Wikisource:Scriptorium/Archives/2010-09

Announcements

Usability Initiative (Vector) Rollout Coming

I'm not sure if this is the right place to post, but I wanted to let people know that the Usability Initiative is continuing the rollout of Vector and the new editing features. The goal of this initiative is to make it easier for new editors to contribute to Wikimedia projects. To date, we've rolled out the changes to approximately 100 projects, including the 10 largest Wikipedias, Commons, and various "back-office" projects such as Meta. We are planning on completing the rollout for the remaining projects sometime in late August.

We wanted to reach out to the Wikisource community for help in identifying bugs and for general feedback about the new features. It would be very helpful for us to have folks test popular gadgets and other customizations on Wikisource for Vector compatibility. Information on how to test may be found on our FAQs page. Please post any bugs either in Bugzilla (please file under "Usability Initiative") or on our bug report page on meta. We also welcome general feedback on our feedback page. We'll be monitoring these pages and look forward to your feedback. Howief (talk) 22:19, 6 August 2010 (UTC)

Wikimedia UK Microgrants

Wikimedia UK has recently launched a Microgrants process - grants of between £5 and £100 to support Wikimedia editing or outreach. There's a really obvious use for these here - if there is an out-of-copyright book that you want to make available here, which can't be found on archive.org/google/etc. but copies of it are available for purchase, then a microgrant can be used to buy a copy. A related possibility is to purchase a scanner to scan in books that you already have. Those aren't the only possibilities, of course. :-)

The microgrants are designed to be quick and easy to apply for - the process is described here, and you can apply online here. You do need to be a member of Wikimedia UK to apply for a grant - but this only costs £5 per year (and you can apply online). You'll also need to make a brief report (1-2 paragraphs) once you've used the grant/completed the project.

I hope that you find these microgrants useful. If you have any queries, please let me know. Thanks. Mike Peel (talk) 09:58, 14 August 2010 (UTC)

Proposals

Link to a whole text version

Hi all,

What about something like that: a whole text in plain text format link in the header? Of course the page's title wouldn't be Zyephyrus' test but a more appropriate one. Users who ask for a whole plain text might either copy and paste this text or save it. Would there be any inconveniences if we did that? --Zyephyrus (talk) 09:41, 14 August 2010 (UTC)

Continuous scroll display

In case I didn't spam the mailing list enough, I wrote a "Google Books"-like scrolling thingy (to use the technical term :-) like so. it.wikisource already uses a template (blue reading guy icon under "Sommario") that links an index to the corresponding toolserver display. The tool still needs lots of work, but IMHO (POV!) it's already pretty neat. There's even "search in this book" functionality. Would there be interest in linking the tool from wikisource index pages, "main namespace" pages, or even individual pages? I'll be happy to accommodate feature requests where possible :-) --Magnus Manske (talk) 08:25, 17 August 2010 (UTC)

Thanks. I was bold enough to add the icon to MediaWiki:Proofreadpage_index_template, (even though it will fail on non djvu indexes).

Perhaps it would be nicer to display it in an action tab

ThomasV (talk) 09:33, 17 August 2010 (UTC)

Cool! Tab might be nice, as it fails to show on some index pages, e.g. Index:Romeo and Juliet (The Illustrated Shakespeare, 1847).djvu. --Magnus Manske (talk) 09:47, 17 August 2010 (UTC)

I have created a template under User:Magnus Manske/Book2scroll that will work on both "Index" pages (linking to page 1) and "Page" pages (linking to the respective page number); no need to fill in any parameters (though one still could). --Magnus Manske (talk) 10:10, 17 August 2010 (UTC)

Author page creation

If people are interested I can install a new version of the javascript helping author page creation (gadget Preload useful templates such as). You can test it by adding a

importScript('User:Phe/Author fill.js');

to you monobook.js or vector.js. This function conflict with the gadget so you must disable it through your preferences. There is two difference with the old code, lastname/firstname/lastinitial are filled automatically, works only for the most common case. The script also check for the existence of a page with the same name on en:wp: and fill the wikipedia_link field, and also the birthear/death field if the relevant cat are present in en: article. I already know it works with firefox, down to 2.0 and the latest version of Opera and Chrome, older not tested. It breaks with IE 6.0. Phe (talk) 12:03, 21 August 2010 (UTC)

While I have used and like the Author page creation that you have set up, I do find it problematic that I now have the option of one or the other. I would suggest that we may be wanting to nullify the Author component of Mediawiki:Gadget-TemplatePreloader.js so that we can have the best of both worlds. Would it be possible to insert your script so that it calls your script and leaves the main ns header completion intact. — billinghurst sDrewth 14:58, 21 August 2010 (UTC)

Sorry, I was unclear, I planned to replace only the author page creation code, and let the other template filling as it, using User:Phe/Author fill.js is only a temporary solution for people wanting to test it. Phe (talk) 15:10, 21 August 2010 (UTC)

Added to the gadget, for any trouble with the author page creation, ping me on my talk page or on IRC. Phe (talk) 12:21, 23 August 2010 (UTC)

Other discussions

Upgrading HotCat

On Commons, HotCat has made it to v 2.2. I was wondering if anyone we could upgrade the version we've got here to match Commons' new version. They add some important functionality, like the ability to edit multiple categories without having to save each time.—Zhaladshar ^(Talk) 14:12, 28 July 2010 (UTC)

Look up ^^^ — billinghurst sDrewth 08:49, 29 July 2010 (UTC)

Dang it! How did I miss that? Thanks!—Zhaladshar ^(Talk) 12:17, 29 July 2010 (UTC)

As the original announcement has now archived, and there has been no complaints, I have presumed that all is okay and I have moved the test version to be the main production version. — billinghurst sDrewth 08:11, 14 August 2010 (UTC)

That is incorrect, objection and void. cygnis insignis 09:08, 14 August 2010 (UTC)

Questions

Header templates

Time to add a splash of colour to this page! On the tail of the introduction of the Portal: namespace, I think we need a {{Portal header}} template which is better suited to the purpose than {{process header}}, which is used for Wikisource operations pages like this one.

In addition, I want to:

• Change {{process header}}'s colour, as it is too similar to {{header}} for my liking. A suggestion is made below as to the colour, but feel free the change it, it is just the first muted colour that I thought looked OK. I think that every namespace's template should be unique to that namespace to instantly provide a cue as the purpose of that page (eg, green->content, blue->portal, etc).
• Make all four types of header take the class from common.css. {{Header}} and {{author}} do this already. {{Process header}} and {{portal header}} should do this too.

Summary of templates

Header (existing)

Header
Background: #E4F2E4, Border: #ADA; (taken from common.css)

Author (existing)

Author
Background: #E4D8D8, Border: #BEA2A2; (taken from common.css)

Process header (existing)

This is similar to header, which may be confusing, and the border is a little bit too bold when compared to the other templates.

