Wikisource:Scriptorium/Archives/2012-10

= Announcements =

Poem extension updated
'''An announcement that actually affects Wikisource! :)'''

On September 19th, an updated version of the Poem extension will be deployed to Wikisource. Most of the changes are back-end, but there is one change that will effect the formatting of poems. Per the fix to, colons inside tags will no longer invoke definition list syntax, but instead will invoke a 1 em indentation via a span tag. This visual difference is minor, but some editors may want to make adjustments accordingly. Kaldari (talk) 17:56, 10 September 2012 (UTC)
 * That makes a lot of sense. Thanks. Chris55 (talk) 19:27, 10 September 2012 (UTC)

Legal Fees Assistance Program
The Wikimedia Foundation is conducting a request for comment on a proposed program that could provide legal assistance to users in specific support roles who are named in a legal complaint as a defendant because of those roles. We wanted to be sure that your community was aware of this discussion and would have a chance to participate in that discussion. If this page is not the best place to publicize this request for comment, please help spread the word to those who may be interested in participating. (If you'd like to help translating the "request for comment", program policy or other pages and don't know how the translation system works, please come by my user talk page at m:User talk:Mdennis (WMF). I'll be happy to assist or to connect you with a volunteer who can assist.) Thank you! --Mdennis (WMF) (talk) 15:34, 5 September 2012 (UTC)

Wikidata is getting close to a first roll-out
(Apologies if this message isn't in your language.)

As some of you might already have heard Wikimedia Deutschland is working on a new Wikimedia project. It is called Wikidata. The goal of Wikidata is to become a central data repository for the Wikipedias, its sister projects and the world. In the future it will hold data like the number of inhabitants of a country, the date of birth of a famous person or the length of a river. These can then be used in all Wikimedia projects and outside of them.

The project is divided into three phases and "we are getting close to roll-out the first phase". The phases are:
 * 1) language links in the Wikipedias (making it possible to store the links between the language editions of an article just once in Wikidata instead of in each linked article)
 * 2) infoboxes (making it possible to store the data that is currently in infoboxes in one central place and share the data)
 * 3) lists (making it possible to create lists and similar things based on queries to Wikidata so they update automatically when new data is added or modified)

It'd be great if you could join us, test the demo version, provide feedback and take part in the development of Wikidata. You can find all the relevant information including an FAQ and sign-up links for our on-wiki newsletter on the Wikidata page on Meta.

For further discussions please use this talk page (if you are uncomfortable writing in English you can also write in your native language there) or point me to the place where your discussion is happening so I can answer there.

--Lydia Pintscher 13:14, 10 September 2012 (UTC)

Distributed via Global message delivery. (Wrong page? Fix here.)

MediaWiki 1.20/wmf12 rollout imminenent
MediaWiki_1.20/Roadmap says that we are due for the rollout of the next microversion upgrade of WMF locally. Beyond the ProofreadPage releases (thanks Tpt), a quick glance doesn't show any intended radical changes for us. ProofreadPage — billinghurst  sDrewth  15:07, 19 September 2012 (UTC)
 * Small improvements in the UI of the page edition and a few cosmetic changes
 * (bug 27558) Add of a support of wiki links to proofreadpage_page_status
 * (bug 36070) Add of support of non-arabic page number in page namespace.
 * (bug 36979) Add of class prp-pagequality-X to all links to the page namespace.

= Proposals =

Better help page index
In recent days I've drawn up a potential help page index which could be placed into Help:Contents after community approval. It is likely to be improved, especially for the position of links: Which ones should have a prominent position? Please improve the draft and post your advice.--Erasmo Barresi (talk) 09:02, 12 September 2012 (UTC)

New parameters to dotted TOC page listing
I have added two new template arguments to this to more finely control the layout of the "Entry" field (aka column 2.) The default behaviour of the template without specifying these is unchanged.

There are occasions (like this) where normal use of dotted TOC page listing looks "ragged." Some alternatives now available are:

Why am I wording this proposal as if it is already done? It is because I made the change yesterday with only minimal consultation (thanks Mpaa!) and then discovered this process after I had done it… Oops! MODCHK (talk) 23:01, 18 September 2012 (UTC)

= BOT approval requests =

HesperianBot

 * In lieu of available crats, I have requested that this permission be applied m:Special:PermanentLink/4082185 — billinghurst  sDrewth  16:07, 1 September 2012 (UTC)
 * I've granted the bot a flag. I'm sorry it's taken so long for this to happen.—Zhaladshar (Talk) 01:58, 4 September 2012 (UTC)
 * Thanks Zhaladshar. Hesperian 00:52, 5 September 2012 (UTC)

=Help=

= Other discussions =

List of interesting books in public domain in US due to non-renewal
Over the past few weeks I've been sifting through this non-comprehensive listing of non-renewed books. Here's the result. Books that had a Wikipedia article, had a highly notable author, went through multiple editions, or struck me as interesting made the cut. The entire class is a big source of potential growth and specialization (as no other organizations really touch them), the only problem being scans (for the same reason). Prosody (talk) 14:17, 26 August 2012 (UTC)


 * While browsing the Wikisource mailing list, I noticed the "Which books to scan to support Wikipedia" thread and thought of your page. If any of these books are held by the University of Toronto (I haven't checked yet) it might be possible to get them scanned by the Internet Archive (albeit for a "small fee").
 * Regarding the list in general: It is interesting and there are books on there I didn't expect. The problem is as you state, actually getting the scans.  University of Toronto Library aside, if we attempt to do it ourselves many of these books are likely to be expensive and/or just hard to get hold of due to their age. - AdamBMorgan (talk) 12:20, 30 August 2012 (UTC)


 * I don't know these modern books just by the title. The title is always designed to catch the eye and interest then you look inside. Too many books have wonderful a wonderful title but are bland when reading. This is just my 2¼ cents on the overall idea and to support "Wikipedia"? Recently we had a script to remove their begging advertisements. Where is the monetary support for Wikisource? WP is doing better than WS isn't it? Yes. &mdash; William Maury Morris II  Talk 13:17, 30 August 2012 (UTC)
 * I don't think we should be judging which books are bland and which ones are not, considering we just had Climatic Cycles and Tree-Growth as the Proofread of the Month. ;) EVula // talk // &#9775;  // 15:59, 30 August 2012 (UTC)
 * AHahahahaha! that's great! You are so right! I humbly bow to your wisdom here. Kindest regards and thanks for the hearty laugh. Cheers! &mdash; William Maury Morris II  Talk 16:13, 30 August 2012 (UTC)
 * So, can we put a proposal in to WP get funding for scanning works? 10:34, 31 August 2012 (UTC)
 * We probably need to create a double-checked short list and work out how the scanning will take place (with an estimate of costs). The first book in the list "My Disillusionment in Russia by Emma Goldman" is not only held by the University of Toronto but its copy has already been scanned and is available at the Internet Archive. We need a specific list with no redundancy (and re-checking the copyright status at this stage would be useful).  I don't know how much scanning costs, either at Toronto or in general, although the Wikimedia Foundation should be able to easily afford the former.  Which is another thing; if scanning via Toronto, will this be paid for by the Foundation, Wikimedia Canada or another group (the chapters have separate bank accounts, if I recall correctly)?  Either way, it would probably help to start with a smaller proof-of-concept project rather than the whole list. User:LA2 is the one who visited the scanning centre (in the mailing list post), he might know more about their requirements. - AdamBMorgan (talk) 11:53, 31 August 2012 (UTC)

