This is the message I sent.
Oliver
+-------------------------------------------+ | Oliver Pereira | | Dept. of Electronics and Computer Science | | University of Southampton | | omp199@ecs.soton.ac.uk | +-------------------------------------------+
---------- Forwarded message ---------- Date: Tue, 8 Jul 2003 17:15:39 +0100 (BST) From: Oliver Pereira omp199@ecs.soton.ac.uk To: Russ McNeil russmcneil@malaspina.org Subject: Use of Wikipedia articles at Malaspina.com
Dear Dr. McNeil,
I discovered your "Malaspina Great Books" website today, and found that you have included quite a few biographical articles adapted from Wikipedia in your database. I'm an editor at Wikpedia, and I should point out that the articles there are covered by the GNU Free Documentation License (or GFDL), which says that if you want to copy material covered by this licence, you need to release your copies under the same licence.
See the "Wikipedia:Copyrights" page at: http://www.wikipedia.org/wiki/Wikipedia%3ACopyrights
and the text of the GFDL at: http://www.wikipedia.org/wiki/Wikipedia%3AText_of_the_GNU_Free_Documentation...
I am not a lawyer, and I'm not sure *exactly* what you have to do to licence your copies of the articles under the GFDL, but you have to at least say on each page that the text is licenced under the GFDL, and you have to link to the text of the GFDL as well. Perhaps you could link to the copy of the GFDL on the GNU website itself:
http://www.gnu.org/licenses/fdl.html
Perhaps it would be worthwhile asking what you have to do on the WikiEN-L mailing list wikien-l@wikipedia.org or just come along to the Wikipedia website itself and ask there.
Oliver
+-------------------------------------------+ | Oliver Pereira | | Dept. of Electronics and Computer Science | | University of Southampton | | omp199@ecs.soton.ac.uk | +-------------------------------------------+
Oliver Pereira wrote:
This is the message I sent.
[Snipped!]
It's perfectly fine that Oliver sent a letter, since he discovered and his letter was just from him. We should send one more letter: a form letter (or something that looks just as official ^_^) that's worked out on [[meta:]] and can be used again.
-- Toby
On Tue, 8 Jul 2003, Toby Bartels wrote:
Oliver Pereira wrote:
This is the message I sent.
[Snipped!]
It's perfectly fine that Oliver sent a letter, since he discovered and his letter was just from him. We should send one more letter: a form letter (or something that looks just as official ^_^) that's worked out on [[meta:]] and can be used again.
I've had a reply from Russ McNeil of Malaspina Great Books. I was going to forward his reply to the list, but then I had the thought that that would probably be an infringement of *his* copyright, which would be quite ironic in the circumstances. :)
Basically, he says that he supports the Wikipedia philosophy, and thought that by acknowledging the source and linking back to the Wikipedia articles he was dong everything he had to. He said he would check to see if he needed to do more. (I think this means he didn't trust my statement that he *did* need to do more!) So I think we should check his site again later to make sure that he's making the necessary changes.
Another question, though: Does someone who creates a derived work from one of our articles have to link back to the actual version of the article that they used to create the derived work, or is linking to the current version of the article acceptable?
Oliver
+-------------------------------------------+ | Oliver Pereira | | Dept. of Electronics and Computer Science | | University of Southampton | | omp199@ecs.soton.ac.uk | +-------------------------------------------+
Oliver Pereira wrote:
On Tue, 8 Jul 2003, Toby Bartels wrote:
Oliver Pereira wrote:
This is the message I sent.
[Snipped!]
It's perfectly fine that Oliver sent a letter, since he discovered and his letter was just from him. We should send one more letter: a form letter (or something that looks just as official ^_^) that's worked out on [[meta:]] and can be used again.
I've had a reply from Russ McNeil of Malaspina Great Books. I was going to forward his reply to the list, but then I had the thought that that would probably be an infringement of *his* copyright, which would be quite ironic in the circumstances. :)
Probably not. If you were writing on behalf of Wikipedia, then his reply to you was received as an agent of Wikipedia, so sharing it with the list would be appropriate.
Basically, he says that he supports the Wikipedia philosophy, and thought that by acknowledging the source and linking back to the Wikipedia articles he was dong everything he had to. He said he would check to see if he needed to do more. (I think this means he didn't trust my statement that he *did* need to do more!) So I think we should check his site again later to make sure that he's making the necessary changes.
