Wikisource:Scriptorium/Archives/2013-11

From Wikisource
Jump to navigation Jump to search
Warning Please do not post any new comments on this page.
This is a discussion archive first created in , although the comments contained were likely posted before and after this date.
See current discussion or the archives index.

Announcements

English Wikisource to trial new CirrusSearch

mw:Extension:CirrusSearch is now live as the default search backend for English Wikisource. Please update here if you have any trouble with it. You can still get the old search engine by adding &srbackend=LuceneSearch to the url of the search results page. --NEverett (WMF) (talk) 17:31, 15 October 2013 (UTC)

For the community to note that from this week that English Wikisource will be set up with a pilot of the new search infrastructure CirrusSearch via the addition of mw:Extension:CirrusSearch. I volunteered our community for the pilot as there are significant advantages for the Wikisources in overcoming existing shortcomings in the present indexing systems, and in testing and refining new search functionality prior to a full roll-out. More information about the project at CirrusSearch and the page contains the process by which the developers will be assessing in the pilot.

The implementation will be that we will have the new index turned on as a secondary backend so that the indexes and indexing will be running besides each other. It is my understanding that new data will be indexed very shortly after each save, unlike the current daily reindexing of new data. I am still seeking further information about how we can best configure tests, and searches to allow those at the front end to configure and see more, and will provide that information when I have it. — billinghurst sDrewth 06:43, 21 September 2013 (UTC)

When it is operational, to get a search to utilise the new backend, you should append &srbackend=CirrusSearch to the url. If there is someone clever enough to design a form that does that, or can show me how to adapt the existing form, that would be fantastic. Best available detail is still in beta documentation from the mailing list at http://lists.wikimedia.org/pipermail/wikitech-l/2013-August/071548.html — billinghurst sDrewth 14:39, 22 September 2013 (UTC)
The features have now been documented at mw:Search/CirrusSearchFeatures. Remember that the page is a wiki and can do with your input. — billinghurst sDrewth 08:26, 24 September 2013 (UTC)
Comparative searches searching for Wolff in the Dictionary of National Biography, 1885-1900
I made a template, {{NewSearch}} to create links to old and new versions of the search. I couldn't see how to incorporate it into the {{engine}} search box template. --Eliyak T·C 16:44, 24 September 2013 (UTC)
I'm Nik (manybubbles on freenode) and I spend all day every day working on CirrusSearch. I have a question and two announcements and I figure this is the right place for them. If this isn't, please move it and let me know what I should do in the future.
  • I'm going to switch CirrusSearch's English Stemmer from the illustrious Porter Algorithm to the newer KStem algorithm. It should make fewer mistakes when finding root words. If you see have any trouble with this or object to the change entirely please just reply here.
  • I'd like to switch CirrusSearch to the primary search. &srbackend=CirrusSearch would no longer be required to get CirrusSearch. You could use &srbackend=LuceneSearch to revert that search to the old search. I'd love for this community to find the time to either bless this idea or find problems that need to be fixed with the new search.
  • CirrusSearch has (today) grown an option to do aggressive word splitting. Currently we're planning on enabling it on mediawiki.org because it is really good for searching for programming terms. Searches for "namespace aliases" will find "$wgNamespaceAliases", for example. Would this be good for wikisource? You can read more about it here.

NEverett (WMF) (talk) 16:33, 1 October 2013 (UTC)

I'm excited to see CirrusSearch made default here at enWS. Because of the way WS is set up, searching transcluded text is really essential. Since CirrusSearch seems to work well based on all anecdotal evidence thus far (including my own), any bugs are a distant concern compared to actually getting the additional relevant search results which were previously unavailable to the average user. As far as the technical details: I am not sure whether or not word splitting is a benefit to WikiSource, and KStem I will leave to people who have knowledge of such things. --Eliyak T·C 00:24, 2 October 2013 (UTC)
I welcome a new and improved search. From initial tests, I'm a bit concerned about the how stemming gets applied and "aggressive word splitting". This may be an unusual case, but see my comparative searches for author Duhring:
  • Duhring MWsearch 83 results, found "DÜHRING, EUGEN KARL" the intended author
  • Duhring CirrusSearch, 50 results, most hits unrelated to the author "DÜHRING", but top results given to "DUHR".
@Nik Can you or anyone with CirrusSearch experience help reduce the impact of word stemming in cases like this? Thanks. - DutchTreat (talk) 01:08, 3 October 2013 (UTC)
Looks like KStem is making a mess of that name both with and without the umlaut. You don't actually have aggressive word splitting turned on. Anyway, you can work around the problem by wrapping the name in quotes. That'll drive the stemmer away. I've filed a bug which will have us sorting results that contain "duhring" above results that just contain "duhr". I _think_ that'll be enough to solve your problem. If it isn't good enough there are other things we could do like switch to a less aggressive stemmer. I'll post on here when I've deployed the bug fix. NEverett (WMF) (talk) 20:46, 3 October 2013 (UTC)
That bug is going to need more time to because it involves some upstream components. I think the answer is no, but I need to confirm: is this a show stopper for switching the new search to the primary? I'd like to do that early next week if possible NEverett (WMF) (talk) 20:51, 9 October 2013 (UTC)
I'd say that the benefits of searching transcluded text outweigh the problems from the bug; especially as our proportion of translcuded works continues to increase. - AdamBMorgan (talk) 12:55, 10 October 2013 (UTC)
Any changes to the default search? I'm having trouble finding hints from the Author namespace: cooper returns only 10 hits, none of them for Author:James Fenimore Cooper. Any insights welcome. - DutchTreat (talk) 11:09, 24 October 2013 (UTC)
Fixed! Not sure what changed, but the search of the Author namespace is working again. Thanks for the help - DutchTreat (talk) 09:40, 25 October 2013 (UTC)

Echo is coming to Wikisource this week

If I correctly read mw:Echo/Release Plan 2013, it would seem that we are going to get the release of mw:Echo notifications this week. For those using English Wikipedia, and meta among other sites, you will have seen in in action. There is a recent blog post from WMF to be read at "Notifications Launch on More Wikipedias". — billinghurst sDrewth 10:11, 19 October 2013 (UTC)

Proposals

Verifiability Requirement

We have talked about it, and hinted that the day is coming for years. I propose that we make it official: Modify Wikisource:What Wikisource includes to require any works added to Wikisource after December 31, 2013 must have an original at Wikimedia Commons. JeepdaySock (AKA, Jeepday) 10:44, 10 October 2013 (UTC)

Unless pagescans are not available? —Clockery Fairfeld (talk) 11:18, 10 October 2013 (UTC)
When are scans not available? (Sorry, don't let that sound abrupt; I'm actually asking; the only examples I have worked on have come from Project Gutenberg.) I think it's a good idea to require scans (or, I guess, some digitally-born equivalent... source material, in other words, against which things can be checked). — Sam Wilson ( Talk • Contribs ) … 11:58, 10 October 2013 (UTC)
I'm opposed. It puts the barrier to entry too high. We need to make it easy for newbs to join up and contribute. We can start educating them on the joys of scans once they are settled in and addicted. Hesperian 12:28, 10 October 2013 (UTC)
I can brook "Strongly encouraged" or "Expected, where available, ..." though not a mandated component. Some modern works are not available as scanned works, and some works as being copy and pastes have very little in the way of verification required. So that puts me a little higher on the expectations chart than Hesperian, though suitably lower than Jeepday's proposal. To note that the policy requirement would impose some burden on those checking edits to grab a file for a user and upload to meet the requirement. While such a requirement is good at times, at other times, impossible.— billinghurst sDrewth 13:24, 10 October 2013 (UTC)
I'm mostly opposed. I would have to give up RC patrolling and leave it to others with time, bandwidth (computer and mind), and country-based access to files. Why? Because for every unsourced work or document uploaded by a new editor I would need to go off and find scan files, upload them to Commons with an adequate description and appropriate Commons categories, add the book template, create the Index and the pagelist, and match and split the mainspace text. By then the editor will have lost interest and wandered off, or have copy/pasted another work and it all starts again. The alternative to this is to mark the work for proposed deletion because it isn't scan-backed, which will then get us the reputation of being overly-fussy.

However, established users should be "strongly encouraged" to always provide scans for the works that they add. That said, I'm also aware that some of the works we host have been proofread by Distributed Proofreaders and then ported over. In the end, I'd rather we had the work, then insist on scan-backing only.