New dynamic layouts
I suggest we add at least two new layouts to the general list. The current options mimic other Wikisources (French/Italian and German, I believe). I think it would help to have two more: (1) Largely similar to the standard layout but with space either side to accommodate side notes; (2) a narrow central column, which I don't actually like but it is a common forced layout and advocates claim it is easier to read. This will allow side notes without disrupting everything else and let users cycle to their prefered layout without forcing their own preferences on everyone. I've only dabbled with personal layouts, so it might help if someone with more experience look at things before they go live (if they go live). So, what do people think of this? Are there any others we might need? - AdamBMorgan (talk) 13:05, 30 August 2012 (UTC)
 * I would like to add more; and take a look at the existing ones. Of course, I come loaded with suggestions and I imagine introducing any would just cause havoc, as I'm sure a lot of people have ideas, which will make the whole process lengthy. Can I suggest that the "narrow central column" be set to be the width defined as "frameless"? I assume it is smaller than the existing 400px Layout 2. I ask this because I personally size images using frameless (and portions of it using "upright") (example.) which would make the text aligned to the picture size. So, if I prefer larger images and change my thumbnail parameter in options, the text and images would all be updated to that width.


 * Just a suggestion :) - Theornamentalist (talk) 20:30, 30 August 2012 (UTC)


 * ABM there is a sandbox for new dynamic layouts, with a gadget for people to turn the test dynamic on and off. Why not put them there one at a time, advertise the test, and take the feedback on the talk page. — billinghurst  sDrewth  07:47, 5 September 2012 (UTC)

British Library project
Hi all,

I've been spending the past few months working with the British Library as the Wikipedian in Residence, and one of the things I've been wanting to look at is explore what I can do for Wikisource. Most of the content work we've done so far has been oriented around writing about the collections on Wikipedia or pushing images towards Commons, but library-Wikisource collaboration seems a natural progression and I'd be interested in any suggestions you might have for what I could do to help the project.

Relatedly, we're holding an open afternoon and exhibition tour on Monday 10th September, and if any of you are in the London area and would like to come along you'd be more than welcome!

Thanks, Andrew Gray (talk) 16:40, 30 August 2012 (UTC)


 * Welcome! While I don't know the details of what you are specifically involved in I still would like to interject a plea. Place illustrated books circa 1750-1890 on Internet Archives. Text alone gets tiresome which Wikipedia doesn't have to deal with to any great extent. The fact is Wikipedia is heavily illustrated with unique lithographs and other visual materials in color and b/w. We here on Wikisource often don't have enough illustrated works -- in my opinion. Counting tree-rings in our last featured works isn't very visual nor interesting. In the U.S.A., kindest regards, &mdash; William Maury Morris II  Talk 16:52, 30 August 2012 (UTC)


 * Some things Wikisource can use are:
 * Scans.
 * More people.
 * More attention.
 * Scans.
 * Good quality scanned texts are useful because that's what Wikisource is built on. On the other hand, if we get a hundred scans, most will sit idle because we only had 317 active users in the last 30 days.  (I think French Wikisource had this problem when they received texts from the BNF.)  Anything the British Library can do to boost either would be nice. - AdamBMorgan (talk) 17:28, 30 August 2012 (UTC)


 * Yes! I concur with AdamBMorgan's enumeration. ;-)  Scans!  Are you taking requests for specific items in the BM collection?  For I should love to get anything to do with the Rationalist Press Association around the turn of the last century (i.e. its founding and early years).  There's lots.  :-)
 * And is there anything we can do to help you?
 * &mdash; Sam Wilson ( Talk &bull; Contribs ) &hellip; 06:10, 31 August 2012 (UTC)


 * I will venture a guess that if you look at our list of "featured texts of the month" you would see what the people here on Wikisource like and want. Then you will know what is desired. Kind regards, Maury ( &mdash; William Maury Morris II  Talk 06:23, 31 August 2012 (UTC)


 * Thanks all. Unfortunately I'm not in a position to request BL digitisation of specific items (not without funding it, anyway), but I have had a couple of (tentative) discussions about reusing already-scanned material. However, I was concerned about the problem Adam notes - one is good, ten is better... but a hundred might be worse than ten, and I don't want to turn up and dump a lot of material on the community without support! Unfortunately the material in question isn't currently hosted anywhere public, so we can't just point to it and say "take what you might need".
 * I will see what I can do about finding some samples which can be made available without complications and uploading them. Am I right in assuming that DjVu is preferred but PDF is acceptable?
 * One final point is that I'm hoping to pick up the WMUK digitisation discussions (see above a few sections about WWI history), and hopefully we should be able to do something on-demand with that if it gets going. I will keep you all posted! Andrew Gray (talk) 22:37, 31 August 2012 (UTC)
 * In non-scanning terms, a "proofread-a-thon" is a possibility. I have to say, proofreading is not quite the same as editing and I've never seen or heard of it done in this way; it might not work in the same way as an editathon.  If necessary, it could be trialed by adding something to a future Wikipedian editathon (to be clear, I've only attended one in the past, so I'm really not sure about the practicality or usefulness of this suggestion).
 * We can also try to focus on one or a few works if there is interest from the British Library. The US NARA project seemed to work quite well.  I can't guarantee anything but if it does work, the British Library could link to the final product or even do something with QRpedia codes. (Ideally, this would be something unavailable elsewhere and not possible without the British Library's involvement.)
 * NB: Regarding scans, you are right about DjVu and PDF. However, if they are simple works of only a few pages, basic JPEGS will work as well.  As long as they are framed properly these can be taken with a basic digital camera.  Also, if there are specific works that can be scanned and you can get an estimate (under £250), we could request a microgrant from Wikimedia UK.  This is probably more likely if it relates to their current World Wars 1 and 2 project (as you mention). - AdamBMorgan (talk) 23:51, 1 September 2012 (UTC)
 * Andrew Gray, in your discussions with the appropriate personnel, it would be most helpful to mention the British Library's lack of online searchable titles by author. (This lamentably lack of feature is also true of the German and French National Libraries). The exceptions (in the English language), are the US Library of Congress and the Australian National Library. — Ineuw talk 00:18, 2 September 2012 (UTC)
 * To add to Ineuw's point, the British Library do not appear to be part of VIAF. VIAF is used for authority control on Wikisource, Wikimedia Commons and German Wikipedia.  I understand English Wikipedia plans to start making more use of it in the near future.  This might add weight to improving their search function and online presence. - AdamBMorgan (talk) 00:41, 2 September 2012 (UTC)
 * I'm one of the organisers of the enwp VIAF plan ;-) The BL does contribute to VIAF, but slightly indirectly - the entry for "Library of Congress" in that page actually represents NACO, a cooperative authority control program headed by LoC and including several other national libraries, including the BL as well as counterparts in Scotland, NZ, etc. I believe this arrangement dates to the mid-nineties, hence why they're not an independent member. Unfortunately, it's not currently possible to resolve VIAF IDs directly into a search of the catalogue - this is something I've brought up with the authority control group here.
 * When you say "online searchable titles", do you mean specifically digital materials searchable & accessible through the catalogue? There is a large chunk of material a) currently accessible through the catalogue, but limited to on-site access only; or b) linking straight to subscription services, & neither is much help for us... I do know there are plans afoot to more closely integrate the digital content with the catalogue (particularly for the MS collection) and hopefully as more material becomes publicly available to off-site readers, this will be linked accordingly.
 * Regarding scans, it looks like I should be able to get a couple of sample works uploaded to Commons shortly - I just need to figure out the logistics of approval. I'll let you know what transpires Andrew Gray (talk) 12:01, 4 September 2012 (UTC)