Process
Background: #DFD, Border: #006400

Process header (proposed)

Process
Background: #DCA, Border: #966

Portal (proposed)

Portal
Background: #DDE, Border: #36A

I have created {{portal header}} as a provisional template for people to play with. However, consideration needs to be given to the structure of the namespace. I see it as a tree, similar to categories, with a portal potentially having more than one parent. For example: Portal:British Museum belongs to Portal:Museums and Portal:United Kingdom. For the time being, while we hash stuff out, it has one "parent" field. Inductiveload—talk/contribs 00:17, 1 July 2010 (UTC)

As an aside, it seems very odd to be dropping font-size to 0.9em in the mainspace header, but not in any of the others. We probably need a generic-header class that they all use, and specific classes that set the background colour and other differences. Hesperian 01:54, 1 July 2010 (UTC)

• Support these changes; I like the concept of different colors for different types of pages. I agree that font size should be consistent across the board. —Spangineer^wp (háblame) 14:18, 14 July 2010 (UTC)
• I changed the process header, colour schemes are bugging me today. It is, perhaps, a bit dark, but less nauseating. Cygnis insignis (talk) 06:09, 10 August 2010 (UTC)

Bot Flag for LcawteBot

Hello there. I'd like to request a bot flag for LcawteBot (talk • contribs). Here's a quick overview of what I'd like to do with it:

• Tasks -
  • Clean sandbox 12 hourly or daily (I personally will go for 12 hourly, although will be happy to go with daily if decided here.)
  • Fix Double Redirects. (Daily or every 3 days, depending how busy I am in real life. May do it with cron if I move to toolserver.)
• Framework and Scripts - Pywikipediabot using script clean_sandbox.py
• Previous Experience - I often run pywikipediabot scripts on Brickipedia, a Wikia wiki, including the script that fixes redirects, I have used clean_sandbox.py on their before, although the site is not active enough/does not use a shared sandbox, so the script cannot be run there.
• Operator - Lcawte, a fair amount of experience running pywikipediabot on a group of wikis.
• Other details - I am going to be running the clean_sandbox script unsupervised, as far as I am aware, as my schedule does not allow me computer access at each 12 hour interval during the day. The redirects.py script I will be running when I am alert at my computer.

Thanks, --Lcawte (talk) 17:03, 11 July 2010 (UTC)

If requested, the bot can also do one off batch category changes say from Foo -> Bar. --Lcawte (talk) 17:20, 11 July 2010 (UTC)

Hello! If LcawteBot is looking for tasks, check out these unassigned bot requests. Cheers, stephen (talk) 01:07, 16 July 2010 (UTC)

I'm just waiting for the bot flag before I start running things.. --Lcawte (talk) 06:28, 20 July 2010 (UTC)

Typically, a bot does a preliminary run at a high interval (60+ second throttle) for a period before the bot flag is granted. Check out WS:BOTS#Community authorisation for more information. stephen (talk) 23:33, 27 July 2010 (UTC)

Hm, forgot about this.. going to run the preliminary run some time this week, when I get round to re-setting the bot up. --Lcawte (talk) 14:05, 20 August 2010 (UTC)

Bot flag for Ryuchbot

Dear editors, I found WS:Bot requests is not the place for bot flag, but this forum is for the flag. Would you please review my request for Ryuchbot:

• Tasks -
  • semi automation for importing OCR texts
  • to delete '\n' and hypens and insert '\n' for new paragraphs for a Page for djvu files.
• Framework and Scripts - Pywikipediabot using script basic.py
• Previous Experience - I am running Request for Comment bot for ko.wp on Toolserver
• Operator - Ryuch, I coded RfC bot for ko.wp with Pywikipediabot.
• Other details - It would be occasional for import. It will run on my computer, the load will not be heavy. At this time, I will change just around 500 pages for once.

happy sourcing! --Ryuch (talk) 00:27, 21 July 2010 (UTC)

no objection or no agreement? --Ryuch (talk) 14:47, 4 August 2010 (UTC)

What is the account for your bot? Before we grant it a flag it usually goes through a probationary period where it performs edits (at a 1min throttle or more) so its edits can be seen by the community and evaluated for any errors. If it passes and the community is happy with it, it gets the flag.—Zhaladshar ^(Talk) 14:56, 4 August 2010 (UTC)

As I requested above, User:Ryuchbot. Thanks. --Ryuch (talk) 15:22, 4 August 2010 (UTC)

searching across page breaks doesn't work

I just discovered the somewhat disturbing fact that our present system of page transclusion makes it impossible to search for phrases that cross a page break.

Go to William Blake, a critical essay/Lyrical poems. Search for "clamour of envy and stupidity". You'll find it, because it appears in whole on Page:William Blake, a critical essay (Swinburne).djvu/141.

Now look at the preceding words, "the conventional". You can see the phrase "the conventional clamour of envy and stupidity" on your screen. So search for it. You won't find it.

It gets worse. Google "clamour of envy and stupidity" site:en.wikisource.org. Two hits, one for the page, one for the mainspace transclusion. Okay, now Google "the conventional clamour of envy and stupidity" site:en.wikisource.org. Nothing. Google can't find Wikisource phrases that cross page boundaries.

Hesperian 01:01, 4 August 2010 (UTC)

I view this as a sub-set of bugzilla:18861 which discusses the lack of indexing of transcluded pages, and instead the indexing of the underlying Page: ns. The Wikisource community should be pushing hard on this particular buglet and asking for an improved indexing that works with transcluded pages. We are such a passive lot. — billinghurst sDrewth 03:02, 4 August 2010 (UTC)

No, this is an unrelated problem. When I say "Search", I'm talking about doing a <ctrl-f> find in your browser. The text "the conventional clamour of envy and stupidity" will be right there in front of you plain as day, but a <ctrl-f> find will not find it, because in fact the page text is "the conventional [125] clamour of envy and stupidity", with the "[125]" floated across into the left margin. In fact Google does index the transcluded page, as can be seen here; but it still won't find anything for the phrase "the conventional clamour of envy and stupidity", because we are jamming a "[125]" into the middle of it. Hesperian 03:20, 4 August 2010 (UTC)

Google Scholar has the same issue with CSS floats and page numbers: [1]. In my personal opinion, Lucene's inability to search transcluded text is a bigger issue here, where nearly all the content is built through {{page}}. stephen (talk) 03:51, 4 August 2010 (UTC)

It's been the same search-result story, to the best of my recollection, with the application of sidenotes. Search results with the word before embedded sidenote; -- without the sidenote. Does anybody else see the same? George Orwell III (talk) 04:59, 4 August 2010 (UTC)

MediaWiki:Proofreadpage_pagenum_template causes this.

According to Google, you cannot prevent a part of a page from being indexed [2]. They suggest to use javascript in order to achieve that.

Concerning firefox's control-f, it tolerates an empty span : "the conventional clamour"

This means that we can use javascript to render the page numbers and page links; the needed parameters should be hidden in the id of the span

ThomasV (talk) 04:21, 4 August 2010 (UTC)

Thankyou for taking this on, Thomas. I see you have removed the page numbers from MediaWiki:Proofreadpage pagenum template. I'm guessing you're working on the Javascript now? Before you get too far into it, I think that, strictly speaking, ids are not supposed to start with a numeral, though apparently all browsers support such ids. Still it might be an idea to prepend a character. Also when you're ready Template:Page will need to be sorted out too. Hesperian 04:46, 4 August 2010 (UTC)

I made a few tests. A minor problem is that page names encoded into ids cannot be unambiguously decoded into URLs. This could be fixed by modifying the extension, to have it pass the correct page URL to the pagenum template. Another problem, which is more serious, is that once page numbers have been rendered with javascript, Firefox's search does take them into account. Thus, the proposed solution would only fix Google's indexing. In order to fix Firefox's search, we could make the display of page numbers optional (this will be complicated for users), or we could try to render page numbers in a separate div (for each transcluded page, read its vertical offset and report it in the div where page numbers are displayed; kind of complicated too, but feasible). ThomasV (talk) 06:56, 4 August 2010 (UTC)

Ok, I went for the javascript solution. This fixes the control-F search with firefox, and the Google indexing. With this option it is also possible to display pagebreaks (for example, I added a function that draws a transparent box over the page when the mouse is passed over a link). I guess some debugging will be necessary ; the page numbers might be misaligned with actual pagebreaks if any other javascript interfers with the page. ThomasV (talk) 12:04, 4 August 2010 (UTC)

Positioning and highlighting are broken at http://en.wikisource.org/w/index.php?title=William_Blake,_a_critical_essay/Lyrical_poems&oldid=1936145. I'm not sure whether this is because of it is an oldid page, or the incomplete template, or something else...? Hesperian 12:27, 4 August 2010 (UTC)

And the hover seems to be masking the click? Clicking on a page number does nothing, but the link is definitely there, because if I right-click, I get a fully functional context menu that allows opening the link in a new tab. Hesperian 12:31, 4 August 2010 (UTC)

