Re: GPL'ed Patent Pool

On Fri, Sep 10, 2004 at 09:05:13PM +0100, Alan Cox was heard to remark:
> On Gwe, 2004-09-10 at 21:40, Linas Vepstas wrote:
> >    RedHat sort-of tried to do this, but its so weak and
> >    half-hearted that its all but useless.
> >
> It should be sufficient given Estoppel if not then someone needs to make
> concrete proposals to Mark Webbink.

OK, I had to read about 'promissory estoppel' on Wikipedia to understand
that RedHat can't just change its mind at a later date.  IANAL, so I
get to wondering if there are mechanisms for breaking promises, or if
I can indeed interpret the estoppel as lasting in perptuity.   

After re-reading that page several times, I got the sense that the 
'promise' does seem to be stronger than what I first sensed.  However,
it still seems weak in a number of ways: Its a promise that redhat
won't hurt me; I might have been happeier with an explicit license
that gave me rights.  IANAL, so I don't know of all the ways that 
such a 'promise' might be attacked, or might be rendered invalid.

The other problem is that it is explicitly interwoven with RedHat;
there is no easy way for another company/indivdual to make the same
kind of promise; I'd really like to see something more general purpose,
something re-usable by other companies or patent-holding individuals.

I'd also like to see something that I could use defensively: so,
for example: when Company A accuses 'gnucash' of infringing on 
thier patent 123, then I can respond, "ah ha, but gnucash implements
redhat patent 456, and company A does not have a license to patent
456".  That is, I'd like to defend myself with a counterclaim of damages
done to me; its not clear that RedHat's 'promise' would allow me 
to make counterclaims against company A.   

Traditional patent pools are used defensively, not offensively;
I'd like to see a defensive pool for GPL'ed software; I don't beleive
that RedHat's promise is enough for this purpose.  But IANAL, and
am on terribly shaky ground here, not understanding the principles
on which this might be grounded.  

I'm thinking of taking this disucssion to the OSI license mailing list,
see what happenes there.


pub  1024D/01045933 2001-02-01 Linas Vepstas (Labas!) <linas linas org>
PGP Key fingerprint = 8305 2521 6000 0B5E 8984  3F54 64A9 9A82 0104 5933

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]