Re: Binary-only firmware covered by the GPL?
I wrote:
> Assuming IBM is the
> original licensor, we are left to infer that the preferred form of
> the firmware for making modifications is the binary as it is shipped.
Nathanael Nerode wrote:
> Right -- provided that they aren't, uh, "lying" about it.
I don't see why IBM has to reveal anything about where the binaries
came from. IBM publishes the binaries under the GPL and thereby
tells others that they can distribute derived works under the GPL
provided they include "the" preferred form for modification. What
'the preferred form for modification' means has always been one of
the difficult questions about the GPL, not least because everyone
has his or her own preference. But in the present context I don't
see how it can include any form that IBM possesses but doesn't
publish. IBM can't reasonably require a licensee to furnish something
if that thing belongs to IBM and IBM chooses not to provide it or
even tell anyone that it exists.
Conclusion: Either IBM doesn't require that, or IBM is being silly
in granting permission to distribute under conditions that it causes
to be unfulfillable.
N.N. continued:
> Given how incredibly unlikely it is that the binary really is the preferred
> form of the people at IBM, and given IBM's reasonable behavior in the past
> regarding open source projects, someone should ask them about it, and if
> they say it isn't really their preferred form, point out the licensing
> problem.
>
> If
> (1) the assembler source code is the preferred form for modification
> (2) IBM doesn't make it available with the binary
> Then the GPL doesn't grant *any* rights to distribute the binary, to anyone,
> period. :-P So effectively, IBM would not have granted any right to
> redistribute the firmware!
That is the "IBM is being silly" interpretation. I don't think that IBM
is being silly. I think IBM doesn't intend 'preferred form for modification'
to include forms of the code that it doesn't publish.
BTW I have written to IBM a couple of times asking for clarification but
I have never received any reply.
For the reasons given above, though, I don't think it is essential to
get a clarification.
I understand the line of reasoning which leads to the conclusion that IBM
has published the mwave firmware under a license that in fact forbids
redistribution. But I hope you will still grant that the clear intent
of IBM was to allow free distribution of the firmware binary. It would
be totally absurd of IBM to publish the firmware on its website, to
declare it to be licensed under the GPL, and then to turn around and
say that IBM interprets 'preferred form for modification' in such a way
that redistribution is prohibited. If we are right in regarding that as
absurd then we are right in treating IBM's license as a license to
redistribute.
My main qualm right now is that I know absurdity is in the eye of the beholder.
--
Thomas Hood
________________________________________________________________________
Yahoo! Messenger - Communicate instantly..."Ping"
your friends today! Download Messenger Now
http://uk.messenger.yahoo.com/download/index.html
Reply to: