Wikisource:Copyright discussions/Archives/2008-06

From Wikisource
Jump to navigation Jump to search



JPS 1917[edit]

The following discussion is closed:
No consensus

I seem to have detected a somewhat unusual copyright problem: the apparent electronic source of the WS edition of the Jewish Publication Society 1917 Bible comes from This electronic edition, though derived from a Public Domain original, claims a new (2002) copyright. In fact, Sacred Text Archive removed this text from its archives because of the "new" copyright and text purity issues. See: for more details. I'd really like to see the existing MM version of the text removed from Wikisource for the same reasons: "new" copyright, and the text corruption by the transcriber. That is, unless by means of electronic comparison (ie Unix diff, etc.) there can be shown to be no differences between the MM STA and a print edition save those introduced by genuine transcription error (which need to be corrected). If anyone has a print copy of the JPS 1917 for comparison that would be good too. —Wikijeff 21:27, 24 June 2007 (UTC)

I don't believe that such a claim of copyright by transcription has any merit. Obviously we need to eventually proofread this version against a print edition, but there is no copyright violation.--BirgitteSB 00:43, 25 June 2007 (UTC)
I think Wikijeff's issue is that the man who did the transcription made alterations (sounding fairly substantial) to the text as he transcribed it. Of course, no one knows who "creative" his alterations were or if they would qualify for a new copyright. I propose, to keep text purity first of all and to avoid keeping this man from demanding licensing fees from WMF secondly, to phase this edition out (going all the way as to delete the current text) and replace it with the accurate, definitely free edition Sacred Texts is doing here.—Zhaladshar (Talk) 01:10, 25 June 2007 (UTC)
From what I read the differences were meant to be quite insubstantial; rather only enough to prove it was "his" transcriptions. It is not actually any additional material, but intentional errors that are small enough not to be casually noticed. The claim of copyright is not for his alterations, which we don't want anyways, but for the entire transcription. These intentional alterations the transcriber made should be cleaned out during proofreading. Of course we should replace the available chapters with the more accurate transcription at scared texts. I just don't think we should delete the rest.
The difference between transcribing from scratch and proofreading an exiting digital text of this size is huge. There is no evidence the introduced errors are more substantial than OCR errors. Do we really want to set the precedent that we accept such ridiculous claims as "copyright by transcription"? And just because we found that a report that such a claim was made to a completely unaffiliated website? Even if we had evidence of there being differences from the 1917 edition we should only delete the parts that are different. This is jumping at shadows and I think the bar should be higher than this.--BirgitteSB 01:50, 25 June 2007 (UTC)
I never said we should agree with "copyright by transcription" as that is clearly not protected by copyright. The FAQ entry at Sacred Texts which addresses this issue definitely indicates that the errors are not just standard OCR errors and are not just errors by transcription
"Not the least of these is that the transcriber deliberately introduced an unacceptable number of errors into the text. These are not accidental errors, which would be acceptable, as they could be fixed incrementally. He edited the text with the intention of creating a version which would vary from the 1917 public domain edition, so that he could claim a copyright on the result."
It sounds material was added by the transcriber that differs from the content of the original publication. This adds not only a level of copyright uncertainty (because his own additions can be copyrighted), although that is only minor to this issue for me, but also a level of impurity which means we have an inaccurate text as of right now with no knowledge of what needs to be changed to bring this in accordance with the original edition. Unless anyone knows how to weed out these errors, I believe we need to phase out the entire document.—Zhaladshar (Talk) 02:13, 25 June 2007 (UTC)
I don't think it sounds like any material was actually added at all. It sound to me like he made alterations, changing a word here or there, just so he could trace the legacy of the etext. I know these are not OCR errors, I meant that I don't think the text with intentional errors is further from the accuracy of the published text than the text with OCR errors. I have no problem with this text being replaced by new OCR, or an etext with a different legacy, etc. I just cannot support it being deleted as a copyright violation.--BirgitteSB 12:59, 25 June 2007 (UTC)

Yes, my primary issue is the fact that the transcriber had the audacity to knowingly alter the Sacred text (Cf. Deuteronomy 4:2). Moreover, he used such alterations as a basis for a "new" copyright is just a case of adding insult to injury. I am for phasing out the existing text in favor of the Public Domain edition located at That is, unless scans are made available from which we can work directly. —Wikijeff 05:43, 25 June 2007 (UTC)

I've found a scanned copy of the 1917 JPS. It is located at I've tried numerous times to upload the Deja Vu file to WM:Commons, but the server keeps resetting. This should help a LOT. Though the text is in columns, so most OCR software may have an issue with it. —Wikijeff 12:01, 25 June 2007 (UTC)
You may have to make deja Vu file for each chapter and upload it that way. There is a file size limit at commons that you are probably exceeding.--BirgitteSB 12:59, 25 June 2007 (UTC)
Jeff, Birgitte's right: the DjVu file is far too large for the upload limit at Commons (which is around 20MB). This file will need to be split (I think doing it by chapter is most logical, but any way is fine) to fit the upload limit.—Zhaladshar (Talk) 14:36, 25 June 2007 (UTC)
Thanks for the heads up with the file cap. I'm looking into what it takes to split a Djvu file. I did some Google searching without much avail. I've found sites for crating Djvu files, just not splitting them. I've posted a question on Commons Help regarding this issue. I hope someone gets back with me soon. If anyone here knows how to split Djvu files under Linux or Windows, please let me know. —Wikijeff 17:32, 25 June 2007 (UTC)
No idea on how to split them. Luckily the files I was working with before were available for download either as the complete works or split up by work. The quickest way to find help would probably be to get on IRC and search freenode for a channel on djvu.--BirgitteSB 17:52, 25 June 2007 (UTC)
I've tried finding a program that will split DjVu files and have found absolutely nothing. The closest thing I've found is that you can individually save all the different pages as separate images, but how to rejoin them into a multi-page DjVu is beyond me. What we might have to do (which sucks), is save all the image pages separately and upload them individually to Commons.—Zhaladshar (Talk) 18:12, 26 June 2007 (UTC)

Hi, sorry I didn't notice this before, and thanks for the link, Jeff. In my opinion there is no reason not to post this text. The website itself says the changes are technical (having to do with paragraphing based on objective sources with no copyright claims).

Now it may be that they did change the actual 1917 text in some places (I suspect in very few, but the images may help in checking some samples of that). WikiJeff, I don't think there is any "audacity" in this or anything wrong at all, after all a translation is just a translation and nothing more, and they might have wanted to improve (in their eyes) the translation. If it was to create a dubious copyright claim then that indeed is audacity, but it still doesn't mean the trick is legally valid.

If, as I suspect, these changes are few and far between, then I suggest using the JPS text until it can be replaced by the new "Sacred Texts" version, book by book. We should just check some samples here and there to make sure that the overall text adheres fully to the original (MM tends to proofread very well). If we find a significant number of discrepancies we should stop, but otherwise the burden of proof should be on them to point out what original contributions they have made. Dovi 20:10, 26 June 2007 (UTC)

Ok, I have given this some thought. I believe Zhaladshar is correct, we can't have inaccurate texts here. But on the other hand, we only have a singular "witness" (Sacred Texts) alleging significant inaccuracies in the texts. The other "witness" (MM) alleging "In addition to converting his text to HTML, we did correct a few typographical errors, based mostly on comparing the various text versions we found, and took out all of the paragraph marks (¶), which were not based on the Hebrew original.". The way I see it: we need another "witness" going either direction on this matter to make a truly intelligent decision on the matter.
So, I propose the following, and I'd like others to way in on this:
(1) We post an text box on ToC page, and each page with text (at this point the whole Torah, and the Song of Songs) indicating that the fidelity of the electronic edition forming the basis of the WS edition is contested or otherwise in doubt.
(2) We request users proof read the existing text against the images of the 1917 text, and provide a link to the Djvu and PDF file that will (hopefully) be on Commons shortly.
(3) We request that users identify which portions of the text they have proofed against the images, and comment on the Talk pages about any irregularity in the text.
(4) We don't add any more Books of the Bible until the existing posts are proofed and normalized against the page images.
(5) We move forward systematically, one book at a time until the entire text is good.
This way we don't have to retype everything. We end up with a known-good copy, checked vigorously and systematically against the scans, the text becomes normalized against the print edition, and most importantly everyone who comes to the text knows it is suspect and in the process of being proofed. If Dovi is right, (and MM has been truthful) then this will be a reasonably simple (if tedious) task. But, if Zhaladshar is correct (and Sacred Text has acted prudently) it should become readily apparent and we can do as he suggests: and dump the whole thing in favor of re-entry. Either way we will know the truth of the matter for ourselves; as well as have our "second witness" for this point of fact. And, this text, sacred to so many, will will be a trustworthy copy given time. Though we should agree ahead of time as to where the line in the proverbial sand is that if crossed, means we need to dump the existing text. Comments please. —Wikijeff 04:12, 27 June 2007 (UTC)

