On Fri, May 21, 2004 at 11:36:17PM +0100, Lewis Jardine wrote: > 4 may be non-free: as RMS understands it, violating the GPL for a given > work terminates any rights you have to it, now and in the future, even > those which would otherwise be granted by having another copy of the > work distributed to you. This may (/may/) violate DFSG #9 (if you were > to violate the GPL in one work, you then couldn't distribute the same > code in a different work), or may (/may/) violate the spirit of the DFSG.
I don't think this violates the spirit of the DFSG; Free Software *is* about balancing the rights of authors with the rights of users, and it's important to protect authors from this particular loophole. In any case, the argument we've seen in practice that gets around this is that a violator need not have accepted the terms of the GPL before distributing, so their distribution is "merely" copyright infringement and not a violation that terminates their license. > 8 bears a lot of resemblance to "don't break the law" clauses, which > used to be considered DFSG free, but which now are increasingly not. By *whom*? I haven't seen anyone offer a solid argument that "don't break the law" clauses are acceptable, I don't believe there's consensus on d-legal that they should be acceptable, and I haven't seen any cases of them actually being accepted into main. Please provide references to relevant mailing list discussions if I've overlooked them. > This option, if exercised, may be counter to DFSG #5, or possibly DFSG > #1, and may possibly fail the 'dissident test'. I think you're probably right that this option, if exercised, would be non-free. However, I have never seen anyone exercise this particular option -- I had even forgotten it was there. Reagrds, -- Steve Langasek postmodern programmer
signature.asc
Description: Digital signature
