible.

I am glad you clarified the one-pass expansion, because I could have sworn I'd read something about there being two passes. Must have been imagining it. (Maybe I'm thinking back to DCL/VMS days.)

Lua: have you any idea how close to this the "divify" code was going? (I can't choose routine names for toffee.) Anyway, unless a new user accessible (i.e. ruling out PHP) mechanism for launching scripts other than {{#invoke:}} comes into being, Lua stuff simply cannot be launched early enough for at least some template expansion to already have happened before control is transferred to the script. And when that happens: Interesting Times.

Thanks for the lessons and reality check. Just wish I'd cocked up earlier and saved myself just a little dignity. Ah well! MODCHK (talk) 07:58, 23 March 2013 (UTC)

What BS! What makes you think that you should have dignity? Where do you think that you are better than the rest of us? Are you aware the higher you lift your standards, the further that you have to fall? /me crawls back into his hole and pulls the rock back across the top. Oh, I came here as something is broken somewhere, not sure about for others, but I have lost toolbars, and regex tools, and I haven't changed any of my scripts this day on site, and to say that I have left a note at Promotable.js — billinghurst sDrewth 11:50, 23 March 2013 (UTC)

I've had stuff intermittently breaking or slow to load over the last two days. A lot of the js and css lives on meta and/or bits; presumably there has been trouble with one of those. Hesperian 11:55, 23 March 2013 (UTC)

Error identified as closer to home and being attached to the keyboard note to user that if you turn off js, turn it back on. — billinghurst sDrewth 11:59, 23 March 2013 (UTC)

Nice tie in with the dignity comment. :-) Hesperian 12:00, 23 March 2013 (UTC)

What can I say: he's a mining-town boy. You know they get coarse when they're feeling rich and relaxed: its something of a tradition or old charter with them. "Dignity" simply ain't in their vernacular; so I wouldn't expect a different response. 'Sides which I wouldn't put it past the cunning sod to have deliberately set up a prat-fall just to take the edge off. Respect? Who needs it? MODCHK (talk) 05:44, 26 March 2013 (UTC)

Raw Images and Mainspace

Hesperian, I have no idea as to how busy you are so please take my question with that in mind. As you travel about from page to page on books marking some with "raw image" could you also add the mainspace? I *think* that is what it's called. It is the area that allows a person to see all of the text and images at once vs page by page. That way it would be much easier (for me) to know where I am in a book and can spot my errors as well as fill in the blanks or needed changes. It is a process that Beeswaxcandle does on his works that I've noted as I validated his works. Respectfully, —Maury (talk) 17:13, 23 March 2013 (UTC)

Hesperian, I have works without images but when I can get someone to work with me on a book (as I do now) I then go to that book to work on those images as I am presently doing. I am fully aware I have images in books that I need and want to complete and I have not forgotten them. Those undone images I think of as being on the "back-burner" to complete when I am working alone with no specific book to work on with someone else who does 1/2 of the work -- usually the text and no images. It is approximately a 50/50 work like this. I will get to the images throughout The History of Yachting as soon as I can. It is one of my favorites on ships. But presently I have a partner and we are working on Central America. This book also has a lot of images. They only show as "proofread" when posted so the fact that there are many images within is not indicated. Again, I am aware of this. I just wanted you to know because you have been working hard with so many raw images making Wikisource better for everyone. Respectfully, —Maury (talk) 14:43, 24 March 2013 (UTC)

Thanks for that Maury. I am not under the impression that my own efforts entitle me to direct the efforts of others. You've been working very hard too. Continue to contribute as you think best; and if that means that some images sit on the 'back-burner' for a while, or indeed that you never clean up another image ever again, that's just fine. Hesperian 23:38, 24 March 2013 (UTC)

Thanks

Thanks for changing my username that fast. I appreciate it. ~ DanielTom (talk) 14:07, 24 March 2013 (UTC)

No probs. Hesperian 14:20, 24 March 2013 (UTC)

Formatting opinion

I am not familiar with these poems other than by word of mouth, and am seeking your opinion as to whether you think I should treat each poem individually by using {{block center|}} or by treating the lot of them as one long poem by using {{block center/s & e}}. The book is short enough that it will only use one Mainspace page, and so it is a matter of do I want all the poems to be block centered and inline? or each individual sonnet, etc. to be block centered in themselves. From what I can tell, the poems are thematically sequential (or however you would phrase that). Thanks, Londonjackbooks (talk) 23:24, 24 March 2013 (UTC)

There's no obvious way to proceed here, is there? I would have come up against the same question. I guess, as a personal preference only, I would treat each poem separately. The only rationale I can give for doing that is that each poem has a numerical heading and a drop cap. That tells me that they are much more than mere stanzas of the whole. Efforts have been made by the typesetter to mark them off as individual poems, and our layout should support that rather than undermining it. Hesperian 23:27, 24 March 2013 (UTC)

Forgot to end nowiki in my post, which made our signatures both render as ~~~~. Then, when I corrected it, "I" signed your post in your stead. Sorry... Londonjackbooks (talk) 23:26, 24 March 2013 (UTC)

Heh, no probs. Hesperian 23:27, 24 March 2013 (UTC)

