Jonathan P Tomer <[EMAIL PROTECTED]> wrote: > it's not gpl compatible in that you can't take a work that's partially > qpl and partially gpl and license it under either one. such is the > nature of the infective copyleft.
It's as much the nature of the QPL as it is the nature of the GPL. > (see below for a short rant on this subject). however it -does- > seem dfsg-free to me (because you only have to give troll a copy of > derivative works if they are otherwise non-free). Actually, it's not whether they're non-free or not, it's whether they're available to the general public. [Also, it's not clear whether or not it's troll you have to give the copy to -- is the library author the initial developer, or would that be someone else.] All in all that aspect of the license is rather ambiguous. But, other than the ambiguity it's DFSG free. > ok, that was the (at least arguably) useful content of this message; > below is an unrelated rant about copyleftage, so if you're pressed for > time you can hit 'd' now. Likewise. > it irritates me that licenses for free software force all the work > derived from the software to be under the same license. Er.. First: this contradicts your statement about BSD software being relicensable. Second: even the GPL doesn't force you to put the derived work under the GPL -- except for the part which is already GPL, the rest of it could be placed under the BSD license. > partly this is because i think it's wrong to force people to make > software free; if freedom really is superior we should let it show > itself in a fair competition. What's unfair about a free software license that's not also unfair about a non-free software license? Or are you suggesting that we work to repeal copyright law? > put more simply, it is wrong to think that software should be more > free than people; if someone wants to make his work proprietary, > that is his option even if i find it a morally reprehensible option. Free software licenses don't prevent someone from making his work proprietary. > however there are lots of people who disagree with me there, and i'm > not going to try to convince anybody here that copylefts are wrong, or > attempt to keep anyone from using them. I guess I'm an example of someone who disagrees with you. See above. > more practically, copylefts like the gnu gpl have a very big problem, > which is that you can't put derivatives under another license, even > a copyleft. this means you can't combine two different copylefted > works into one big work, which causes unnecessary competition > between differently licensed sets of free software that shouuld be > co-operating. Yes you can. See above. > richard stallman proposes to solve this problem by having all other > free licenses contain a paragraph allowing the licensee to relicense > derived works under the gpl. this is quite frankly stupid; if everyone > thought the gpl was an acceptable way to license the software it > would all be gpl'ed in the first place. similarly we can't just hope > that all licenses will contain a list of all the other licenses you > can substitute. That's not what Richard was suggesting. He was reminding people that the author of a GPLed work can resolve the conflict between the GPL and a non-GPL free license simply by granting additional permissions beyond those granted in the GPL. > so i think (and i may propose to the gnu project one of these days) > that a proper copyleft should allow the licensee to sublicense > the product under any license that preserves the rights and > restrictions that are important to the license (including of > course the sublicensing clause). Er.. the GPL already has this. > that way, if you have to take two differently copylefted programs > and combine them you can just take all the required restrictions and > rights from both of them and put them in one license. of course if the > two licenses had conflicting clauses you still couldn't combine them, > but that's much rarer. Great idea. -- Raul