I would be very happy to see insistence on a source being provided with every work added. We've got too many unsourced works. The 1590 in Category:Texts without a source are only some of them. We have no real idea whether we should be hosting these, but that's another issue. Beeswaxcandle (talk) 08:32, 11 October 2013 (UTC)

I'm torn. I would like to emphasise scan-backed works but I don't want to alienate or scare off new users. Most people don't seem to realise there's a difference, both readers and editors, which can be a problem for the project. (Some new users who do get the proofreading aspect don't seem to get the transclusion stage, but that's a different matter.) Scans serve as a verification and assurance of reliablity, not just a base for the text, so I would prefer even Distributed Proofreaders-sourced works get split to scans eventually. Perhaps, following Beeswaxcandle, we can change it to a user-based requirement: eg. all autopatrollers after x, all users with n edits/active since x date, or something similar. - AdamBMorgan (talk) 10:18, 11 October 2013 (UTC)
All of these concerns are valid. Verifiability becomes a balancing act between the desire to have a high quality library and the need to keep the library fully staffed by an all volunteer team. The next step then is considering that established volunteers, make a concentrated effort find and make available scans for all the works on Wikisource. Hence the modified proposal below. 09:39, 12 October 2013 (UTC)


One (somewhat oblique to the issue) consideration is that Commons policy exceeds the legal requirement that works be in the public domain in the United States, where the servers are hosted; they also require that works be in the public domain in the country of origin. This is simply a policy of the Commons project, not a legal requirement nor the position of the Wikimedia Foundation. We are under no obligation to adopt it. Indeed, we have sometimes chosen to host scans here that would be deleted were they hosted at Commons. Hesperian 12:28, 10 October 2013 (UTC)

We could change that part to simply "must be based directly on page scans" regardless of the location (within Wikimedia) of the scans. 'Born digital' documents would need to be considered as well, and I admit "page scans" doesn't cover everything we do; like transcripts of audio files, for example. - AdamBMorgan (talk) 10:18, 11 October 2013 (UTC)
Concur with Adam & Hesperian, we already have most (all?) of this in place but it is a small percentage so I did not include it in the proposal for the sake of brevity. Jeepday (talk) 09:39, 12 October 2013 (UTC)

Modified Proposal

Leveraging on the established policy at Wikipedia:

Verifiability means that people reading and editing Wikisource can check that the work comes from a reliable source. The work is physically present in a Library, and/or an electronic version(s) is/are available online.

At the point of new work being begun established volunteers can check for the existence of scans (multiple potential sources) or presence in a library card catalogs (many are available online) and include that on the works talk page. This is not significantly different then our current process for validating that new works otherwise meet WS:WWI. Should we be unable to find it, it is unlikely that it would meet existing WS:WWI, we can than approach the new volunteer early in the process. Jeepday (talk) 09:39, 12 October 2013 (UTC)

Everything that we have should be sourced, though I don't favour anything stronger than "strongly encouraged" with regard to the proposal. I have no issue with us pushing harder on contributors to properly source their contributions. We have verification for a purpose, and we use it. If a work cannot be verified by a second party, then its full verification will not be there. If we think that a work is completely dodgy then we can propose its deletion. That said, I would think that (encouraging) adding a verification process to {{textinfo}} that relates to a library catalogue entry, or WorldCat has value. Commons has their Institution namespace, and their {{IA}} template and if we can ease the way to make these things easier, all the better, and all that necessary. — billinghurst sDrewth 12:13, 12 October 2013 (UTC)
I think the best approach at this time is to always require a source of some kind (web link or library catalog number), provided either via a new parameter in {{header}} (did I really just suggest that?) or a new template {{source}} which will display the link in a visible location on the work page for works not backed with scans. If it's done with a new template, the template should be preloaded (like {{header}}, but for all base pages and without a gadget). It should somehow be easy to locate pages that are not using this template (not sure how to accomplish that - can pages be automatically categorized?) A warning should come up if the template is not used or not filled out. Also, we would not allow users to be autopatrolled if they do not regularly provide sources. --Eliyak T·C 02:19, 15 October 2013 (UTC)
It's already a part of the {{textinfo}} template and is only required for works that aren't scan-backed as the templates for the files that are uploaded to Commons have a source field. It's sometimes inaccurately or non-usefully filled out, but that is a different problem. Beeswaxcandle (talk) 05:39, 15 October 2013 (UTC)
I meant that a template should be used on the main work page, and should be preloaded and mentioned in the edit page notice, more obviously emphasizing the sourcing requirement for newbies. --Eliyak T·C 16:48, 15 October 2013 (UTC)

BOT approval requests

Help

Other discussions

History of the Anti Corn Law League VOLII

This appears to be the second volume of a work already on Wikisource:-

http://books.google.co.uk/books?id=hZkBAAAAQAAJ&printsec=frontcover#v=onepage&q&f=false

Is it reasonable for someone to consider working on the second volume as well? ShakespeareFan00 (talk) 18:38, 27 August 2013 (UTC)

Sorry; I'm a bit confused - what are you asking for exactly?

Are you asking for the GoogleBooks PDF to be processed into workable .DjVu file for Indexing or asking folks if they interested in working with you in transcribing/proofreading the second volume in general? -- George Orwell III (talk) 22:47, 21 September 2013 (UTC)

Belated reply, I am asking if anyone can track down a non-google source, If there are people interested in transcribing of course,I have no objections to them starting on it, as long as the style is matched to the first volume ShakespeareFan00 (talk) 16:40, 15 October 2013 (UTC)

Anyone using custom edit tool buttons?

To analyze some problems, I am collecting info from users who have custom edit tool buttons/bars defined in either their vector.js or username.js and your participation is much appreciated. There is a section on MY TALK PAGE where you can post info, in case you don't wish to clutter this page.

SCREEN IMAGE IN EDIT MODE.

This is my own info:

  1. Skin:Vector
  2. Toolbar loads normally:No. Must click on the 'edit' button several times while in edit mode which essentially recreates the page.
  3. OCR page image loads normally:No
  4. If there is a problem, what is the page count of the file being edited?:600-900pp
  5. Browser, version and OS:Firefox 23.0.1 in Win XP.

I pay well for the info and you can assume that the cheque/check is already in the mail. Thanks in advance. — Ineuw talk 21:38, 28 August 2013 (UTC)

Hi. I am also having intermittent problems with loading of toolbar (FF 23.0.1). I thought I solved it by using mw.toolbar.addButton but now it fails again. It is very annoying.--Mpaa (talk) 16:16, 29 August 2013 (UTC)
"mw.toolbar.addButton" no longer works. Please see my User:Ineuw/common.js. It works, but in most instances, I must click the edit button while in edit mode before it appears.— Ineuw talk 17:51, 29 August 2013 (UTC)
Now I basically copied your commons.js set-up, I even tried to copy exactly your content. No success. In addition, also all the clean-up scripts (the Hesperian-like, so too speak) are not working. I am getting frustrated at this intermittent behavior. Looks like chasing a ghost ... Maybe better and wait if they will come back same way they disappeared :-( --Mpaa (talk) 19:36, 29 August 2013 (UTC)
What happens when you eliminate the toolbar code altogether? Do the scripts function then? My common.js is very simple and still has issues.— Ineuw talk 20:04, 29 August 2013 (UTC)
Now I copied Hesperian Commons.js for the side scripts and that it is OK. Vector.js still does not load anything.--Mpaa (talk) 20:07, 29 August 2013 (UTC)
I don't think it's the skin. It happens with MonoBook as well. — Ineuw talk 21:48, 29 August 2013 (UTC)

┌────────────────┘
Thinking that the loading of 12 button images affects toolbar loading, I removed the images from the code but it doesn't change things.

  1. Accessing a page in read mode (before editing) the page image loads with a 3 second delay.
  2. Opening the page in edit mode - there is a 4-6 second delay for the page image to appear.
  3. Then must click on edit button several times to display custom edit toolbar.— Ineuw talk 18:33, 30 August 2013 (UTC)

A solution of sorts to the edit button issue

For those who use Firefox and used a custom edit toolbar to insert character and text and have problems with the toolbar, I found the following solution.

  1. Installed this Addon
  2. Installed this user created "Pie" button from the Custom buttons Forum page This includes a number of useful characters as well.
  3. Replaced some of the character list in the Initialization code to include my needed extended Latin characters.
  4. My changed code is stored here. — Ineuw talk 18:45, 4 September 2013 (UTC)

mw taketh and giveth back the custom toolbar buttons

For those who are interested, it seems that the mw developers backtracked on their stand and repaired the problem of the legacy custom toolbar buttons because they are working again.— Ineuw talk 16:28, 23 October 2013 (UTC)

Transcribing audio

Over at Index:Ian Charlton.ogg, we are trying to transcribe an audio file. If the index has /1 , /2 , /3 , etc subpages, proofread page tries to extract a jpg from the ogg file, and fails, resulting in a big black box.