New help pages wanted
A problem with Wikisource at the moment is relative lack of help pages. New users don't know how to start and it isn't easy to find useful information. Some things aren't explained anywhere except Scriptorium archives and maybe so user talk pages. So, we need to make new help pages (or see if old ones need updating). If you want to help, there's a page here to co-ordinate this, or you can work on help pages on your own. - AdamBMorgan (talk) 00:27, 1 September 2012 (UTC)

Page numbering and layout toggles missing for me
Is it just me, or have we lost page numbering on the left hand side of our transcluded Proofread Page works? Plus I no longer have the toggle that rotates through page layouts. not been doing as much in main ns so not been paying attention to page numbers. — billinghurst  sDrewth  11:14, 1 September 2012 (UTC)
 * Happened to me as well. Have you updated to the latest version of the script "Without Text" (see some posts above)? It was changed lately due to some issues. In my case, once updated to latest modifications everything got back to normal.--Mpaa (talk) 11:32, 1 September 2012 (UTC)
 * BLING! ✅ — billinghurst  sDrewth  11:39, 1 September 2012 (UTC)

Where the heck has the OCR button gone?
Sorry, I seem to remember this has already come up, but I can't find the discussion. Where the heck has the OCR button gone? Hesperian 12:52, 1 September 2012 (UTC)


 * What do you use it here for? All of my Optical Character Recognition is done off-line. You can have mine since I have no use for whatever it does here. &mdash; William Maury Morris II  Talk 14:19, 1 September 2012 (UTC)
 * I don't see it either. It is turned on. Also cannot see it at frWS and their's is turned on too. Maybe a MW issue in the scripting on oldWS. — billinghurst  sDrewth  14:28, 1 September 2012 (UTC)

I am pleased mine is gone. There are fewer buttons and less confusion for new people. I would like to see the following gone and in the distant future possibly some other buttons could be used for a true customized editor bar. It could also help new people (somewhat offensively so-called "newbies") because it would be less confusing. Respectfully, &mdash; William Maury Morris II  Talk 15:41, 1 September 2012 (UTC)

embedded file - (picture frame) file link - (yellow trumpet) Mathematical formula (La(Tex) - symbol is Square root of n signature with time stamp - symbol is scribbled pound - symbol is # small caps - symbol is

In reply to "What do you use it for?", mostly I prefer to fix a file offline rather than generate OCR on the fly, but in cases where OCR has failed on a single page of a document that otherwise has OCR embedded, as for example on Page:U.S. Department of the Interior Annual Report 1871.djvu/24, then I like to press the OCR button. Hesperian 23:37, 1 September 2012 (UTC)


 * @Hesperian, it (the OCR button) disappeared with the last wikimedia update. @Maury, the pound and small caps buttons you copied from my js, so you can just delete those from your js. The other 4 buttons are standard across all wikimedia implementations and should remain for consistency. Beeswaxcandle (talk) 23:39, 1 September 2012 (UTC)
 * @Maury use the old toolbar and decide what you see, create your own, remove the unneeded. — billinghurst  sDrewth  01:59, 2 September 2012 (UTC)

Script helper to populate Index:
had a script at frWS that he designed to assist to extract data from Commons:File:... and populate Index:... Following a request he has done some tweaking of the file, and this has allowed for me to have a local customisation. At some point we can gadgetify it, though others should test it first. To do that, add the following lines to Special:MyPage/common.js // test of local implementation of Gadget-Fill_Index.js     mw.loader.load('//en.wikisource.org/w/index.php?title=MediaWiki:Gadget-Fill_Index.js&action=raw&ctype=text/javascript'); Please reply here to the functionality of your tests, and your opinion of whether this would be a useful gadget. Thanks. — billinghurst  sDrewth  15:45, 1 September 2012 (UTC)


 * Is there anything else that needs to be done beside importing the .js? [Re]creating Index:Beowulf (Wyatt) x-xi.pdf (a file recently merged into it's parent source file and no longer needed as a stand-alone Index) for example doesn't come up any different than before adding the above to my common.js. -- George Orwell III (talk) 19:04, 1 September 2012 (UTC)
 * I recreated the page with the script for me. I do note an issue, which I need to explore, but think that it means that we have someone pointing here rather than using the . If that is not what you received, just checking that you purged your common.js file. — billinghurst  sDrewth  01:52, 2 September 2012 (UTC)

Migration from Category:yyyy works to year in header
Hi. In order to move on with the clean-up of the Undated works Category, I could continue the task to replace with the year parameter in the header (disclaimer: I will not check if the current categorization is correct …). Is there anyone objecting that I should not proceed? Feedback welcome.--Mpaa (talk) 17:40, 1 September 2012 (UTC)
 * I think this sounds like a good idea and support the move. The information should be kept in the header anyway.—Zhaladshar (Talk) 18:15, 1 September 2012 (UTC)


