>> Hmm...does the first bullet point rule out the GPL (or any >> non-permissive, BSD-style license)? > > No.
I wouldn't have thought so. At least, assuming that the lifting of the language from the OSI implies they intend the same interpretation as the OSI -- and hence are OK with the GPL. [But Andrew seems to be of the opposite opinion? "Yes, I would have to say it does rule out those restrictive licenses" -- Andrew, do you mean to include the GPL as one of "those restrictive licenses"? Or am I making a goof here?] > (Did you mean "non-BSD-style license", there? It's hard to call (new) > BSD "non-permissive".) Indeed I did. Wasn't finished with my coffee yet. :) > A "distribution" is simply a aggregate collection of programs; it has > nothing to do with linking or derived works in terms of the GPL. Your > "most natural" reading is not the accepted one. :) So I must conclude. Though I would think that outside the context of talking about "Linux distros," the "accepted" reading is not necessarily more natural than one in which a "distribution" refers to something more like a tarball of a package. I guess I am mostly pointing out that, absent a definition of "distribution," we should probably hope that whoever is enforcing this policy is less prone to that interpretation than pre-coffee me. Richard