Malaspina Great Books and Malaspina College did come up before, and Ed Poor appears to have written to them in October 2002 about the Ernest Hemingway article to mention the GFDL requirements, and the attributions were modified accordingly by Malaspina. My impression is that Malaspina is quite willing to treat the matter with appropriate consideration.
Ec
Oliver Pereira wrote:
Another question, though: Does someone who creates a derived work from one of our articles have to link back to the actual version of the article that they used to create the derived work, or is linking to the current version of the article acceptable?
I'm in no way an expert, but looking at the practical implications of such a requirement, it would seem that such a requirement would make the GFDL useless for most purposes. Hardly any webmaster would archive every past version of a page.
Timwi
Timwi wrote:
Oliver Pereira wrote:
Another question, though: Does someone who creates a derived work from one of our articles have to link back to the actual version of the article that they used to create the derived work, or is linking to the current version of the article acceptable?
I'm in no way an expert, but looking at the practical implications of such a requirement, it would seem that such a requirement would make the GFDL useless for most purposes. Hardly any webmaster would archive every past version of a page.
Well, the GFDL doesn't actually require that the original work be provided at all; only that the derived work itself be provided in "transparent" form. With normal text this just means that you have to provide the derived work; with wiki-text, since "transparent" is derived as the wiki-markup version, someone who makes an HTML/ASCII/whatever derived work would have to provide the wiki-text, which is sometimes onerous to make them do. Thus you can just let them link to wikipedia, where the text can be found. I think reading the GFDL it actually is the exact work you derived from that legally should be provided, but as all past versions are archived by wikipedia anyway, simply linking to the wikipedia page should suffice. If someone wanted to be pedantic they could link to the specific page in the history instead of the "current version," but I think that would have adverse consequences (like people reading out of date articles). At most I think it might be reasonable for them to cite the date they took it off wikipedia, and then someone can search through the page history themselves to find that version (citing the date you used an online source is a pretty good idea anyway).
-Mark
Delirium wrote in part:
Well, the GFDL doesn't actually require that the original work be provided at all; only that the derived work itself be provided in "transparent" form. With normal text this just means that you have to provide the derived work; with wiki-text, since "transparent" is derived as the wiki-markup version, someone who makes an HTML/ASCII/whatever derived work would have to provide the wiki-text, which is sometimes onerous to make them do.
Since HTML and ASCII are themselves transparent (at least HTML is unless you deliberately obscure it), how do you know that part of the derivation process wasn't to translate the soure from PediaWiki to HTML? Then they just have to provide the HTML of the derived work -- which is exactly what their HTTP server does.
OTC, the link back to Wikipedia is required, as I understand it, to give /credit/ for the work, which the GFDL also insists upon. It could be avoided by copying all of the credit info to the new site (which might include copying contact info from user pages, I'm not sure). Simpler just to link back to Wikipedia.
-- Toby
Toby Bartels wrote:
Since HTML and ASCII are themselves transparent (at least HTML is unless you deliberately obscure it), how do you know that part of the derivation process wasn't to translate the soure from PediaWiki to HTML? Then they just have to provide the HTML of the derived work -- which is exactly what their HTTP server does.
OTC, the link back to Wikipedia is required, as I understand it, to give /credit/ for the work, which the GFDL also insists upon. It could be avoided by copying all of the credit info to the new site (which might include copying contact info from user pages, I'm not sure). Simpler just to link back to Wikipedia.
After reading through the GFDL again I'm more clear on this now: HTML is explicitly mentioned as an acceptable "transparent" format, so to fulfill that requirement all they have to provide is the webpage itself; they do not anywhere have to provide access to the original unmodified work.
The only further requirement is that they list all authors of their modifications (e.g. list themselves), and list the names of five authors of the original version (or all the authors if fewer than five). Contact information is not required, just the names (presumably if the authors originally published under a pseudonym/username, that name would suffice as well). This is a bit vague though -- it seems to indicate that essentially all they have to do is preserve the credit in the author list of the original document: but Wikipedia documents do not have an author list. One can be inferred by looking at the page history, but there is no explicit list anywhere of "these are the authors of this document." The GFDL isn't clear on whether it's the modifier's responsibility to construct such a list of the original document does not contain one; I would lean towards not. If not, then all the deriver has to do is provide their webpage and say "this is licensed under the GFDL." They don't even have to say "from Wikipedia," as Wikipedia does not own any of the relevant copyrights.
Of course I may be wrong in that interpretation -- perhaps they have to list the copyright holders even if the original document does not, but that seems like it'd be an odd requirement.