Thanks, seems to be working well now. The only issues I see are: (1) highlighting finishes at the actual page break, but starts at the beginning of the line the page break is on; and (2) there is an extra space inserted at each page break. Hesperian 23:39, 4 August 2010 (UTC)

Within the body of a work it overrides the gadget Wikipedia:Tools/Navigation_popups, such that any link with the body has no pop-up facility. If it is the greying that is the issue here, I so no specific need to identify the page line by line where it starts/ends. If that is seen as a feature, can we then have it as an alternate gadget rather then the overriding default. — billinghurst sDrewth 00:27, 5 August 2010 (UTC)

I need to investigate why nav popups do not work with these links, but I do not think it is related to the highlighting. It could be because the links are created with javascript ThomasV (talk)

hmm, I was wrong about that ; I have partially fixed popups, in the sense that they work until the first highlighting action is done. so there is definitely an interaction between popups and highlighting :-) ThomasV (talk) 07:05, 5 August 2010 (UTC)

ok, this is fixed : it works well with popups now ThomasV (talk) 07:51, 5 August 2010 (UTC)

←With this update, the behaviours of Template:Page and <pages /> are now pretty divergent. Will Template:Page be caught up, or made depreciated? Prosody (talk) 20:40, 6 August 2010 (UTC)

• I updated Template:Page ; however, this template is deprecated ; you should only use if if you have a good reason (and references spread over multiple pages are not a good reason). ThomasV (talk) 17:56, 9 August 2010 (UTC)

• As discussed below, fixing this problem seems to have introduced a new one that affects hundreds of pages in the main namespace. Tarmstro99 (talk) 20:27, 9 August 2010 (UTC)

I love the new functionality where if you mouse over the page number in the left margin, it highlights the corresponding text that was transcluded from that page! Makes it very easy to spot exactly where pagination occurred in the original source, which can be very important when citing to a work. Bravo! Tarmstro99 (talk) 14:05, 10 August 2010 (UTC)

Bot flag for User:BotMultichill

Hi guys, looks like the Wikisource authors can use an interwiki update. I would like to run the interwiki bot here. My bot is flagged on a lot of wiki's and a global bot. I would like to have a botflag so I can run here too (and not flood your recent changes). I already did a couple of test edits. Multichill (talk) 18:29, 10 August 2010 (UTC)

Looks okay to me, and great that you are bringing your toys here. Appreciated. Care to give a little background or link to similar operations, as per Wikisource:Bots. I would also recommend that you give it a burst of (throwing out a figure) 50 pages, and I think that it would be sweet. We probably also want to look at borrowing your bot so we can transfer legacy files to Commons, if that is okay with you. — billinghurst sDrewth 04:32, 11 August 2010 (UTC)

Background? I run lot's of bots. Have of 3 million edits, see my contributions for examples. I'll do some more edits here.

If you have free files here than these should be moved to Commons. Just tag them with {{Move to Wikimedia Commons}}. Multichill (talk) 16:11, 12 August 2010 (UTC)

Did some more edits. Multichill (talk) 22:25, 22 August 2010 (UTC)

Looks totally fine to me. Note to 'crats: from my viewpoint I am happy for a bot flag, checked last 20 edits for zero errors. — billinghurst sDrewth 12:50, 23 August 2010 (UTC)

URL upload bot

Everybody, I'd like to introduce the newest member of the team, courtesy of Magnus Manske—the URL2Commons upload bot! This bot greatly simplifies and accelerates the transfer of files from the IA, or anywhere else (even from WS to Commons), and it can handle batch uploading. No more downloading a 100MB file, and re-uploading to Commons. Just a heads up to let you all know this is now possible. Inductiveload—talk/contribs 03:45, 11 August 2010 (UTC)

WOOHOO! Should we also be asking for a special bot for the occasions where we need to take FROM Commons to here or our sister sites where a file does not meet the dual criteria of Commons. — billinghurst sDrewth 04:27, 11 August 2010 (UTC)

page numbers rendered in javascript

As mentioned above, page numbers are now rendered in javascript. I am planning to move this into the core of the ProofreadPage extension, after a probation period. It would be great to hear from users who use "exotic" browsers (IE, Safari are exotic for me ; my operating system does not allow me to test them) if everything is OK for them since the change. ThomasV (talk) 21:29, 12 August 2010 (UTC)

Somewhat broken under IE7. Page links reappear after purging the page 2x initially only to have the highlight "lock up" (stays constantly highlighted) on the 1st hover-over. After that intial visit & first highlight, it becomes hit or miss no matter what combination of purge, refresh, and the like is tried. Links do work but appear a good 2 inches plus from null left-margin as well. George Orwell III (talk) 05:59, 13 August 2010 (UTC)

Under IE7 I get correctly placed links, and highlights that turn on and off just fine. However the highlight is a black rectangle that completely obscures the text, ranging from the start of the line containing the point where highlighting ought to start, to the end of the line before the line containing the point where highlighting ought to end. Hesperian 00:12, 13 August 2010 (UTC)

I found this page on transparency and IE. I modified the script accordingly. It would be great to know if it really works with IE7 and with IE8 ThomasV (talk) 15:58, 13 August 2010 (UTC)

Same behavior although now when the once black now aqua-blue highlight "locks up", you can still read the text. The link does go dead while that section is highlighted as Zhaladshar describes until another mouse-over of a different page link occurs (then of course that section remains highlighted instead even after the pointer leaves the link). I don't think having the associated page link go dead upon mouse over is so bad - it would be better if engaging the highlight depended on hovering the pointer over the page link rather than a simple mouse-over pass. George Orwell III (talk) 16:17, 13 August 2010 (UTC)

• Camino, Safari, and FF on mac OSX is fine, cheers. cygnis insignis 04:23, 13 August 2010 (UTC)

• Under IE7 page links do appear after a page refresh; highlights as described above by Hesperian. Under Opera, occasionally one or two page links will appear at the top of the work. For example, in Insects, Their Ways and Means of Living/Chapter I, page links for page 17 and 24 overlap with the link for page 1 at the top of the work. Page highlight starts and ends exactly one line directly above before it should. On pages where the link appears at the top of the work, the highlight for the page previous only covers (directly above) the first line of that page.--T. Mazzei (talk) 05:22, 13 August 2010 (UTC)

• IE6 fails to show page numbers, so that means that it doesn't show page breaks, and I notice that it fails with {{float right}} to, so it is less than optimum, though it doesn't seem to break. — billinghurst sDrewth 07:08, 13 August 2010 (UTC)
  • Doesn't work on Lynx 2.0 either. ;-) Hesperian 08:29, 13 August 2010 (UTC)

It worked last night on Chrome. The only issue I had was on Essays in librarianship and bibliography, where sometimes the page numbers would disappear on me after clicking a page and then hitting the back button.

Also, right now the page numbers are pushed into the sidebar area. I don't know if that's because "indented-page" is broken or because of the javascript (all the page numbers show up and in the right spot when the prose class is used, though).

On the Essays in librarianship and bibliography page, there is an issue with highlighting of the table of contents. If you hover over page "xiv", it highlights "Photography in public libraries" and "225" and everything below (except the Index). What it should highlight is "Subject-indexes to transactions of learned societies" as well as the Index and everything in between. In other words, it's not highlighting two links that are on the actual page "xiv."

