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.
Hi, Splarka created LinkPatroller.js on request (hooray Splarka!), as I was discovering that, when a user makes a spate of known good edits, it is annoying to have to open all of the diff windows to find the [Mark patrolled] link. This Gadget adds "(patrol)" to unpatrolled edits on Special:Recentchanges.
Not the Wikipedia Weekly is planning a special recording this week that could send more readers to Wikisource: we're creating a Wikipedia article that relates to content hosted at Wikisource and expanding it enough for the "Did you know" section of Wikipedia's main page. Wikisource editors are welcome to join.
Here's an example: yesterday a new article for the Irving Berlin song "I Want to Go Back to Michigan" ran in Wikipedia's the "Did you know" section. The article contains a link for Wikisource's content: lyrics, full sheet music, and a restored period recording. Another new article for "That_Mysterious_Rag" should run in a day or two with a link to more content here at Wikisource.
Now Wikisource also has full sheet music for 22 songs by ragtime composer Author:James Scott (musician). Not the Wikipedia Weekly will be writing an article for one of those songs, putting another link to Wikisource content one click from WP's main page. Recording time is 20:30 UTC on Friday, September 26. We'd love to have you with us. Best, Durova (talk) 10:51, 24 September 2008 (UTC)
Progress report: we had a great time, and the result is a new article "Frog Legs Rag" that has a link to the Wikisource sheet music and has entered the queue for the main page at "Did you know?" Best, Durova (talk) 22:15, 27 September 2008 (UTC)
From time to time people have talked about a need for some sort of basic social contract. I put something together at User:BirgitteSB/Trust and Duty for consideration. Please feel free edit these items as you like. We also might want to consider completely different formats than what I used here. Compose a new format in userspace and provide a link here for easier comparison between formats.--BirgitteSB00:57, 15 September 2008 (UTC)
I strongly disagree. While the document is full of positive sentiments, I cannot in any way support such autocratic notions as "duty to conform" and "duty to submit." Eclecticology (talk) 20:29, 21 September 2008 (UTC)
I have to agree with Eclecticology on the issue. While they are noble goals, WS should not be demanding conformity and stating that all editors have a duty to conform to accepted standards. Some great improvements have come to WS in the last few years specifically because of editors' refusal to conform to current standards; Wild Wolf's categorization of authors comes to mind. SherurcijCollaboration of the Week:Author:Albert Schweitzer20:36, 21 September 2008 (UTC)
I was thinking of the guy that was reverting the {{header}} template way back when and insisting that the texts he worked on not use it. But please remove what you disagree with and see if we can get down to the priciples that do have agreement.--BirgitteSB18:05, 22 September 2008 (UTC)
One person being contrary does not warrant drafting and implementing some sort of communal contract. If someone's being a dick and can't work in a collaborative environment, obviously that's an issue, but there's no reason for everyone else to do affected in any way. EVula// talk //20:24, 22 September 2008 (UTC)
Those are fine improvements that Sherurcij added, respect is a better word then duty in this setting. To EVula I would say the communal contract already exists, there is no community in the universe that I am aware of that does not have some form of behavioral expectations, spoken, unspoken or written; we are just writing them down. Jeepday (talk) 23:50, 22 September 2008 (UTC)
Unwritten rules are generally best left that way; once those rules are codified, all sorts of new problems can arise, interpretations can conflict, they can be mis-applied, etc. EVula// talk //03:45, 23 September 2008 (UTC)
They're certainly good sentiments and we should try to live by them, but I fully agree with EVula that they work better as unwritten rules. Ideally we wouldn't need to worry about them, but if we were blocking someone for being unable to collaborate, I wouldn't particularly like to be pointing them to that page in the process. Giggy (talk) 06:56, 23 September 2008 (UTC) Incidentally, seeing Birgitte write "social contract" reminded me of Moulton... and that's a bad thing.
I like respect alot better. I don't see this as "reasons to block someone" at all, but rather "what to expect here". Keeping these rules unwritten doesn't help newcomers develop reasonable expectations and leads to problems unless we develop some other sort of program. Writing down the rules however is the least time intensive thing we can do in this regard. We don't have anything on general conduct. Keeping these expectations unwritten here has only led people to link to the polices at en.WP. We need something local. I would rather try something different than en.WP and worry about scrapping it locally if proves ineffective than letting people link authoritatively to non-local policies that are known not to be effective. BTW I have no idea what the negative conotation to social contract is or who Moulton is. A social contract is just the expected behaivors that a breach of will cause a negative reaction in the community. This contract exists whether we write it down or not. Writing it down and discussing it however can only prevent misuderstandings.--BirgitteSB17:42, 23 September 2008 (UTC)
I agree that "respect" is a better word in the context, and I share her ignorance of the reference to Moulton. I also agree with Sherurcij that these themes might best be placed in a welcome template. The matter of "reasons to block someone" has more to do with the nature of wikilawyering than with the intentions that underlie this document. If you adopt rules (or whatever else you call them) the wikilawyers are sure to flock to them. It is naïve to believe that they can easily be withdrawn if they prove unsuccessful. By the time we realize that the whole issue will have developed into a major controversy, and we already experience the unwillingness of some to get rid of even uncontroversial material when it has long since outlived its usefulness.
I am fully aware that some reach to WP when they perceive a policy vacuum. We can politely tell them that we have not adopted the policy then in question, and suggest that they participate in a discussion about how the current issue might be resolved, with no guarantee that if and when it arises again in the future it will be solved in the same way. A vain attempt to write "something different" from WP is precisely to make the same mistake as WP.
The recent Poetlister situation was painful for the community, and well it should be. Fortunately, we do not face such problems frequently, but it would be unfortunate if, when such a situation arises again in the future, we could apply a cookie-cutter solution that circumvents the painful soul-searching aspects of the dispute. We need to be constantly reminded that both the quick to punish and the quick to forgive are essential participants in the social contract.
Good principles are not enhanced by verbosity. When we try to explain such concepts as "assume good faith," "respect one another" or "anyone can edit." we invariably commit the sin of introducing new words which must themselves be explained. When is a policy a policy? What are common practices? When is an edit trivial? When do we have consensus? These are all among the questions that are begged by this document.
The idea of a "social contract" is founded on implicit understandings, not explicit written ones. It is not a matter of "expected" behaviours, but of ones that are commonly accepted. The negative reaction is not a part of the contract; it is a by-product. Only the principles that define the values are a part of the contract. Eclecticology (talk) 07:52, 24 September 2008 (UTC)
Just to clarify I never suggested labeling this "policy" or anything in particular. Frankly I want to try something as different as what en.WP has done as possible. And I really hope for others here to help me see "outside the box". Which is really why I did not propose exactly what to label this. I didn't want hamper any creative ideas someone might come up with. But I would rather have it on some sort of reference page than just in a template though.
Wikilawering is not a good reason to not make things explicit. It is a good reason to keep things simple and give them context and not to tolerate wikilawyering. I don't see the problem with wikilawyering here. And I don't find them "verbose" either. Policy is policy when it is clearly labeled as such (i.e. WS:WWI). Common practice is self-explanatory. If you mean you are unsure if X is a common practice, ask in the Scriptorium. Same with trivial decisions. We regularly see such questions here of how should I arrange x and What kind of numeral do I use. And mostly we tell them to do how they like because it is trivial. Consensus is pretty clear when we have it, and if we don't have it the editors doing the work can do what they like. {{header}} has consensus, but there are issues that don't. None of this is different than what we already do. We do not even have the principles of "assume good faith" and "anyone can edit" written here. And I can show you actual examples from en.WP of those concepts being wikilawyered on those three words alone without any added verbosity. Since we have no problems of wikilawyering here, no tolerance for it, and the context given with "trust" part of the document works to prevent it; wikilawyering is not a convincing concern to do not do this.