Here is an example of the problem: https://en.wikisource.org/w/index.php?title=Page:Ian_Charlton.ogg/1&action=edit

One way to work around this is to initialise each subpage with the player into the header of the page, like Page:Ian Charlton.ogg/2.

Then, in order for the player to appear in edit mode I have created a JS hack User:John_Vandenberg/proofread_audio.js . Would love some feedback on this approach, or alternative suggestions on how to undertake transcriptions of audio. John Vandenberg (chat) 11:24, 18 September 2013 (UTC)

I've not much to add but have you looked at the Timed Text feature on Wikimedia Commons? It's not quite the same but covers a similar area. - AdamBMorgan (talk) 20:48, 19 September 2013 (UTC)
Thanks for pointing me at that. I was looking for it a few days ago and couldnt find it. I have started commons:TimedText:Ian_Charlton.ogg.en.srt. The player has problems with more than two lines of text.
I have a gut feeling that Wikisource can provide a better environment for transcription of audio than Commons. The biggest problem I see is that one SRT file will be problematic if there is crowdsourcing of transcription - e.g. edit-conflicts, and the 'chunks' for SRT are single sentences, instead of sizable segments of the text. A JavaScript editor of the SRT format would be cool.
Wikisource could use templates to export an SRT from our Page: text, however the timing of the text might be best left out, as it needs to be carefully timed to maximise understanding. John Vandenberg (chat) 02:04, 20 September 2013 (UTC)
I have improved User:John Vandenberg/proofread audio.js so that it understands start= and end= (using URL fragments, and tested in Firefox 24 and Chrome 27. I would like to install it as a default gadget, so that the State Library of Queensland knows it will work for anons who they send to transcribe via their Pitch In! program. John Vandenberg (chat) 08:27, 29 September 2013 (UTC)

Reader boost due to Wikipedia's Twitter account

I'm not sure how new this actually is but it was new to me. I noticed a large spike in readers (well, page views) for the last work I transcluded (The Fangs of Tsan-Lo, 194 views on 25 Sep, up from 61 the previous day). After a little searching, I found it was tweeted by the Wikisource twitter account, which was then re-tweeted by the Wikipedia twitter account. I believe the latter, with its 220,865 followers, is the reason for the spike. Other recent tweets include Best Russian Short Stories (363 views on 24 Sep, up from 70 the previous day and falling to 'just' 136 the day after), Workhouse nursing (110 views on 18 Sep, up from 11 the day before), Putting a Stop to Modern-Day Slavery (212 on 18 Sep, up from 29 the day before), and The Legend of Sleepy Hollow (1864) (115 on 11 Sep, also up from 29 the day before). It happens for other projects too: Wikivoyage:Tokyo spiked to 158 on 21 Sep, from 49 the day before and 34 the day after it was tweeted. The Wikisource tweets are obviously driven by the New Texts section on the main page. There is no real moral to this story, other than a reminder to add works to New Texts, I just found it interesting. - AdamBMorgan (talk) 20:06, 26 September 2013 (UTC)

Yepp, I was one of those who clicked :-) I think it is a good way of promoting works.--Mpaa (talk) 20:15, 26 September 2013 (UTC)
Cool! And sorry to be a noob, but where do you get the statistics? I'd love to check out some of the works I've helped transcribe. — Sam Wilson ( Talk • Contribs ) … 05:28, 27 September 2013 (UTC)
Go to the page you're interested -> History -> Page view statistics link in upper-part of the page.--Mpaa (talk) 05:50, 27 September 2013 (UTC)
Hm, can't find anything there (e.g.). Have you got some gadget turned on or anything? — Sam Wilson ( Talk • Contribs ) … 08:36, 27 September 2013 (UTC)
Not that I am aware of, related to this. I see in your page: External tools: Revision history statistics • Revision history statistics • Revision history search • Page view statistics —unsigned comment by Mpaa (talk) 09:26, 27 September 2013.
It's in the blurb at the top, at the end of the second line. The full blurb as I see it is:

For any version listed below, click on its date to view it. For more help, see Help:Page history and Help:Edit summary.
External tools: Revision history statistics • Revision history statistics • Revision history search • Page view statistics

It links to an external tool: http://stats.grok.se/en.s/201309/History_of_West_Australia (note that this was written for Wikipedia, so if you change the date, for instance, you need to remember to put the ".s" back into the URL, otherwise you will be looking at Wikipedia stats). - AdamBMorgan (talk) 11:04, 27 September 2013 (UTC)
This is strange. I dont' see any of that at all! I just have the 'browse history' fieldset, and some info about diff selection and the legend. Ah well, now I know where the stats are, I've bookmarked that. Thanks! — Sam Wilson ( Talk • Contribs ) … 23:03, 27 September 2013 (UTC)
Ah, but it's there when I'm not logged in! — Sam Wilson ( Talk • Contribs ) … 23:08, 27 September 2013 (UTC)
In many cases where you "see" one thing when not-logged-in and something different when you are logged-in, the likely cause is skin related somehow. The current default skin for User:s not logged in is Vector, so you might want to verify/test your current skin selection(s) in your User: preferences first to see if that changes anything while logged in. Make sure you scroll to the bottom of the edit history list as well - there was one point in time under Monobook where the links where forced to the end for me. -- George Orwell III (talk) 23:47, 27 September 2013 (UTC)
I'm seeing exactly the same as Sam. I've tried changing to Vector, but to no avail. I can only assume that there's a setting hidden somewhere in plain view that I can't find. Beeswaxcandle (talk) 08:08, 28 September 2013 (UTC)
What about now? I edited the rather bloated MediaWiki:Histlegend to call Template:Histlegend instead - which is now a table-ized conversion of the old content. -- George Orwell III (talk) 22:31, 28 September 2013 (UTC)
Hi. Previous look&feel was nicer. Can you mimic that? Bye--Mpaa (talk) 10:21, 29 September 2013 (UTC)
No, still the old version. Beeswaxcandle (talk) 22:40, 28 September 2013 (UTC)
Aha! It's a language thing. I had my language set to 'en-gb'; as soon as I put it back to 'en' all is well. MediaWiki:Histlegend/en-gb needs to be the same as MediaWiki:Histlegend. :-) And there was me thinking I was going crazy... — Sam Wilson ( Talk • Contribs ) … 23:00, 28 September 2013 (UTC)
Well spotted! I've changed the en-gb version and it works. Beeswaxcandle (talk) 23:09, 28 September 2013 (UTC)
Thanks. Works now. Don't forget our Canadian friends at en-ca though. — Sam Wilson ( Talk • Contribs ) … 23:30, 28 September 2013 (UTC)
Done. Good thing there aren't separate ones for Australian, New Zealand, RSA and Indian English as well! Beeswaxcandle (talk) 23:46, 28 September 2013 (UTC)

Draft Wikisource survey

There is a poll being drafted on Multilingual Wikisource which will probably be going out to every Scriptorium within a week or so. This is part of the Wikisource Vision Development project, with Aubrey and Micru, that was funded by a grant from Wikimedia. The point of this is to survey the Wikisource community, learn how to make the project better and find out what Wikisourcers want. You might want to have a look at the draft, make corrections or suggestions, and so forth. - AdamBMorgan (talk) 17:46, 27 September 2013 (UTC)

I have set up a sitenotice for this survey[1]. --John Vandenberg (chat) 00:43, 22 October 2013 (UTC)

please lend your support to this new Individual Engagement Grant project: Towards a PlanetMath Books Exchange

Please lend your support to this Individual Engagement Grant meta:Grants:IdeaLab/PlanetMath_Books_Project proposal I've put together with PlanetMath contributor Raymond Puzio. Inspired by the PlanetMath Exchange project, our aim with this proposal is to improve the PlanetMath platform and make it easy to produce mathematics textbooks -- for export to places like WikiBooks and WikiSource. Your endorsement of the grant proposal would mean a lot! And any comments prior to the Sept. 30, 2013 deadline will help us improve the proposal. --Arided (talk) 22:39, 27 September 2013 (UTC)

Numerical cuneiform

Would someone please transcribe the symbols on this page? You can find some of them here. Thanks. —Clockery Fairfeld (talk) 05:35, 28 September 2013 (UTC)

