Patent grant no good [LWN.net]

Patent grant no good

Posted May 26, 2005 11:43 UTC (Thu) by MathFox (guest, #6104)
In reply to: Patent grant no good by mjr
Parent article: A toy and a promise from Nokia

I agree with you that the press release is mostly a publicity stunt. When Nokia starts distributing Linux, they have to abide the GPL. The GPL forbids to add restrictions on use and distribution of the software, so Nokia effectively gives up the right to enforce their patents on the Linux (and Gnome) code the moment they start distributing their toy.

Third, even if some patented thing in Linux was covered by this license, its use is only authorized in the Linux kernel as published on kernel.org (not even vendor- or self-patched versions), which is pretty useless.

By GPL, the non-enforcement of patents carries over to anyone receiving its code via Nokia (or kernel.org). So the only place where problems could crop up is in the patches that a distributor applies.

Anyway, I think that it is good to applaud this as a good first step from Nokia and mention that there could be similar patent issues with Gnome.

(Log in to post comments)

Patent grant no good

Posted May 26, 2005 11:55 UTC (Thu) by mjr (subscriber, #6979) [Link]

When Nokia starts distributing Linux, they have to abide the GPL. The GPL forbids to add restrictions on use and distribution of the software, so Nokia effectively gives up the right to enforce their patents on the Linux (and Gnome) code the moment they start distributing their toy.

This is true as such, but there are caveats: The device likely doesn't include all of the kernel code commonly used on workstations and servers, and the Gnome framework is a stripped-down version too. Later versions of either aren't covered at all for new infringements. And then there's the fact that they have an army of lawyers to pull an SCO and go "oh, there was that code in there, we didn't intend to license that freely around; we'll just rework the code for our next version, that's fair, innit?" I'd be wary of these implicit patent grants, especially now that they've gone on record with an explicit but limited grant that they can point to and go "look, this is what we actually meant".

IANAL. I don't know that the implicit grant (with the caveats mentioned above) isn't solid, I just don't count on it, and wouldn't recommend it to others either. Professional opinions are welcome.

Patent grant no good

Posted May 26, 2005 14:48 UTC (Thu) by MathFox (guest, #6104) [Link]

I agree with you that the implicit patent licence by GPL only applies to the GPL code that Nokia distributes. But it also covers the "we didn't intend to license that" code; reworking allready distributed code is like bolting the stable door when the horse is gone.
I don't like the "revocable" part in their license either, but I think that in practice revocation will not happen. It now is in Nokia's interest to keep good relations with (at least) the Linux kernel and Gnome communities. Creating any Open Source patent trouble will cloud those relations.

IANAL either, but lawyers told me that the GPL implicit patent license is pretty solid.

Patent grant no good

Posted May 26, 2005 16:13 UTC (Thu) by mjr (subscriber, #6979) [Link]

But it also covers the "we didn't intend to license that" code; reworking allready distributed code is like bolting the stable door when the horse is gone.

Mainly I was concerned that they might be able to claim that they didn't know about some bit of code being in the kernel while distributing it and wouldn't have done it had they known, ie. pull an SCO. Not that SCO's doing a good job at pulling an SCO...

IANAL either, but lawyers told me that the GPL implicit patent license is pretty solid.

This is a nice tidbit, and alleviates my concerns a bit (as far as they relate to code being distributed by patent holders). Thanks for sharing. May I enquire in which country this was, and did they comment on how things are likely to be internationally (and have they got expertise on that)?

Patent grant no good

Posted May 27, 2005 15:35 UTC (Fri) by giraffedata (subscriber, #1954) [Link]

There is no implied patent license. I've seen the FSF's explanation of the patent ramifications of GPL, and it consists mainly of Section 0. But Section 0 is a scope clause. It says what things the license (along with its conditions) does and doesn't restrict. One thing it doesn't restrict is how you can use the program. That's a far cry from affirmatively giving you the right to use the program unrestricted. The clause is, "Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, ..."

FSF also has a complicated explanation of how a not-freely-licensed patent might conflict with GPL by requiring the licensee to place restrictions on modifying the program. I agree that a patent license might conflict with the GPL. But others do not, and no license at all definitely does not.

For the sake of the next argument, lets imagine that there's another section in GPL that says, "I permit you to use the program in any way you please."

That still wouldn't effect an implied patent license from redistributors. Nokia could (in various ways) nullify that clause in its redistribution. The only effect would be that Nokia would be redistributing without a license, and that means Nokia would owe damages, including royalties, to the copyright owners. That's not the same thing as the recipient of the code from Nokia having a legal right to use Nokia's patents. The recipient still owes Nokia for patent infringement.