While a social contract is founded on implicit understandings, writing them down does not make them no longer a social contract. The negative reaction upon being breached is the only way to know for certain what the social contract was when it is not written down. And that make for a harsh learning curve that writing things down can prevent.--BirgitteSB13:48, 24 September 2008 (UTC)
Things like "how a page should be formatted" works better as a guideline (or, to use another phrase, put it in a Manual of Style) than being put in some sort of implied "contract". Anti-collaborative behavior is certainly block worthy if it's disruptive (on every edit page, there's the phrase "If you don't want your writing to be edited mercilessly [...], then don't submit it here"); I still don't think we need to codify it any further than that (and if a newbie doesn't "get" that after a few helpful "hey, stop being a dick" messages on their talk page, they never will). I think Sherurcij's idea of working some of our more "behavioral" matters into {{welcome}} is a good idea; that's the most efficient way of informing new editors of how things work without listing out every single little fact (which then opens the door to wikilawyering, as I was mentioning above). Perhaps I'm just responding so negatively to the term "social contract"; it just sounds so high-and-mighty, and just doesn't quite sit well with me. *shrug* EVula// talk //14:46, 24 September 2008 (UTC)
Your first sentence makes me think you are really misunderstanding this. Maybe you could try reading it again with fresh eyes. Nothing in the current version is new or different to how we already operate. And this can hardly be described as "listing every little fact." If you disagree with something in the text, change it, remove it, or put it on talk page if you need help tweaking it. If you agree with everything but don't want it written down, we will simply have to disagree there. You needn't refer to it if you don't like it written down. But I don't understand why you wish to prevent people who want something to refer from having this if you agree with what is there. We have absoultely nothing locally to refer to genreal conduct and just telling people "not to be a dick" is not effective. It sometimes works when the person is being told this by someone they consider a friend or otherwise respect, but I have never seen it work when it is just some random admin. I would oppose making a local version of m:Don't be a dick because of it's ineffectivness and it tendency to rasie a person's hackles rather than bringing them into the fold. So please don't use that link in dealing with problems here; we have not adopted it locally.--BirgitteSB17:38, 24 September 2008 (UTC)
Uh, I wasn't suggesting that the "don't be a dick" essay be taken up here; I was using a very broad, general example of a statement, as I wouldn't expect any admin to use that phrase when dealing with a contentious editor. EVula// talk //18:46, 24 September 2008 (UTC)
As for my issue with the page as-is, I don't like that it comes across as some sort of pact; wikis are very much as "come and go as you please" type projects. Laying out some general behavioral guidelines is one thing, but this... I'm just not comfortable with (as I've said). As per your suggestion, I'm looking it over with a fresh set of eyes, and honestly, the "social contract" concept is just plain creepy to me. I'm all for making sure everyone's on the same page, but I think what you're hoping to accomplish with this page could be done far more effectively without the forced emotional gravitas that you're trying to assign it by making it such an ephemeral, "this is who we are" type thing. If someone violates policy, that's a more directly actionable offense than a vague, nebulous "you violated our social contract", which by its nature has no consequence other than taboo. Obviously, if everyone disagrees with me and this becomes policy, I wouldn't dare do anything to violate any of the general concepts (actually, I won't do that regardless of whether it becomes policy, primarily because everything it says is common sense-type stuff). I just disagree with the presentation and actual function that it fulfills. EVula// talk //19:00, 24 September 2008 (UTC) (and, for what it's worth, that first sentence seems out of place because it went hand-in-hand with a different sentence that i removed before hitting save. my bad.)
Other discussions
Questions
Using ProofreadPage with EB1911
I think it would be a good idea to make use of our excellent scans of EB1911 by using them with ProofreadPage. I believe there are two ways this could be accomplished: using djvu files or using <span class="hiddenStructure" id="pageURL">. Someone made a torrent of all 29 volumes, converted to djvu. If someone could download these and, starting with just Volume 1, split it up into chunks under 20 MB, we could use these just as we would with any other text. I would do this myself, but I have a satellite internet, and I would be heading towards trouble if I downloaded a 2 GB torrent. The other way would be to create pages with arbitrary names and link to the individual TIF files, as done at this example page. However, I can't get ProofreadPage to work with TIF files. Does anyone know how to do this? If there is no way to do it, djvu files may be the best option, unless someone can alter the extension to work with the TIFs. Psychless01:43, 15 August 2008 (UTC)
TIFF is not a standard web format and take way to much place. So the TIFF files should be converted to PNG, or better to DJVU. May be I can help for the conversion. Yann06:07, 15 August 2008 (UTC)
These tiffs are supported by a special extension on Wikisource, developed by Tim. A third option is to migrate that extension to integrate with the proofread page extension. The extension already supports both PNG and TIFF; we could have a user preference to control whether the TIFF is used. John Vandenberg(chat)06:49, 15 August 2008 (UTC)
The above is certainly one way to approach proof-reading, but not the only way. To the extent that I have worked on EB, I work on article size chunks, and find it easier to compare with my dead-tree copy of the books. Eclecticology12:46, 15 August 2008 (UTC)
DJVU files of EB1911 already exist on the internet. I believe these were put together by downloading each individual TIFF file, converting them to PDF then to DJVU, and merging them together.
I'll look at the two extensions again.
This way breaks the proofreading up into smaller chunks and makes it easier to compare to the scans. Hopefully this way will encourage more people to contribute, because they can just click a page and proofread it.
Then, why not use these DJVU files? [1] The easiest would be to ask a developer to upload them, or wait that the upload limit to be increased. Yann16:14, 15 August 2008 (UTC)
I assume those DJVU files are the same ones I linked to above. I doubt the upload limit will be increased to over 80MB anytime soon, so we need a developer to download the torrent and upload the DJVU files, presumably bypassing the limit? Psychless17:53, 15 August 2008 (UTC)
After doing some more searching, I found the djvu files on archive.org: [2]. I can download one volume at a time just fine. So, if no one minds having each volume in 4-5 parts, I'll split it up and upload it to the commons. Psychless18:01, 15 August 2008 (UTC)
I would certainly support using the archive DJVU files. It would be good if the DJVU could be kept together rather than split up. Does anyone know how long it will take to get a developer to upload them? Suicidalhamster (talk) 19:30, 15 August 2008 (UTC)
It has not happened for a very long time. The last time I asked, for JPS1917, it looked like it was going to happen until a senior dev put a spanner in the works. In this case we are lucky, because the scans are already on the servers. Maybe these images can be imported as normal "Image:" pages. Uploading 20MB slices of these volumes as djvu files would be messy, and waste disk space. John Vandenberg(chat)12:06, 16 August 2008 (UTC)
The scans, in TIFF format, are on the servers. However, we can't upload them to the commons and we can't use them with ProofreadPage. Unless we can get them to work with the extension, djvu files are the best option. Psychless00:02, 17 August 2008 (UTC)
I apologize for not bringing this up earlier, but I got Magnentius to upload his/her djvu files of EB1911 to archive.org. The page for them can be found here. These are much smaller in file size than the ones I found earlier. I uploaded Volume 1 to the Commons as two files (superior to five). I created index pages for them at Index:EB1911 Volume 1 A-Alava.djvu and Index:EB1911 Volume 1 Alb-Androphagi.djvu. If everyone agrees we should start with these two indexes, I will start transferring the text to these pages when I'm done with my current project. I will mark it all as not proofread, since varying styles are used on different pages and some may not have have been proofread. Psychless18:29, 30 August 2008 (UTC)
These djvu files are a waste of diskspace, they are not optimal for transcription, and Distributed Proofreaders already have EB1911 set up and the images are cut up for optimal proofreading (one column per image). Distributed Proofreaders have already finished a number of volumes. Unless we are going to establish a better setup than is already in place at EB1911, we should be going to Distributed Proofreaders and helping them complete the transcription and proofreading. Please do not rush this; please do not upload any more small slices of EB1911 as DJVU files. I have spoken with Tim Starling about EB1911, and I think that if we can provide him with DJVU files that are lossless copies of the TIFF scans that he has put together, he will replace the existing TIFF scanset with the DJVU files, thereby saving diskspace rather than wasting it!! Please note that the TIFF files he has produced have been cleaned - uploading a different set of scans is a regression. John Vandenberg(chat)02:49, 31 August 2008 (UTC)
A quick search reveals there are no EB projects one can actually work on on Distributed Proofreaders. About 10 chunks of Volumes 5-6 are in a bureaucratic limbo state, and Volume 1 and slices of Volume 2 and 4 have been posted to PG. I believe we should try to get an effective setup here on Wikisource. As noted here, the djvu files I split were lossless copies of Tim Starling's TIFF files. If the TIFF files are replaced (presumably by uploading the djvu files to "upload.wikimedia.org"), it will make using them much more difficult. We'll have to choose arbitrary page names and use <span class="hiddenStructure" id="pageURL">.
I still believe using the djvu files with ProofreadPage is the best option. I don't know how we could get one column on each page, though; since having two columns apparently makes proofreading more difficult, please tell me how this could be done. Psychless17:42, 31 August 2008 (UTC)
Manuscript sketch for Beethoven's Piano Sonata No. 28, fourth movement
Hi, coming here with a question about music. I recently located a high resolution scan of an original Beethoven draft manuscript and did a restoration on the document. Over on en:wiki Shoemaker's Holiday and I created tandem featured picture and featured sound nominations for the manuscript and the finished sonata. This prompts a couple of questions. First, does Wikisource have the capacity to host musical scores? And if you do, is there anyone who can read Beethoven's handwriting? Shoemaker was going to do a midi for the manuscript, but had a hard time making out exactly what Beethoven had written. If Wikisource regulars are interested in this sort of thing and can move forward I'd be glad to collaborate. My talent is for finding these documents and digitally restoring them--and yes I've found others for a variety of composers. Transcription and (probably) coding would require an editor with different skills. Best wishes, Durova18:09, 25 August 2008 (UTC)
After a collaboration project discussion, we bit the bullet and tried sheet music at the beginning of the year with various results: Wikisource:Sheet music. See the talk page for the discussion since the collaboration started. If you find a musical notation for any work on Wikisource, it would be great to have the image added to our page, and then that page can be added to the slowly growing Wikisource:Sheet music collection.
bugzilla:189 would be very useful, but there are ways around it. The biggest hurdle is finding people who are willing to work in this area, and work around problems until we figure out better solutions.