Further, I have seen many, if not all, subsequent publications of the sonnets without the roman numerals (and drop caps). If I treat them as separate poems, should I then also create first line redirects to each poem? E.g., Long, long ago, when all the glittering earth would redirect to anchored poem I. Or not worry about it? Londonjackbooks (talk) 23:34, 24 March 2013 (UTC)

More info than you care to know (just finishing a thought), but Masefield's Good Friday and other poems (also 1916) contains most of the above sonnets in basically the same order, but some are missing, and new ones are injected throughout. No roman numerals/no drop caps... But I do believe that I should treat them as separate poems as you suggest. I think I will also create those first line redirects. I also came across one of the poems listed in a [non-Masefield] publication with a title. I will do as you suggest. Thanks for your input, as well as for being an unwitting sounding board! Londonjackbooks (talk) 00:41, 25 March 2013 (UTC)

I like how thorough you are; I would not have thought of creating redirects from the incipits. Hesperian 00:45, 25 March 2013 (UTC)

Workflow conflict

Hi,

Please refrain from creating temp .png files for Indexes that you've determined to have structural issues from now on. You created the .png's for Index:Through_hell_with_Hiprah_Hunt.djvu and then later determined that some pages & images were missing from position /132~/133 on forward. I fixed the structural issue by creating a fresh Djvu on IA from a recent digitization by Google Books. Now every image page (save the last two /145 & /143) is still detecting the .png files based on the old broken page progression 7 naming.

I realize all I need to do is move the File: to the new page number but I stopped because I realized the source for those .png files is now superseded by the new DjVu from GooBoo via IA as well. Its only about 6 or 7 images in question near the end of the source file in this case but I hate to think what would happen if a low /position number was missing then replaced - all these .png file names would be off by the number of missing/inserted pages made to fix the source file at that point.

Again: found structural issues with a source file? ---> Please no touchy or makey anything in related File:, Index: and/or Page: namespaces until resolved.

I uploaded those images weeks ago; I only paged out that index yesterday. As far as I can tell, the only way to avoid this happening again in future is for me to page out each index before I upload any images. There's a backlog of over 500 un-paged indices in User:Hesperian/Script. I'm picking the juicy bits out of that but I have no appetite for the mammoth task of paging them all out. So I guess that means no more image uploading at all.... :-( Hesperian 10:13, 26 March 2013 (UTC)

I guess not. Why adjusting the pagelist to insure completeness is not a strict requirement to create an Index: page in the first place is beyond me. The "I'll get around to doing that someday" or "let somebody else interested in the work do it" is, was and forever will be a joke of a rationale behind supporting to keep avoiding such requirements imho.

Delete the 500 Index pages? Its not like the ones really important to someone won't eventually get re-created. The rest are useless padding, no? -- George Orwell III (talk) 10:56, 26 March 2013 (UTC)

Heh, I don't think the community would approve that. I wish there were an index status for "Index not checked". We could make that the default status for new indices, and make it a rule that an index not be promoted to "To be proofread" until it had been paged out. It would be a trivial matter to demote those 500-odd index pages. I suspect the result would be an effective piece of social engineering: people would be motivated to get their precious indices promoted, and as a result a great many indices would get paged out. Meanwhile I would only upload images for indices that had been paged out. The big question is, is this possible for little old me / us, or does it require a developer? Hesperian 11:19, 26 March 2013 (UTC)

I'd rather see that than keeping doing what we've been doing until now. I think you can do it.

In the interim - should I just finish moving the .png files in that Index: ? Looks like Theornalmentalist is creating his own JPG files and putting them on Commons anyway.

... And what about the volume 1 & volume 7 of those flora-related Indexes you marked for structural issues? They seem to be your uploads -- got replacements? pointers? or should I just insert blank place-holders for now? -- George Orwell III (talk) 11:31, 26 March 2013 (UTC)

You can leave the png moving to me if you want. It's my problem and I'm sure you have plenty to be going on with.

Re Flora of Australia, I can get the pages. If they aren't at the Biodiversity Heritage Library then I know where I can get my hands on a hardcopy (facsimile). If I upload the page scans can I leave it to you to do the DjVu jiggery-pokery? Hesperian 11:54, 26 March 2013 (UTC)

That will work. I'll put blank place-holders for now to secure the proper page positions and for a contiguous numbering. When you locate the missing pages, I can just convert/swap them in for the blanks. You have to remember that I'm in the U.S. and my "search results" won't always show everything the rest of youse guys see as far as possible replacement sources go. Not much if anything at all in full view coming back for the Flora series for example. -- George Orwell III (talk) 12:27, 26 March 2013 (UTC)

I appreciate the work you've both done on the "new" status. Particularly as it will make it easier for me to find these Indexes. I was trying to find them through Special:NewPages and getting frustrated. Could I ask one of you to update the table at Help:Index pages#Progress? Beeswaxcandle (talk) 04:01, 27 March 2013 (UTC)

Done. Hesperian 05:27, 27 March 2013 (UTC)

The above text is preserved as an archive of discussions at User talk:Hesperian. Please do not modify it. Subsequent comments should be made on User talk:Hesperian. No further edits should be made to this page.