ietf
[Top] [All Lists]

Re: IPR Disclosures for draft-ietf-xrblock-rtcp-xr-qoe

2013-09-17 10:56:14

From 29 years experience in ATIS, CCITT, CEPT, ETSI, IETF, ITU, TIA and
other standards organizations and extensive experience with standards that
do have associated IPR it is apparent that asking for confirmation at
multiple points in the standards development process IS necessary.

For example:

(a) The ITU requires that IPR holders make statements prior to the
publication of a standard.  A top 5 telecom equipment provider submitted a
proposal to add an a new capability to an existing (IPR free) standard and
did NOT state that they had IPR related to this - in fact the inventor on
their patent was the person who wrote and submitted the contribution. Some
years after the standard was published, when their patent was granted, they
started writing to implementers to demand that they take out a license.
Intentional abuse of the standards development process DOES happen.

(b) We often hear from large organizations that, as they have thousands of
patents, they can't possibly know whether they have patents related to a
standard or not.

Repeatedly asking questions about IPR disclosure IS important as it does
make it harder for IPR holders to claim that they did not know.

It should be noted that the duty to disclose IPR is NOT ONLY for the authors
of a draft, and the IETF "reminder" system seems to be focused solely on
authors. The duty to disclose IPR lies with any individual or company that
participates in the IETF not just authors.

Alan Clark


On 9/16/13 1:00 PM, "John C Klensin" <john-ietf(_at_)jck(_dot_)com> wrote:



--On Monday, September 16, 2013 19:35 +0700 Glen Zorn
<gwz(_at_)net-zen(_dot_)net> wrote:

... 
The wording of this question is not a choice. As WG chairs we
are required to answer the following question which is part
of the Shepherd write-up as per the instructions from the
IESG http://www.ietf.org/iesg/template/doc-writeup.txt:

(7) Has each author confirmed that any and all appropriate
IPR
disclosures required for full conformance with the provisions
of BCP 78
and BCP 79 have already been filed. If not, explain why.

We have no choice but to relay the question to the authors.

I see, just following orders.

For whatever it is worth, I think there is a rather different
problem here.  I also believe it is easily solved and that, if
it is not, we have a far deeper problem.

I believe the document writeup that the IESG posts at a given
time is simply a way of identifying the information the IESG
wants (or wants to be reassured about) and a template for a
convenient way to supply that information.  If that were not the
case:

 (i) We would expect RFC 4858 to be a BCP, not an
Informational document.
 (ii) The writeup template would need to represent
community consensus after IETF LC, not be something the
IESG put together and revises from time to time.
 (iii) The various experiments in alternative template
formats and shepherding theories would be improper or
invalid without community consensus, probably expressed
through formal "process experiment" authorizations of
the RFC 3933 species.

The first sentence of the writeup template, "As required by RFC
4858, this is the current template..." is technically invalid
because RFC 4858, as an Informational document, cannot _require_
anything of the standards process.  Fortunately, it does not say
"you are required to supply this information in this form" or
"you are required to ask precisely these questions", which would
be far worse.

From my point of view, an entirely reasonable response to the
comments above that start "As WG chairs we are required to
answer the following question..." and "We have no choice but to
relay..." is that you are required to do no such thing.  The
writeup template is guidance to the shepherd about information
and assurances the IESG wants to have readily available during
the review process, nothing more.   I also believe that any AD
who has become sufficiently impressed by his [1] power and the
authority of IETF-created procedures to insist on a WG chair's
asking a question and getting an answer in some particular form
has been on the IESG, or otherwise "in the leadership" much too
long [2].

In fairness to the IESG, "Has each author confirmed..." doesn't
require that the document shepherd or WG Chair ask the question
in any particular way.   Especially if I knew that some authors
might be uncomfortable being, in Glen's words, treated as
8-year-old children, I think I would ask the question in a form
similar to "since the I-Ds in which you were involved were
posted, have you had any thoughts or encountered any information
that would require filing of additional IPR disclosures?".
That question is a reminder that might be (and occasionally has
been) useful.  A negative answer to it would be fully as much
"confirming that any and all appropriate IPR disclosures..."
have been filed as one whose implications are closer to "were
you telling the truth when you posted that I-D".  I think Glen's
objections to the latter are entirely reasonable, but there is
no need to go there.

Finally, I think a pre-LC reminder is entirely appropriate,
especially for revised documents or older ones for which some of
the listed authors may no longer be active. I assume, or at
least hope, that concern is were this item in the writeup
template came from.   Especially for authors who fall into those
categories, asking whether they have been paying attention and
have kept IPR disclosures up to date with the evolving document
is, IMO, both reasonable and appropriate.  Personally, I'm
inclined to ask for an affirmative commitment about willingness
to participate actively in the AUTH48 signoff process at the
same time -- non-response to that one, IMO, justifies trimming
the author count and creating a Contributors section.

It seems to me that, in this particular case, too many people
are assuming a far more rigid process than actually exists or
can be justified by any IETF consensus procedure.  Let's just
stop that.

best,
   john

[1] pronoun chosen to reflect current IESG composition and with
the understanding that it might be part of the problem.

[2] Any WG with strong consensus about these issues and at least
20 active, nomcom-eligible participants knows what to do about
such a problem should it ever occur.  Right?