In Category:Index you might find a few djvu files of sheet music; one I can recall is Index:College Songs (Waite, 1887).djvu, which is used for Oh My Darling, Clementine in a clever way (too clever perhaps). On archive.org there are many more to be found and imported; e.g. 2,500 "songs". Finding works via archive.org should also be on your radar, especially as it comes with the added benefit that archive.org has already done the OCR and embedded the text into each page of the djvu, which means we can import the text onto the correct page here.
It has been a while since the general sheet music collaboration; perhaps we should focus on a single composer for a CotW and see what we come up with.
Forgive me, because I've been out of touch and haven't properly glanced over referenced documents, but I'm guessing that the Lilypond extension isn't currently functioning? I remember looking into it last year some time (or '06, I forget), and pretty sure it wasn't functional then... Jude (talk) 08:47, 26 August 2008 (UTC)
Well, considering we've obviously decided that sheet music could be considered within our scope, perhaps we need to be asking the devs to enable this extension on Wikisource (or at least find out if they think it's a viable extension and would enable it). My two pence, I guess. Jude (talk) 12:32, 26 August 2008 (UTC)
Tim Starling said that Lilypond won't be installed on Wikimedia projects. I understand that this extension is quite buggy. I have installed it on Wikilivres to see what are the problems. It is true that there are serious performances issues, and some other problems. Yann16:42, 26 August 2008 (UTC)
Unfortunate. I seem to recall some discussion of it with him on IRC at some point in time, just couldn't remember the exact outcome. Jude (talk) 04:05, 27 August 2008 (UTC)
Okay; there are obviously easier things to transcribe than this particular manuscript. I'm still figuring out how to get started, so if anyone is bold enough to lend a hand there's an interesting little commons-en:wiki undertaking that might be a good place to start: we've got a recording of John Philip Sousa's "The Stars and Stripes Forever" by Sousa's own band with Sousa conducting, dated 1909. Matching that, I've located an edition of the sheet music for the piece from the Library of Congress dated 1898. The sheet music with cover page is about 6 pages long. Due to other commitments it'll probably be about a week before I have time to do a digital restoration on the image. Think that'll fit within your capabilities on Wikisource? Durova15:42, 27 August 2008 (UTC)
As an irrelevant aside: I'd just like to say that I think that hosting music on WS would be a very necessary step forward, and a very good thing. We should focus on developing this area in my opinion. —Anonymous DissidentTalk12:43, 31 August 2008 (UTC)
<lilypond>
\version "2.3.25"
tune = \relative c'' {\clef treble
\key g \major\time 6/8
e16( dis e fis g b, c d e gis, a b)
c16( e, f g! a c, d e f g a b)
}\score{\new Staff \tune\layout{ raggedright = ##t }}
</lilypond>
Playing around with User:Jayvdb/code I found that adding either <poem> or <pre> to that template caused it to fail (param 1 would be discarded), so I am just using a div. It doesnt look pretty, but it does the job for the moment.
Is there some lilypond syntax that would cause problems being embedded within a template call? I suspect there is some future problem due to this Commons template diff.
Following up: we've now started en:Wikipedia:WikiProject Media Restoration and in its first days it already has seven members. Would love to move forward with cross-project collaboration, doing manuscript and musical score hostings here and linking to your pages from the captions at encyclopedia articles. We're actually aiming for a four WMF site collaboration between Commons, Wikisource, Wikibooks (for our tutorials), and Wikipedia. For now the project under en:wiki's aegis, but as this grows we might migrate it to Meta. Best, Durova04:18, 5 September 2008 (UTC)
It looks like you can use an extension tag as long as you don't need to pass arguments to it (I'm not sure what's wrong there). I've added the poem tag to your template. -Steve Sanbeg (talk) 21:37, 11 September 2008 (UTC)
Moving along--we've got six Irving Berlin songs covered this way now:
Why does the DNB project not put its biographies under subpages? Organizationally, the subpage method is obviously superior, and one-sentence biographies like this are clearly not "works". And the biographies are not even easier to find under the current method. A search for "Munson, Lionel" would return virtually the same results whether the biography was titled Munson, Lionel (DNB00) or Dictionary of National Biography, 1885-1900/Munson, Lionel. Psychless17:55, 31 August 2008 (UTC)
What makes the current format obviously superior is that it will allow for easier connection on a long-term basis between articles about the same person in different encyclopedic works, periodicals, and other collections of mutually unrelated materials. In the case of the DNB it is to be noted that each biography has its own identified author, and can stand alone in its own right. The Munson example is bad good one because it is not a substantive article; it merely refers the reader to a different article under a different name for Munson. Such non-substantive articles are being maintained for the sake of continuity, and to allow for the fact that the person could be sought under the alternative name. Note too that we now receive hints when a word is entered into the search box. What possible benefit would there be by starting a search with "Dictionary of ..." when we know ahead of time that it will produce thousands of results?
There is also the benefit that putting "(DNB00)" makes for a considerably shorter title than putting "Dictionary of National Biography, 1885-1900/", and if we are to cross-link articles that factor becomes significantly more important. While even I have some reservations about using this kind of abbreviated disambiguator, it is hard to have a clear evaluation of that without looking at the scope of different sources that may need to be distinguished.
In the past it has been suggested that the question of multiple articles about the same person or subject could be solved with a disambiguation page. It is an option worth considering, but unless we have the manpower willing to actually assemble such a large number of disambiguation pages on a current basis it will never happen. Can we even foresee which pages will require disambiguation? Table-of-content pages, on the other hand, are much easier to produce: just go through the encyclopedic compendium and list the article titles as they arise in whatever format has been agreed.
To be really useful Wikisource must look forward in the way it organizes its content so as to be most useful to the user. To be competitive with others with relatively massive resources, it must also present material with value-added put together by human input. That human input is in short supply. That requires working smarter, and recognizing that the way these old and often obsolete materials were put together may have suited the demands of the time when they were produced, but are of no benefit now when we know that competing materials were also produced. Eclecticology19:50, 31 August 2008 (UTC)
How, exactly, would two articles from different collections be related under your system? The mere fact that they start with the same letters don't make them related. An Old Acquaintance and An Old Desire aren't related because they start with "An Old". There would have to be a "topic" page listing different articles on that topic. This topic page would link to the two articles, regardless of their titles. As you noted, these wouldn't be maintainable now. Therefore, the current method offers no organizational benefit (like having no binding on a book, pages strewn everywhere), while the subpage method keeps the articles in a hierarchy.
An encyclopedia article isn't a standalone work, even if it does have its own author. The author was hired by the encyclopedia company to write an article specifically for that encyclopedia, and traditional encyclopedia articles aren't republished in other works like poems are.
The length of the title is irrelevant compared to the other issues, since it doesn't really affect the user, who just clicks links. Just use copy and paste. Psychless22:41, 31 August 2008 (UTC)
Two articles would not be related just because they come together, but they would come together because they are about the same person. Sure we could have all kinds of topic pages and categories to bring similar or identical topics together, but we do at least agree that this would not be maintainable in the foreseeable future. What's the benefit of the hierarchy if it's only order for the sake of order? What's the benefit of looking at old encyclopedias in isolation from the rest of the world? It's not as though we were expecting people to read them from cover to cover as might have been the case with the paper volumes.
Why can't an encyclopedia article be a standalone work? To me the key factor for distinguishing standalone works is the mutual independence of the material found in neighboring articles. Alphabetical proximity does not imply relationship. Separate authors are only one additional factors. The employment relationships between the writers and the editor or publisher mean nothing; that's on a par with saying that because you had a certain set of classmates through school you must forever be linked with them in biographical records. It's true enough that encyclopedia articles are not anthologised to the same extent as poems; we now have the opportunity to change that.
# Munson, Thomas Volney (CAB03)
# Munson, Lionel (DNB00)
# Munson, James Eugene (EAm06)
I don't believe that we were looking to reinvent the wheel. Not sure how a search result would display further down a hierarchy. I would think that an article in EB and DNB to same person, title and dates of life, though that becomes more about article naming styles. Happy to learn as Wiki side is newish to me. :-) --Billinghurst08:32, 1 September 2008 (UTC)
Those three examples were created by Eclecticology, not "others".
I cannot disagree with that definition of a standalone work, so I withdraw my argument. Encyclopedia article pages should be named "Name of article". If there is more than one encyclopedia article/work with the same name, the pages should be disambiguated by adding, for example, (DNB00). Continuing the example, (DNB00) should not be added if there is no other work by the same name. I would also advocate a similar system be used for periodical publication articles. Psychless02:18, 2 September 2008 (UTC)
Thank you for giving this some thought. I have already been advocating for the same thing on periodicals. There may still be some things where the hierarchical format is unavoidable, but this is mostly broad-based jumbles of material affecting the publication as a whole.
Although I have been using the disambiguator with everything in DNB, its not without reservations. In preparing an article for Wikisource I like to make Wikilinks to other biographies cited in the article, but it's difficult to know ahead of time which ones will require disambiguation. There's also still some benefit to maintaining a thread of continuity to the encyclopedic work with the "previous" and "next" references. This would be less important for the periodicals. On the other hand having master articles for a subject that either list the articles or have multiple articles on the same page is the other possibility that I have experimented with. I'm not convinced with either possibility being the clear solution. How to effectively interweave these various encyclopedic works remains an interesting challenge. Eclecticology06:32, 2 September 2008 (UTC)
Yes, previous and next links should definitely be added.
I'm still convinced that the method I defined in my previous message is effective. For the links, just link to the article title. When the entire work is complete, a bot would run through all the articles in it and make a list of the links to disambiguation pages in the text. Then it could automatically fix them, reporting any it can't fix (links to other works, not encyclopedia articles). Or an editor could manually fix it, if he or she prefers to do that. It's not an elegant solution, but the system creates the best end-result for the user. Having multiple articles on a page would violate the principle of stand-alone works more than putting them in a hierarchy would. Psychless21:42, 2 September 2008 (UTC)
I'm inclined to approach all these multiple possibilities with an open mind, rather than hard-wiring any of these options. I still like the multiple articles on a page idea, but not so much as to take a hard stand on it at this point when all the possibilities are not yet clear. I suppose if we are going to be using bots it would be just as easy to have one that removes unnecessary "(DNB00)s" as to have one that adds them, so we don't need to come to a final answer on this. Proofreading or providing some Wikilinks of whatever kind suits the editor strikes me as adding more value for now. Our holdings for the Catholic Encyclopedia appear complete, but I have serious concerns about the reliability of our text. Eclecticology04:31, 3 September 2008 (UTC)
I agree with Jude on this, it's not only the standard; but it seems to be the standard for good reason. Moving a page will automatically move its subpages...moving "X" won't move "12345 (X)". We do the same with newspapers such as New York Times. The trailing parentheses are used to disambiguate between two standalone works by different authors, such as The Education of Women (Defoe) and Education of Women (Selassie). But works published as part of a larger work (articles in a newspaper, articles in an encyclopaedia) certainly belong on subpages. In addition, DNB800 will hurt us on search engine returns, whereas spelling out the name of the publication will aid us. SherurcijCollaboration of the Week:Author:Charles Spurgeon06:12, 3 September 2008 (UTC)
As I already stated, what makes an article standalone is the mutual independence of its contents from that of neighbouring articles. To say that there is a standard as you describe it is mere rhetorical trickery. Where is it shown that such a standard was so completely adopted; you seem to be mistaking sclerotized inertia for a standard. Of the four encyclopedic works mentioned only the Catholic Encyclopedia even pretends to be complete.
Where is the "good reason" for this so-called standard? Trailing parentheses are indeed used to disambiguate, but where is it stated that the basis for such disambiguation is limited to authors? The ease of moving pages or the impact on search engines seem to satisfy people who write for the benefit of computer systems; I prefer to write for humans. In the format "Smith, John (DNB00)" there are no subpages to move. (There may be if that article is very long, but that situation is the exception.) The majority (but not all) of our handful of New York Times are indeed hierarchized, but that doesn't prove anything; adopting that as a standard for periodical articles is even more ridiculous than for encyclopedia articles.
For now I don't expect any broad movement to move hierarchic articles to a disambiguating system, but I de expect that the advocates of such a hierarchic system accept that there are alternatives that others may use. Eclecticology18:33, 3 September 2008 (UTC)
It starts to get messy if everybody chooses to use their own alternative naming system, difficult to find articles and difficult to curate them. The "good reasons" I believe I already outlined. While "Smith, John (DNB00)" may not have subpages that need to be moved, if the DNB is in fact renamed for some reason, whether to include the subtitle, correct capitalisation, the community votes it should be DoNB, whatever, then "Smith, John" and the other 14,000 entries all will need to be moved. That can't be done if you're just using trailing parentheses, it's why we have subpages. It's not ridiculous to use hierarchized naming structure for perioidical and encyclopaedias; they are two areas that are absolutely certain to have overlap with other texts on the project ("Obituary", "Himalayans", "Churchill, Winston" are all going to have a half-dozen or more texts with the identical name) - so we put the periodical or encyclopaedia's name in the text as the parent text - just like we put the author's name in Letter to his mother or Wikisource:Resignation letters. Your route is actually encouraging us to rely more on having to direct users to disambiguation pages...that seems like the opposite of progress, we should be making it so that readers have a minimum of trouble finding their article. SherurcijCollaboration of the Week:Author:Charles Spurgeon18:41, 3 September 2008 (UTC)
One can hardly say that the hierarchical system is there to make articles easy to find; if anything, it makes that more difficult. A second-rate encyclopedia like The New Student's Reference Work already has an article on Pope Adrian IV, but the only way you're going to find that is by knowing ahead of time that it's there and looking through the alphabetical listings. The 1911EB will also have an article, so will the DNB since he was the English pope, and so too will many other encyclopedic works. Where is the effort to bring all these together for comparison? A disambiguation page is only good if it's there. The useless author page won't help. The notion that we might someday want to change "DNB" to "DoNB" is absurd since it's an artificial tool in the first place; if we are going to have a vote on that we are in worse trouble than I could ever have imagined. If we want to look up the obituary for John Thompson when he dropped dead in front of the queen, we aren't going to try to guess which publications carried the obituary, nor will it be helpful to look under the generic "obituary" we are going to want to look for John Thompson, and with a common name like that we will also want to make sure we have the right John Thompson. So without disambiguation pages, how will your system find what we are looking for? Eclecticology00:56, 4 September 2008 (UTC)
Disambiguation pages are actually really easy to navigate. I've seen an average fourteen-year-old completely unfamiliar with Wikipedia navigate a disambiguation page in a couple seconds. A user would probably get through a disambig page faster than they would a bunch of search results.
The base of your argument is that it would be an inconvenience to move them. First of all, not all the pages would have trailing parentheses. Only articles that need to be disambiguated would have them. It's an exaggeration to say that 14,000 entries would all be added, all disambiguated with trailing parentheses, and after that is done, the community decides to move them. The parentheses would probably be done right the first time, and any mistakes would be caught early on. Anyways, bots can do mass moves.
If you say we should use a hierarchy for articles to avoid disambiguating them, couldn't the same argument be made for poems? "There are many poems named Spring. It is an inconvenience to disambiguate them. Therefore, we should put them in a hierarchy with other poems from the collection they were published." But we've decided that poems are standalone works and not related to other poems in a possible hierarchy. The same principle applies for periodical/encyclopedia articles. They are "mutually unrelated" material, and therefore are standalone works and should have their own pages.
I must disagree with the multiple articles on a page idea. Finding an article on that page is slightly more difficult than clicking a link for the user. Also, the articles, besides being on the same topic, are unrelated. But this is a minor disagreement, and perhaps if progress is made, we can do a few examples to compare. Psychless01:04, 4 September 2008 (UTC)
I have no idea who, if anyone, is suggesting multiple articles belong on a page. However, if you want to find all the works about Pope Adrian VI, then you look at his page, that is the purpose of the space. See Author:Sylvester II for example. Because people who want to find a variety of works on the same person should be looking at the person's index, or through categorization, not through introducing a third way of trying to sort things. SherurcijCollaboration of the Week:Author:Charles Spurgeon04:55, 4 September 2008 (UTC)
So what you are suggesting is that we should have author pages even for people who were never authors? Sylvester II's scholarship was exceptional for popes of that era. Eclecticology06:51, 4 September 2008 (UTC)
Speaking only of poems for now, many of them first appeared in magazines, then they appeared in editions of the selected or collected works of the poet, and if they were really popular they appeared in school anthologies. They are standalone works, but it would be uneconomical to publish a book containing only a single sonnet. That could put the dedicated hierarchist into a quandary. Eclecticology04:35, 4 September 2008 (UTC)
Whether encyclopedic articles could possibly stand alone is not relevant to the fact that they do not stand alone. The analogy to poems is a faulty one. Poems are frequently independent of collection; the same poem is seamlessly included in many editions, and are intended as a works in themselves. They were not intended to be published in a collection, they just were for the sake of convenience (if a poem is specifically intended to be part of a collection, it should be in the collection on Wikisource). Encyclopedic pages were never intended to be standalone works; they're articles in an encyclopedia, frequently with see-also references to other articles in the encyclopedia. As such, they should be subpages of the encyclopedia.
Furthermore, subpages have a wide range of benefits: list pages or search inside a work, simplicity of editing and maintenance (standard naming means no juggling too many links with conflicting naming styles based on suggest convenience and disambiguation or lack thereof), high search engine rank and user-friendliness due to the presence of the work title in the document title, simple navigation and interlinking within a work ([[../article]]), more uniform presentation across Wikisource, et cetera (these are just a few benefits off the top of my head).
Personally I think that the benefits and logic of subpages far outweigh those of ad-hoc naming. —{admin} Pathoschild 01:02:53, 05 September 2008 (UTC)
Well said. Another auxiliary benefit of subpages is that User:JVbot/patrol whitelist supports entries based on a prefix, as that is a structural element used throughout English Wikisource.
There are many incomplete projects that consist of multiple articles (e.g. Easton's Bible Dictionary (1897), A Short Biographical Dictionary of English Literature, Dictionary of Greek and Roman Biography and Mythology,...), but they all use this same structure, except pages created by Eclecticology, and this new DNB project. It is worth pointing out that Latin Wikisource has recently adopted a "/" based hierarchial naming convention, and has applied it throughout. A few of the Wikibook projects have recently adopted it as well, and audited all of their pages to follow the new standard.
English Wikisource did have consensus on this, and Eclecticology was given the opportunity to experiment, but has yet to convince me and others of any benefits in the naming system that Eclecticology has been pushing for. Both approaches have the same problems :- Article names about the same person often have different names in different encyclopedia; MediaWiki search results will be rubbish either way; disambiguation pages will be required either way. The difference is that subpages are the structural mechanism chosen to be used on Wikisource and other wikis, and is increasingly being supported by the software itself and tools that are built by contributors. John Vandenberg(chat)01:40, 5 September 2008 (UTC)
Clearly we are going to continue to have differing points of view about whether encyclopedia articles are standalone. For me it will remain an obvious fact that they do stand alone. Are we somehow questioning that Wikipedia articles are stand alone. These articles are most often written by individuals who work independently of each other, and their coalescence in published volumes only comes after the fact. "See also" references can be made in either proposed systems, but even more useful are links that would make a comparison between encyclopedias in the treatment of the subject of interest.
The matter of poems was raised because it was proposed that periodicals follow the same hierarchical structure, and a significant portion of independent poems have their first appearance in periodicals.
How is "list pages" any less of a benefit in a disambiguation based system? A system of categories will work just as well for listing all the articles in a given encyclopedic work. Putting the article title as the first element will allow the same "list pages" function to show more of the related articles together in the absence of those disambiguation pages that nobody has the time to create. Whether we are better off with the specified Google search is doubtful. Who is going to search for a specific article that way when he knows that there will be so many thousands of results?
The Orwellian list of other "benefits" could all just as easily apply to any system, and could be used just as easily to belittle any alternative proposals to them. I see no importance whatsoever to being concerned about search-engine rank; we're not running a popularity contest. If the material is good viewers will come. Why is the regimentation of uniform presentation so important across all of Wikisource? Progress does not happen by keeping us mired in that rut.
Comparing your subpage system with "ad-hoc naming" creates a false comparison. I can just as easily argue that the disambiguation system that I have been working on is superior to ad-hoc naming.
The JVbot whitelist is essentially a personal tool. I don't see how accommodating a personal tool can be viewed as a community benefit. Sure the listed works use the sub-page system ... so what? That does not mean that other projects can't use different approaches. We should welcome other approaches, not just mine but any other with a reasonably logical basis.
What Latin WS does is it's own business, and I support their autonomy. How many Latin language encyclopedias are their likely to be? Wikibooks is not likely to host previously published encyclopedic works or periodicals, and only these types of work are being disputed here. We have no overlap in these areas, so Wikibooks is irrelevant here.
Some kind of consensus may have existed at one time but that is never a concept that should be set in stone, and I don't buy into the argument that the accumulated inertia of the sub-page system should determine everything that we do. I don't accept that just because such a system works well even in a majority of its application it must work for everything. Sure the difficulties arising from persons and other concepts with variant names will be a problem for everyone, but let's not make that problem more serious that it really is. Advocates of any system will look for suitable solutions to that problem.
The experiment was about the multiple pages on one. I still think it's workable, but I'm currently holding it in abeyance. It requires adding new material from too many encyclopedias at the same time. I do intend to continue developing the DNB articles in the way that I have been, and perhaps do so for a few others. It's important to remember that only two large classes of articles are in dispute. Eclecticology13:57, 5 September 2008 (UTC)
More than whether encyclopedia articles are standalone (they have a strong relationship IMO), the issue is whether using subpages has more benefit than disavantages. I was not convinced up to recently, but there are two new important features which add some weight in the balance: the gadget preloading header script, and the possibility to moveall subpages at one. So I think that using subpages for encyclopedia articles would be beneficial. I am lobbying the French WS for moving to subpages. I think most peaople there agree with my opinion. The problem is in the implemention, but it's always easier if done from the start. Yann18:35, 5 September 2008 (UTC)
That depends on whether people want to use preloaded header scripts, and moving all subpages at once does nothing unless you have subpages. Eclecticology19:45, 5 September 2008 (UTC)
I apologize for flip-flopping on this issue, and perhaps my input isn't that important, but for the record, I think encyclopedia articles are too connected with each other to be standalone works. I still believe most periodical articles should have their own pages, but I support the hierarchy method for encyclopedia articles per the reasons outlined by Pathoschild. Psychless01:46, 6 September 2008 (UTC)
I have some time this weekend so I would like to take a series of 25 articles and show you all a proposal of what I imagine as a "best practice" and then have a discussion with concrete examples. Is there any particular texts in this discussion which have page scans? Cause if I am going to describe a "best practice" I should start with scans.--BirgitteSB16:40, 5 September 2008 (UTC)
It has just occurred to me that having the actual text separated and placed into another namespace and transcluded into texts kills the default search. A simple and incomplete fix would be to have search include Page: namespace by default, but that still isn't really satisfactory, as the search would bring up the pages rather than the text. Any ideas? Prosody (talk) 02:31, 7 September 2008 (UTC)
Why the footer? Could there be a link at the top of each Page namespace page back to the chapter that the page is in? Wherever it is, I think such a link is a great idea. — Sam Wilson ( Talk • Contribs ) … 09:37, 23 September 2008 (UTC)
leaked CIA document
I have uploaded the table of contents of a leaked CIA document released from Wikileaks.
I uploaded a leaked US report on the w:First Battle of Fallujah to Wikisource some months ago; and ended up deleting it out of concerns of US law (I can just picture being held up at the border "Sir, is this your Wikipedia account?"). Here however, I think there's a beautiful loophole. Ignoring the first 12 pages which are copyright of the US gov't and thus CAN be controlled; the speeches themselves are not -- only the translations can be controlled; thus publishing the translations are a crime at worst. However, if we use the translations to springboard our own English translations and publish those, there is no legal precedent suggesting that derivative works of classified works are controlled; and there's certainly no copyright concern. It would be more akin to publishing say, a summary of the works like WN already did. But you are broaching dangerous Pentagon Papers territory, and it can be a minefield. SherurcijCollaboration of the Week:Author:Albert Schweitzer13:01, 15 September 2008 (UTC)
Sounds like the Department of Homeland Paranoia has people spooked. What a change from the Vietnam War days when I snuck "Make love, not war" bumper stickers across the border into Detroit. :-) Eclecticology (talk) 18:32, 15 September 2008 (UTC)
"translations to springboard our own English translations"
While it may not be an issue in the present matter, this could more generally be problematic. Such a derivative translation would be a derivative work, and could be subject to copyright treatment as a derivative work. If a work is public domain in its original language, but the existing translations are copyright it is better to create a whole new translation to avoid the risk of being influenced by a copyright one. Signs of being influenced can be quite subtle.
Machine translations (bloody awful as they may be) are the product of a mechanical process and thus lack the originality required for being copyrightable. If we use those, wiki processes can then be applied to produce a sensible result. Eclecticology (talk) 18:32, 15 September 2008 (UTC)
I put this into a sub-heading because it was really about a derivative (?!) issue. The idea of a "classified" translation is bizarre. Documents are classified because of the information they contain. If the document in the source language is in the public domain only a very inaccurate translation would have such new information as to require being classified. Eclecticology (talk) 00:04, 16 September 2008 (UTC)
Aren't all documents prepared by the CIA in the public domain?
I'd welcome correction on this -- but aren't all documents prepared by the employees of US federal agencies in the public domain, the moment they are created? They may also end up being classified as a secret, or be on topics that are "born secret". But it my understanding that, once leaked, they should be considered in the public domain, just like any other work of a US Federal employee, in performance of their duty.
About the border guard scenario, I have considered whether to say something like:
It is true, I did help make some information that was already on the public record more widely available. But I think you are mistaken if you think I made the public less safe by doing so.
I have studied the public record on the work of the USA's counter-terrorism analysts. We have chosen how to allocate our precious and limited counter-terrorism resources based on those analyses. And the public record shows their analyses were fundamentally flawed. The record shows US counter-terrorism efforts included essentailly zero sanity-checking. Suppressing information that is in the public domain, that reveals weaknesses in the analysis may embarrass the analysts, and their bosses. But it absolutely did not make the public less safe -- the exact opposite. Because we need to understand what the real threats are, and, if necessary, change how we allocate our counter-terrorism resources. We should stop allocating resources to threats that weren't real threats, and reallocate those resources to the real threats.
Or maybe it would be best to keep one's mouth shut, if what one has to say would be complicated.
