Hi John,
please note that the vast majority of IETF IPR disclosures promise patent
licenses only on technologies described in an I-D in the event that the I-D
becomes an IETF "standard". This avoids the problems you mentioned nicely,
I think, at the expense of some uncertainty when people implement I-Ds that
do not become standards, and some uncertainly during the prototyping/early
deployment phase---which, so far, most if us can comfortably live with.
In short: The licensing promise kicks in only once an I-D becomes an "IETF
standard" (interpreted by most of us as an RFC is published). In any other
stage there is no licensing promise. Some statements are more carefully
worded about the subtleties of "standard" then others... but most seem to
have the same intention.
Enough (hardly qualified) people mix up copyright related issued with patent
license promises already---I don't think I need to mention names. Let's not
fall into this trap.
Regards,
Stephan
On 8/14/08 8:33 AM, "John C Klensin" <john-ietf(_at_)jck(_dot_)com> wrote:
--On Thursday, August 14, 2008 11:15 AM -0400 Powers
Chuck-RXCP20 <Chuck(_dot_)Powers(_at_)motorola(_dot_)com> wrote:
I think that Stephan raised some very good points as to why
allowing some IPR disclosures to be removed actually makes
sense. Since quite often IPR disclosures are made for a
specific ID in a specific working group, if that WG ultimately
does not choose that technology (and the ID expires), I am
curious as to what the value would be of keeping that IPR
disclosure on file forever? If narrowly worded (as many are),
it would not be applicable to any other ID submission or
working group, and would therefore have little use but to add
to the growing list of disclosures in the IETF IPR database.
I would be curious to hear the reasoning for keeping these on
file, apart from 'historical record', since I am not convinced
the IETF IPR database is the right place to hold onto IPR
disclosures simply for historical purposes that only apply to
technology that will never see the light of day in an IETF
standard, since the IETF doesn't see any value in keeping the
IDs that they applied to in the first place.
Chuck,
As a long-term advocate of taking the provisions that I-Ds
expire after six months --at least to the extent of having _all_
rights in them revert to the author(s)-- I think what you are
saying above is profoundly sensible.
However, the IPR WG, in its wisdom, has concluded, in some
phrasing that was changed fairly late in the game, that the IETF
Trust should get enough rights in I-Ds to authorize all sorts of
subsequent uses of them and their content, with no time limit.
That phrasing passed through IETF Last Call and IESG signoff and
is the context in which the Trust is now writing rules. It
seems to me that, if the IETF (through the Trust) is going to be
in a position to grant rights to use material in I-Ds forever,
and if rights to use code in I-Ds (even for the first time)
don't expire after six months or some other closed period, then,
logically, we are obligated to keep the IPR disclosures forever.
I suppose that, if I were a paranoid lawyer (and IANAL, even if
I'm paranoid about these sorts of things) and giving advice to a
participant in the IETF, I'd recommend that an IPR disclosure be
filed on every single I-D, indicating that any licenses I might
grant were good for only six months after the posting of the
last I-D in the relevant series unless it were approved for RFC
publication. What that would do to the system we seem to be
making for ourselves would be interesting, at least.
john
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf