Messages in this thread | | | Date | Tue, 22 Dec 1998 10:28:23 -0500 (EST) | From | Eric Lee Green <> | Subject | Re: GPL and s/w patents [was Re: Article: IBM wants to ...] |
| |
On Mon, 21 Dec 1998, Jim Freeman wrote: > On Mon, Dec 21, 1998 at 05:17:16PM -0600, Michael Elizabeth Chastain wrote: > > IBM is not saying "we want a BSD-like license where we can hoard our > > enhancements." They just don't want to lose our patents in other products > > if they write linux kernel code and give it away. > ... > > IBM is not looking for a license that permits them to hoard their code. > > They are looking for accommodation so that they can distribute Linux but > > not surrrender all their patenent rights everywhere. > > IBM doesn't lose any patent protection if they give away code that > implements patented processes/methods - after all, they have patented > (ie: opened to public inspection) them in return for exclusive control > over the USE of said processes/methods. In fact, making an implementation > thus available furthers the opportunities for them to impose license/use > fees and restrictions on anyone who USES the patented method implemented > in such GPL'd code.
Except that the GPL prohibits them from doing this. See the following paragraph in V2 of the GPL (under which the Linux kernel is licensed):
"Finally, any free program is threatened constantly by software patents. We wish to avoid the danger that redistributors of a free program will individually obtain patent licenses, in effect making the program proprietary. To prevent this, we have made it clear that any patent must be licensed for everyone's free use or not licensed at all."
It is pretty clear that if you wish to obtain a patent on any portion of the Linux kernel, the GPL requires you to license that patent for everybody's free use or not license it at all. (i.e., not enforce it at all). I think what IBM's lawyers are wondering about is this: What happens if the patent was obtained for non-GPL code and is then licensed (for free use) for GPL code? Must it then be licensed for free use by EVERYONE?
Right now it's not a problem. The authors of contributions to the Linux kernel are the ones legally responsible for obtaining licenses and assuring that they meet all the legal requirements of the GPL, and IBM is happy that way 'cause who's going to sue Linus Torvalds because IBM licensed him (for free use in GPL'ed software) a technology to use in the Linux kernel? But: if IBM itself contributed a piece to the Linux kernel, AND part of that piece was covered by an IBM software patent, AND IBM licensed that patent for free use in all GPL'ed software -- what is IBM's legal exposure then? Can they be sued for violating the GPL (since said patent is licensed for free use only in GPL'ed software, not in ALL software as required by the GPL)?
Remember, software patents are a racket. IBM uses them to force other companies to trade their own patents with IBM. And vice-versa. At this point it seems to be a no-win game, but it's like nuclear disarmament -- if one side (IBM) unilaterally disarms, it puts them at the mercy of all other parties, who could literally hold them for ransom. Nobody seems to actually make MONEY off of these software patents, but everybody is reluctant to give them up, for fear that someone else won't give theirs up. Expecting IBM to thus effectively give up their patents (by licensing them for free use by everybody, not just for free use in GPL'ed software) is thus not a realistic expectation. The LPGL is a similar recognition of reality, so there is precedent for the FSF to make such adjustments pursuit of their goal of free software everywhere.
The real deal, of course, would be to get a law through Congress eliminating mathematical algorithms as patentable material (remember, all computer programs are in fact mathematical algorithms, as proven by Turing and Von Neumann many years ago). Fat chance. The NSA alone would fight this the same way they fight relaxing encryption export requirements, because the RSA software patent is a keystone of their effort to discourage strong encryption. But it is worth a try. Who knows, we might get lucky.
-- Eric Lee Green eric@linux-hw.com http://www.linux-hw.com/~eric "Linux represents a best-of-breed UNIX, that is trusted in mission critical applications..." -- internal Microsoft memo
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |