Re: Article: IBM wants to "clean up the license" of Linux

Albert D. Cahalan (acahalan@cs.uml.edu)
Sat, 19 Dec 1998 21:11:21 -0500 (EST)


George Bonser writes:
> On Sat, 19 Dec 1998, Chip Salzenberg wrote:
>
>> Naming aside, my point is that open source software is always, well,
>> open -- open to be picked up and carried by new interested parties.
>> Even more, the GPL subset of the open source pool can't spawn private
>> forks; this is even stronger insurance.
>>
>> So I don't worry about corporate interference. Not yet, anyway.
>
> Yes, I think that according to the GPL, any change that IBM makes in
> existing code would have to be also GPL. Now if they write their own
> separate piece, say a module, I think they can use any license they want.
> OSS does not distribute the source for their commercial sound stuff, do
> they?
>
> QUESTION: Say IBM wrote their own mmap.c replacement with no GPL code in
> it. Can they distribute a binary kernel image made with that mmap.c
> without distributing the source to that small program? I mean, can that
> particular mmap.c have a non-gpl license? In other words, they distribute
> only the source for the GPL parts of the code?

That is not the problem at all.

The problem: if IBM supplies 24x7 support for Linux, they will want
to tweak the code to help their customers. No problem, and it falls
under the GPL of course. What if IBM uses an IBM patent? It would be
very bad (from their point of view) if that somehow gave everyone
rights to the patent.

IBM needs a way to tightly bind a patent license to the source code.
Without that, they are afraid to contribute. IBM might even want to
specifically tie patents to Linux. (not also gcc, HURD, emacs...)

It is unfortunate that the FSF doesn't hold any patents. I don't think
that they like patents at all, but they need some so that they can trade.
(if anything, patent issues could give GPL code an advantage over BSD code)

-
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/