[Ksummit-discuss] [TECH TOPIC] Firmware signing

[James Bottomley]

On Thu, 2015-07-30 at 09:08 +0100, David Woodhouse wrote:
> On Wed, 2015-07-29 at 16:39 -0700, James Bottomley wrote:
> > On Wed, 2015-07-29 at 18:32 +0100, David Woodhouse wrote:
> > > ... as well as contradicting its explicit statement that its intention
> > > is to "control the distribution of derivative OR COLLECTIVE WORKS".
> > 
> > That quote loses the last important piece; that paragraph, which I'll
> > quote in full
> > 
> >         Thus, it is not the intent of this section to claim rights or
> >         contest your rights to work written entirely by you; rather, the
> >         intent is to exercise the right to control the distribution of
> >         derivative or collective works based on the Program.
> > 
> > says that no additional rights over pieces that were written by another
> > (in this case a firmware provider) and are not based on the Program are
> > claimed.  That rather supports the idea that the extent of the license
> > attachment is limited to derivation.
> 
> The more conventional interpretation of that paragraph — the
> interpretation which *doesn't* require us to believe that the GPL
> wasted all those words *explicitly* talking about things which are "not
> derived from the Program, and can be reasonably considered independent
> and separate works in themselves", only to contradict itself and say
> "haha just joking; *ALL* aggregation is fine" — is that this is the GPL
> clarifying how it operates within the constraints of copyright law.
> 
> There is a common — though bogus — complaint about the 'infectious'
> nature of the GPL, which goes along the lines of "how can they require
> that I publish the source code to the bits that *I* wrote. Copyright
> law gives them no rights over *my* code".
> 
> The paragraph you cite above is more reasonably interpreted as a
> clarification that the GPL isn't claiming any *rights* over your own
> separate works, and that it merely operates by withholding permission
> to use the *GPL'd* part in that context.
> 
> But again, I'm not requiring that you publicly accept this point of
> view. There is plenty of scope for debate, and it's not impossible that
> a court *could* uphold your interpretation and effectively just delete
> the paragraphs of the GPL that you don't like.
> 
> All I'm asking is that you stop making the bogus claim that yours is
> the *only* possible interpretation. It isn't even the sanest one.

OK, let us suppose for the sake of argument that this is correct and the
GPL does manage to get extended to non derived included projects.  Even
in that case, we're not causing any corporate legal jeopardy because of
the principle of estoppel.  Estoppel says we cannot accuse someone of
breaching our licence for something we also did.  So if we ship the
firmware with the kernel, anyone else also shipping firmware with the
kernel is automatically innoculated against accusations of license
breach for that action.

James

-------------- next part --------------
A non-text attachment was scrubbed...
Name: smime.p7s
Type: application/x-pkcs7-signature
Size: 5819 bytes
Desc: not available
URL: <http://lists.linuxfoundation.org/pipermail/ksummit-discuss/attachments/20150730/7c474bc7/attachment-0001.bin>