-Mark
Delirium wrote:
This is a bit vague though -- it seems to indicate that essentially all they have to do is preserve the credit in the author list of the original document: but Wikipedia documents do not have an author list. One can be inferred by looking at the page history, but there is no explicit list anywhere of "these are the authors of this document."
Well, we've always treated this as "It's in the 'Page history'". But it probably would be simple enough to have a link to 'Authors' on each article, and when you click on it, you get a list of all unique id's that worked on the article.
Presumably, at some point as we draw closer to 1.0 and the possibility of print reuse, we'll want to have a setting in preferences for people to (optionally) put their real name for attribution purposes. And then on this 'author' page would show that as well.
--Jimbo
On Thu, 2003-07-10 at 17:26, Jimmy Wales wrote:
Delirium wrote:
This is a bit vague though -- it seems to indicate that essentially all they have to do is preserve the credit in the author list of the original document: but Wikipedia documents do not have an author list. One can be inferred by looking at the page history, but there is no explicit list anywhere of "these are the authors of this document."
Well, we've always treated this as "It's in the 'Page history'". But it probably would be simple enough to have a link to 'Authors' on each article, and when you click on it, you get a list of all unique id's that worked on the article.
Presumably, at some point as we draw closer to 1.0 and the possibility of print reuse, we'll want to have a setting in preferences for people to (optionally) put their real name for attribution purposes. And then on this 'author' page would show that as well.
I strongly think we're better off formalizing a policy in which particular author attribution is not required. That is, by contributing to Wikipedia, you agree to be attributed as one of the "Wikipedia Contributors" or somesuch.
The Cunctator wrote:
I strongly think we're better off formalizing a policy in which particular author attribution is not required. That is, by contributing to Wikipedia, you agree to be attributed as one of the "Wikipedia Contributors" or somesuch.
Well that certainly much more closely matches our social custom, in which articles aren't "owned" by anyone, and we value all sorts of contributions to the project without specifically privileging 'authorship'.
But, how can we reconcile your suggestion with the FDL?
--Jimbo
Jimmy Wales wrote:
The Cunctator wrote:
I strongly think we're better off formalizing a policy in which particular author attribution is not required. That is, by contributing to Wikipedia, you agree to be attributed as one of the "Wikipedia Contributors" or somesuch.
Well that certainly much more closely matches our social custom, in which articles aren't "owned" by anyone, and we value all sorts of contributions to the project without specifically privileging 'authorship'.
But, how can we reconcile your suggestion with the FDL?
For past contributions, we probably can't. But for future contributions, we'd ask them to assign authorship rights to the Wikimedia Foundation. Is this what you want, Cunc?
-- Toby
Toby Bartels wrote:
Jimmy Wales wrote:
The Cunctator wrote:
I strongly think we're better off formalizing a policy in which particular author attribution is not required. That is, by contributing to Wikipedia, you agree to be attributed as one of the "Wikipedia Contributors" or somesuch.
Well that certainly much more closely matches our social custom, in which articles aren't "owned" by anyone, and we value all sorts of contributions to the project without specifically privileging 'authorship'.
But, how can we reconcile your suggestion with the FDL?
For past contributions, we probably can't. But for future contributions, we'd ask them to assign authorship rights to the Wikimedia Foundation.
Another alternative may be that as authors we effectively appoint Wikimedia as our copyright agent, but with this may come Wikimedia's '''obligation''' to globally defend those rights.
Most of the discussion on this range of issues leaves the impression of being "stuck" with the GNU-FDL without any right to amend it to suit our own purposes. When the whole project began basic copyright decisions obviously had to be made just to be able to function. Most of us in allowing our efforts to be covered by GNU-FDL are more concerned with the spirit rather than the letter of that document. I certainly do not agree that FSF should have the right to amend the document as it applies to us without the amendment having been fully discussed in one of our appropriate forums. I don't recall any such discussion taking place in November 2002 when version 1.2 was adopted.
In this context I think that the establishment of Wikimedia provides an opportunity for us to develop our own home grown system that takes into consideration the particular problems that Wikipedia has encountered since its inception. If the GNU-FDL (1.2) is to be a basis for that then Wikimedia must retain all rights to make subsequent amendments.
To date the Wikimedia Foundation does not have a published constitution or any kind of democratically chosen by-laws or internationally representative Board of Directors. When it is finally elected that Board should give serious consideration to what kind of copyright licensing rules etc. should apply in the future before everybody's rights are massively transferred from Wikipedia (or whomever) to Wikimedia.