According to Google Books it seems to be present in several texts on the author, many of which are by thoroughly legit Lovecraft biographer and critic S. T. Joshi. I'd advise to check one of the books themselves, just to be certain. Prosody (talk) 01:00, 17 September 2008 (UTC)
It's always a concern when the only thing submitted by an IP is something that stirs controversy. I have no reason to dispute that it was written by Lovecraft, but where did our contributor get it from. The sources seem to be secondary ones that quote Lovecraft. Was it part of a larger work? Was it first published in Lovecraft's lifetime, or was it posthumously attributed to him, and perhaps still protected by copyright? This is probably the kind of thing where we should be more strict about sources. Eclecticology (talk) 03:29, 17 September 2008 (UTC)
The web reference [4] list "From New Nation, No. 4 (Autumn 1983) pp. 20-21." as the source, as Author:Howard Phillips Lovecraft lived to 1937 and there is no other evidence we can not assume that it is an original work published by him prior to 1923, so it is either an original work by an unknown author in 1983 or it potentially HP Lovecraft without clear PD until 2037, any either case pending more evidence it would need to be deleted as a copyvio. Jeepday (talk) 00:53, 18 September 2008 (UTC)
This isn't WS:PCR :\ Anyways, I don't think there's any call to delete it just yet; there does seem to be tradition tying it to him - and I see a reference he wrote it in 1912 - though no sense of publication is given. But deleting because somebody can't prove a negative (that it's not copyrighted) is a backwards way of thinking; we should delete because we can prove the positive (that it is copyrighted). SherurcijCollaboration of the Week:Author:Albert Schweitzer00:57, 18 September 2008 (UTC)
What's PCR? The presumption is that a piece of work is copyright in the absence of evidence to the contrary. In most of what we accept that presumption is easily overthrown. Here we don't even know if it was in fact published by Lovecraft at all; if so the problem would be a much easier one. In cases of serious uncertainty the burden of proof falls back on the contributor to tell us where he got this. Admittedly, it tries my patience when an anonymous individual has used this as an opportunity to pseudo-legitimately say "nigger". Eclecticology (talk) 06:55, 18 September 2008 (UTC)
Still dubious - hplovecraft.com gives only a title and no text; the book cited was not published until 2001. Why the disconnect between the date of authorship and the earliest reported publication? BD2412T14:42, 18 September 2008 (UTC)
Probably because Lovecraft didn't bother or didn't succeed in getting it published, and it lay in a dusty set of papers until someone printing a collection of bills signed by Lovecraft noted it? I would certainly assume it's not PD.--Prosfilaes (talk) 01:14, 30 September 2008 (UTC)
Roman or Arabic chapter numbers?
I'm copying and pasting Barchester Towers from The Literature Network, and then using my printed copy (which is based on the first edition) for comparison. If the original edition of 1857 used I, II, III, IV, V, etc. for chapter numbers, should I stick to that, or is it okay to use 1, 2, 3, 4, 5, etc? Thanks. Stratford490 (talk) 09:02, 18 September 2008 (UTC)
Opinions on this are likely to be mixed. The choice in the original book was more than likely that of the publisher than of the author. While I personally would tend away from using Roman numerals in what I contribute, when it comes to what other contributors do it's their choice. At this moment the TOC page has Arabic numerals, but if you change that and then do the work of loading up the whole novel, few people are going to have the energy to reverse your effort. That would require too much work. Eclecticology (talk) 16:02, 18 September 2008 (UTC)
I was the one who created the TOC page, yesterday, and would have no problem changing the Arabic numerals to Roman if that's the correct thing to do. The question is whether The Warden should be changed as well. (That's the only other Trollope novel currently hosted by Wikisource, and is the prequel to Barchester Towers.) My printed copy of The Warden (which claims to follow the first edition) has Roman; the Wikisource version has Arabic. If I'm not mistaken, changing it would mean not only modifying the TOC page, but also moving the page of every chapter to a new title. I like things to be consistent, and would do it if I thought this had general approval. Stratford490 (talk) 16:30, 18 September 2008 (UTC)
It's curious that in the TOC page for The Warden Chapter 8 has the numeral in both forms. That tells me that you aren't the first to struggle with this problem. Your analysis of what would need to be done to change The Warden is correct, and as I said above, few people would have the energy to change it back. Waiting for general approval on this kind of thing is unlikely to produce a straight answer. If you have undertaken a project in relation to Trollope and his novels (about which I know virtually nothing) the overview that you give to that project is bound to dominate the project. This is not so much because it is right or wrong, but because you are the one who has chosen to do the work. Future complaints about your decisions will be as valid as the work that those future complainers are willing to do. Whatever you choose to do (or not do) about this with The Warden is correct. Eclecticology (talk) 17:23, 18 September 2008 (UTC)
Thank you. I'll see how I get on with Barchester Towers first. If/when I finish all remaining Barsetshire novels, I may feel I've earned the right to tamper with the choice of the editor who originally upoaded The Warden. :-) Stratford490 (talk) 17:40, 18 September 2008 (UTC)
I prefer chapter numbers in arabic numerals; they're just easier to follow and don't require that extra (albeit slight) translation effort. That said, I've never changed them to be different from the reference scans; but I certainly wouldn't mind if anyone else did. — Sam Wilson ( Talk • Contribs ) … 07:54, 20 September 2008 (UTC)
Old English versions
I had translated, from French to English, two documents by Lower Canada Patriot Robert Nelson. I have since found the English version of the same two documents as they were originally published back then (in 1838):
Obviously, these documents are more interesting than my own original translations and I would like to add them to Wikisource.
How should I proceed to add the these two documents in? Should I move the current (newer) translations to a different name? If yes, are there guidelines for doing this? -- Mathieugp (talk) 18:41, 18 September 2008 (UTC)
It's an interesting question to which very little serious attention has been given. Fundamentally, it's important that we have both translations. The 1838 translator may have a very different perspective on the document from yours as a 2008 translator.
A canonical title such as Address to the people of Canada is just fine, for all translations of the same document. (Though I leave aside for the present the generic risk that that title may be confused with someone else's address from some other time.) Having settled on a canonical title, it should suffice to distinguish by a minimalist disambiguator such as "(Smith translation)" based on the name of the translator. "(Wiki translation)" could be used for a translation that is original to this project.
This is not to criticize your claim on the translation; that approach has been recommended by some. But ultimately an on-wiki translation that cannot be improved by others strikes me as very unwiki.
If we are to seriously recommend new translations (sometimes necessary when all extant translations are tied up in copyrights) I believe is a two column solution with the original language in one column and corresponding paragraphs in the target language face-à-face in the other column. Eclecticology (talk) 21:03, 18 September 2008 (UTC)
I think it's more important to consider whether a policy will attract editors and produce useful content than to dismiss it as unwiki. Note that not all new translations are going to be on-wiki; some may merely be uploaded after translation, just like old translations.
New translations are necessary for the 99% of works that were never translated in the first place. A two column solution is fine for a very literal translation, especially one designed for students of that language, but makes it very hard to print or reuse or even read the English translation, and isn't really appropriate for smooth reading translations.--Prosfilaes (talk) 14:53, 20 September 2008 (UTC)
What makes a translation unwiki is that it discourages further editing. A signed translation implies taking ownership of that translation, and for Wikisourcerors who would certainly respect previously published translations even to the extent that they may be erroneous this is problematic. Editing a signed Wiki translation means that it is no longer that named person's translation. Wiki translations necessarily bring POV issues into play. Whether the translator's work habits involve doing things on-wiki or uploading them later doesn't matter; the translation still must link back to the original language of the work.
Your 99% figure makes sense, but a two-column solution is not just for language students. It is as much for anyone who may be concerned with the multiple interpretations that can apply to a sensitive passage. Smooth reading should not be affected if you keep focused on the one column. Perhaps the techies can develop a technique for extracting the one column for those who don't want to know where it comes from. Eclecticology (talk) 00:00, 21 September 2008 (UTC)
I agree with you about labeling translations, but disagree about displaying them in two columns. As long a comparative version is one click away, there is no need to force the main text to be displayed in that way. Put up the translation by itself and have a either an interwiki to the original language or use Proofread Page extension with the links to individual pages.--BirgitteSB11:59, 21 September 2008 (UTC)
It works to a point, but not on a one-click basis in all skins. One has to know to add "?match=fr" (for French) to the address of the page. It also doesn't help unless the original version is already on wiki. It also does not help for keeping corresponding paragraphs aligned. While I'm not saying that it should apply in all cases, there are definitely some where the two column solution is clearly beneficial.
