On 5/17/05, Michael Poole <[EMAIL PROTECTED]> wrote: > Raul Miller writes: > > I don't really know what a "pure copyright license" means, > > and I'm not arguing that that is the case. [I am arguing, > > elsewhere, that there are some reasonable moral standards > > associated with legal conduct in these areas, and that a > > person could be sued for neglecting those issues. But > > that's tangential in this context.[ > > > > I am arguing, here, that the scope of that license does not > > include making unlicensed collective works. > > This is circularly true. No license licenses (permits) acts that it > does not license.
Actually, it's circularly bullshit. The scope of that license does include the making and distribution of "collective works", courtesy of both the "mere aggregation" clause (as ably construed by Humberto) and the simple fact that people in the computer industry lump a bunch of software installers on the same media all the time, without any special license. When that's the prevailing and expected practice, it's easily construed as an implied term in the contract, in the absence of clear language to the contrary. And if you have any worries about the (uncopyrightable because obvious from the text of Quagga alone) collection Quagga + NetSNMP + libssl, you certainly needn't worry about it in the US given the way those components are actually packaged (dynamic linking); that combination only occurs at run-time, and is exempt from the entire "copyright infringement" calculus per 17 USC 117. > I do not read his argument as meaning that. I read his argument and > case citations as meaning that, when a contract exists, a court would > have to determine the proper forms of liability and relief by looking > at the contract first. The court also has to look at the contract to determine the scope of the license. That scope is determined by what license is granted when all conditions of all return performance, etc. are met. For example, the scope of a typical license to publish involves editorial corrections (subject to reasonable approval by the author); if the book goes out with corrections that weren't run by the author for approval, then the author has a possible cause of action for breach of contract, but not for copyright infringement, because the publisher's action was within the scope of the license. This seems to be another part of the agenda behind all of that "GPL is a creature of copyright law" bullshit -- an attempt to get the covenants of return performance rolled into the scope. > My layperson's interpretation of the rationale is that since the > contract is specific to the parties, it defines their rights and > responsibilities -- within the bounds of the contract -- rather than > using those defined by statute. I think that's right, modulo rules about what terms can and can't be construed into a contract (see discussions about copyright assignment, license termination, and agency to sublicense). IANAL, etc. Cheers, - Michael