@file upload limit: this may helps ;) Lugusto 00:55, 27 June 2007 (UTC)

Thanks Lugusto, I just emailed Eric. I requested he upload the Djvu and the PDF version of the JPS 1917 text to Commons. I'll let everyone know when it's up by posting here, and on the text's Talk page. —Wikijeff 03:04, 27 June 2007 (UTC)

I think Jeff's suggestion is reasonable and thank him for his efforts. Dovi 03:00, 28 June 2007 (UTC)

I think Jeff's suggestion is good, too. I do hope I'm wrong about this, and all we have to do is minor typo errors to restore it to the original edition. :) By the way, how is it Erik can upload documents past the limit while no one else can?—Zhaladshar (Talk) 14:10, 28 June 2007 (UTC)
Zhaladshar, according to the post given by Lugusto, he [Eric] is a (2006) member of the Wikimedia Board. Apparently he has direct FTP access and shell accounts into the relevant Wikimedia servers. I still haven't heard anything from him yet, though. —Wikijeff 16:02, 28 June 2007 (UTC)

As I'm sure this is not the last time a source document's fidelity will be at issue, I've created a new template {{fidelity}} for texts with fidelity issues. All such documents are in a Category called "Doubtful Fidelity". This should help in dealing with future issues (hence the name) like this one. I know this discussion will be Archived given time. How do I create a permanent link to it, on Talk:Bible_(Jewish_Publication_Society_1917)? —Wikijeff

There's no real easy way to do it, but I would imagine this discussion will be archived sometime in July, so the link Wikisource:Possible copyright violations/Archives/2007/07#JPS 1917 should eventually work.—Zhaladshar (Talk) 02:59, 29 June 2007 (UTC)

Has there been any progress on the above proposal? If not, I suggest we delete the work and replace it with an edition which does not carry a claim of copyright or contain deliberate errors. —{admin} Pathoschild 23:25:30, 17 October 2007 (UTC)

I would still oppose outright deletion but not overwriting with a new edition.--BirgitteSB 15:34, 24 October 2007 (UTC)
I also oppose the deletion of what we have. I have requested the djvu file to be uploaded onto commons so that we can vet the text and remove any alterations. John Vandenberg 06:38, 7 November 2007 (UTC)
The upload of this djvu file did not go ahead as planned, however proof readers can still download it onto their own PCs. John Vandenberg 03:23, 18 December 2007 (UTC)
I can maybe shed some light on this. Over the past four years, I've done a few etexts for PalmBible+, a Bible study program for PalmOS. I came across this version two years ago. I never did much with it due to technical issues that are not germane to this post. It does appear the JPS1917 text is copyrighted and changes have been made (how many I don't know, ISTA's claims not withstanding). In 2003 the SWORD Project (open source bible developers) dealed with a accusations from another programmer over the use of copyrighted texts; the JPS1917 was mentitioned:
He did a bunch of modules for OLB, like the JPS 1917, Roderham, LXXE (Brenton), Lamsa, etc. We took all of his works straight from OLB (except LXXE & Lamsa). We also got permission from him. (source)
On my handheld, I have another Bible Prorgram called BibleReader. The JPS1917 is featured. In the about section, there is a copyright notice. I have screenshots of it. If any of the folks involved want to see them, email me.
As for additions to the etext JPS1917, the Online Bible Software Site (TOLBSS), there is a essay dealing with the JPS1917 and how it came to be as an etext. Most of it seems to be a reprint of the JPS's preface from 1916. There is an addition by the transcriber. It reads in part:
In 1995, [DELETED], began to transcribe this edition into computer format (ASCII). Advantage was taken of this opportunity to correct a few typographical errors. One change was made in the text of the translation. In Numbers 29:28 the word "single" was replaced with the word "sin." The context supports this change. (source)
In conclusion, it's fair to say that the JPS1917 is copyrighted. As for additions to the etext, it seems like some were made, but as I said beforehand, I'm not sure how many were. As the case may be, I support the plan wikijeff as laid out. On a separate note, the sources have the name of the transcriber of this text. Since ISTA and other users in this thread have not named him, I will refrain from doing it as well. Hoshie 11:10, 13 January 2008 (UTC)
No, it's fair to say that JPS claims that the JPS1917 is copyrighted. That's not the same thing. One can easily see why JPS would want the text to be copyrighted, and one can sympathize with a wary corporation wanting to avoid using a text for which copyright is claimed (even if the claim is fallacious). But the original text cannot be copyrighted in the U.S., and minor alterations that were made in order to more strictly adhere to the original are not "creative content", and are not subject to copyright. There is nothing stopping a company from incorrectly claiming copyright on a public domain text--companies do this all the time--but there's absolutely no reason we should honor a claim that is obviously false. —Quadell (talk / swapmeet) 00:18, 3 February 2008 (UTC)