That is the current style decision of the Wikisource community; all texts are Titleby Author, and that pretty much been the way it's been for ages. If you're wanting to suggest the community change the way the template presents the title, you'd best do so under proposals, above. Jude (talk) 07:28, 19 September 2008 (UTC)
That's called rule by inertia, or the fastest way to be stuck in a rut with blinders on. If he can plough through the occult mysteries of the templates, he should develop one that best suits his purposes, and that he will use himself. This does not imply that the whole rest of the community should make the same change. Eclecticology (talk) 20:59, 19 September 2008 (UTC)
One option is to assign both a class name (the default style of which being to match our existing standard), which each user can then override in their own Monobook.css file. EVula// talk //21:19, 19 September 2008 (UTC)
@Stratford: What style would the "by Author" part be in? If it were bold, it would be distracting, and if both were italics, it would blend with the title to much. Psychless22:58, 19 September 2008 (UTC)
Every aspect of the template is tagged and can easily be changed per-user. Just to show what is possible, see a screenshot of this extreme header makeover. Cascading stylesheet support was added in the last header design overhaul; I have no objection to changing the style again if there's consensus. —{admin} Pathoschild 06:29:19, 20 September 2008 (UTC)
Each user having their own version of the template defeats the entire purposes of having a standard system of headers, which is the only reason I didn't suggest it myself. Jude (talk) 03:30, 20 September 2008 (UTC)
Indeed, it would make the cool stuff we can do with metadata (example) impossible. —{admin} Pathoschild 06:29:19, 20 September 2008 (UTC)
You're entitled to your opinion that doing this tech-cruft is "cool." Others have other priorities, including deviation from "standards" to accommodate changing circumstances. No-one should need a computer science degree to do this. Eclecticology (talk) 23:30, 20 September 2008 (UTC)
Yes, that's the whole point. We should not expect every editor to do the mundane work of indexing each work or author they add. This is easily programmable. When adding a new text, editors should only need to edit the text itself. A bot can extract the metadata (the {{header}} or {{author}} values) and index the page accordingly. How often do you remember to add a new work to indexes by genre, by country, by era, by author, et cetera?
There is no need to put "standards" in quotation marks (it's a real word). I fail to see any relevance between applying our standards and accomodating changing circumstances. Can you give an example? —{admin} Pathoschild 01:26:29, 21 September 2008 (UTC)
It's not that I forget to add works to the mentioned Wikisource: pages; I just don't bother. Those pages are essentially unmanageable, so why bother? I do make more use of categories for this. So just yesterday (and this perhaps addresses both of your paragraphs) I undertook to divide up the author category pages where the number of entries in the category exceeded 200. Just changing the last initial parameter from "M" to "Ma" does this when the relevant sub-categories have been set up. The proper new category does show up in the list of categories at the top of the page, but the author index indicator in the header goes red. Indeed, when I worked through all the "Ma" authors I found about 30 who were in the categories but not in the indexes. These index pages require manual maintenance, and only those who consider that kind of maintenance important are going to do it. What has your programming done to add them? There are incompatibilities between a top-down list system, and a bottom-up category system. Both are important, but they serve different purposes. Eclecticology (talk) 21:32, 21 September 2008 (UTC)
Indexes are manageable by bots. If you add a new author whose family name is "May", it doesn't require human intelligence to add them with their dates to Wikisource:Authors-M among other indexes. The whole purpose of metadata is to allow editors to describe the work in only one place (the work itself), and software can take care of everything else. This is similar to the way indexes in published books are created; people no longer manually keep track of page numbers for each keyword, they simply highlight or list them for the software which automatically maintains the index.
Do you have any concrete arguments against describing authors and works in one place, or only vague complaints about incompatibilities and accommodating changing circumstances? —{admin} Pathoschild 21:53:25, 21 September 2008 (UTC)
You fail to note the inconsistencies in your arguments. On the one hand you argue that the data can be put in only one place and Big Brother "software can take care of everything else." At the same time expect new authors to be added manually to a second index place. Which is it? The 30-author discrepancy does not imply that all 30 of those authors were added by the same contributor. Eclecticology (talk) 22:12, 21 September 2008 (UTC)
Who said anything about manually editing indexes? You add works to the main namespace, authors to the author namespace, and a bot takes care of the rest. Who adds the work or author is irrelevant. —{admin} Pathoschild 01:44:33, 22 September 2008 (UTC)
Ah!! I see now. We don't manually edit the indexes. The bots have not been editing the indexes. So we at least agree that the indexes are irrelevant. Eclecticology (talk) 02:45, 22 September 2008 (UTC)
We agree we shouldn't need to manually update index pages, but they are useful (or could be, if they were comprehensive). I'm writing a bot to maintain them automatically, but that requires a standard header template. —{admin} Pathoschild 13:56:03, 22 September 2008 (UTC)
I've been away for a few days. Thanks for all the comments, though they got a bit too complicated for me to understand after the first few lines. :) Stratford490 (talk) 09:11, 26 September 2008 (UTC)
Archetypal author page?
I am either blind, useless or it doesn't exist. [All quite possible today] I am looking for the guidance on what an author page should contain. Failing that could someone please point me towards what would be considered an archetypal or extensive Author page so that I can more replicate what currently exists, rather than fluffing around like I have been. :-) Thx. -- billinghurst (talk) 07:17, 19 September 2008 (UTC)
It was the body of the page that was interest, not so much the head of the page, I had managed that. Apologies if I wasn't specific enough. What is in? What is out? Preferred headings? Linkings, etc. If it is simply "go for your life", that's fine. -- billinghurst (talk) 10:48, 19 September 2008 (UTC)
Divisions of works (poems, novels, etc) should be subheaders of 'Works', and the dates in parentheses are optional. 'External links' are generally not used, although 'Sources' sections are common when some works are not on Wikisource yet. The license is one of those listed on Help:Copyright tags. —{admin} Pathoschild 15:57:35, 19 September 2008 (UTC)
Quick follow-up question with regard to {{license}}. Above it discusses using those at Help:Copyright tags, and when cruising, I found at Category:License_templates that it suggests using Author-PD-old.... Has there been a change in philosophy? Alternatively, is there something that needs to be added as guidance to Wikisource:Style guide#Author_pages?
In response to Sherurcij and Birgitte above, I looked into how {{PD-GovEdict}} would impact crown copyright as well as other governments works.
per W:Wikipedia:Public_domainW:Wikipedia, and the W:Wikimedia Foundation, its legal body, are based in W:Florida, W:United States. Although legislation is sometimes unclear about which laws are to apply on the Internet, the primary law relevant for Wikipedia is that of the United States. For re-users of Wikipedia content, it is the laws of their respective countries.
per W:Berne Convention for the Protection of Literary and Artistic Workscopyright of works of authors from other signatory countries (known as members of the Berne Union) in the same way it recognises the copyright of its own nationals. For example, French copyright law applies to anything published or performed in France, regardless of where it was originally created.
Conclusion - as long as the WS sever is in the US, {{PD-GovEdict}} trumps local law for publishing on WS but each countries own law is different. I have not addressed W:Rule of the shorter term as it applies to non government works here. I would suggest that if we bring "undeletion of all the Commonwealth laws" to WS:DELETE#Undeletion_requests it might be helpful to have a template that specifically states that w:Crown copyright applies in the country of origin. Jeepday (talk) 13:38, 21 September 2008 (UTC)
I would suggest a slight change to {{PD-GovEdict}}. Since it should only be used when we either don't know the local copyright or need to override it, I think the default should say "While this text is not copyrightable in the US; they are assumed to be copyrighted outside the US". And "assumed to be copyrighted" should customizable to say "covered by Crown Copyright" or any other more exact message.--BirgitteSB13:55, 21 September 2008 (UTC)
"Crown copyright" is not monolithic. Its meaning varies from one Commonwealth country to another. You can't use that term if you don't know what the law is in that country. Eclecticology (talk) 21:53, 21 September 2008 (UTC)
"Crown copyright is relative to only a half dozen or so of the 160 some Berne Convention parties. How about customizable to say "covered by local <insert country name> Copyright", no entry gives "covered by local Copyright" while entry gives "covered by local Senegal Copyright". Jeepday (talk) 00:00, 23 September 2008 (UTC)
I consider {{PD-GovEdict}} a general tag for the USA, but more specific tags should be considered. We have {{Legislation-CAGov}}, {{Legislation-SGGov}}, and {{UK-Crown-waiver}} now. In addition to mentioning that the USA Copyright Office considers then non-copyrightable, I would like to suggest a list to show which countries and areas honor or dishonor these official work copyright, like a list at commons:Commons:Reuse_of_PD-Art_photographs#Country-specific_rules. For example, as a Chinese-speaking admin, I have heard that constitutions and laws are not copyrightable in Taiwan Area ({{PD-ROC-exempt}}) even if a foreign government copyrights it. In this case, works covered by {{Legislation-SGGov}} cannot be copyrighted in Taiwan Area despite Singaporean governmental copyright.--Jusjih (talk) 01:21, 25 September 2008 (UTC)
Disambiguating authors
I have looked through the Style Guide, WS: and templates and, from what I can find, they only discuss works. Is there guidance for disambiguating Authors? Thx. --billinghurst (talk) 16:34, 23 September 2008 (UTC)
While it would be nice to have Jr./Sr. or middle names available as easy disambuguators, failing this years may be the next most reliable, but it will not always be available. As the project grows we fill find new ambiguous situations where we have no basis for differentiating them except the articles themselves. We need to remember that we are documenting more than just whole books. Eclecticology (talk) 00:11, 25 September 2008 (UTC)