 * Great Idea!!!! but the issues surrounding implemention still haven't been resolved/addressed. Please see...
 * Wikisource_talk:Maintenance_of_the_Month; and
 * Template_talk:Header
 * ...until those issues are at least addressed, if not resolved outright, I'm opposed to this being done across the board. Bad enough recent runs have made dozens of American legislative works the apparent products of "stuttering" politicians (i.e. Judicial Improvements Act of 1990 (1990) ). -- George Orwell III (talk) 18:42, 1 September 2012 (UTC)
 * George, can I say in an Australian context that there can be later versions of acts marked with a year. For instance the Quarantine Act of 1908 has many versions up to the current day, so I don't see that such an appendage is necessarily wrong. — billinghurst  sDrewth  01:56, 2 September 2012 (UTC)
 * Show me the instances of passed legislation all excatly named/titled Quarantine Act of 1908 and I'll show you why its absolutely "wrong". You are mixing 'published as passed' with the 'ongoing codification' of the law. I'm not opposed to the auto categorization functionality via the header template parameter - just need a way to prevent the redundant (if not misleading)display of it for certain instances. -- George Orwell III (talk) 03:54, 2 September 2012 (UTC)
 * The legislation history is at Comlaw. The Bills are what is put before Parliament and can propose/changes single/multiple acts, changes then put before the Executive Council, and the Acts are continually republished under the title "Quarantine Act 1908."  It may not be what happens in US law, however, I can say that it is not incorrect for Australian legislation to have a publication year different from the legislation.  Similarly, we have Ombudsperson reports for years (named in titles), that are published in the subsequent year.  That is not to say that we cannot remove the year from the title field when it is not part of the title, and clearly is meant to indicate the year of passing, we will need mindfulness. — billinghurst  sDrewth  13:58, 2 September 2012 (UTC)
 * It's exactly the same thing - you still are confounding codification with simple publication. Here Title 42 - United States Code remains constant; There Quarantine Act 1908 remains constant. These are the ongoing codifications of a particular area of law. To "see" what was actually changed in a law you need the bill making the amendments - striking this part; adding that part; repealing the other -- these would be the bills in the far right hand column at your link or, in our case, something like Patient Protection and Affordable Care Act (i.e. changes being made to Title 42) here in the U.S. Frequently, these bills are named things like Amendments to the Public Health and Service Act of 1993 (1993) here just as it would be for Act No. 158 of 2007 (2007) over there (that tends to get to be too far down in the weeds for the layman after staring at it for awhile & partly why we use an additional "accounting system" for legislation beside the designated title(s) in the form of Public Law 111-148; exactly the same as found under the Patient Protection and Affordable Care Act).  Again, all I'm looking for is either keeping the manual YYYY categorization viable, much like it has been a norm across the wikisphere, or let the year be hidden in certain instances if that is the only way to cat the year in moving forward. -- George Orwell III (talk) 15:30, 2 September 2012 (UTC)
 * I have implemented something that will hopefully solve this problem. I've added some stuff to the header template.  The parameter "noyear" will stop the year being displayed in the header and the parameter "noyearcat" will stop the header template attempting to categorise a work by year (replacing my previous addition of "no_year" for the same, which would have been confusing if it stayed).  There is also "override_year" which will put any text in the year part of the header but will no categorise it; this is an older addition.  I've also mostly copied my subroutine from the author template to categorise works by year (overcoming the 1000 AD problem, where works before this date were ignored, and adding more flexibility).  I may have to add to this as I've already seen "Medieval" used in one work and the "YYYY-YYYY" format in a few others.  Hopefully this will allow Mpaa to continue; users can still add works by year categories manually, of course, and they will probably be caught in future sweeps like Mpaa's current work. - AdamBMorgan (talk) 01:07, 3 September 2012 (UTC)
 * How to tell where 'year' should be displayed and where not? Any clue? A couple of common instances: Category:UN Security Council Resolutions in 2000, or Category:Presidential radio addresses of George W. Bush, or Category:1985 court decisions (pls note that there are pages with and without here). Common sense, maybe?--Mpaa (talk) 18:36, 3 September 2012 (UTC)

Not sure if you are being illustrative or rhetorical with that question, but as far as American legal or legislative types of works go - Always hide the year. A loose rule of thumb for us (or U.S.) is if a work ends in "... Act of 1992" then its almost a lock that the bill/ruling/order/opinion/etc. in question was published in that same year (if only as a stand-alone if not in a compilation) and displaying the Cat year would be redundant. As for other nations, I can only look to the closest body of legal-ish work to the U.S. (South Africa) here on en.WS and I see that [//en.wikisource.org/w/index.php?limit=250&tagfilter=&title=Special%3AContributions&contribs=user&target=Htonl&namespace=0&tagfilter=&year=&month=9 User:Htonl] has already determined that hiding the year is best in their case as well. For the rest, I guess trial and error might be the only way to find out for sure. -- George Orwell III (talk) 21:36, 3 September 2012 (UTC)
 * I would say pragmatic :-) Just trying to sort out things before acting and to avoid rework. To keep life simple avoiding to add too many parameters when not strictly necessary. Just as an example, my approach would be to add year without bothering about display in Category:Presidential radio addresses of George W. Bush, even if the original editor did not inserted it. I will keep away from laws instead, tough ground :-) Court decisions give me a bit of headache but I saw that a mixed approached is used there, do not know if that was a non-wanted consequence of previous runs of this swap process.--Mpaa (talk) 22:01, 3 September 2012 (UTC)
 * Both the bulk of the addresses and the court decisions are from an era prior to such concerns over uniform title formatting or structured works categorization like we seem to have now so I can sympathize with you on that point -- there is no real standard for those. The real issue with both bodies of work is they should be falling into their own sub-directories for any given year since they consistently run into the dozens if not hundreds every year. Putting them as individual Category:Works of YYYY only serves to clutter up the category list from more pronounced literature and the like published in that same particular year (For example if the Executive Orders issued in 1917 were listed one each as a work of Category:1917 works, the list would push over the 200 default unnecessarily and seem to give them undue weight in relation to those other works). Ultimately, this sub-directorization is really what is needed for works of those types and I don't know if the year parameter can produce that result efficiently never mind if displaying the year is in question for them or not. -- George Orwell III (talk) 23:23, 3 September 2012 (UTC)
 * The more I look into this problem, the more complex it gets due to very different situations. I agree with you that for such bulks Category:Works of YYYY only clutters. Much better approach IMHO is to use noyearcat together with year (since it is decided that the year info shall reside in the header).--Mpaa (talk) 20:55, 4 September 2012 (UTC)
 * So the dictionaries, encyclopeadias, compilations and other like bulks should have an extra header work-load compared to straight books or stories? No need to answer that but I did bring it up because this circular-patchwork again goes back to moving away from the current table based header to an all div based one once and for all. A new header where the standard layout is kept but clearly defined alternative blocks that can be swapped in, out or added as needed are also an option for just such cases. The header and its options would be much like the all-div based header(s) being used on the French wikisource for some time now. I tried to do this ( see Template:Header/testcases2 & Template:Header/dyn ) but I never could "finish" it in time for a real test-run & formal proposal before something else took my attention away (or before additional changes were made to the existing basic header). -- George Orwell III (talk) 21:44, 4 September 2012 (UTC)
 * The vast majority of the South African laws are simple as-enacted texts that don't involve consolidation of subsequent amendments; and they all have titles that end in ", 19xx" so setting the noyear parameter was an obvious choice. For the few that do involve consolidation, I am still pondering whether they should be included in the category for the year of the original text, the year of the latest amendment, both, or neither. But I set the noyear parameter anyway since there's still no need for the year to appear twice (or two years to appear) in the header. - Htonl (talk) 13:52, 4 September 2012 (UTC)


