Spencer,
Let me offer one small variation, extracted from comments I've
made in an off-list discussion. I promised myself I wasn't
going to write this note, but circumstances change...
--On Friday, August 15, 2008 10:41 AM -0500 Spencer Dawkins
<spencer@xxxxxxxxxxxxxxxxx> wrote:
I'm trying to summarize in a way that might someday be
translated into IETF process, and not Yet Another Great
Discussion On Ietf@xxxxxxxx That Doesn't Result in Change...
I agree with John's suggestion that IPR disclosures be handled
as "moderated postings", and I assume that we could actually
come up with a short canonical list of things to look for.
This would help with cruft on the front end.
We might also want to distinguish between IPR disclosures for
CURRENT I-Ds and the complete (non-cruft) list of IPR
disclosures on any I-Ds that have ever been submitted. This
would help with cruft on the back end, and make it easier for
working groups to consider IPR disclosures relevant to
technologies included in their drafts, which I understand to
be the primary reason why we ask for disclosures in the first
place.
Here is where I've disagreed with the IPR WG (or at least the
interpretation of consensus and documents from that WG), and
this is exactly one of the reasons.
Whether we make the documents readily available or somehow try
to un-publish them or otherwise pretend that they don't exist, I
think we would be much better off taking the "expire in six
months" statement seriously wrt I-Ds and IPR. I've been
reminded that a large fraction of the serious IPR postings that
contain releases, licence grants, or license guarantees do so on
an "if the technology is standardized" or equivalent basis.
Those statements are worthless wrt I-Ds; anyone implementing
technology and shipping product from an I-D that might be
covered by the disclosure is assuming all of the risks of having
to withdraw a product and/or quickly negotiate a license under
unfavorable conditions should the IETF decide to not standardize
that particular technology.
That situation exists regardless of what the IETF does. Even
insisting on complete and open implementation rights (not even
RAND), as has periodically been proposed, wouldn't help if
licenses are granted only for implementation, deployment, and
use of standards. So there is no point wasting time discussing
it, and it is not my intention to do so here.
But it seems to me that we can keep the IETF out of all sorts of
messes by taking the position that, when an I-D reaches that
six-month limit, any rights granted by the author other than for
IETF use in developing standards revert to the author. From a
patent standpoint, that means that the old disclosures are the
old disclosures, even if they state licensing terms that are
guaranteed to never be meaningful, but we don't accept new
disclosures because anyone doing an implementation against an
expired I-D should be clearly on their own as far as the IETF is
concerned.
Even from a copyright standpoint, if someone, for uses other
than IETF standards development, pulls text or code out of an
expired I-D that has not progressed into RFC publication, let
the acknowledgments (if any) be to the authors, not the IETF
(which effectively decided it wasn't interested in the document
by letting it expire and may have made no contribution to it at
all). Anything that was done with IETF permission during the
time the draft was active stays done and valid, of course. But
things done post-expiration, and disputes about whether or not
something had expired when the text was taken and used, are
between the authors and the entities doing the copying (and
their respective lawyers, if appropriate). It is, IMO,
undesirable and inappropriate for the IETF [Trust] to become
involved in any of those issues for expired drafts -- it isn't
necessary for the standards process and it isn't necessary for
anything else that is the IETF's business or problem.
Just my opinion, of course.
john
_______________________________________________
Ietf@xxxxxxxx
https://www.ietf.org/mailman/listinfo/ietf