On Thu, 12 Feb 2004, Luchezar Georgiev wrote: > On Thu, 12 Feb 2004 09:10:33 +0100, Roberto Mariottini wrote: > > > I understand you Lucho, but Bart is right: if it's not possible, then just it > > isn't. > > So, we are (OK, I am) in the trap of our own license! The GPL, instead > of giving us freedom, takes it away from *us*, the *developers*! Isn't > that absurd?! :-(
no it's not absurd, to the contrary, it's the whole point of the GPL! Otherwise a BSD or MIT/X11 style license should be used. There are lots of things the developers cannot do, for instance distributing a closed-source program that derives from the FreeDOS kernel is not allowed either. But please note that the GPL only affects distribution, not usage. This is also why I think restricting military usage is pointless: the really bad guys will ignore such usage restrictions anyway. > Besides, nobody has yet written a *real*, convincing argument *why* we > have a "combining" and not "aggregation" between the different parts of > a compressed executable. This doesn't matter for the GPL. Aggregation only applies to having differently licensed seperate programs on one CD or in one zipfile, for instance. The only exception that may possibly apply is the "compiler" exception but I highly doubt that because a) an FSF spoke person did not mention it. b) the UPX authors clearly don't think so either. They clearly make an exception in the UPX license for non-GPLed binaries. Because otherwise, because of the GPL-ness for the stub the whole binary would need to be covered by the GPL! p.s. i'm deliberately ignoring upx-ucl vs. upx-nrv here -- it's besides the point since the stub they insert is the same. > So I'm not convinced yet that using aPack would > be illegal EVEN THOUGH I register my own copy No. *Using* apack is not restricted at all. You may apack all your FreeDOS executables. It's the *distribution* that is (likely) restricted by the GPL. You may even (subject to APACKs license) distribute: * kernel.exe * exeflat.exe * apack.exe then allow the user to apack and then exeflat the kernel. Now aggregation applies. Sounds ridiculous? Maybe. But you have to realize that the FSF has an agenda too, which likes to minimize the use of closed source software. Which such a technique it becomes crystal clear to the user that a non-free packer is involved. And, moreover, you give the user extra freedom by being able to use *any* packer, including upx, pklite and what have you. What speaks in favour of Steffen who just refuses to distribute pre-packed command.com's (neither do I pack command.com in the kernel binary zip by the way). Bart *PS. I really object to comments about Microsoft being "MASTERS" and lawyers being "monsters". I think this is highly inappropriate and does not contribute to the discussion at all. Many lawyers (some being pro-bono), after all, have helped people who were accused for crimes they were not involved in. ------------------------------------------------------- SF.Net is sponsored by: Speed Start Your Linux Apps Now. Build and deploy apps & Web services for Linux with a free DVD software kit from IBM. Click Now! http://ads.osdn.com/?ad_id=1356&alloc_id=3438&op=click _______________________________________________ Freedos-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/freedos-devel