 * Reminder that evolving works like bills that have NOT been passed are are excluded from WS. JeepdaySock (talk) 10:47, 4 September 2012 (UTC)

Book uploads and link requests
Just managed the IA processing, the commons uploads, and the linking of the following titles based on requests by our editors. Admittedly, being fairly new at complying with others' requests, I ran into unknowns with two of the titles. Knowing more would have helped me complete the tasks smoothly i.e: being told what Commons/Wikisource title of the upload is needed and whether it's a replacement or a new book.


 * Index:Our Sister Republic - Mexico.djvu This is a new book added to our collection.
 * Index:Stories by Foreign Authors (French).djvu This is an addition to a series, but I didn't know the series number.
 * moved to the correct numbering--Mpaa (talk) 12:28, 5 September 2012 (UTC)


 * Index:Rivers, Canals, Railways of Great Britain.djvu This is a corrected replacement.

On the commons, a category is created based on the title and the appended word (book) as shown below, and are also used to categorize the images, maps, etc. of the books, so that the source of origin is known:


 * commons:Category:Our Sister Republic - Mexico (book)
 * commons:Category:Stories by Foreign Authors (book) cancelled ✅
 * This is not needed, there is one in place already commons:Category:Stories by Foreign Authors--Mpaa (talk) 12:28, 5 September 2012 (UTC)


 * commons:Category:Category:Rivers, Canals, Railways of Great Britain (book)

— Ineuw talk 08:51, 5 September 2012 (UTC)


 * Well done, Ineuw. We who do not know all on what to do respect those who show kindness and help us too. My "Sister Republic" request looks good but the category has nothing of value that I know of. There is just one image of a book and I have already downloaded and worked on the images. The other two book requests show in their category, "There are no pages or files in this category." I suspect all three categories will "populate" with time and show all images. Respects, Maury ( &mdash; William Maury Morris II  Talk 09:24, 5 September 2012 (UTC)
 * Mpaa, I canceled the commons category, but based on past experience, someone will change it :-) — Ineuw talk 13:49, 5 September 2012 (UTC)

If you know how to create a DjVu and/or PDF...
...can you please explain the process on Help:Digitising texts and images for Wikisource? I have mostly re-written this page but I have little experience of actually creating finished files (my approach is to upload them to the Internet Archive and let them do the hard work). Thanks, AdamBMorgan (talk) 19:31, 8 September 2012 (UTC)

AdamBMorgan, I don't see where anyone is going to answer your question unless, perhaps, you mention some names. I post this now in hope that someone will reply to you. I do know that people have different ways of doing what you ask. George Orwell III and Ineuw are both very good at handling the files, stripping out watermarks, et cetera but they use different programs. One uses Adobe Acrobat which is expensive and the other uses programs that are free such as Infrared(sp?) IrfanView--to my knowledge. Peace and prosperity, &mdash; William Maury Morris II  Talk 20:30, 9 September 2012 (UTC); strikethrough and correct spelling of program added &mdash;  William Maury Morris II  Talk 09:30, 10 September 2012 (UTC)


 * Back up. Let me be clear - I don't create DjVu's or PDFs from scratch as outlined on the Help: page being cited. The most that I can do with any level of comfort is further manipulate files already created or convert existing files from one format to the other. All that jazz about scanners, cameras and pre-processing is not the same as what I'm doing. The most that I can accomplish is to tidy up an existing file before conversion to another format for somewhat improved end-results or to prep a file for creating/inserting a workable quality text layer needed for the ProofReading process. What Adam is asking for is far beyond the busy work that I can fumble through. -- George Orwell III (talk) 21:46, 9 September 2012 (UTC)


 * I stand corrected on what you work on, George. You have gifted skills so I did think you can, or have done, all of it. I would bet that you can. Kind regards, &mdash; William Maury Morris II  Talk 09:46, 10 September 2012 (UTC)


 * Well, Adam, as you will remember, I did SHSP vol. 7 by scanning every page into images from the book because IA did not have that volume. Then I placed the images into a .PDF file (portable document file) and then I uploaded that to Internet Archives (aka archives.org) They then processed the text. I had to straighten some of the images, clean some. I scanned 2 pages at a time in one experiment and individual pages in another experiment. You and I worked on that together so you will probably remember it. I think you had me upload the finished file from IA to WS and then you with your wonderful skills took over from there. I think I then transcribed that text. I would have to look back at it to see how much transcription I did or did not do. Respectfully, &mdash; William Maury Morris II  Talk 09:46, 10 September 2012 (UTC)


 * Thanks (and yes, I recall SHSP 7), I'll look for some Adobe Acrobat and DjVu Libre help from which to borrow information. - AdamBMorgan (talk) 20:34, 10 September 2012 (UTC)

CSS field on the Index page
Can anybody point me to an example where the CSS parameter of an Index page is properly utilized and works as desired on the content it targets? -- George Orwell III (talk) 23:56, 8 September 2012 (UTC)

Transclusion
Are all finished works ("Done") on WS transcribed? Are they supposed to be? Is that a difficult process to learn--transcribing works? I thank whomever answers. My present concern is the United States Naval Darien Expedition of 1854 with Lieutenants Strain, Truxton, and John Minor Maury. I thank whomever, if anyone, answers. &mdash; William Maury Morris II  Talk 20:22, 9 September 2012 (UTC)
 * William, I think you mean transcluded. No it's not complicated and there's now even a help page that describes it: Help:Transclusion. If you haven't done it by tomorrow, I'll give a hand! Chris55 (talk) 22:34, 9 September 2012 (UTC)
 * Thank you, Chris, and yes, I did mean "transcluded". Kind regards and have a wonderful day, &mdash; William Maury Morris II  Talk 09:16, 10 September 2012 (UTC)

The overall progress of a project
Following a side issue that came up in this discussion I thought of a fairly straightforward fix to the command which would solve the problem of a misleading progress representation for a work occurring on the main page, in a similar way to the recent header parameter.

