Re: [rtcweb] Allegations of misconduct (was: Re: H.264 IPR disclosures (or persistent lack thereof))

[Harald Alvestrand <[e-mail]>]

On 12/17/2013 06:11 AM, Stephan Wenger wrote:
> Hi Harald,
>
> I have stopped responding to this thread (and the recent MTI codec
> discussions in general), because it veered into uninformed evangelism and
> advocacy territory long ago, and I¹m not interested at all to discuss at
> that level.  However, you are one of the people with some authority in
> interpreting the IETF IPR policy.  I can¹t ignore your email and let your
> unfounded allegations of misconduct (and that is what the end of your
> email, cited below, constitutes, despite its nicer wording) stand without
> response. 

Stephan, no allegiations of misconduct was intended. If I ever intend to
accuse you of misconduct, I assure you that you will have no doubt
whatsoever that that's what I'm doing!

At this time, I was not revisiting the areas where we disagree, and
where I wish IETF practical consensus were different than it seems to
be; I was merely addressing the *new* wrong idea that whether a draft is
"adopted by the WG" makes a difference to whether or not the rules say
that one has to disclose IPR.

With the number of ideas I disagree with floating around in the IPR
area, I have also stopped responding on people who repeat common
misunderstandings; the reason I chose to respond on this one was that it
was a new one.

But since we're here..... repeating the plot for the benefit of new
readers....

>
> Your reading of the IPR policy is not inline with mine.
>
> Note that this email has nothing to do with details such as WG draft
> status, and a lot with the broader issue of disclosure obligations against
> any IETF contribution.  I agree with you that WG status of an ID has
> nothing to do with the definition of IETF contribution.  That is why I
> changed the subject line.  However, I am in full compliance with the
> IETF¹s 
> IPR rules.  Let me explain my reasoning why I don¹t make an IETF IPR
> disclosure
> against IPR I am admittedly reasonably and personally aware of that
> pertains to VP8 and H.264, and my contributions to rtcweb.
>
> VP8: I don¹t have to disclose, as I have not Contributed to VP8.
>
> H.264: I don¹t have to disclose, because while I have contributed to
> H.264, included patented technology, I have not done so in an IETF
> Contribution.

Here's an area where we are known to disagree. I think that a
contribution that says technology X MUST be used by technology Y should
trigger a disclosure obligation on technology X for anyone who
contributes to the discussion on whether or not to adopt that statement,
as long as that information may contribute to letting the WG make an
informed decision.

But it's clear that I'm neither in the majority nor in conformance with
recent practice in the IETF on this viewpoint, so this is my version of
"And I also believe Carthago should be destroyed"

(see https://en.wikipedia.org/wiki/Carthago_delenda_est for the reference).
<https://en.wikipedia.org/wiki/Carthago_delenda_est>
>
> Non-normative video encoder tricks: I don¹t have to disclose, because
> regarding those few tricks I have disclosed in an IETF contribution (for
> example the loop filter things around H.261), I¹m not reasonably and
> personally aware of IPR that would be necessarily infringed and which any
> of my employers/sponsors/clients has the right to license.  I agree that
> if I were aware of such IPR, my level of Contribution (describing actual
> technology in technical terms, not merely pointing towards another
> document) would have triggered a disclosure obligation.
>
> Normative references to H.264 and VP8: I don¹t have to disclose, because
> the IETF IPR rules do not require me to disclose against merely
> normatively referenced specs (be they IETF or non-IETF specs), even if I
> contributed the normative reference in an IETF contribution.  In that
> regard, let me note three further points.
>
> (1) None of the IETF policy documents contain an explicit requirement for
> disclosure against normatively referenced specs.
I agree.
> (2) At least in the area of IPR encumbered media codec specs being
> normatively referenced by IETF documents (mandatory or optional), there is
> no history of IPR disclosures.  The majority of IETF documents that
> normatively reference media coding specs are RTP payload formats--an area
> I¹m very familiar with.  There is not a single IPR disclosure against the
> media codec technology itself that I¹m aware of in any disclosures related
> to payload formats.  That is despite the fact that the majority of payload
> formats are authored by people familiar with the coding technologies, and
> the IPR-ridden nature of most media codecs.  OTOH, there are tons of
> payload formats with no IPR disclosures against them, that normatively
> reference coding technologies that are well known as being encumbered.
> I¹m also not aware of IPR disclosures against IETF RFCs that normatively
> reference encumbered IETF specs, for example in the field of security.
> Clearly, if there were a doubt in the interpretation of the policy docs
> themselves, one would probably look at past interpretations.  Not that I
> personally see a need to consult past conduct, though.
This is an area where I think that it matters if it matters....