Speaking only for myself, I can think of a few survival manuals that would be cool. I can also imagine that some gear heads would like to see the manual for the 1941 Bantam. Jeepday (talk) 23:52, 23 September 2008 (UTC)
I think our policy is pretty much "If it's public domain, we want it!". Though substitute "public domain" with "free and previously published" and that's probably a bit closer to the truth. Jude (talk) 23:57, 23 September 2008 (UTC)
Perhaps I wasn't specifically clear; our copyright policy demands that it be public domain (which are most works by Federal or Federal government employees), but our scope of works reaches only to that previously published. Of course, published can be interpreted very loosely; your public domain short story (or equally free licensed) that has only ever been published on your website, we do not want. Lost letters of somebody (or even letters of interest by a historical figure), despite being unpublished, would also be desired. Jude (talk) 09:33, 24 September 2008 (UTC)
I don't know very much about this project, so here's another question: the original manuals generally have lots of illustrations. Is there a way that I could upload scans of the images to commons and then incorporate them in the text, or would you guys recommend a text-only version? Cerebellum (talk) 12:17, 24 September 2008 (UTC)
No, please upload images. We appreciate them, and they add immense detail to a text. If possible, (perhaps in collaboration with John) you could also upload actual scans of the text itself (thus allowing for proofreading and DJVUification etc). Jude (talk) 08:48, 25 September 2008 (UTC)
While there is no formal or required review process, you might be interested in rewarding your effort by adding the text to the top of the list at {{New texts}}; which will cause it to be featured on the front page for a week or three. More readers will invariably also mean more people scanning it for errors. SherurcijCollaboration of the Week:Author:Isaac Brock17:08, 25 September 2008 (UTC)
Do you mean ⇔? It appears when I view the page. The link you provided appears as though the vertical align in the Spanish side is set to center, rather than to top... Jude (talk) 03:05, 26 September 2008 (UTC)
It appears when I click the link, too. And, yeah, the DoubleWiki has some really bad matching issues, in part because different wikis format the pages so differently from each other that we confuse the software when it tries to match the actual text of one translation with the actual text of another.—Zhaladshar(Talk)03:29, 26 September 2008 (UTC)
If it's bad with a relatively short poem it gets much worse with a long text. In French prose in particular the French text is usually longer than the corresponding English. I suppose it could be helped if the two languages could be scrolled separately. Eclecticology (talk) 17:22, 26 September 2008 (UTC)
Got it. The problem was that the headers section was the first line of the edit box. There should be a blank link between the top of the edit box and the ==Vegetables== line. Why this is, I don't know.—Zhaladshar(Talk)14:55, 27 September 2008 (UTC)
Thanks. I think it might have something to do with the page template because if you just use the transclusion code, it works without the blank line. Psychless16:38, 27 September 2008 (UTC)
The guidelines are probably based on the fact that most works have numbered sections. But some works, like cookbooks and encyclopedias, don't number their sections. For example, we don't call 1911 Encyclopædia Britannica/Aaron's Rod "Article 15". Therefore, I would propose changing:
Subpage titles should be separated from the parent title by a forward slash ([[Title/Chapter 1]]). Sections should be numbered, not named ([[Title/Chapter 1]] and not [[Title/The Dog Returns]]). The section name should reflect those in the original work (Chapter II, Chapter 2, Act 2, et cetera).
to "Subpage titles should be separated from the parent title by a forward slash ([[Title/Chapter 1]]). Sections should be numbered, named ([[Title/Chapter 1]] and not [[Title/The Dog Returns]]), unless the sections were not numbered in the original work. The section name should reflect those in the original work (Chapter II, Chapter 2, Act 2, et cetera)."
There are works which are intended to be read sequentially that do not use numbered sections, and there are works that contain distinct entries yet number them sequentially. Do we want our guideline to defer to the printers choice, or do we want to impose a more rigid guideline that imposes structure in deference to the author/printers choice?
An alterate proposal is "Works that are intended to be read from start to end should use page names that include the sequence, whereas works that contain many distinct entries which are still valuable in isolation of the rest of the work should be named according to the entry name." i.e. if order of subpages is important, the order should be obvious from the page name.
The sections in The Pilgrim Cook Book are all distinct, and there is no numbering in the index, so the order used by the printer is primarily of use in the study of "cookbook publishing", and I cant see much value in the ordering beyond that. That begs the question then, ... is Boston Cooking-School Cook Book intended to be read from start to end? [6] My answer to that vague question is that the order in it is also mostly useless. The first few chapters have a bit of prose to them, and there is likely to be a little bit of interdependence between those chapters, but I cant see any need to retain order for the benefit of the reader. That leaves us with the question of whether the imposition of using numbered chapters as the pagenames is still desirable for the benefit of the editors, bots and admins.
365 foreign dishes is an interesting case - the order is vital, as the selection for each month is a "recommendation". However in this case, using the month name for the subpage name does provide the obvious ordering needed, albeit an obtuse ordering that wont be used often, and an ordering that bots wont instantly understand.
We deal with such a wide variety of texts. I think we need to write into policy fixed rules for the simple cases which cover 99% percent of works, and then have guidelines for the rest. John Vandenberg(chat)03:10, 30 September 2008 (UTC)
The most important rule is to trust the judgement of the contributor. I don't give a damn if bots can recognize it, because I don't add things for the benefit of bots. I agree that the 365 suggests a certain type of order, and that's fine, but what's obtuse about it?
Guidelines say "should" and that's perfectly OK. "Should" is not equivalent to "must." Whether someone prefers to use Arabic or Roman numerals is his choice; whether he adds or omits the numbering is his choice; whether he chooses to add the chapter title to the number is his choice. It's important to trust good independent judgement. If that judgement is too eccentric the option to edit mercilessly is still available. Eclecticology (talk) 04:46, 30 September 2008 (UTC)
DefaultSort
Would it be possible to get the relevant DEFAULTSORT help & template pages moved over from source. (By the way, a request, could we please have a substantive and permanent link in the ToC to the Scriptorium archives, it took me forever to find the little link at the right top of the page when I went looking to see where DEFSORT may have been discussed previously. Thx)
We don't have good guidance on sort, and the pedant in me would like to see some of that advanced. We seem to have quite a variety of uses to sort in Author pages, some successful, some interesting, especially with relating to titles. Here is one that I prepared earlier Author:Herbert Edward Ryle that has sorted well at Category:Authors-R. --billinghurst (talk) 04:09, 30 September 2008 (UTC)
Doctrines to be rejected
Thanks for your comments. I'm pretty new to this, so I'm not sure how to do a lot of things. I did actually set up an account, but I was assuming a confirmation would be sent to my email address and I don't think I got one and I'm not sure what name I created, so I will create another.
Re the doctrines to be rejected, properly it should be merged with the Birmingham Amedended Statement of Faith, because it is normally considered part of it. The reason I altered the doctrines to be rejected is because the version that was there is a doctored one and not the original one. I can see why it has been altered, because the author of it is trying to clarify what he thinks it should say rather than what it actually does say.
He also repeats the phrase 'we reject' when it is normally found only at the beginning not under each one and splits the last one into two with slight additions. The words heathen, pagans and young children is altered to convey what the author believes is the underlying reason and to make it seem less lacking in grace, but in so doing it is not accurate to the real version. It would be like me or you editing a book to make it express what we think it should say and leaving the impression that that is what it always did say.
It's important that the correct version is here because its related to how the Christadelphian body is structured and officially belief and adherence to the statement of faith is a condition of membership and you can be disfellowshipped for not believing the points in it. —unsigned comment by81.96.196.186 (talk) .
Categorization authors to works
I am looking for guidance on Categorization (cannot find sufficiently discussed elsewhere) to the existing Category framework in place. Where works (eg. encyclopaedia & dictionary) have multiple contributors (> 100) is there the possibility of having a category that can be applied on each Author page that would link back to the generic Work itself. My reason for asking is that I am creating and pre-populating the author pages, though the specific articles cannot be added until they have been transcribed, hence may sit empty for a period. -- Billinghurst (talk) 01:46, 18 September 2008 (UTC)
The more I think about it the more I believe we could be making better use of {{TextQuality}}. At present it is putting things into four categories, but a 75% category with more than 3300 items tends not to be used. In fact any category with more than 200 items has diminished usefulness because it requires users to look at a second page. A more project oriented use of the text quality template could be warranted. Thus instead of {{TextQuality|50%}} we could have {{TextQuality|DNB50%}}. The effect would be to put all relevant items into new categories that could be perused by those interested in the project looking for something to do. In addition, when used in a project, the definitions of the percentages could vary to suit that project. Eclecticology (talk) 16:24, 18 September 2008 (UTC)
I would recommend a small change to the intermediate category: "Authors in..." rather than "Authors by..." "By" leave me with the impression that they are editing the whole encyclopedia. Eclecticology (talk) 07:22, 21 September 2008 (UTC)
I had thought that, then thought that it could be interpreted as these "these authors are in this encyclopedia ...", I had thought "... contributing to ..." which gets long, then I went with the default in the category. I will go with whatever is the universal opinion. -- billinghurst (talk) 07:47, 21 September 2008 (UTC)