What we need is an extra parameter to (say base) which tells the extension to print the color bar which applies to the whole work rather than just the pages transcluded by the particular command. This would always be used on the base page for a project but not on subsidiary pages. Are people happy for me to propose this to the maintainers? Chris55 (talk) 09:56, 10 September 2012 (UTC)


 * There are maintainers?! (Just kidding... mostly, I think Amire80 has adopted it recently.) I am OK with a change like this.  However, it would be better if it automatically detected that it was on a basepage rather than rely on user input.  It just wouldn't be consistent in the latter case and automatic detection would apply to all the existing instances as well.  I'm curious if any other wikisources will need to be consulted or not (French would be polite, at least, as—I think—the home wiki of the extension). - AdamBMorgan (talk) 11:52, 10 September 2012 (UTC)

"Source" and "View source" on protected texts - maybe a little confusing?
For protected texts that are derived from scanned sources, there are two "source" links. The "Source" tab on the left lets users view the original scans, while the "View source" tab on the right lets users see the wikitext of the page. This can cause confusion for people who aren't familiar with Wikisource. I'm thinking that it might not be a bad idea to change one or both of them to avoid ambiguity (e.g., "Source" to "Original" or "View source" to "View raw text"). Thoughts? --Ixfd64 (talk) 21:57, 10 September 2012 (UTC)


 * I'd like to keep "Source" for the link to the Index page (although, if we do change it, I think "Scans" would be more intuitive). It makes more sense when we say things like "this work has no source" if the tab is also called "source."  I support changing "View source" to "View raw text" or "View wikitext" to reduce ambiguity.  While this tab is common across Wikimedia, it is the least important of the two and we rarely protect pages anyway. - AdamBMorgan (talk) 11:26, 13 September 2012 (UTC)
 * Given a choice, I prefer view raw text over view wikitext. View Wikitext sounds like a finished work. People are used to view source just as they are familiar with cite your source. &mdash;  William Maury Morris II  Talk 12:02, 13 September 2012 (UTC)
 * What about "wikicode" which is a mix of "raw code" and "wikitext"?--Erasmo Barresi (talk) 13:04, 13 September 2012 (UTC)

A proposed addition to the ts template
In a previous discussion about table style parameters, I neglected to mention that the use of the ts template eliminates the ambiguity of where the word "style=" is optional or required. However, the most important information that have always been missing is WHERE the parameters are usable. In the table?, the row? or the cell?

Finally had the opportunity to create a copy of the table HERE, with the proposed expansion of three additional columns to indicate where the HTML shortcuts apply. Some of the information is unknown to me because I never used the parameter. I ask anyone who is interested, to make necessary corrections by changing "y/n" or the "?", and indicate in this post whether I can proceed with the replacement of the previous table in the template documentation. Thanks.— Ineuw talk 21:46, 11 September 2012 (UTC)

ScanSet going away
In 2005, before support for multi-page files was introduced, I wrote an extension called ScanSet to provide access to a scan of the 1911 Encyclopaedia Britannica, uploaded directly to the servers in the form of 5000 PNG files. You can see them at User:Tim Starling/ScanSet PNG demo. The images are black and white (thresholded), so not as high quality as the current DjVu scans hosted on Wikisource at Index:EB1911 - Volume 01.djvu etc. We don't want to support the ScanSet extension anymore. So I'm going to remove the extension and the hosted images files. I'll put a tarball up on download.wikimedia.org in case anyone wants the old images for some reason. -- Tim Starling (talk) 21:05, 12 September 2012 (UTC)
 * What about User:Tim Starling/ScanSet TIFF demo? JeepdaySock (talk) 10:26, 13 September 2012 (UTC)
 * Jeepdaysock, what do you mean? I looked at those works and they look both good and important to me but then "youse guys", as they say in South Philly, know more about these things than I. Still, I give my opinion on those works. Respectfully, &mdash; William Maury Morris II  Talk 10:39, 13 September 2012 (UTC)
 * Tim is saying that User:Tim Starling/ScanSet PNG demo will no longer be available because the software that makes it so you can see it is being discontinued. I am asking if User:Tim Starling/ScanSet TIFF demo will remain, these are(were?) higher quality scans of the EB911 to look at when the lower quality ones were not sufficient to tell what was printed. It has been a while (years?) since I worked with EB911 so maybe only the DJVU are needed now. JeepdaySock (talk) 11:00, 13 September 2012 (UTC)
 * Oh, .png vs .tiff vs .djvu &mdash; William Maury Morris II  Talk 11:08, 13 September 2012 (UTC)
 * Both PNG and TIFF will be gone. -- Tim Starling (talk) 00:14, 19 September 2012 (UTC)

It's gone now. The files are archived at http://dumps.wikimedia.org/scanset/Encyclopaedia_Britannica_1911.tar -- Tim Starling (talk) 01:18, 19 September 2012 (UTC)

Needed: Metadata
I think such a page would be very useful, similar to http://en.wikipedia.org/wiki/Wikipedia:Metadata and http://commons.wikimedia.org/wiki/Commons:EXIF. I am however not familiar enough with Wikisource to begin to create one. The reason I think it is useful is because I've been talking to an organization about sharing their collection with us, and I was asked "what kind of metadata Wikisource has". And sadly, I have to tell them "I don't know", because Wikisource has no page to help me out here... :( I hope we can be better prepared next time? --Piotrus (talk) 21:35, 13 September 2012 (UTC)


 * Maybe [//wikisource.org/wiki/Wikisource:ProofreadPage/Configuring_index_pages#Metadata This Page] can lend some insight on what is considered Metadata for Wikisource? -- George Orwell III (talk) 22:15, 13 September 2012 (UTC)


 * Interesting, particularly the part about Dublin Core. The person I was talking to was interested in this, through I am not very familiar with that. I wonder if we need a page on Dublin Core? Or at least a redirect for it? --Piotrus (talk) 00:15, 14 September 2012 (UTC)
 * In order to export metadata from Wikisource, there is a set of html classes that is used by templates of a lot of Wikisources (including English Wikisource). Some tools already use it like Wsexport that use it in order to add metadata to exported books. Template:Author implement also the hcard microformat.
 * Creating a page about metadata in Wikisoutce is a very good idea and I think hat oldwikisource may be a better place to create it because it isn't an English-only problem. Tpt (talk) 18:15, 14 September 2012 (UTC)
 * I've created a beginning of list : oldwikisource:Wikisource:Metadata. Please, expend it ! Tpt (talk) 19:10, 21 September 2012 (UTC)

Trancluding tables
This is formated as a table layout- http://en.wikisource.org/wiki/Chronological_Table_and_Index_of_the_Statutes/Index_to_the_Statutes

However to have not scre up on translcuding it I had to add some additional code in the underlying template (An additional |- before the table a template I was using formats.