It is also difficult to click "Subject-indexes to transactions of learned societies" after highlighting it. All the other links still work fine, but the browser doesn't seem to treat that one element as a link anymore.—Zhaladshar ^(Talk) 14:17, 13 August 2010 (UTC)

Zhaladshar, I can see it too with the first and last link, though if you hang over the link to [page iii] it resets the space, so it is a little artefact. — billinghurst sDrewth 06:14, 14 August 2010 (UTC)

Zhaladshar : thanks, I fixed the "difficult to click link" after highlighting ThomasV (talk) 07:02, 14 August 2010 (UTC)

I'm seeing page numbers in the sidebar (as struck above).[3] Hesperian 10:19, 14 August 2010 (UTC)

Other parts of the work have an "indented-page" class, it should need none. Prose class is causing the page numbers to be indented from the margin, diff. cygnis insignis 11:14, 14 August 2010 (UTC)

wheel zoom

On a related note, for side-by-side editing in Opera, while selection zoom still works, mouse wheel zoom is broken. Wheel zoom did work briefly after the change to the new (much less intuitive) zoom/pan controls, but appears to have been broken sometime since.--T. Mazzei (talk) 05:22, 13 August 2010 (UTC)

• • Actually, wheel zoom does not appear to work on IE7 either. Or am I doing something wrong--left-click to get crosshairs ("zoom mode"), then wheel zoom or selection zoom should work?--T. Mazzei (talk) 05:26, 13 August 2010 (UTC)

thanks for the comment. wheel zoom has been disabled by me, because of a bug. I will put it back after the next code update. In the mean time, you can activate it as an individual setting, see in Common.js. ThomasV (talk) 05:30, 13 August 2010 (UTC)

Bot flag for Diego Grez Bot

I want to request bot flag for user:Diego Grez Bot on Wikisource. It would mainly do interwikis and fix double redirects, and AWB per request. Flagged on many wikis, such as french wikipedia, english wiktionary, wikinews, and many others. Does this need a vote? :) Diego Grez (talk) 22:40, 15 August 2010 (UTC)

I think it does. Anyway, I support since the bot is already flagged in many wikis and the task isn't controversial at all. Jafeluv (talk) 00:49, 16 August 2010 (UTC)

It needs to address the process as described at Wikisource:Bots. It is not a high hurdle. Somethings are clear, and somethings are not. Double redirects, no brainer, should be fine, whereas with interwikis the issue is that it is not a direct one to one relationship for works in the main namespace (eg. translations), whereas in the author namespace, there is a corresponding one to one, so we would like to know what the bot is doing in that space, how it is controlled, how is the decision made, etc. — billinghurst sDrewth 01:29, 16 August 2010 (UTC)

The bot would only work on Author, Category and Wikisource's namespaces. The bot is mainly supervised by myself, and if it fucks up (it has done idiocies previously :-P) I quickly resolve them. It works with pywikipedia for both tasks. Diego Grez (talk) 01:39, 16 August 2010 (UTC)

What to do with Author:Clarence Young

I doing one of my occasional forays into continuing the preparatory work in Wikisource:WikiProject CrankyLibrarian and have come across a link for Author:Clarence Young. The difficulty is that w:Clarence Young indicates that we are not actually talking about an author, instead it is a publishing syndicate that published under that name, hence it doesn't fit under our traditional Author: namespace. Would fellow contributors prefer to see such a collation to be put into the Portal namespace, ie. Portal:Clarence Young, or do they see that it is Author: namespace. We lack clarity in our guidance in that regard for the Author: namespace. — billinghurst sDrewth 00:58, 16 August 2010 (UTC)

This is going to be an issue for me as well in the near future. I was planning to use the Author: namespace, to be honest; I hadn't thought of a portal. John Peter Drummond was a house pseudonym for the publisher Fiction House. I've scanned one of their pulps with a Drummond-attributed story and, once I've finished processing the scan into a DjVu (which will take a while as I'm not too good at scanning properly yet), I'll probably have to use Drummond as the author in the header as I have no idea who actually wrote that particular story. However, someone else may come by later on and add the correct author information, which should eventually make the pseudonymous Author page redundant.

So far, I've attributed texts to the real author ("Grant Stockbridge" was mostly Author:Norvell Page, "Sam Walser" was Author:Robert Ervin Howard etc). I haven't created Author pages for the pseudonyms which, although it would be easy with the direct pen names, would get complicated with shared house names. We don't have Author:Boz (Charles Dickens), although we do have Author:Mark Twain (a redirect to Author:Samuel Langhorne Clemens), so there isn't an established pattern beyond which name is most famous.

It would probably make most sense to use redirects in the Author: namespace for pseudonyms like Mark Twain but portals for shared identities like Clarence Young. Ideally, all pseudonyms will be replaced with the correct author eventually, so Author pages for them ultimately be empty except for the header. We might as well start with a portal and not create "temporary" author pages, even if the ideal is never really reached.

Short version: I think a portal is probably best. - AdamBMorgan (talk) 12:35, 16 August 2010 (UTC)

Personally, I think if we have a name on a set of books, it will be most convenient for everyone just to treat that as an author, even if it's a collective pseudonym. Our readers might well expect to find it in our author list, etc. Excess pedantry does not serve us here, IMO.--Prosfilaes (talk) 19:47, 16 August 2010 (UTC)

I totally support Prosfilaes. We should do what is pragmatically most useful to readers.--Longfellow (talk) 20:17, 16 August 2010 (UTC)

I tend to agree with Prosfilaes; I view the portal namespace as the place for subject-based collections of materials, while the Author namespace can be defined widely enough to include groups/organizations. —Spangineer^wp (háblame) 23:16, 16 August 2010 (UTC)

Well without getting deeper into the weeds about the legitimacy (or the lack thereof) in placing entities or organizations under the Author: namespace to mimic individuals - what about some type of soft-redirect briefly explaining any odd attribution nuances, such as this apparent collective-pseudonym issue, before sending the reader off to the Author's page?

We might have it do double-duty by supplementing "the oddity" in brief snippet with optional/additional links to all the appropriate or related Wikisource:, Wilipedia: or Portal: pages dealing with the attribution issue at hand. This could also help cut down on the urge to repetively explain this group or that collective in the Notes: field for each & every single article attributed to that particular "author" or "Author:" too. George Orwell III (talk) 01:13, 17 August 2010 (UTC)

I think the Notes section of author pages and mainspace pages is the place to clarify authorship if such clarification is needed. In the case of Author:Clarence Young, I'd say that the notes on the individual works should say that they were ghostwritten but published with "Clarence Young" as the author, and Author:Clarence Young should explain it similarly in its notes section. I don't like the idea of requiring an additional click to get to the desired page (with a soft redirect). —Spangineer^wp (háblame) 16:24, 17 August 2010 (UTC)

Found an earlier deletion discussion: Wikisource:Proposed deletions/Archives/2007/04#Author pages for organizations —unsigned comment by Billinghurst (talk) .

Rather weird reading something you forgot that you wrote, and disagreeing with it. Anyway, my opinion is that, wherever possible, we should name the true author (even if it was originally created anonymously, with a pseudonym, ghostwritten, etc.), with possible references to any better-known authors associated with the work (such as with ghostwritten works). Where it isn't possible, use the "fake" author, in the author namespace. Thus, Author:Alexander Hamilton, not Author:Publius, is the author of Federalist 30. However, Author:Centinel and Author:Cincinnatus would be the authors of [4] and [5], respectively. I think an organization as an author fits into this model. The portal namespace should generally be for collections of works on particular topics.

All this said, I reserve the right to change my opinion in three years. =) —Spangineer^wp (háblame) 15:53, 18 August 2010 (UTC)

What if it is a collective authorship? Of course the author page should say what is known about the real authors, and if they have their own pages there should be x-refs.--Longfellow (talk) 19:57, 18 August 2010 (UTC)

More about Author: namespace vs Portal: namespace

We have been having a little upswing in pages like Portal:Parliament of South Africa being created, and a number [Author:United States …]. Now I wasn't here at the time of the discussion for the creation of Author: namespace, however, from my understanding, the Author: namespace is meant to be solely for people author's not for organisations. It would seem opportune to have that discussion, get some clarity, and if it is determined that such pages are not for Author: namespace, then getting Portal namespace sufficiently organised for the move. — billinghurst sDrewth 01:23, 16 August 2010 (UTC)

Quick grab of some links

• 2010-07#Expand_scope_of_Author_namespace
• 2007-09#Author_namespaces
• 2008-11#Author_pages
• 2009-03#Publisher_on_Author_pages
• 2006-07#Author_and_Portal_namespaces
• 2005-09#Author_pages

Some who were here in early days may have better pointers — billinghurst sDrewth 01:44, 16 August 2010 (UTC)

My time here has been far less than most of the daily participants I've come to appreciate over the months but it didn't take me very long to see the Portal (or Projects-side??) hasn't evolved anywhere near as much as author-prose-Page: "side" has. This is understandable given the limits that copyright places on the ability to host certain types or collections of works on WS. Still, after following some of your efforts in the Author: namespace just this past weekend, I was surprised to find no real Portal "structure" whatsoever to export errant pages to without compounding the overlaps or injecting other just as troubling issues by hastily trying to do so. I did get a chuckle out of NASA and Alaska having their own Author: pages though. George Orwell III (talk) 02:12, 16 August 2010 (UTC)

I have been planning major shifting work to Portal: from Wikisource: (and Author: where required), but I haven't got to it yet. I think we still need to have a good think about how the Portal: namespace is arranged. Will it be a tree system like Category:, and if so, what is the best way to navigate it (will we have parent links in the header, or what)? This is the ideal time to do that, as the Portal: namespace is still small enough to be managed easily. Inductiveload—talk/contribs 03:11, 16 August 2010 (UTC)

Absolutely, partly why I have prompted the discussion on the principle, rather than just did. For portal a search is easy, it is the browse that adds relevance. That said, it deserves its own discussion header, rather than subsidiary to here. — billinghurst sDrewth 03:49, 16 August 2010 (UTC)

Organisations such as NASA might count as authors as they have a corporate identity (which I think means that, among other things, the organisation can own copyrights and other author-based privileges). Apart from that, which method would be the most intuitive? Separating people-authors from organisation-authors, in Author: and Portal: respectively, might complicate the way information is stored in Wikisource and could confuse new users. Then again, new users may well just click on links and not really think about the concept of namespaces at all. - AdamBMorgan (talk) 11:56, 16 August 2010 (UTC)

Corporate identity and being an author is different, NASA acts rather as editor or publisher than as author. Another example, not yet existing in Author namespace: Author:Geological Society of London look like an error to me. Phe (talk) 12:23, 16 August 2010 (UTC)

It might be easier to accept the current situation and expand the scope of the Author: namespace to cover "Corporate Authors" such as governments, organisations and companies (maybe with an additional, slightly different header, or a modification to the existing template). A risk of this is texts being listed separately by author, editor, publisher and so forth; so, limit it to the attributed author of a text. Alternatively, someone could create a new namespace for non-authors (although this may just complicate the system). - AdamBMorgan (talk) 12:42, 17 August 2010 (UTC)

I moved content to Portal:Florida Legislature, but I propose it be moved back to the Author namespace; it is an authority.

I noted that references within some of those documents were not linked, but they were also linked from Portal:Florida. Access to texts by creation of new pages and content should forwarded to, and realised at, the big and popular sister: "for notable content, categories, infoboxes, images, cited facts, and browsing or general navigation, see w:Florida Legislature." I doubt users (readers) are accessing texts here via portals, they would use a search here or see it turn up on goggle. The solution of Users (anyone, 9-90) creating 'portals' means some has to devise it, create it, check it, maintain it, argue about it, arrange it, categorise it, and inevitably conflict with some other notional arrangement. Creating new content is problematic, whether it is bias or benign, and pointless when users can already create their own 'pages' with a search.

Limiting the 'page' of search results by author, or any authority, would be a crucial feature.The alternative is to create wikipedia-lite in the portal namespace, not simple and a tangent to the purpose of the site. Libraries had subject indexes because they didn't have search engines, and searching by subject elsewhere is still pretty flaky - even with 'advanced searches'.

We provide searchable content, if we need a subject index to that we should adopt a published and authoritative one. cygnis insignis 06:30, 16 August 2010 (UTC)

Perhaps a tree based on the Dewey Decimal classification? That would be a standard library indexing system (published and authoritative) with three levels of classes/portals shown in that wikipedia list alone. A fourth level beneath that could be added for an extra level of detail if and when it is required. For instance, Dewey classification 556 (Earth sciences of Africa) would be:

• Portal:Science ( = 500 Science)
  • Portal:Earth sciences ( = 550 Earth sciences)
    • Portal:Earth sciences of Africa ( = 556 Earth sciences of Africa)

The downside is fitting our existing system to the Dewey system, and the creation of up to 900 portals that would be required to fully implement it. - AdamBMorgan (talk) 11:56, 16 August 2010 (UTC)

The Dewey system is still under copyright I think. Hesperian 12:01, 16 August 2010 (UTC)

We are looking to classify using the system and no different from 10-100k+ libraries around the world, many of who are online through Worldcat, how would that be seen to be a breach of copyright? From memory it is a 19thC system, so couldn't we implement something pre1923 (in line with the bulk of our works). While it wouldn't be 100% perfect, I would think that it would be a vast improvement on where we are now. — billinghurst sDrewth 12:21, 16 August 2010 (UTC)

I hadn't thought of copyright (and I'm not even sure if you can copyright the implementation of a system like this). Wikipedia says that the latest version was released in 2003, so that might be off limits (if there is a copyright issue). Author:Melville Louis Kossuth Dewey has a link to a Project Gutenberg version of Dewey's original. The most obvious difference is that "000 - General" is now "000 – Computer science, information & general works." - AdamBMorgan (talk) 12:51, 16 August 2010 (UTC)

You certainly can copyright a categorisation system like the Dewey Decimal System, and I believe the 10-100k libraries that use it pay OCLC for a license. But obviously a public domain version would be acceptable here. Hesperian 13:19, 16 August 2010 (UTC)

What about Library of Congress Classification? No copyright issues there. Also, I don't think we should create 900 unused portals, nor limit ourselves to integer-level portals (if DD is used): there will always be some portals with far more material than others, so creating and dividing should be done as needed. —Spangineer^wp (háblame) 15:54, 16 August 2010 (UTC)

I had to read up on it but the LCC should work just as well. The requirements stated so far are just 1) published, 2) authoritative and 3) public domain. If this is going to be the basis for the portals' structure, I'd suggest a poll in the Proposals section of the page.

Most of the 900 Dewey portals could remain as red links until they were actually used for something; I doubt we would ever really need a Earth Sciences of Africa Portal. While LCC is smaller, it is still made up of 21 classes (top level portals?) and 223 sub-classes (mid level portals?), without going into the numeric part of the call numbers for more detailed subjects (it probably isn't worth implementing these on Wikisource). Also, both classifications may end up being smaller in practice on Wikisource: foreign language literature should be out of the scope of this project and, depending on how strictly we want to adhear to the system, LCC on Wikisource probably won't need three separate classes for History (including two for just America). (Incidentally, if we do use one of these systems, are we going to include a direct reference to them in the portal or just steal their structure and headings for our own use?) - AdamBMorgan (talk) 17:53, 16 August 2010 (UTC)

Personally, I think the advantage of portals is that they are flexible and allow material about specific topics to be brought together. Personally I wouldn't want to sacrifice portals like Portal:Kentucky and Virginia Resolutions just to implement a standard system. I like using a system to set up organization, but not to determine what gets a portal and what doesn't. As for referencing the system, I think that makes sense, especially on the higher-level portals (which I would expect to primarily consist of links to lower-level portals). —Spangineer^wp (háblame) 20:48, 16 August 2010 (UTC)

I'd say the existing portals will fit comfortably as the lowest level of the whatever structure (if any). The LCC system would be easier in this regard:

• Portal: History of the Americas (Class E: History of the Americas)
  • Portal: Kentucky and Virginia Resolutions (class E doesn't appear to have sub-classes)

The DDC system will also work but will involve more intermediary steps:

• Portal: History (Class 900 History)
  • Portal: History of Modern North America (Division 970 Modern North America)
    • Portal: History of Eastern Modern North America (Section 974 Eastern)
      • Portal: Kentucky and Virginia Resolutions

LCC is simpler but risks becoming overwhelmed if there are a large number of subjects/texts in the subclass. DDC can better handle larger numbers of child portals or texts but it's a lot more complicated. Each system also has gaps we could use for new, Wikisource-specific categorisation if we need them. - AdamBMorgan (talk) 12:42, 17 August 2010 (UTC)

If it's decided to extend the Author: namespace, the French solution is a possible way: fr:Catégorie:Auteurs-A Author: prefix is removed in category; fr:Auteur:William_Allingham, title is removed for all Author:* page (Another solution is to remove only the Author: prefix in Author:* page). This way Author: namespace can be extended making it a sort of transparent namespace for readers. I see two other solutions: 1) don't care and get inconsistent title from time to time (author:something which is not an author) 2) rename the namespace with a more generic name and keep Author: as an alias for backward compatibility. Phe (talk) 13:13, 16 August 2010 (UTC)

I quite like not showing the Author prefix in the categories regardless of expansion or not. It makes the categories appear a lot more user friendly. - AdamBMorgan (talk) 12:42, 17 August 2010 (UTC)

• I am leaving town and can't contribute to this conversation at the moment. But I wanted to point you all to what has already been created and, I am rather certain, abandoned with regards to LCC. If you haven't noticed it before, it is at Category:Library of Congress Classification. --BirgitteSB 18:07, 17 August 2010 (UTC)

OK. The more I read back on this, the less I understand. Is there or is there not a near certain rule of thumb on WS that the Author: namespace is for listing the works authored (or created/written if you prefer) by an individual? Seems to me like there was once a steadfast rule that has erroded somewhat along the way and now I'm not so sure what to think. George Orwell III (talk) 20:16, 18 August 2010 (UTC)

• Regarding the structure of the Portal: namespace, see Structure Portals by Dewey Decimal or Library of Congress Classification in Proposals. - AdamBMorgan (talk) 17:57, 23 August 2010 (UTC)

Bot flag request for Phe-bot

I would like to request a bot flag for Phe-bot. The bot is based on pywikipediabot. It doesn't use standard script but custom script built on the top of Pywikipedi. Actually I want to update sub-page of User:Phe/Author. I've already a (misnamed) w:fr:User:Badmood bot with more than 400K contributions. On en:ws I don't plan to do mass contributions, I'm more focused nowadays to data gathering, analysis and creation of maintenance page like User:Phe/Author. Typical contributions of this bot can been seen here. For the curious the bot code is here. Later it'll do more works like trying to match red author link to existing page (User:Phe/Author possible match), doing some sanity check on Index: (User:Phe/Check index) and Author: page (User:Phe/Check author). Phe (talk)

nod Having seen it in action, and benefited from its output, it gets my nod. — billinghurst sDrewth 12:14, 16 August 2010 (UTC)

support Its recent output caught my eye as well. Seems like it ran fine. George Orwell III (talk) 18:59, 16 August 2010 (UTC)

support --Zyephyrus (talk) 19:30, 16 August 2010 (UTC)

support ThomasV (talk) 08:58, 18 August 2010 (UTC)

Author:John Thompson, U.S. Army

A portion of Author:John Thompson, U.S. Army does not display properly and I can't say why. There is another Author:John Thompson who is an actual author. Fred Bauder (talk) 21:20, 16 August 2010 (UTC)

Lucky for us, the Canadian Prime Minister Thompson had 2 middle names so I expanded his Author: page to include them - freeing up the "simple" Thompson name for your U.S. Army Officer. You'll need to re-add him to the Author index. If you find out his birth-year, year of death or his middle name/initials, you should go back and add them to his Author: page to help avoid confusion and editing problems for the next guy who may have the same problem you just had. George Orwell III (talk) 22:03, 16 August 2010 (UTC)

We should be disambiguating the authors at Author:John Thompson using {{disambiguation}} and adding {{similar}} to each author page. It has generally been agreed that the means to disambiguate is by (XXXX-YYYY) in the page name, though we have allowed liberal use of text variants as redirects. — billinghurst sDrewth 11:46, 17 August 2010 (UTC)

I just took the name already given in sisterlinks and checked-to-be-working direct to that name too. My mistake for assuming. George Orwell III (talk) 12:54, 17 August 2010 (UTC)

Copyright check

Thought I should double check copyright status of Index:Aboriginal welfare 1937.djvu. The National Library of Australia says its out of copyright [6] does this mean in US as well? Also assuming it's ok, should I be bothering with columns as in the original? They will look rather odd when it's collated into continuous scrolling pages.Misarxist (talk) 11:09, 17 August 2010 (UTC)

As a government work of 1937, it would have come out of copyright in 1987 in Australia (then 50 year rule for crown copyright), and subsequently out of copyright in US. As a note, if it can be hosted at Commons, then it can definitely be hosted here. — billinghurst sDrewth 11:41, 17 August 2010 (UTC)

Don't bother with columns. The formatting will produce a good result on its own, better than the original. cygnis insignis 12:58, 17 August 2010 (UTC)

Nupedia

moved back to the Scriptorium 19:36, 12 November 2010 (UTC)

wikt:MediaWiki:Gadget-WiktSidebarTranslation.js

It would be elegant to import this gadget, especially for those who have never installed any additional fonts. It just translates the interwiki links into English, you can test it by ticking it here. JackPotte (talk) 09:13, 19 August 2010 (UTC)

Speedy delete

• Wikisource:Books/Annals of the World/1
• Wikisource:Books/Annals of the World/2

Not needed, delete to get rid of clutter. Don't delete parent page. Thanks, Arlen22 (talk) 17:47, 20 August 2010 (UTC)

Done. You can also add {{sdelete}} to pages to request speedy deletion. —Spangineer^wp (háblame) 19:26, 20 August 2010 (UTC)

I noticed I failed to link those two pages from Wikisource:Books. If the point of Wikisource:Books is so that people can print these books, then we'll need to keep the individual volumes, as PediaPress--and I suspect most normal printing systems--just can't handle anything as big as the one volume version.--Prosfilaes (talk) 01:35, 21 August 2010 (UTC)

I successfully created the pdf of the combined volumes, but didn't attempt to print. If you feel like the split volume makes more sense, feel free to undelete. —Spangineer^wp (háblame) 01:57, 23 August 2010 (UTC)

Why can't I get the scanned pages to show up?

In trying to proofread pages, I can’t get the scanned version of some pages to load. This is a recent problem. The scanned version of the page shows up as black. Am I doing something wrong or is this another problem outside my control? Another editor (talk) 20:14, 22 August 2010 (UTC)

I'm seeing this also occasionally and mostly recently. Moondyne (talk) 01:37, 23 August 2010 (UTC)

Seems to be problems at Commons. Seems all quite temperamental at the moment, and for the past few days. — billinghurst sDrewth 03:18, 23 August 2010 (UTC)

English Wikipedia has a watchlist message up: 'Wikipedia is currently experiencing technical difficulties with the display of images: media file thumbnails may appear empty, or not at all. While technical staff work to resolve the issue, please do not remove "missing" media files from articles. Your patience and cooperation is appreciated.' Hesperian 05:20, 23 August 2010 (UTC)

Thanks for your answers. (I though it was just me and I was doing something wrong!) Another editor (talk) 12:59, 23 August 2010 (UTC)

Incorrect licensing in PressPedia Books

Template:PediaPressTicket I created this. However, I notice to issues. First, the contents page lists the pages from WS as articles. Secondly, and more importantly, the licensig information is incorrect. How should this be rectified?--Forty two (talk) 06:08, 24 August 2010 (UTC)

A ticket has been filed for this issue.--Forty two (talk) 06:46, 24 August 2010 (UTC)

The Giaour it.source version

I just added the interlink to it.source Italian translation of The Giaour; there's too it:Il corsaro waiting the the interlink to The Corsair.

it:Il giaurro uses a test implementation of a bold DIY alternative of ref / references tags, take a look if you like this kind of strange it.source things. --Alex brollo (talk) 08:38, 24 August 2010 (UTC)

Footnotes in Encyclopedia Britannica

Hi, I'm wondering how to best handle footnotes in an Encyclopedia Britannica 1911 article. (The page is here and the article is here.) In the original, the footnotes are at the bottom of the page. However, since we publish by the article rather than the page here, I think it looks strange to have footnotes sitting in the middle of the article. I'm sure this has come up before, but I had trouble finding any similar cases in other EB articles. Any recommended approach? -Pete (talk) 14:54, 24 August 2010 (UTC)

Add <ref></ref> as used at the other place, the note between these tags at the end of the line (not at the bottom). Add {{smallrefs}} in the footer (using [+] Button) to have them appear in Page:ns and at the bottom of the main article. All the footnotes are gathered and renumbered when transcluded. cygnis insignis 16:22, 24 August 2010 (UTC)

• I have done the four notes at that page as an example, but I'll leave the other pages for you to do to get the feel of it. Ignore the broken separate reference to [1] for now - that is a technical problem (see below). Scratch that, I'm talking rubbish, it just doesn't support a name beginning with a digit. You can also see on that page how to make more than than one reference to the same footnote using the <ref name=name></ref>. Inductiveload—talk/contribs 04:31, 25 August 2010 (UTC)

I ran to see doc for <ref name=name></ref>.... VERY interesting, and (my blame) new at all for me. Has the name=name introduced recently, or it was a known extension syntax from the very beginning....? :-( --Alex brollo (talk) 11:40, 25 August 2010 (UTC)

The name=name functionality has been there from the beginning of the <ref> syntax, or close to it, if I recall correctly. The group=group functionality is newer though. —Spangineer^wp (háblame) 13:26, 25 August 2010 (UTC)

Thanks all for the help -- I use the "name=name" feature on Wikipedia, so that makes sense. Sounds like it must be temporarily broken here though? At any rate, this looks like a much more elegant solution to the problem than I had expected. Thanks. -Pete (talk) 15:40, 26 August 2010 (UTC)

Why do you say that? It looks like it's working to me at Page:EB1911 - Volume 20.djvu/280. Or am I missing something? —Spangineer^wp (háblame) 15:55, 26 August 2010 (UTC)

Ah, I see my comment was out of date -- I was responding to an inline comment by Inductiveload which he then deleted: http://en.wikisource.org/w/index.php?title=Page:EB1911_-_Volume_20.djvu/280&diff=prev&oldid=2022304 So, looks like it's all good! -Pete (talk) 17:41, 26 August 2010 (UTC)

HathiTrust

According to http://www.resourceshelf.com/2010/07/09/hathitrust-digital-library-adds-new-content-accessible-to-all-users-nearly-100000-volumes-digitized-by-internet-archive/ members of HT partner institution seems to have access now to PDFs of Google-digitized PD works. If the work is PD in the US it can be uploaded to the Internet Archive, if it is PD in the US and in its country of origin it can be uplöoaded to Commons. It would be nice to have contact persons who are members (or friends of members) of HT partner institutions. Greetings from Germany --FrobenChristoph (talk) 12:38, 10 July 2010 (UTC)

I'm not sure flooding Commons with PDFs we have no plans of working on is a good thing. In any case, Americans have decent access to the PDFs of Google-digitized PD works anyway; what's the goal here that we would need Hathi Trust access for?--Prosfilaes (talk) 14:01, 10 July 2010 (UTC)

To be as impolite like this " Americans have decent access to the PDFs of Google-digitized PD works anyway": Könntest du bitte in Erwägung ziehen, dass es noch andere Menschen auf der Welt als US-Bürger gibt (z.B. UK-Bürger)? Es ist ein legitimes Ziel von Commons, edukative Inhalte zu hosten, wozu auch PD-Werke zählen. Commons wird von miesen Gallica-Scans geflutet und fr-WS muss diesen Müll transkribieren, was als großer Fortschritt gefeiert wird. Von daher verbitte ich mir diese Formulierung "flooding Commons".

Nachdem ich HT vermutlich besser kenne als ihr alle zusammen, darf ich darauf hinweisen, dass bei Google die ganzen Werke von ca. 1909-1922 NICHT einsehbar sind, während sie in HT abrufbar sind. Zusätzlich gibt es nicht von Google digitalisierte PD-Bücher in HT --FrobenChristoph (talk) 17:00, 11 July 2010 (UTC)

Translation (feel free to make it better): "Would you please consider that there are people in this world who are not US citizens (e. g. UK citizens)? It is a legitimate goal of Commons to host educational content, among which PD works should count. Commons is overflowed by bad Gallica scans, and fr-WS must transcribe this garbage; this is celebrated as progress. Hence, I ask you not use the expression "flooding Commons".

Since I presumably know HT better than all of you, may I point out that NOT all works from about 1909 to 1922 are available at Google [Books], while they are available at HT. Further, there are PD books at HT that haven't been digitized by Google." Kåre-Olav (talk) 22:13, 11 July 2010 (UTC)

Yes, there are people in this world who are not US citizens, but there's a lot more of us then there are members of Hathi Trust, and we're usually pretty willing to upload a file from Google Books on request. A number of us US citizens regularly upload such books to the Internet Archive. It is a legitimate goal of Commons to host educational content; that doesn't mean that it is particularly advantageous to upload a bunch of files to Commons that the Internet Archive hosts, especially as IA is just better at hosting books then Commons is. If HT has books not found on Google Books, then that's a good thing, but if you know HT better than all of us, maybe you should have introduced it better, and probably you'll have to find someone with access.--Prosfilaes (talk) 22:51, 11 July 2010 (UTC)

Can I put it with an outcome focus. If there is a book available and it is legal for us to utilise the source, we are happy to do so, whether it be via HT, Google, Archive.org, or a scan of great aunty Bertha to us doesn't matter. Where scans have restricted accessibility, then requests can be made here or find someone in irc://irc.freenode.net#wikisource (immediate requests) or at Wikisource:Requested texts (whenever round tuit). We generally will only upload texts in which we have some prospect to undertake work, rather then an upload first, and value judgement second. If you have suggestions for works please utilise th

←

Err... not sure if this falls under the same POV here or not but we were actually looking for a way to download several of the new volumes -- finally made available as Full View on HathiTrust instead of the Snippet (or Useless) View they have been hosted as for years now on Google Books -- containing large swaths of the missing Executive Orders on WS. Unfortunately, nobody has figured out how to obtain these volumes as single PDFs rather than the 1 PDF per 1 volume page display HathiTrust has now. Example. We would thern find the best way to convert the PDF to DJVU to upload and source the same large swaths of EOs here. These volumes not only cover a few years of Executive Order issuances in one pop but the Presidential Proclamations as well. I'd appreciate further discussing this if it led to solution for us or not. George Orwell III (talk) 01:54, 21 July 2010 (UTC)

Again some facts and suggestions:

