I just had an interesting experience with an apparent copyright violation.
The text in [[Marina Tsvetaeva]] is very similar and in many cases exactly the same as text in an external website. One user deleted the text in the article citing a copyright violation on our end. Even though I saw obvious similarities between our version and theirs I wasn't so sure /we/ were the ones who violated copyright because the article existed in pretty much the same form since before the move from the UseMod wikiware back in February.
My suspicion that we in fact were not violating any copyright was confirmed when user:sjc claimed authorship and stated that the text had been in the public domain for 12 years.
I am not sure if the external website used the public domain version by sjc or one of our earlier versions, but as Wikipedia grows we will increasingly find apparent copyright violations on our end that are in fact just the opposite: other websites taking our text and using it in a way that is not compatible with our license (there is also the issue of public domain text too.... Remember, cite your sources!).
What should we do in such cases? Furthermore, should we make a rule that if a Wikipedia article "fails" the Google test AND is over a set number of weeks old, then we should give the original contributor the benefit of the doubt and /not/ delete the text?
We almost lost a great article on Marina Tsvetaeva and I don't want to lose any others due to this type of mixup.
--mav
Daniel Mayer wrote:
My suspicion that we in fact were not violating any copyright was confirmed when user:sjc claimed authorship and stated that the text had been in the public domain for 12 years.
What should we do in such cases? Furthermore, should we make a rule that if a Wikipedia article "fails" the Google test AND is over a set number of weeks old, then we should give the original contributor the benefit of the doubt and /not/ delete the text?
This points to a need for a policy on procedures when someone suspects a copyright violation. I suspect that this issue will arise more frequently as Wikipedia becomes bigger. It doesn't do to have members deleting articles on the mere suspicion that it may violate copyright.
Eclecticology
Ray Saintonge wrote:
This points to a need for a policy on procedures when someone suspects a copyright violation. I suspect that this issue will arise more frequently as Wikipedia becomes bigger. It doesn't do to have members deleting articles on the mere suspicion that it may violate copyright.
Well, we can start by experimenting with potential policies on [[Duesseldorf]] and [[Great Sandy desert]], both of which are recent additions to the wiki which include text that is duplicated verbatim on/from an external website without a citation.
My first step was to ask in the Talk: page about source & copyright status, noting the similarity with given URLs.
-- brion vibber (brion @ pobox.com)
Ray Saintonge wrote:
This points to a need for a policy on procedures when someone suspects a copyright violation.
Here is one method: Do nothing until the copyright owner presses charges. This way, we can all focus on writing new articles and avoid wasting time on playing police. Maybe this is too easy. Does it need to be more compliated?
Lars Aronsson wrote:
Here is one method: Do nothing until the copyright owner presses charges.
Well, "pressing charges" is a bit late in the game. Under the DMCA, an ISP has a "safe harbor" against copyright violations by users, so long as there is a "takedown" policy in place. Upon the receipt of a proper written complaint, I'm required to take down the disputed material. If the person who posted it wants to make a proper written response claiming to have the right to the material, then I can put it back up pending those two parties suing each other, but my liability is limited.
However, we have other concerns that extend beyond mere legal liability.
First, we want to have a solid reputation for the intrinsic reason that we want our encyclopedia to be a good one, well respected, and useful. We don't want to get the reputation of "the so-called encyclopedia that really just stole articles from all over the web."
Second, we want potential licensees to have a level of comfort about the content. Print publishers will not have the protection of the DMCA ISP safe harbor provisions, and so they will desire to have a stronger defense than a laissez-faire attitude on our part could provide.
--Jimbo
Jimmy Wales wrote:
Lars Aronsson wrote:
Here is one method: Do nothing until the copyright owner presses charges.
Well, "pressing charges" is a bit late in the game. Under the DMCA, an ISP has a "safe harbor" against copyright violations by users, so long as there is a "takedown" policy in place. Upon the receipt of a proper written complaint, I'm required to take down the disputed material. If the person who posted it wants to make a proper written response claiming to have the right to the material, then I can put it back up pending those two parties suing each other, but my liability is limited.
Of course, in the general case, waiting until someone presses charges is waiting too long to do anything, though there may still be anxious plaintiffs who try to by-pass earlier steps. Hopefully those cases would be thrown out of court for failing to give a proper notice to the ISP in the first place. It does seem to me that "proper written complaint" involves more than a complete stranger appearing out of nowhere and saying "That's mine!"; there has to be a modicum of evidence that he has the right to make that demand.
Steve's "Tsvetaeva" article appears to present quite a different situation. I have no reason to doubt that he had the original copyright on the material, and that at some point he chose to abandon that copyright - all this over a twelve year period. We are really dealing with the situation where something which has gone into the public domain has ben used or incorporated into someone else's work, and a blanket copyright on that user's application has been alleged to apply on a part of that work that was no longer copyright.
This is a risk of success. As Wikipedia develops a positive reputation for its content we can expect a lot more of this to happen. In practical terms there will always be a time lag while things unfold, if for no other reason than that it takes time for our unknown plagiarist to "research" and put his material together. Logs and other proof can be kept. Under U. S. copyright law what I write today goes into the public domain on Jan. 1, 2098. This letter is covered by GNU-FDL, but if my great-grandson chooses to dispute the fact in the year 2070, what evidence will remain? (assuming that I had not raised the subject in the letter) What confidence can we have that Jimmy will be able to argue the point at that time?
Even if every Wikipedia article has GNU-FDL statement attached there is no practical way of enforcing it upon all users. At some point I read something from Lee where he had imbedded a watermark into a picture that he uploaded. It would probably be helpful to embed a watermark that claims coverage under GNU-FDL in every picture that is uploaded; something of the sort could probably also happen with material in other special media such as sound. Text material will remain a problem. It would also be an interesting experiment to put watermarked and undisputedly legal but very desirable illustrations into some of the most popular articles, and try to track what happens to them on the net.
It is without regret that I predict that the copyright industry will eventually become the victim of its own recent successes. Disney's mouse has been eating cheese for 75 years, but wholesome family values have inhibited any suggestion that in all that time he ever had to relieve himself in the can. A mouse that is so full of shit will just have to blow up sooner or later.
Eclecticology
Of course, in the general case, waiting until someone presses charges is waiting too long to do anything, though there may still be anxious plaintiffs who try to by-pass earlier steps. Hopefully those cases would be thrown out of court for failing to give a proper notice to the ISP in the first place.
Problem is that happy sequence will cost you a couple of thousand bucks.
Fred Bauder
Hmm. Well, in a way I think we should be much more proactive in checking copyright. It will also help to highlight deficient articles (the Tsvetaeva one which was considered initially problematic was a stinker).
I have done a major refactoring of the Tsvetaeva article today; it will be interesting to see whether this one gets "borrowed" at any point. There were a number of factual inaccuracies in the original article, + a lot of hearsay evidence and these have now been corrected. It is also an altogether more balanced and encyclopaedic piece, concentrating on the specifics rather than my own take on things.
----- Original Message ----- From: "Lars Aronsson" lars@aronsson.se To: wikipedia-l@nupedia.com Sent: Wednesday, July 31, 2002 2:31 PM Subject: Re: [Wikipedia-l] "Copyright violation" question (just the opposite of what you think)
Ray Saintonge wrote:
This points to a need for a policy on procedures when someone suspects a copyright violation.
Here is one method: Do nothing until the copyright owner presses charges. This way, we can all focus on writing new articles and avoid wasting time on playing police. Maybe this is too easy. Does it need to be more compliated?
-- Lars Aronsson (lars@aronsson.se) Aronsson Datateknik Teknikringen 1e, SE-583 30 Linuxköping, Sweden tel +46-70-7891609 http://aronsson.se/ http://elektrosmog.nu/ http://susning.nu/
[Wikipedia-l] To manage your subscription to this list, please go here: http://www.nupedia.com/mailman/listinfo/wikipedia-l
I have substantially refactored and will continue to refactor this article until it does justice to a poet I rate in the very highest rank. It was a straight cut and paste job from my original handout, and I never really bothered to do much to it other than to expand my notes about Marina Tsvetaeva's poetry. It looks very very jaded now, particularly in the light of Viktoria Schweizer's biography. Some of it is just //wrong// e.g. the Lily Feiler comments, the dates of Tsvetaeva's mother's TB, etc. We need some way, however, of benchmarking our stuff which we import from our private work, because otherwise this type of issue will surface from time to time. Maybe a source comment on the /Talk page would be sufficient so that people are aware of who's done what.
----- Original Message ----- From: "Daniel Mayer" maveric149@yahoo.com To: wikipedia-l@nupedia.com Sent: Wednesday, July 31, 2002 7:37 AM Subject: [Wikipedia-l] "Copyright violation" question (just the opposite of what you think)
I just had an interesting experience with an apparent copyright violation.
The text in [[Marina Tsvetaeva]] is very similar and in many cases exactly the same as text in an external website. One user deleted the text in the article citing a copyright violation on our end. Even though I saw obvious similarities between our version and theirs I wasn't so sure /we/ were the ones who violated copyright because the article existed in pretty much the same form since before the move from the UseMod wikiware back in February.
My suspicion that we in fact were not violating any copyright was
confirmed
when user:sjc claimed authorship and stated that the text had been in the public domain for 12 years.
I am not sure if the external website used the public domain version by
sjc
or one of our earlier versions, but as Wikipedia grows we will
increasingly
find apparent copyright violations on our end that are in fact just the opposite: other websites taking our text and using it in a way that is not compatible with our license (there is also the issue of public domain text too.... Remember, cite your sources!).
What should we do in such cases? Furthermore, should we make a rule that
if a
Wikipedia article "fails" the Google test AND is over a set number of
weeks
old, then we should give the original contributor the benefit of the doubt and /not/ delete the text?
We almost lost a great article on Marina Tsvetaeva and I don't want to
lose
any others due to this type of mixup.
--mav
[Wikipedia-l] To manage your subscription to this list, please go here: http://www.nupedia.com/mailman/listinfo/wikipedia-l
wikipedia-l@lists.wikimedia.org