Re: Hardware licensing (fwd)

[Ian Carr-de Avelon <avelon@emit.pl>]

Pat wrote:
>I guess one thing I haven't really read about in this proposal
>is how you guarantee acceptance of your license terms.  Are
>you proposing that all OpenNDA source code repositories be
>protected with a click-through license agreement, or how do
>you plan to try to enforce that a valid contract has occurred?
>IBM gets real pen and ink signatures on their NDAs, but that's
>a major hassle for any kind of hobbyist, distributed group.
I have left off dealing with this until anyone thinks it may be worth 
a try. Implementation always has its own problems.
I think that as we move towards electronic signitures this can be
done quite painlessly, but if the legal advice was that paper signatures
are necessary, I don't think it is too big a problem. Remember they only
have to sign Part 1, which binds them to give hardware designs the same
status which the law gives programs through copyright. All the Part 2s,
saying that if you use any of this you have to be prepared to tell other
designers the whole story, or that they must never tell anyone about this, can
be different. I imagine that we would have a central registry with paper
signed Part 1s and those signers would get some form of electronic
signature which would give them access to our repositry. Big companies
would see that they could let people who had signed our Part 1 into 
their own repositries, without needing their own paper, as long as we let 
them verify the signature with us (maybe they might put some funds in?).
Sudenly it is not just Mr Smith who will sign, but say Intel, because 
the whole of IBMs NDA archive becomes an extranet for them. Essentially
the problem which we, as a group wanting Open Source Hardware, have is the
same problem the whole industry has. If we step back from the impression
that we are different because we want control of our designs to do 
different things, and realise that we are united in wanting that control,
we have some very powerful allies. Once access to so much information
is governed by the signitures we issue we probably don't even need the 
courts. If say Conectix rip off someone elses design (big guy, little
guy, it does not matter) we can revoke their signature and they become
the only company which does not have access to all the designs in the
world. The community can govern itself because not being in it is sufficient
punishment to act as a deterent.
  

>Even an NDA from IBM, as discussed above,
>will do IBM no good if someone can show that they legally
>received the information without agreeing to the NDA,
I hope that this can be partly dealt with by the fact that
most major players will have signed the Part 1 at one time 
or another, an that that can apply to other designs which are
released under an NDA but then find themselves on an open 
access site.

>Another potential issue with the OpenNDA is that, since you
>haven't really disclosed anything to the public, there is
>not necessarily anything keeping someone else from (either
>legitimately, or illegitimately, looking at your sources)
>discovering the same innovations.  If a patentable invention
>happens to be among those innovations, you may lose the ability
>to use that particular innovation in your hardware design,
>and you don't even have the defense that you have published
>prior art, because it was only licensed to those agreeing
>to the OpenNDA, not truly published.
But only the inventor can patent (this may be different in some
contries) so the unpublished prior art is still prior art and
the central repositry can act as a date stamp. This is partly
where we really started, we just then hit the problem that
puting your design in the repositry may mean that you have given
it away to everyone who works closed source as well.
Yours
Ian