I fail to see why adding the additional |- should be needed and think that there is a possible bug in the transculsion code.

Can someone here please write a guide on how you are SUPPOSED to to have tables over multiple pages that do translucde without error or 'unnecessary' coding please? Sfan00 IMG (talk) 00:47, 14 September 2012 (UTC)


 * See Help:Page breaks for guidance. Beeswaxcandle (talk) 07:19, 14 September 2012 (UTC)

Comment period on the Wikimedia United States Federation
There is a proposal for an an umbrella organization for chapters and other groups in the US called the Wikimedia United States Federation. A draft of the bylaws is now up at meta. There will be an open comment period on the bylaws 17 September, 2012 to 1 October, 2012. The comments received given will be incorporated into the bylaws and they will be put up to a ratification vote from 8 October, 2012 to 15 October, 2012. --Guerillero (talk) 21:49, 17 September 2012 (UTC)

Rule problem
Re Page:A Voyage to Terra Australis Volume 1.djvu/1, why won't my rules centre?


 * I don't know the answer but I at least forced them to center. You will probably have to change it although I don't know why since they are centered now. Kind regards, &mdash; William Maury Morris II  Talk 04:33, 18 September 2012 (UTC)


 * I noticed this happening on another page earlier today. I couldn't figure out why they weren't centering on that page either. --EncycloPetey (talk) 04:52, 18 September 2012 (UTC)


 * Sorry to differ. There are too many embedded templates with centering within centering. As it is, the rule centers automatically. So, it's best to enclose the whole page with a single HTML .— Ineuw talk 05:12, 18 September 2012 (UTC)


 * Below shows left aligned to me.


 * Moondyne (talk) 05:17, 18 September 2012 (UTC)


 * You are right. Checked this page of an earlier proofread where the line centered naturally. As soon as I validated the page, it shifted to the left. The change must be affected by the latest software update 1.20wmf11 of Mon, 17 Sep 2012 16:18:18 +0000 (17:18 +0100). I think it's a correctable error in the CSS file. There should be a notice soon here from someone in the know. :-) If not, there will be many messed up pages. — Ineuw talk 05:34, 18 September 2012 (UTC)


 * AdamBMorgan's rule is also thrown to the left. [] &mdash; William Maury Morris II  Talk 17:02, 18 September 2012 (UTC)


 * Hi folks. Don't want to throw more fuel on a fire, but I just changed a fair number of constructs (e.g.) into  equivalents as I noticed they were (now) all rendering as whitespace. Did this sort of thing work O.K. previously? MODCHK (talk) 08:17, 18 September 2012 (UTC)
 * Also chart has apparently lost control of its vertical spacing. Fortunately I don't believe it is currently used in wikisource. The wikipedia version still seems normal as of right now. MODCHK (talk) 00:49, 20 September 2012 (UTC)


 * IMHO; when new programming changes the behavior of a tag/template, don’t update the tag/template, fix the programming. If you adjust the tag/template, it will lead to an endless circle of changing tags/templates. Jeepday (talk) 11:49, 18 September 2012 (UTC)
 * Jeepday is correct and we should not modify these errors caused by software update but wait for the software being corrected . . . and this includes yours truly.— Ineuw talk 18:38, 18 September 2012 (UTC)

Of the three ways to achieve a simple 10em centered line... 











...only the last one is technically still somewhat in synch with the specification for HTML 4.01. None of them are HTML5 compliant, as the specification now recommends to set the parameters in CSS and diferentiate them by class or id assignment. It looks like a recent change in the wmf code has revealed a long existing conflict with the w3 specification is all that happened here and certain browsers are more sensitive to the change than others (my ancient IE6 still centers all three variants given above for example.)

Long story short - this template, among several others, has only worked so far by chance and on borrowed time. The simple formatting of a paragraph level thematic break ( or line ) was never meant to be template-tized never mind wikicoded the way I've always understood the development of the HTML spec. to be in moving forward - too many deprecated options, values, parameters, etc., nowadays. To say that conforming to the standard(s), as late as that may be now from our perspective, requires wmf developers to re-instate what amounts to bad code is never a good solution nor a long lasting one given the state in specification development (imho). -- George Orwell III (talk) 01:25, 19 September 2012 (UTC)


 * In other words, there is no longer any way that will generate a simple 10em centered horizontal rule. --EncycloPetey (talk) 01:32, 19 September 2012 (UTC)
 * All three examples above do not center for you? -- George Orwell III (talk) 01:37, 19 September 2012 (UTC)
 * Correct. Not one of these is centered; all are aligned to the left. FWIW, I just bought a new Mac and am running the latest Safari, but the problem was not in evidence for the first days that I had my new computer.  Rather, it started after a few days, at the same time that others began noticing the problem. --EncycloPetey (talk) 03:02, 19 September 2012 (UTC)
 * It is most likely not you but the latest upgrade to the wmf code - which takes place roughly twice a month now. The only thing that leaps out at me is the rather short < !DOCTYPE > tag. I'm pretty sure this was better defined in previous versions (i.e strict, transformational, etc.) and the lack of this specificity makes for an obvious loss of which file specification to adhere to throughout the page that follows. (I've been wrong about things like this many times before, however). If this is the case, it is a big step in the wrong direction as far as we are concerned here locally as well as how the HTML specification implementation goes wiki-wide in particular. -- George Orwell III (talk) 03:32, 19 September 2012 (UTC)

Just to let you know that I'm also encountering the same problem on some pages, like Anti-Human Trafficking Act, BE 2551 (2008) and others, even ... and template:center have been applied. But on the existing pages without new edits, rules still centre. --Aristitleism (talk) 20:17, 19 September 2012 (UTC)


 * Worked out the SOLUTION for a HTML5 single rule but no luck so far with the double rule. — Ineuw talk 04:06, 20 September 2012 (UTC)
 * Seems to have corrected the double line as well in the above link.— Ineuw talk 04:44, 20 September 2012 (UTC)

Unwarranted changes in the software updates
The above posted rule problem is just another unnecessary tinkering with the software and is nothing more than irresponsible children's play. These software changes which affect our work are nothing more than the following the latest trend implemented by free software organizations like Mozilla, etc., just to keep themselves entertained. In other words they are bored and have nothing better to do.