It's not perfect since there's no vertical U U in Unicode, they expect us to just use two U characters, which ends up being horizontal. Dunno if that's worth making an image out of or what. Prosody (talk) 06:44, 28 September 2013 (UTC)
Thanks a lot, Prosody. However, since then, an IP has come along and replaced the characters with SVG images. Any thoughts? —Clockery Fairfeld (talk) 07:04, 28 September 2013 (UTC)
Like I said, it might be better, at least for that one ( vs 𒌋𒌋𒑗). And we should probably be consistent. Whatever we go with, there's another one on the following page that I did that would have to be consistentized along with those. Also, I made a typo originally (wrote "𒌋𒌋 for 23, 𒌋𒌋𒑗 for 30" where I should have wrote "𒌋𒌋𒑗 for 23, 𒌋𒌋𒌋 for 30." If someone else ends up being the one to revert back that state, it needs to be fixed. Prosody (talk) 07:23, 28 September 2013 (UTC) (correction 07:24, 28 September 2013 (UTC))
Pardon me @Prosody, I was the IP who had an edit conflict. Feel free to revert; however what font do you have installed to display these (i.e. Prosody's) glyphs? They only show up as hex-in-a-box images for me. 121.218.242.182 08:43, 28 September 2013 (UTC)
The font should be supplied by the Universal Language Selector for anyone viewing the text on the web. The language Akkadian is being set by the {{cuneiform}} template (eg. {{cuneiform|𒌋𒌋𒑗}}), which triggers the Selector to display the Cuneiform/Akkadian font (in this case, actually called "Akkadian"). Are you using Internet Explorer 7 or lower? A look at the Selector page shows that it is incompatible with those browsers. - AdamBMorgan (talk) 14:34, 28 September 2013 (UTC)
And now DanielTom's come along and replaced them with images from the original text. Is it okay? Because I have absolutely no idea what the standard policy is for these matters. —Clockery Fairfeld (talk) 10:12, 28 September 2013 (UTC)
I don't think we have a policy on this. I prefer to use {{lang}} or one of the specific template to set these things, and let the software handle the rest, but that is just a personal preference. Images might be a problem for some users, especially downstream users like downloaders and republishers, but for normal web-based purposes, just being consistent throughout a single work is the closest we have to an actual Wikisource policy. - AdamBMorgan (talk) 14:34, 28 September 2013 (UTC)
Thanks, all. We've decided to stick to the images from the original image. —Clockery Fairfeld (talk) 14:55, 28 September 2013 (UTC)

Test layout border

May I ask who applied the test layout border to the main namespace pages? — Ineuw talk 10:04, 28 September 2013 (UTC)

Pardon me, but what exactly do you mean by that, Ineuw? —Clockery Fairfeld (talk) 10:13, 28 September 2013 (UTC)
Just noticed the heavy two tone border around the pages of the main namespace. . . . named "test layout" and I was wondering who is doing the tinkering.— Ineuw talk 10:48, 28 September 2013 (UTC)
An important request to whoever applied the border frame test scheme on the main namespace articles. Could you kindly change it from being the default to the last of the possible layouts? I find it visually very disturbing. It looks as if I was attending a funeral. Thank you.— Ineuw talk 13:08, 29 September 2013 (UTC)
Not seeing this at all! Is it on all pages for you? Maybe the tinkering is in one of the gadgets? --Eliyak T·C 18:06, 29 September 2013 (UTC)
I'm not seeing anything either. There is a gadget called test layout (MediaWiki:Gadget-pr test layout and MediaWiki:Gadget-pr test layout.js) but that hasn't been changed recently. A few users appear to have personal layouts with that name but that shouldn't affect you. Can you describe the effect a little more? Is this a dynamic layout? Does cycling the layout make it go away? Does it happen on both transcluded, proofread texts and copy & paste texts? Try blanking your personal CSS and Javascript (separately, to determine which one might be at fault). Don't worry, you can just undo the edit like any other afterwards. - AdamBMorgan (talk) 18:15, 29 September 2013 (UTC)
It is gone now. It showed around PSM main namespace articles, and forgot to take a screen shot. Its name was "test layout" and was the 1st item in the cycle list. When cycling through the layouts it returned. I looked at all the gadgets and found nothing that referring to "layout", or something that was checked. But, will look over all the Preference pages as I am not familiar with all the settings, and not wanting to appear as if I am hallucinating. Perhaps someone from the "night crew" (the southern hemisphere) was experimenting. (I am from the northern hemisphere) :-)— Ineuw talk 19:09, 29 September 2013 (UTC)
Ha! 'Night crew'! I love it. :-) Three cheers for UTC+8! — Sam Wilson ( Talk • Contribs ) … 02:41, 30 September 2013 (UTC)
@ABM MediaWiki:Gadget-pr test layout.js was the code - but it's a mystery to how and why it was activated, unless it's because I selected the sandbox gadget? - Strange. — Ineuw talk 11:28, 1 October 2013 (UTC)

Spacing in between maths symbols

Is spacing required in between mathematical symbols? That is, in an equation five plus three equals eight, should it be given as 5 + 3 = 8 or 5+3=7? —Clockery Fairfeld (talk) 15:08, 28 September 2013 (UTC)

I think for ease of reading, it's better to have the spaces.—Zhaladshar (Talk) 16:45, 28 September 2013 (UTC)
Okay then, thanks. —Clockery Fairfeld (talk) 17:00, 28 September 2013 (UTC)
Also, 5+3=7 should never be used, due to inaccuracy. Adam Cuerden (talk) 05:51, 30 September 2013 (UTC)
Yikes! I didn't notice that. Thanks. :) —Clockery Fairfeld (talk) 13:08, 30 September 2013 (UTC)
For comparison there is always the "internal" standard exhibited by <math>\scriptstyle {5+3=7}</math>, i.e.: .
<innocent> well the system neither adds spaces nor complains about the apparent inequality…!</innocent> 121.217.32.78 20:51, 30 September 2013 (UTC)

Paragraph in references

Hello together, why is it not possible to have paragraphs (by inserting empty lines) in references similar to the normal text? See example: test. In the HTML view the empty lines are just ignored and only one text block is shown. Even more interesting is the behavior in case when the text is exported by "Download as PDF". There, the paragraphs are created but not in a correct way (please try yourself). The usage of <p> helps for the HTML view but not for PDF. The usage if <br /> works but does not generate real paragraphs but only newline in HTMP and PDF. --Aschroet (talk) 13:22, 2 October 2013 (UTC)

This is what I would do test1, but not sure if that helps or not. JeepdaySock (AKA, Jeepday) 10:35, 3 October 2013 (UTC)
I use <p>. Please see User:Ineuw/Sandbox7 but I don't know how this appears in HTMP or PDF. — Ineuw talk 13:58, 3 October 2013 (UTC)

Tech News: 2013-40

Latest tech news from the Wikimedia technical community. Please inform other users about these changes. Not all changes will affect you. Translations are available.

New features

  • Special:Listfiles can now display old versions of files a user has uploaded. A new link was also added (Special:AllMyFiles) that gives a list of all files the current user has ever uploaded (that haven't been deleted).
  • There is now an Android application to contribute to translatewiki.net from mobile devices. [2]

VisualEditor news

  • VisualEditor now has a new toolbar with drop-down menus for advanced tools. [3]
  • Many bugs were fixed, some related to copy-and-paste. [4]
  • You can now move references, list of references, templates and other elements with the mouse ("drag-and-drop"). [5]
  • You are invited to comment on designs for the interface to add references in VisualEditor.

Future

  • Developers are looking for ideas of small technical projects that new developers could work on. Please add your ideas. [6]
  • Developers are looking for wikis who would accept to try using secure links (HTTPS) for all users. [7]
  • You can join an IRC discussion about "Beta features", a tool to try new features, on October 3. [8]
  • You can join an IRC discussion about Flow, the new wiki discussion tool, on October 17.

Tech news prepared by tech ambassadors and posted by Global message delivery • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

20:01, 2 October 2013 (UTC)

" Wikimedia Foundation ERROR ! "

I —Maury (talk) 12:15, 3 October 2013 (UTC) sure do get tired of that message along with--

"If you report this error to the Wikimedia System Administrators, please include the details below.

Request: POST http://en.wikisource.org/w/index.php?title=Page:Southern_Historical_Society_Papers_volume_02.djvu/406&action=submit, from 208.80.154.9 via cp1004.eqiad.wmnet (squid/2.7.STABLE9) to 10.64.0.134 (10.64.0.134) Error: ERR_READ_TIMEOUT, errno [No Error] at Thu, 03 Oct 2013 12:08:05 GMT"

Notifications

Notifications inform you of new activity that affects you -- and let you take quick action.

(This message is in English, please translate as needed)

Greetings!

Notifications will inform users about new activity that affects them on this wiki in a unified way: for example, this new tool will let you know when you have new talk page messages, edit reverts, mentions or links -- and is designed to augment (rather than replace) the watchlist. The Wikimedia Foundation's editor engagement team developed this tool (code-named 'Echo') earlier this year, to help users contribute more productively to MediaWiki projects.

We're now getting ready to bring Notifications to almost all other Wikimedia sites, and are aiming for a 22 October deployment, as outlined in this release plan. It is important that notifications is translated for all of the languages we serve.

There are three major points of translation needed to be either done or checked:

Please let us know if you have any questions, suggestions or comments about this new tool. For more information, visit this project hub and this help page. Keegan (WMF) (talk) 18:27, 4 October 2013 (UTC)

(via the Global message delivery system) (wrong page? You can fix it.)

Speak up about the trademark registration of the Community logo.

Hi all,

Please join the consultation about the Community logo that represents Meta-Wiki: m:Community Logo/Request for consultation.

This community consultation was commenced on September 24. The following day, two individuals filed a legal opposition against the registration of the Community logo.

The question is whether the Wikimedia Foundation should seek a collective membership mark with respect to this logo or abandon its registration and protection of the trademark.

We want to make sure that everyone get a chance to speak up so that we can get clear direction from the community. We would therefore really appreciate the community's help in translating this announcement from English so that everyone is able to understand it.

Thanks, Geoff & Yana 19:59, 8 October 2013 (UTC)

Paragaph for topic

Hardly ever read WS; never edited it despite doing Commons and en for several years but I looked in The American Cyclopædia (1879)/Madison (post village) and noticed that it covers two different towns far apart and there's no separate paragraphs for each so I went to fix that and, umm, do I know that's proper house style and practice? No, I don't. So, is there a page that covers such questions? Jim.henderson (talk) 13:10, 11 October 2013 (UTC)

The style should match the original. There should be a page scan attached to show this but apparently not in this case. If it did have a page scan, it would look like this. As it happens, the original publisher didn't separate the sections with paragraphs so Wikisource shouldn't either. (NB: WS:STYLE would cover style guidance in general; like WP:STYLE on en.wp.) - AdamBMorgan (talk) 13:36, 11 October 2013 (UTC)

Alex brollo editing tools no longer work

User:Alex brollo bequeathed some interesting .js text editing tools which I tested previously. Now that I want to use it, they no longer function. Unfortunately, he is not currently available at it.WS, so I am asking here if there has been changes to our software that may be the cause of this disruption.— Ineuw talk 16:13, 11 October 2013 (UTC)

Thanks Ineuw, yes, my tools have been wasted by http: ->https: shift; obviously, a my fault. I fixed them and I'm rewriting the whole stuff; the idea is, to centralyze basic scripts into oldwikisource, adding into projects one script only, listing the set of appropriate tools. Se WIP into User:Alex brollo/common.js (calling oldwikisource general scripts) and User:Alex brollo/PersonalButtons.js (listing tools usable here). I'm testing them on Index:De re metallica (1912).djvu. --Alex brollo (talk) 10:28, 15 October 2013 (UTC)
Alex, I tested your links and the sidebar tools work fine but the User:Alex brollo/PersonalButtons.js doesn't show up - or work. Is the "newButton" function a replacement for the "mw.toolbar.addButtons" function? — Ineuw talk 17:52, 15 October 2013 (UTC)
Ok: I need a socket-puppy account to test the whole set of scripts; and I'll take a look to your pages.
No, newButton() is an "original" idea, I'll save a screenshot NOW to show results. Please come into User:Alex brollo/My tools doc and its talk page; I don't want to floot community space here. --Alex brollo (talk) 18:41, 15 October 2013 (UTC)
This conversation is continued on the User talk:Alex brollo/PersonalButtons page.— Ineuw talk 06:42, 20 October 2013 (UTC)

Copyrighted material found in Armistice Day

This compilation, Armistice Day, from 1927, was validated with a lot of good effort and is right now listed as a new text on the Main Page. But I've found a selection that is still copyrighted through renewal (R88408).

How do we handle this? Is it possible to delete those page(s) and redo the DJVU? ResScholar (talk) 18:03, 15 October 2013 (UTC)

Blank (revdel) the pages, but keep the numbering.. ShakespeareFan00 (talk) 19:37, 15 October 2013 (UTC)
Which pages? (So I can blank my own efforts.) ShakespeareFan00 (talk) 19:38, 15 October 2013 (UTC)
OK I've blanked the pages of the contested portion. If an admin would like to apply revision deletions.ShakespeareFan00 (talk) 19:51, 15 October 2013 (UTC)
Let me know if you find other renewals. ShakespeareFan00 (talk) 19:52, 15 October 2013 (UTC)
On a related note- Index:Canadian poems of the great war.djvu should be checked for renewals as well. ShakespeareFan00 (talk) 19:37, 15 October 2013 (UTC)
Removed from Main Page ShakespeareFan00 (talk) 01:01, 16 October 2013 (UTC)
Given this all, it may be worthwhile to try to identify all post-1923 compilation works hosted here and verify that all material in them is public domain. Prosody (talk) 01:52, 16 October 2013 (UTC)

ShakespeareFan00 has nominated the Armistice Day djvu file for deletion on Commons. What will that do to the validated pages? ShakespeareFan, if you want to give up on finishing Armistice Day, that doesn't mean everybody else does. I told you at WS:CV, we only need to look up the copyrights for the works in the acknowledgements section--about 40.

We could put a copyright notice on the file page for a short time, make the copyright searching a Wikisource project and then patch the djvu file with blank pages. Except I remember needing to go into DOS to get my DJVU converter to insert pages, so maybe someone else can do it faster and better. ResScholar (talk) 09:20, 16 October 2013 (UTC)

November is typically a Validation Month for POTM, I see no reason why this November can't be a 'status' verifcation as well as a transcription validation. If this is an agreeable situation, I'll withdraw the DR at Commons. ShakespeareFan00 (talk) 13:12, 16 October 2013 (UTC)
You are welcome to post a clarification at Commons, the issue is NOT the whole file, it's the scans of 'specfic' post 1923 publications, which could be patched by redacting in the DJVU. There wasn't a seperate 'copyright redaction request' at Commons, so a deletion request seemed to be the appropriate process. I can imagine Commons will close it as keep very quickly as a keep.ShakespeareFan00 (talk) 11:09, 16 October 2013 (UTC)
If the file is a deletion at Commons, but fits within our copyright criteria, then the file should be tagged with {{PD-US-1923-abroad-delete}} which should then have the djvu file transferred here. — billinghurst sDrewth 11:43, 16 October 2013 (UTC)
I let ShakespeareFan know his CV/DEL nomination Canadian Poems of the Great War may be applicable to that tag. ResScholar (talk) 07:40, 17 October 2013 (UTC)
AS I understood it Canadian Poems of The Great War IS already localised. It is a 1918 work of Canadian origin so is PD in the US anyway, so that portion of the Deletion issus is resolved (detagging) ShakespeareFan00 (talk) 11:11, 17 October 2013 (UTC)
Garvin's contribution should technically be out of copyright in it's origin country (1935+70)=2005, but as this post dates URAA, I am now wondering if portions of this were also restored. ShakespeareFan00 (talk) 11:16, 17 October 2013 (UTC)
A good question (except Canada is a Life+50 country) but probably not relevant to U.S. law, since the U.S. doesn't follow [usually lengthen] foreign copyright terms [to their source country's even if] unless there's simultaneous publication in the U.S. (in which case the work gets the full 95-year copyright term). ResScholar (talk) 19:18, 17 October 2013 (UTC) clarif. 19:49, 17 October 2013 (UTC) corr. 20:37, 17 October 2013 (UTC)
Commons Nomination on Armistice Day scans Withdrawn. ShakespeareFan00 (talk) 20:01, 17 October 2013 (UTC)
See correction above if it matters. ResScholar (talk) 20:37, 17 October 2013 (UTC)
This is giving me a headache: - Page:Armistice Day.djvu/442 - I can't find a Renewal under the Drama sections of the CCE. and the 1927 version on Google Books notes it's a reprint.. When was this first published? ShakespeareFan00 (talk) 20:40, 17 October 2013 (UTC)
It seems that was a reprint taken from a work made earlier in that same year (January, 1927see the Details section).