Keep Unless someone can document a reasonable quantity of corrected (or added) errors, the project as it stands is defensible as a work in progress. If someone from the JPS discovers it online, the most they can do is request to have their "corrections" removed, which is ostensibly what interested parties are in the progress of doing anyway. To be safe we can say, "if there's no progress on this project in another 6 months or so, we can revisit whether those parties are making a sincere effort to ensure copyrights are not being violated in the course of using it as a work in progress." ResScholar 08:24, 1 March 2008 (UTC)

  • After having been open for almost a year and having gotten no consensus in that time, I'm closing this.—Zhaladshar (Talk) 15:06, 10 June 2008 (UTC)


Gorbachev's resignation speech, 25 December 1991[edit]

This speech is mentioned as translated by CNN. If this is the case, CNN has certainly a copyright on the translation. Yann 07:59, 8 May 2008 (UTC)

Here is the NYT text. It doesnt mention CNN. I suspect it was actually translated by someone else, and these news organisations merely republished it, possibly as fair use. It might have been translated by a US dept. John Vandenberg (chat) 23:09, 11 May 2008 (UTC)
Delete. ‘Following is a transcript of Mikhail S. Gorbachev's resignation speech in Moscow yesterday, as recorded through the facilities of CNN and translated by CNN from the Russian’. --Benn Newman 12:44, 8 June 2008 (UTC)
CNN holds copyright on translation, so delete. giggy (:O) 02:31, 9 June 2008 (UTC)
  • Deleted Yann 12:14, 9 June 2008 (UTC)

The Secret of Eternal Youth[edit]

The following discussion is closed:

The author is still alive. A permission is needed. Yann 12:21, 12 June 2008 (UTC)

The author has also put this content on :
which is again free of any copyrights. Rohitom 12:41, 12 June 2008 (UTC)
There is no mention of copyright on these pages. Copyright is given by default. Puting his works on the web is not a sufficient proof. If you have a mail from the author giving permission, that could be OK. A copy of the mail needs to be sent to Yann 12:49, 12 June 2008 (UTC)
Is there any other way out?--Rohitom 13:12, 12 June 2008 (UTC)
Please Reply--Rohitom 05:24, 13 June 2008 (UTC)

Only the copyright holder can release the copyright, and they must explicitly release it, or explicitly relicense it under GFDL or Creative Commons. John Vandenberg (chat) 06:04, 13 June 2008 (UTC)

From the main page of the web site "All information contained within this Website is copyrighted and all rights are reserved." I'm going to delete now as a clear copyright violation as I can't see a release specific to this text. May also have issues related to Wikisource inclusion criteria. FloNight 14:17, 23 June 2008 (UTC)

Deleted. FloNight 23:21, 23 June 2008 (UTC)