• Google doesn't show most European books ca. 1910-1922 as full view even for US citizen; HT does.
  • I can attest to this as those involved in the Executive Order WikiProject (& are U.S. citizens for the most part) have witnessed the importation of PD Google books to HT and have seen the long-time snippet views become full-view at last (at least as far as U.S. government publications go circa the ~1920s on forward). George Orwell III (talk) 21:24, 22 July 2010 (UTC)
• According an announcement of HT http://www.hathitrust.org/updates_june2010 affiliates (students and staff) of HT member institutions will have the possibility to download whole PD books (if they are Google-scanned, otherwise the feature is for all users)
  • After diving into the new HT repository for a couple of weeks now, it appears that the length of the work and the originating source (University of Michigan, University of Chicago, etc.) dictates the availability as a single PDF or as series of 1 page PDFs. Regardless of type or subject-matter and as best that I can tell), if the work is in the Public Domain per U.S. law(s) the work is viewable in Full View (still not true for same exact works hosted on Google Books though they are shared with HT now. George Orwell III (talk) 21:24, 22 July 2010 (UTC)
• We need a script downloading all single pdfs from a PD book in HT and merging them into one single PDF (which can be converted in a Djvu and uploaded in the Internet Archive or eventually Commons).
  • Unfortunately, creating such a script and subsequent conversion is well beyond my skill-set --- though I'd be willing to whore myself out to download whatever non-U.S. residents may need in return for getting what we need at the same time if and when somebody does come up with such a script/method. George Orwell III (talk) 21:24, 22 July 2010 (UTC)
• We need users who have access privileges to HT fulltext-PDFs and are willingly to give us the PDFs. --FrobenChristoph (talk) 16:16, 22 July 2010 (UTC)
  • I realize that works of the U.S. Government do not share the same copyright and related Public Domain issues but if the work is legitimately proven to be compliant with WS copyright policy I'd be willing to give downloading them a try with understanding that I too may be restricted to what is available to me as I am neither a student or staff of the Trust. Leave a short list (links would be helpful) of specific works you'd like to see obtained over on my talk-page instead of this forum. George Orwell III (talk) 21:24, 22 July 2010 (UTC)

HT is building the PDFs on the fly - unfortunately they do not know how large a PDF would be. http://babel.hathitrust.org/cgi/pt?id=uc2.ark:/13960/t1dj5fv94 has 524 pages and the PDF is VERY LARGE 273 MB (the same edition as PDF in the IA [7] has 23 MB) --Historiograf (talk) 20:46, 27 July 2010 (UTC) = FrobenChristoph

HathiTrust, again: HathiHelper

http://de.wikisource.org/wiki/Wikisource:Google_Book_Search#HathiTrust_PDF-Download Who cannot understand German is politely requested to ignore the language and only to use the links --FrobenChristoph (talk) 22:10, 20 August 2010 (UTC)

Is anybody here with an university account from the institutions listed at http://babel.hathitrust.org/cgi/wayf? --FrobenChristoph (talk) 22:14, 20 August 2010 (UTC)

Hathi-Trust

Actually you can download full volumes with using the following URL

• http://babel.hathitrust.org/cgi/i/imgsrv/imgsrv/pdf?id=mdp.39015009712970 (replace the ID mp... with the ID of the book you want to download). Please make the downloaded PD volumes available in the Internet Archive --FrobenChristoph (talk) 01:40, 25 August 2010 (UTC)

Thanks - it seems to work!! The resulting PDFs are rather large but it is better than nothing. George Orwell III (talk) 06:26, 26 August 2010 (UTC)

Translating

Am I to understand that one is allowed to do original translations of historical, non-original works, on Wikisource? (I read somewhere that, among other things, this site was for "translations of source texts" - can't remember exact words, though). Gott wisst (talk) 08:18, 25 August 2010 (UTC)

Wikisource:Translations is the best starting place. Ideally the translations that Wikisource hosts would be published translations, though there has been a convention to allow community translations to be hosted. — billinghurst sDrewth 10:30, 25 August 2010 (UTC)

You are correct, anyone can translate anything, according to the proposal at the English language wikisource - others only accept published material. Most works here are published texts, as with other libraries, but some religious texts have been pseudonymously translated without review. cygnis insignis 12:53, 25 August 2010 (UTC)

There is somewhere a comparative list of Wikisource policies, and most of them permitted original translations, with the German Wikisource being the only one I recall.--Prosfilaes (talk) 16:30, 25 August 2010 (UTC)

In all of this, when I first came here I was surprised that we actually had WS-based translations, for as I envisaged them, they are original works, and hence maybe community translations of our materials may be better housed at Wikibooks. That said, it never greatly concerned me, so I just left it alone. — billinghurst sDrewth 13:51, 27 August 2010 (UTC)

Reminder: you cannot publish a translation of copyrighted work unless you have permission from the copyright holder. -Arch dude (talk) 22:14, 26 August 2010 (UTC)

Key missing authors

Following on from a discussion at WS:FTC where some key scientists were found not to have author pages, I grabbed a list of the biographies in Wikipedia:Wikipedia:Core topics - 1,000, stripped out the modern people (for whom public domain or suitably licensed material would be scant) created a few redirects, and eliminated the blue links. I ended up with a list of 50 prominent historical people that don't have author pages here. Personally, the most surprising redlink is Author:Socrates—I know he didn't write but an immense body of literature has been written about him. I was also very surprised about Author:Jesus being a redlink. Anyhow, if anyone is interested in creating some of these, the full list is at User:Hesperian/Sandbox. Hesperian 13:40, 27 August 2010 (UTC)

Problem with a page that is gibberish

I proofread this page: http://en.wikisource.org/wiki/Page:In_the_Roar_of_the_Sea.djvu/159 but instead of saving my editing, it saved as gibberish. Is there a way to delete the page and start over with the page scan? Otherwise, I don’t think I can fix it. Thanks, Another editor (talk) 15:54, 27 August 2010 (UTC)

You can usually get the text from the source of the file, this one is fixed now. cygnis insignis 19:23, 27 August 2010 (UTC)

Thanks for fixing it. How do you get the source? It isn’t one of the tabs at the top. (And it wasn’t in the history.) Another editor (talk) 19:26, 27 August 2010 (UTC)

One of those labels that means many things, in this case it means the archive.org address that should appear on the File's description "source = " at commons. cygnis insignis 20:13, 27 August 2010 (UTC)

Announcement/invitation

Hi all, I'm wearing my Wikimedia Foundation hat today. You may be aware of our Bookshelf Project: creating collection of materials that help people understand the Wikimedia world and how to start contributing to our projects.

We believe that screencasts—video demonstrations of how to use software—are a great way to reach a wide audience. We have seen some cool examples emerging in the Wikimedia volunteer community and from other sources. But there are some pretty big challenges involved in producing a good screencast, especially if you're on a tight budget and/or want to use free and open source software.

So, we want to create an online tutorial (using screencasts) that helps people get started on producing their own screencasts. And we want to work with the passionate and knowledgable volunteers in our community to produce the best results and lay the groundwork for ongoing resources (perhaps as a WikiProject) for Wikipedians who want some support in producing screencasts.

Interested? Please visit the Screen Sprint page on the Outreach wiki for more info, and an application to submit by Monday, August 30.

-Pete F (talk) 23:36, 27 August 2010 (UTC)

Seeking advice on how to format poems

I have numerous occurrences in PSM which require the poem format, one of which I selected and formatted in a variety of ways, but unsure which would be the preferred way to do it from a strictly technical viewpoint. Would someone care to look at THIS PAGE and advise of the preferred way? Thanks in advance. - Ineuw (talk) 16:35, 28 August 2010 (UTC)

In works where the poem appears among standard text, I have generally seen it appear as a centred block and slightly reduced in size, so I wrap it in {{float center|<poem style="font-size:smaller">TEXT TEXT TEXT</poem>}}. — billinghurst sDrewth 17:14, 28 August 2010 (UTC)

Thanks. I will have to play with it because the text is small caps. - Ineuw (talk) 18:03, 28 August 2010 (UTC)

Sure, I was providing the model that I used, not the dictate that it was the exact methodology. You can just use {{float center|<poem style="font-variant:small-caps">TEXT TEXT TEXT</poem>}} or {{float center|<poem style="font-size:smaller; font-variant:small-caps">TEXT TEXT TEXT</poem>}}. — billinghurst sDrewth 00:39, 29 August 2010 (UTC)

Billinghurst - This is great. My gratitude. I didn't know that CSS style properties can be embedded in a template, or that HTML can be extended with <poem></poem> - Ineuw (talk) 03:14, 29 August 2010 (UTC)