Can't say that is the first publication date for sure but I believe that's a moot point anyway given the "fact" no renewal was recorded for this particular play (& I looked for one too) so, IMHO, its fair to say the contribution is PD by no-renewal. -- George Orwell III (talk) 23:11, 18 October 2013 (UTC)

Thank YOU One and All! ShakespeareFan00 (talk) 01:25, 19 October 2013 (UTC)

California Highways journals

I have found copies of official journals published by the California Highways and Public Works department, from 1924 to 1966: http://libraryarchives.metro.net/DPGTL/Californiahighways/. However, I'm still trying to figure out the rules on copyright and notability here. I am unsure as to the status of works by the California state government, but my guess is that because most of them are old enough and do not appear to have a copyright notice, they are public domain. Is that the case, and if so, are they notable enough to be included here? --Rschen7754 03:28, 18 October 2013 (UTC)

I also found w:en:Wikipedia:Public_domain_status_of_official_government_works#Template:PD-CAGov, which seems to indicate that California state government works are PD. --Rschen7754 06:13, 18 October 2013 (UTC)
Yes, they are part of our brief and as far as I can tell they are PD for inclusion here. There are also some multi-volume scans for them at the Internet Archive in the DjVu format. These have a reasonable OCR-text layer for proofreading against. No matter which you use for downloading, the images may be best grabbed from IA as they're available in jp2 format. Beeswaxcandle (talk) 06:52, 18 October 2013 (UTC)
Thanks! If I used the multi-volume scans, would I need to worry about splitting it up into different volumes somehow? --Rschen7754 04:14, 19 October 2013 (UTC)
We can split them up in the pagelist on the Index (see Index:Tracts for the Times Vol 1.djvu for an example where I've done this) and at transclusion. e.g. California Highways/Volume 1, California Highways/Volume 2, &c. Other than that nothing is required to split them up. Beeswaxcandle (talk) 04:39, 19 October 2013 (UTC)
Okay, I've got Index:California Highways and Public Works Journal Vols 8-9.djvu. --Rschen7754 10:37, 19 October 2013 (UTC)

http://dumps.wikimedia.org/other/pagecounts-raw/

"Who came up with this stuff anyways?"

The above is improper English, British and American. "anyways" is not needed and neither is "stuff". —Maury (talk) 04:02, 18 October 2013 (UTC)

Copyright of 1949 work

Discussion moved to Wikisource:Possible_copyright_violations#Copyright_of_1949_work Jeepday (talk) 12:33, 20 October 2013 (UTC)

Index proofreading status errors

Is it just me, or do other people have trouble now with index pages like Index:The American Indian.djvu and Index:Mike (Wodehouse).djvu? The proofread status of various pages is no longer being indicated for the individual pages. --EncycloPetey (talk) 18:14, 19 October 2013 (UTC)

I've been getting that, too. No idea why. I thought it was just my browser having issues.—Zhaladshar (Talk) 20:19, 19 October 2013 (UTC)
Its a result of a recent "drop-out" related to caching & css the best that I can figure. A purge or two (maybe three) of your various browser/page caches should restore the status colors back to normal. -- George Orwell III (talk) 23:32, 19 October 2013 (UTC)

Wikicode Issue with US Founding Documents pages

I don't know quite what's going on, but for some reason the pages for the United States Declaration of Independence, Articles of Confederation and Perpetual Union, and United States Bill of Rights all have some sort of weird syntax error in the header, involving an auto-included category for the year of the work in question. Curiously, this doesn't happen on Constitution of the United States of America.

Even more annoying is that, if you go to edit the page and then press "show preview", there is no formatting error anymore! I haven't seen any clear difference in the code for the header between the pages which would point to a cause for the problem in the first place on all but the Constitution, and also how it goes away in edit previews. If someone with more experience with wikicode and the like could check it out, that'd be great. Rnddim (talk) 22:44, 19 October 2013 (UTC)

Its a known caching issue that affected some number of pages using the Header template during a short range/period of time not too long ago (This caching issue was/is also affecting the status colors from initially appearing on Index: pages too).

A "purge" or two (or three) of the page should refresh it back to the normal state & enabling the 'purge tab gadget' is the easiest way to accomplish that, imho. -- George Orwell III (talk) 23:20, 19 October 2013 (UTC)

As this is coming up a lot, and it will probably be useful anyway, I've proposed that we make the gadget default. It's a very small thing that would help solve these problems and others in the future. - AdamBMorgan (talk) 22:15, 21 October 2013 (UTC)

Viet Nam Era song modified: "Where have all the COLORS gone - Long time passing. . . long, long, ago...

Quote from Hywel Dda's talk page

"I didn't mean to save that! Hit the wrong button while previewing! But currently, as we discussed before, I don't see those colors on the index page (no grey/yellow/green/blue pages). Damned annoying."

"The color problem has hit others here too. It will be worked out within a decade. They're all government workers. <smile> --Maury"

As Hywel has stated, he and I both have encountered a problem with colors on pages. Why is this happening? I see all blue on Index pages but after working for some time the correct colors appear. When an Index page is partly proofread the color is yellow, validated is green and so forth. But when we first start to edit, I see all light blue numbers and according to Hywel above -- he sees, well, I really don't know what he sees but he states what he doesn't see. Would someone please explain this and especially to Hywel on his talk page? He is a very smart editor with an excellent education and a famous father smarter than he. Better yet, will someone just please fix the situation? —Maury (talk) 07:06, 21 October 2013 (UTC)

See #Index proofreading status errors section above, where George gives a brief explanation. Beeswaxcandle (talk) 07:12, 21 October 2013 (UTC)

Tech News: 2013-42

Latest tech news from the Wikimedia technical community. Please inform other users about these changes. Not all changes will affect you. Translations are available.

New features

VisualEditor news

  • You can now create and edit references inside media captions. [10]
  • You now need to press the "delete" key twice to delete a template, reference or image; the first time, they only become selected, to avoid accidental deletion of infoboxes and similar content. [11]
  • When you resize images, you will now still see them, and their size will also be seen in the center. [12] [13]

Future

Tech news prepared by tech ambassadors and posted by Global message delivery • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

09:17, 21 October 2013 (UTC)

Index:Book Of Halloween(1919).djvu

Anyone want to try and get this done in 11 days?

I've made a note of the Acknowledgments on the talk page, per what was done elsewhere. ShakespeareFan00 (talk) 11:18, 21 October 2013 (UTC)

And Page:Book Of Halloween(1919).djvu/189 seem to be out-of order in the scans.
There is now a list of quoted poems typed up - Page:Book Of Halloween(1919).djvu/218

Much appreciated if anyone can let me know if there are any poets in there that are problems with respect to life+70 origin countries. ShakespeareFan00 (talk) 18:19, 21 October 2013 (UTC)

None jumped out at me when I scanned the page. A few spot checks didn't find anything either. James Matthew Barrie was the closest I found; as a 1937 death his work would have entered the public domain in the UK in 2008. This is a special case, however: Peter & Wendy is under a unique, perpetual copyright in the UK, but that only applies to royalties and only to commercial publication. It is not valid in the US but I'm not sure what Commons' position is on it if you are asking about that. - AdamBMorgan (talk) 22:52, 21 October 2013 (UTC)

Formatting & images help needed

I'm working on Index:De re metallica (1912).djvu. Please take a look at Page:De re metallica (1912).djvu/145; it contains an excellent image from a series uploaded into Commons by commons:User:TCO.

  1. original image include original caption. Have I to edit the image cutting the caption away?
  2. Presently I leave the caption into the image, but I added the caption as alt parameter of the image. Does this idea compy en.source recommended style?
Reply
  1. We remove the text titles & captions and type it in. Of course text in the image remains as is. — Ineuw talk 10:03, 23 October 2013 (UTC)
  2. There are numerous methods (like yours) for image display and currently the best is {{FreedImg}} aka {{FI}}. This is a very versatile template. Please see the template layouts HERE.— Ineuw talk 10:03, 23 October 2013 (UTC)

Another question (be patient if I didn't browse carefully the style guidelines). Text include many latin words, and æ - œ characters are not used, being expanded into ae - oe. Can I use expanded form of such characters?

