From: "Jimmy Wales" jwales@bomis.com
This need not be a matter of technical legal copyright problems! Our standards go far beyond that of mere obediance to the law. Even if there were some legal gray area here (and there is not, IMHO), it's still plagiarism.
I agree with all of Jimbo's point. The dichotomy commercial v. non-commercial is not really applicable because of the GFDL. As well, even if one were to use the fair use defense under US law, this fair use may not be fair use for downstream licensees. The problem is that fair use does not primarily depend on the third sector (what some people call non-profits as being the "third Force in democractic societies); fair use is a specific use defense. Thus it might be allowed in Wikipedia, copying a whole text might not be considered a copyright infringement, but someone downstream that does not have notice of that fact is going to infringe and may not have the same fair use defense as Wikipedia will be found an infringer.
While anyone who releases a text has the burden to prove that they did not engage in an infrnging act (thus the burden IMHO should be on the editor contributor and the downstream licensee) Wikiepedia may not have any direct absolute liability because "Wikipedia" as such, did not contribute the text (a reason that the editors should retain authorship attribution and copyright IMO) the goal is to create a base of knowledge that can be used, not sidelined because the due dillegence could not be done to show who actually owns the co-author copyright on any Wikipedia article thus effectively making the GFDL license scheme useless except to non-income producing downstream licensees ( not-for-profit publishers will probably be considered no different than commercial publishers, the only difference is that the money in a not-for-profit does not go to the shareholders (there are none) it goes to' salaries and the assets of the corporation, there is still plenty of income from the exploited intellectual property.
Thus, copyright violation vigilance will do at least two things, first make it easier for third parties to do their due dillegence, second make it harder for Wikipedia (be that the volunteers or the not-for-profit structure in Wikimedia) to be held liable for failure to take reasonable steps to prevent any infringement (or defamation/personality rights issues from cropping up). This is important when dealing with non-US nationals that browse Wikipedia, Wikipedia's problem will not be copyright so much as it will be the problems that might arise with defamation and privacy/publicity rights; in those cases the person that posts the tortious material is not shielded by US law. the law in the place of the defamation applies (and that can be any country in the world).
We should also remember that copyright law itself provides the best "out" for this kind of problem. sec. 102(b) of Title 17 USC states:
"(b) In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery, regardless of the form in which it is described, explained, illustrated, or embodied in such work."
Thus any material that is rewritten to the point where the only resemblance with the original material are ideas, concepts, facts, that will never be a copyright infringement.
As far as history pages are concerned my current thinking on copyright infringements appearing on these pages is to invoke sec. 108 of the Copyright Act. Thus while a current page version is released under the GFDL, prior versions of an article should have the famous 108 notice so that if some one has downloaded copyright infringing material at least Wikimedia will not be responsible as it is acting as a non-commercial archive (this cannot apply to the current version of any page as it would be in violation of the GFDL).
Alex756