On 3 December 2010 14:14, Anthony <o...@inbox.org> wrote: > > Okay, true. I still think it accomplishes something very important > which is the status quo under CC-BY-SA. OSMF doesn't get any special > rights which, for instance, a fork wouldn't have.
Ah, I see, and I'm fairly sure that wasn't what those I've spoken to want to achieve, but I may be wrong and I'm happy to be corrected. > > You must know more than I do, because I don't think you can speculate > on the intent of a phrase without at least knowing who added it. It That's what a court would have to do of course, but it may be that we disagree on something else (see below). > very well may have been added precisely for the effect of making > everything effectively PD. I know that's the only reason I supported > CT 1.2, though I wasn't dumb enough (this time) to point it out. > Yes. I am fairly clear that some people don't want to make OSM effectively PD - they do want to restrict its usage that is why ODbL is being used (otherwise why bother with it)? > > It's not superfluous, because the obligation on OSMF is to license the > contents *as part of a database*. The ODbL applies to *the database*, > not the contents. (In some/most/all jurisdictions, if you don't agree > to it, you can probably ignore it, because there aren't any rights in > the database. But probably in at least some jurisdictions you can't > ignore it, due to sui generis database rights and/or sweat of the brow > copyright in the database itself). Yes, it makes the DbCL part > superfluous, but as I've explained before the DbCL, if *it* does not > make the work effectively PD, is itself superfluous. And if you look > at the history, the DbCL language was added at the same time the "and > any party that receives your contents" was taken out > (http://www.osmfoundation.org/index.php?title=License/Contributor_Terms&diff=231&oldid=204 > which, by the way, was *after* the vote). Well, that's not the whole story (from a legal analytic point of view) - but I'm sure you know that. The extra problem is that it may be (it seems likely to me) that the only sui generis database rights (or for that matter database copyright, though that seems like a long short to me in the jurisdictions I know) that OSMF is likely to have now is that which it is licensed to sublicense under the contributor terms. If all the CT's give to OSMF is also given to everyone (who receives the contents, which is of course anyone relevant) then the ODbL is completely superfluous since ODbL can't restrict that which is already permitted by another means (or can't usefully do so) even in countries with a sui generis right. If that's the intent then there's no point at all in messing around with paragraph 3 which can just be removed. > > In any case, even if the requirement were superfluous (which, as I > explained, it isn't), I don't see any alternative explanation. > A mistake? Someone infelicitously drafting the licence? It does happen you know :-). > (*) It has been pointed out previously that we should probably > *require* OSMF to release the database under a free license, rather > than merely *allow* them to, but as it stands they may, but don't have > to. Hmmmm, at the moment paragraph 3 requires OSMF to use or sublicense the contents, which isn't quite the same thing as release I'll grant you, but there is some obligation in there. If I were instructed by OSMF I'd probably suggest that wasn't what was wanted. Its an open ended commitment, which is always a bad thing. In general one wants to at least make the obligations mutual. The obligation is qualified by 1(b) though. -- Francis Davey _______________________________________________ legal-talk mailing list legal-talk@openstreetmap.org http://lists.openstreetmap.org/listinfo/legal-talk