Reply
We use ligatures faithfully as is in the original.— Ineuw talk 10:03, 23 October 2013 (UTC)
Thanks. tl|FreedImg is really interesting! --Alex brollo (talk) 12:25, 23 October 2013 (UTC)

If you like mediaeval latin, please take a look to Page:De re metallica (1912).djvu/31 and review my "transcription style"; I used there original characters æ, œ; I saved original u for v; and I expanded abbreviations for que, since I didn't find a unicode equivalent. Please tell me if I more or less complied en.source style rules for ancient texts. --Alex brollo (talk) 21:53, 21 October 2013 (UTC)

This is not an ancient text. This text does originate from a Mediaeval Latin source, but it was printed in 1912 and edited to norms of that time. That is, some of the orthography does not match mediaeval norms; it has been edited to publication norms for English publishers circa 1912. The ligatures mentioned above, for example are not always present in medieval Latin texts, but show up in 18th and 19th editions of these Latin texts printed in western Europe. The medieval text itself would have been written in a scribal hand, and the orthography would be interpreted by a modern editor for publication with moveable type. So, any changes you make to the 1912 copy would be further interpretations. Such interpretation might take it further from the original, but certainly would change the published interpretation. So, as advised above, we try to reproduce, as accurately as possible, the published copy. To preserve a medieval text, we'd need a scan of the medieval document, and this is usually more difficult to acquire. --EncycloPetey (talk) 01:41, 25 October 2013 (UTC)
Really, Page:De re metallica (1912).djvu/31 seems somehow similar to la:Pagina:Agricola De re metallica.djvu/11, which comes from original edition, 1556. No matter. The only ligature/abbreviation into the text is that for suffix -que, I didn't find an Unicode character for it. Please tell me, how I have to render it. --Alex brollo (talk) 22:44, 25 October 2013 (UTC)
Insert
How many people can or even want to read "mediaeval" latin? We have many immigrants from many nations in the USA (do you where you live?) that read, or need to learn to read, on English Wikisource; English, Spanish, French or German or Basic Latin. —Maury (talk) 16:21, 23 October 2013 (UTC)
I found a trick.... to "cut away" caption using a div with some css so that caption is hidden. But perhaps is confusing. --Alex brollo (talk) 23:06, 21 October 2013 (UTC)
Another question, can I set a specific css for a work into main namespace? I've some tables to format, and much code would be avoided if I could set some simple classes to refine table cells. I found a suggestion to avoid css into Index namespace, and to use instead, css into main namespace, but I didn't find details about how to do it. --Alex brollo (talk) 09:09, 23 October 2013 (UTC)

Index:Book Of Halloween(1919).djvu

Proofread - 1 pass.. (apart from the index.) Anyone care to validate? ShakespeareFan00 (talk) 19:06, 22 October 2013 (UTC)

There is a User List of subscribed Validators around here somewhere. It was created by Chris55 (an invisible administrator who I don't think contributes here) and I recently unsubscribed from it since it was apparently useless. People who are willing to validate will just validate. Their names on a list has served no purpose to my knowledge and perhaps many are not even aware of it. I have learned that validations are a trade. "You help me and I help you in return of the favor" - which includes you ShakespeareFan00 although I did not initially intend to point to you. I was answering your question. I think the lack of validators is a problem here on en.ws. Kind regards, —Maury (talk) 02:53, 23 October 2013 (UTC)

USER STATS

Here is where we can see who's who and what they do. —Maury (talk) 02:43, 23 October 2013 (UTC)

User Stats:

https://en.wikisource.org/w/index.php?title=Special:ActiveUsers&offset=&limit=500

Source tab in the main ns is gone

There used to be a "Source" tab on top of the main namespace articles. Is it possible to restore it?— Ineuw talk 10:31, 23 October 2013 (UTC)

I'm still seing it as normal. It is possible that the introduction of Echo has caused some problems, but not so far for me (I'm using the default Vector skin, which might be relevant). - AdamBMorgan (talk) 12:12, 23 October 2013 (UTC)
It's gone from my views. I checked across a spectrum of main namespace pages, not just PSM, and it's gone. I also use the Vector skin. I looked at my User:Ineuw/common.css but everything there relates to the Page:namespace.— Ineuw talk 16:10, 23 October 2013 (UTC)
Are you still seeing the problem? I've looked around, and I am seeing the "Source" tab everywhere, including in PSM. This may be an issue affecting just you (transient or otherwise), or else it has affected you first (and the rest of us are not yet affected, but have something to look forward to). --EncycloPetey (talk) 01:26, 25 October 2013 (UTC)
If this still happens after bypassing your browser cache: Which browser is this about? --AKlapper (WMF) (talk) 08:29, 25 October 2013 (UTC)

Why teh search function no worky?

Seems to be a problem here... -- Cirt (talk) 18:01, 24 October 2013 (UTC)

You'll need to be more specific. As far as I'm concerned, the search function never really worked here. --EncycloPetey (talk) 01:17, 25 October 2013 (UTC)
It was completely broken for me as well earlier, now it is working. --Eliyak T·C 04:42, 25 October 2013 (UTC)
Was wonky because of bugzilla:56070. Hope that answers your question. :) --AKlapper (WMF) (talk) 08:30, 25 October 2013 (UTC)
Seems to be a bit better now. Thanks everyone for the responses! :) -- Cirt (talk) 15:50, 25 October 2013 (UTC)

Links to Wiktionary (I've removed some)

Friends, I ask your forgiveness of boldness in my recent edits to Twilight (Frankau) in which I removed links, no doubt painstakingly put there by a conscientious proofreader or two, with which there was no greater problem than that they were distracting my reading of the novel! Am I right in doing this? I read in the Style Guide that links are considered a form of annotation, and in this case I certainly feel that to be true: being told the definition of 'morass' was certainly not the author's intention, I think. So I removed the links, and reading flows more easily. And yet! And yet, I wonder if I'm out of line. Can anyone help? Thank you, — Sam Wilson ( Talk • Contribs ) … 01:13, 26 October 2013 (UTC)

The answer will probably vary by contributor. Wikilinks are a form of annotation, but they are an explicitly allowed form of annotation. However, there aren't many guidelines on the use or restriction of such wikilinks. For some works, we definitely want the wikilinks, but others probably have links beyond what is needed.
In my editing of Tom Jones, I added quite a few Wikipedia links, author links, and a smaller number of links to Wiktionary. In this instance, the work was written circa 1749, so (1) a greater number of terms will be unfamiliar to modern readers, (2) a number of cultural references will be beyond modern knowledge, (3) the author assumes the reader will have studied Latin and Latin literature, which although true for an educated reader of the time, is probably no longer true, and (4) the author makes many catty references to personages and locations that will be wholly unknown to a modern audience. In a situation like that, wikilinks are useful for understanding what the author is saying.
Likewise, I know another editor has been creating a detailed original translation into English of a classic Chinese novel. The work has all the Chinese vocabulary linked from the original text to Wiktionary, so that a reader can examine the text for himself.
All that said, I can't say, without knowing which page(s) have been altered, whether the links were useful in the work you were reading. I have certainly seen instances where there was no logical reason to add wikilinks. --EncycloPetey (talk) 01:26, 26 October 2013 (UTC)
My own opinion is that there's no right or wrong answer to how much to link. It depends on what the target audience is of both the original work and of our representation. For example, in Tracts for the Times I'm linking the Bible references as I assume that the reader won't have their Bible open beside them while they read. But in Index:A Critical and Exegetical Commentary on Judges.djvu, which is a Bible commentary, I'm not linking the references because the work will be used with an open Bible. In other examples, when I'm working on a Stratemeyer Syndicate book I'll link to Wiktionary for words that are now little-known because the audience will be younger readers, but I don't do the same in Charles Dickens as the target audience is adults. In scholarly or scientific works such as The Mediaeval Mind or Manual of the New Zealand Flora there is more linking happening because that is how academics will use the works. Lastly, in A Dictionary of Music and Musicians there is a lot of linking between entries and to the DNB00, which enhances the usefulness of a Dictionary. Beeswaxcandle (talk) 01:51, 26 October 2013 (UTC)
To the first mentioned work, I would think that the linking to Wiktionary would be minimal, and would only really be used for archaic words, or so rarely used and unable to be determined by regular etymology. I am a lot less likely to wikilink a work of fiction. Otherwise, I am pretty much in alignment with BWC and that is pretty much my observation of how others work. There are examples of earlier works that were overlinked (measure by our current approach). — billinghurst sDrewth 08:43, 26 October 2013 (UTC)
Thank you all for your thoughts. I am quite in agreeance! I think my reaction to these links was definitely due to this being a novel—in every non-fiction work I've proofread here there have been links, and almost all have been quite appropriate. Actually, very few have been to Wiktionary, which is why these links (like 'philandering', or 'morass', or 'uncongenial') stuck out like sore thumbs... it's just such a slippery slope, for if these words are to be defined, why not others also? :-) Anyway, I shall carry on proofreading and follow my insticts in these matters. (Of course, there's an option that I've not yet explored, and that's the removal of hyperlinks from the epub files I'm reading! A good idea for reading, but not for proofreading like I'm trying to do with this work.) — Sam Wilson ( Talk • Contribs ) … 00:41, 27 October 2013 (UTC)


I would like to see a mainspace ban on all 'naked' inter-project links. When readers click on a wikilink that looks much the same as any other, they do not expect to be taken offsite. Think w:Principle of least astonishment. If you want to link to Wiktionary definitions of little-known words in a work, add to the header's notes field

"See Wiktionary for definitions of floccinaucinihilipilification and antidisestablishmentarianism."