when a group decides to standardize (or review) an RTP payload format,
the group has already decided to encapsulate that encoding. It doesn't
matter for the choice of RTP payload format what the IPR on the
referenced technology is. Therefore, since disclosures won't make a
difference, and it's real work to make one, it's reasonable to say
"don't bother".

The RTCWEB case is different; we can choose whether to require the use
of the technology or not. Therefore, the information matters, and when
the information matters, it should be disclosed.

But as I said above - I don't believe the IETF can be said to agree with
me here. Carthago should be destroyed.


> (3) The requirement of disclosure against normative references is a
> heavily debated (and, in at least one case: litigated) subject outside of
> the IETF, with company positions at least entrenched as in the (relatively
> speaking: minor) question of an MTI video codec in rtcweb.  I do not
> expect a discussion of this subject on this list as likely to be
> particularly productive.  I mentioned the subject only to provide you with
> an insight into my rationale, not to trigger a broader discussion.  (If
> such a discussion were wanted ipr@ietf would probably be a better venue).

Agreed. The latest spate of messages on this subject in that list, which
occured in April 2013, didn't seem to me to reach a conclusion.
>
> Best regards,
> Stephan
>
>
>
>
>
> On 12.16.2013, 05:43 , "Harald Alvestrand" <[e-mail]> wrote:
>
>> On 12/16/2013 12:09 PM, Bjoern Hoehrmann wrote:
>>> * Harald Alvestrand wrote:
>>>> On 12/16/2013 01:50 AM, DRAGE, Keith (Keith) wrote:
>>>>> Of course if you really do want an IETF IPR disclosure under way, then
>>>>> submit a draft that makes both codecs mandatory and asks for it to be
>>>>> a
>>>>> WG item. The codec declarations will have to be made against that
>>>>> document. But do you really believe that will produce any surprise
>>>>> declarations.
>>>>>
>>>>> That draft WG document at the moment does not exist.
>>>> Keith, it is completely unclear to me why such a draft would cause more
>>>> disclosures to come forth, given that we have one draft proposing
>>>> declaring H.264 CBP mandatory that caused no IPR disclosures, and
>>>> another draft proposing declaring VP8 mandatory that, again, caused no
>>>> IPR disclosures.
>>> It looks to me as though the expectation is that once a document is made
>>> an official Working Group draft then others than those contributing the
>>> draft have to make disclosures, and since as far as I can tell the draft
>>> to make VP8 mandatory is not an official Working Group draft, making it
>>> one would then cause more disclosures to come forth, as you put it. That
>>> expectation might not be justified but the interpretation of the comment
>>> above seemed straightforward.
>> This may be an expectation that some people have, but it has no
>> grounding whatsoever in the IETF IPR disclosure rules, where the gating
>> factor is that it has been "contributed", and that others "realize" that
>> the IPR they know about applies to the contribution.
>>
>> RFC 3979 section 6.2.1 is pretty explicit. Nowhere in the document is
>> the concept of "working group draft" mentioned.
>>
>> It's been pointed out any number of times that this is what the rules say.
>>
>> I won't dictate anyone's actions - there are plenty of reasons (some
>> good, some bad) that people might conclude that the right action for
>> them is to act in ways that are not in accordance with the rules - but
>> the rules say what they say - and that does NOT mention "working group
>> draft adoption"; the rules apply to ANYTHING that is contributed.
>>
>>
>>
>>
>>
>>
>> _______________________________________________
>> rtcweb mailing list
>> [e-mail]
>> https://www.ietf.org/mailman/listinfo/rtcweb


-- 
Surveillance is pervasive. Go Dark.