Eclecticology
Delirium wrote in part:
The only further requirement is that they list all authors of their modifications (e.g. list themselves), and list the names of five authors of the original version (or all the authors if fewer than five).
Only 5? Very well. Probably still easier to link to us, however.
Contact information is not required, just the names (presumably if the authors originally published under a pseudonym/username, that name would suffice as well).
What I'm worried about is that if, say, mav submits an edit, then we not only log this under the name "Maveric149" but also link to mav's user page, which lists his real name. Can mav reasonably expect, then, to receive credit as "Daniel Meyer", a name that perhaps he uses professionally? ("contact info" was too broad a term to use.)
This is a bit vague though -- it seems to indicate that essentially all they have to do is preserve the credit in the author list of the original document: but Wikipedia documents do not have an author list. One can be inferred by looking at the page history, but there is no explicit list anywhere of "these are the authors of this document."
I'd argue that standard practice on a wiki is that if you want to know who edited a particular page, then you look at the page history. Understanding that, derivers must look there.
They don't even have to say "from Wikipedia," as Wikipedia does not own any of the relevant copyrights.
I'll agree with this.
-- Toby
I'm in no way an expert, but looking at the practical implications of such a requirement, it would seem that such a requirement would make the GFDL useless for most purposes. Hardly any webmaster would archive every past version of a page.
Timwi
Civil Law is not meant to be practical. It's meant merely to be a roadsign that says, "our stuff, our rules" -- so that later it can look like a breach of contract, and money can be restituted from the prostituting party.
Ironically, the notion that "your not supposed to make $$ off of X" (as a claim) is kinda antithetical to the whole notion of lawsuits, which tend to be about "only Ims supposed to make money off X". And in the end, regardless of GNUFDL, a civil win would be represented (symbolically) in dough. Otherwise why would anyone bother forking out for a lair?
-S-
__________________________________ Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com
People who are using the license are establishing custom and practice which is also relevant. I cannot imagine archiving on my site the past history of a Wikipedia page, the link back to Wikipedia gives access to that. As to law being practical, we all try.
Fred Bauder
http://wwww.internet-encyclopedia.org
From: steve vertigo utilitymuffinresearch@yahoo.com Reply-To: wikien-l@wikipedia.org Date: Thu, 10 Jul 2003 01:54:10 -0700 (PDT) To: wikien-l@wikipedia.org Subject: Re: [WikiEN-l] Re: Use of Wikipedia articles at Malaspina.com
I'm in no way an expert, but looking at the practical implications of such a requirement, it would seem that such a requirement would make the GFDL useless for most purposes. Hardly any webmaster would archive every past version of a page.
Timwi
Civil Law is not meant to be practical. It's meant merely to be a roadsign that says, "our stuff, our rules" -- so that later it can look like a breach of contract, and money can be restituted from the prostituting party.
Ironically, the notion that "your not supposed to make $$ off of X" (as a claim) is kinda antithetical to the whole notion of lawsuits, which tend to be about "only Ims supposed to make money off X". And in the end, regardless of GNUFDL, a civil win would be represented (symbolically) in dough. Otherwise why would anyone bother forking out for a lair?
-S-
Do you Yahoo!? SBC Yahoo! DSL - Now only $29.95 per month! http://sbc.yahoo.com _______________________________________________ WikiEN-l mailing list WikiEN-l@wikipedia.org http://mail.wikipedia.org/mailman/listinfo/wikien-l
Stevertigo wrote in part:
Ironically, the notion that "your not supposed to make $$ off of X" (as a claim) is kinda antithetical to the whole notion of lawsuits, which tend to be about "only Ims supposed to make money off X".
That's not the motivation of the FSF people, of course -- yet even so, they don't endorse that notion in the first place! Commercial application is irrelevant to the GNU licences. And Wikimedia seems to be in line with that too.
And in the end, regardless of GNUFDL, a civil win would be represented (symbolically) in dough.
I'd expect any lawsuit initiated by the FSF, at least, to have a cease-and-desist effect as its main goal. Although Wikimedia is not a GNU project (as we know), there are probably enough people here against profiting that we'd also confine ourselves to that goal in our suits. (Not that such a suit would be very practical in the first place with such a large number of plaintiffs to manage. Or could we initiate a class action?)
Otherwise why would anyone bother forking out for a lair?
???
-- Toby