I would make the developers sit here and create and/or edit for a few weeks at a time, in other words use their software productively. I am willing to bet that they would be more sensitive to the users' frustrations with these unwarranted and unnecessary changes. This applies from wrecking user defined toolbars, to the reduction of the GUI window in edit mode, to save 10 pixels in window height for whatever idiotic reason. — Ineuw talk 05:58, 18 September 2012 (UTC)

Something has changed - deprecated parameters
Seems like the coders to be aren't kidding about finally falling in line with the HTML 5 file specs. Many familar parameters are now automatically dropped or changed to their semi-equivalents. For example, an input in the typical edit box of... ... upon save produces an HTML of...
 * Note the  in the top opening table-tag becomes   once rendered as shown on the bottom;
 * The  in the top table-row-tag becomes   upon final rendering as shown on the bottom; and
 * All the table cell widths,  are converted to   CSS styling in the bottom final rendering.

This probably means many of our current template's coding that use these now deprecated parameters are also broken for the majority of editors out there; we just haven't discovered them yet. What exactly to do about this is anybody's guess at this point. -- George Orwell III (talk) 03:10, 20 September 2012 (UTC)
 * Yeah, we just had a run-in with this at Wikiquote, where it broke a template on the main page. It's a fairly simple fix, but it might take a while to catch everything. EVula // talk // &#9775;  // 03:54, 20 September 2012 (UTC)


 * In general, the best thing to do is to leave things as they are and wait until all the codes will be updated in the months ahead. This turn of events will certainly make the day for those who like writing and running bots.


 * I am very upset because Wikimedia knew their path of progression for quite a while. They had the "jump" on what must be changed. and regardless of their being, or not, aware of the consequences, they were negligent and uncaring in either case where our efforts are concened. Their decision to implement HTML5 wasn't made yesterday and I don't remember any posts about it. It irritates me to no end that we are pushed to be the beta testers for the software, considering that we are forced to test version 1.20wmf12 when the general public still may be using version 1.15 - 1.16.


 * Proofreaders in general, who are NOT INTERESTED in coding, are constantly being forced into wasting their time figuring out what to do, and it's NOT their responsibility to worry about such matters as I have been forced to do now. I spent a good part of these past 24 hours just trying to figure out how to center the rule under HTML5 HERE. The problem seems to have been resolved for single rule but not for the double rule. — Ineuw talk 04:02, 20 September 2012 (UTC)
 * Personally I need to see some examples of the problems caused, not just look at the random html, so we can create solutions. Hopefully the simpler issues will be fixing templates, rather than globally and strenuously bot'ing. — billinghurst  sDrewth  02:12, 21 September 2012 (UTC)

GLAMwiki Toolset Project
dear all,

as some of you may already know, the GLAMwiki Toolset Project, GLAM/Toolset project, is a collaboration between Wikimedia Nederland, Wikimedia UK, Wikimedia France and Europeana, with the goal of providing a set of tools to get materials from GLAM institutions onto Wikimedia Commons in a way that reuse can easily be tracked, and that Commons materials can easily be integrated back into the collection of the original GLAM or even other GLAMs.

as part of our initial goal of creating a GLAM Upload System, we are looking to gather Wikimedia community input on the proposed architecture and technologies. if you have time and interest, please take a look and let us know your thoughts, GLAM/Toolset project/Request for Comments/Technical Architecture.

with kind regards, dan Saw this and thought that it may be of interest to some, especially where some of the imports may be works that we are looking to transcribe. — billinghurst  sDrewth  01:23, 21 September 2012 (UTC)

[Wikitech-l] #switch limits
Over the last week, we have noticed very heavy apache memory usage on the main Wikimedia cluster. In some cases, high memory usage resulted in heavy swapping and site-wide performance issues.

After some analysis, we've identified the main cause of this high memory usage to be geographical data ("données") templates on the French Wikipedia, and to a lesser extent, the same data templates copied to other wikis for use on articles about places in Europe.

Here is an example of a problematic template:

https://fr.wikipedia.org/w/index.php?title=Mod%C3%A8le:Donn%C3%A9es_PyrF1-2009&action=edit

That template alone uses 47MB for 37000 #switch cases, and one article used about 15 similarly sized templates.

The simplest solution to this problem is for the few Wikipedians involved to stop doing what they are doing, and to remove the template invocations which have already been introduced. Antoine Musso has raised the issue on the French Wikipedia's "Bistro" and some of the worst cases have already been fixed.

To protect site stability, I've introduced a new preprocessor complexity limit called the "preprocessor generated node count", which is incremented by about 6 for each #switch case. When the limit is exceeded, an exception is thrown, preventing the page from being saved or viewed.

The limit is currently 4 million (~667,000 #switch cases), and it will soon be reduced to 1.5 million (~250,000 #switch cases). That's a compromise which allows most of the existing geographical pages to keep working, but still allows a memory usage of about 230MB.

At some point, we would like to patch PHP upstream to cause memory for DOM XML trees to be allocated from the PHP request pool, instead of with malloc. But to deploy that, we would need to reduce the limit to the point where the template DOM cache can easily fit in the PHP memory limit of 128MB.

In the short term, we will be working with the template editors to ensure that all articles can be viewed with a limit of 1.5 million. That's not a very viable solution in the long term, so I'd also like to introduce save-time warnings and tracking categories for pages which use more than, say, 50% of the limit, to encourage authors to fix articles without being directly prompted by WMF staff members.

At some point in the future, you may be able to put this kind of geographical data in Wikidata. Please, template authors, wait patiently, don't implement your own version of Wikidata using wikitext templates.
 * While we are not on the server cluster, we should note the instruction about nested #switch: commands — billinghurst  sDrewth  06:28, 21 September 2012 (UTC)

Page:Cairo_Declaration_on_Human_Rights_in_Islam.djvu/13
Can't wrap template opening and closing... Suggestions? Sfan00 IMG (talk) 20:33, 23 September 2012 (UTC)
 * See Template:Block center for one possibility. Hesperian 05:55, 24 September 2012 (UTC)

Obama's UN Speech
http://www.whitehouse.gov/the-press-office/2012/09/25/remarks-president-un-general-assembly?utm_source=wh.gov&utm_medium=shorturl&utm_campaign=shorturl

Is this the sort of item that should be on Wikisource? Sfan00 IMG (talk) 21:37, 25 September 2012 (UTC)


 * Sure. Anything put up on Whitehouse.gov is covered by its own CC-Whitehouse.gov license as well. -- George Orwell III (talk) 22:54, 25 September 2012 (UTC)


 * Now at Remarks_by_the_President_to_the_UN_General_Assembly Although it might need renaming Sfan00 IMG (talk) 17:59, 26 September 2012 (UTC)