This applies to Wikipedia too. Even other-language Wikisources. Hesperian 03:11, 27 October 2013 (UTC)

Those are presently a slightly different color than WS links. More/different styling could be added to make the distinction clearer, e.g. something like the external link arrows. Prosody (talk) 04:00, 27 October 2013 (UTC)
I have a different point of view from Hesperian. This is a wiki, and it should be fairly obvious that a link is to relevant information though out of the normal linear supply, be it another page here, or another site; one still hits the back key to return to your point of origin. As long as links are kept to the minimal and the highly relevant and helpful, that to me is a better balance. Remembering to add a note to main ns some time after a Page: ns page is proofread is too hard for me. — billinghurst sDrewth 10:04, 27 October 2013 (UTC)
As an alternative that combines both perspectives, is there a way in wikiland of forcing links to open in a new tab/window? I do this when developing training modules in MOODLE and it seems to work there. Beeswaxcandle (talk) 19:23, 27 October 2013 (UTC)
There most surely is—but regrettably it is parameter "$wgExternalLinkTarget" in the PHP, so it is a bit hard to control from inside the wiki. I guess somebody will have to put together a Bugzilla entreaty… (ask for it to be set to the value: "_blank"!) 121.217.2.111 09:07, 28 October 2013 (UTC)
Before touching external link targets, I would suggest that we carefully think about the usability of such a change. When windows open behind links, it can be confusing for users on how to go backward, especially on mobile browsers. New windows work well for dialog boxes, but for straight browsing, I think we should try to avoid them. Thoughts? - DutchTreat (talk) 11:18, 28 October 2013 (UTC)
Hmm. There would also be problems when someone hasn't permitted window popups from their browser, mobile or desktop. OK, not such a good idea. Beeswaxcandle (talk) 07:06, 29 October 2013 (UTC)

Add custom character sets to WikiEditor

After a couple of hours of "monkey-see, monkey-do" experimenting, I finally managed to figure out how to add a section to WikiEditor's main toolbar with the ability further to customize your own page in the booklet of character sets. -- George Orwell III (talk) 01:17, 28 October 2013 (UTC)

  MOVED to Tools and scripts/Wikieditor tweaks

Please use the talk page there for discussions. Prost. -- George Orwell III (talk) 01:22, 28 October 2013 (UTC)

Phe-bot (match and split) out of order

I was doing a match and split job, and Phe-bot matched fine at 7:56 UTC today, but at 8:01 UTC the bot refused to split! It just froze, and the toolserver page where the job queue report resides was not accessible. About eight minutes later the toolserver page became available again showing zero jobs, with the original page still saying it was "splitting..."

I moved the job to the sandbox, and it did the same thing. I tried to redo the match, but the bot won't match anymore either and freezes the same way when it had just performed the exact same job half an hour earlier just fine. ResScholar (talk) 08:57, 27 October 2013 (UTC)

Good thing I yelled: I just tried it one last time, and now it's working. Needless to say, if it malfunctions for you too, leave a note in this section. ResScholar (talk) 09:03, 27 October 2013 (UTC)

Old US State Department country guides

OK, so maybe this is a stupid question, I don't know. But the US State Department has, apparently, ceased production of its old book-length country guides, which as government documents are probably PD for us here. Does anyone know of a location which still carries the last edition of them, possibly in a form which could be converted for use here? Particularly for the smaller countries of Asia and the Pacific, and some of the countries in South America and Africa where English isn't spoken, and which thus get less attention in the US, the guides relating to them might be particularly useful sources for us here. John Carter (talk) 17:04, 27 October 2013 (UTC)

USA Major State Department Publications mentions a Country Commercial Guides. Is this what you are looking for? Solomon7968 (talk) 18:46, 27 October 2013 (UTC)
You can also try HERE - I found the DoS 1981 Pakistan travel advisory there debunking one of those Obama myths claiming he was really a Pakistani citizen and traveled under a Pak. passport, and so on and on and on and on ...... George Orwell III (talk) 19:14, 27 October 2013 (UTC)

"Broken/"?

Can anyone explain to me why these mainspace redirects exist?

  1. Broken/AWB
  2. Broken/Author
  3. Broken/Copyright
  4. Broken/CotW
  5. Broken/Deletion
  6. Broken/Includes
  7. Broken/Links
  8. Broken/OTRS
  9. Broken/S
  10. Broken/Scriptorium
  11. Broken/Style
  12. Broken/TM
  13. Broken/Template
  14. Broken/Templates
  15. Broken/Wikisourcers
  16. Broken/id:184488
  17. Broken/id:186185
  18. Broken/id:186187

Hesperian 05:11, 28 October 2013 (UTC)

Repository/listings of broken links created after the (re)introduction/renaming/reasigning of a namespace from one purpose to another... or for another? ...at some point early on in WS history? I say whack em all - let the NSA sort it out.

I'm guessing you saw that stat. about en.WS having accumulated over 100K of redirects in its first decade, eh? -- George Orwell III (talk) 06:04, 28 October 2013 (UTC)

It does seem like some kind of internal reassignment. I don't believe they were created at that name, yet there is no history of a move.

No, but I believe it. About half of our redirects are case law citations. We have tens of thousands of Supreme Court cases uploaded by bots, and virtually every one has three citation redirects to it; e.g. 238 U.S. 174, 35 S.Ct. 820, 59 L.Ed. 1254 all point at the same unremarkable case. I've done my bit: I reckon I've deleted over ten thousand redirects over the last few years: mostly subpage redirects created when a text is moved with "move subpages" checked. Hesperian 08:19, 28 October 2013 (UTC)

I empathize & appreciate that perspective.. but only to a point. Not to chase anyone into early retirement & therapy or anything over this but any of the superior alternatives to the oodles & oodles of "citation redirects" were not available nor ever a development priority during that time so it is what is unfortunately; & again - one man's unremarkable trash may just be another's unexpected fortune.

At least they are in a standard format that one day can be furthered into more useful data - if the project gets some new blood in it or - just as easily scripted away as they were when scripted to life I suppose.

They're really no more annoying than some of the other stuff one might encounter if they are around long enough if you stop to think about it (the Jjaarbinxi-to-English Dictionary's every word basically all pointing to Wiktionary's file-not-found page or the 4 or 5 decades of the Official Proceedings of the Lower-Uncton Chimney Sweep Guild setup-in-full but amazingly still untouched for example :). -- George Orwell III (talk) 09:17, 28 October 2013 (UTC)

I'm not bothered by all those case citation redirects. There are an awful lot of them but they are valid.

Actually I'm pretty sure I've uploaded a full set of raw page scans for the Proceedings of the Lower-Uncton Chimney Sweep Guild plates. :-P Hesperian 11:29, 28 October 2013 (UTC)

Tech News: 2013-43

Latest tech news from the Wikimedia technical community. Please inform other users about these changes. Not all changes will affect you. Translations are available.

New features

  • The "Toolbox" section in the site sidebar is now called "Tools" in English. You can do the same in your language by editing the interface text on translatewiki.net. Someone else may have already done it. [16]

VisualEditor news

  • On wikis with VisualEditor, you can now use it on pages in the File, Help and Category spaces. [17]

Problems

  • On October 22 (UTC), an error in the site settings caused *.wikimedia.org sites (like Meta-Wiki and Commons) to redirect to wikimediafoundation.org for a few hours. [18]

Future

  • MediaWiki 1.23wmf1 was added to test wikis on October 24. It will arrive to non-Wikipedia wikis on October 28 and all Wikipedia wikis on October 31 (calendar).
  • In the next days, servers in San Francisco will start providing (cached) content to users located in Oceania. If you are in that area and notice problems, please tell us. [19]
  • You will soon be able to test new features easily using the "Beta Features" view. VisualEditor will be in the list on sites where it works and isn't automatically enabled. Another example is a set of changes in the article text style.

Tech news prepared by tech ambassadors and posted by Global message delivery • Contribute • Translate • Get help • Give feedback • Subscribe or unsubscribe.

09:46, 28 October 2013 (UTC)

Call for comments on draft trademark policy

Hi all,

The Wikimedia legal team invites you to participate in the development of the new Wikimedia trademark policy.

The current trademark policy was introduced in 2009 to protect the Wikimedia marks. We are now updating this policy to better balance permissive use of the marks with the legal requirements for preserving them for the community. The new draft trademark policy is ready for your review here, and we encourage you to discuss it here.

We would appreciate if someone would translate this message into your language so more members of your community can contribute to the conversation.

Thanks,
Yana & Geoff

Retrieved from "https://en.wikisource.org/w/index.php?title=Wikisource:Scriptorium/Archives/2013-11&oldid=13656374"