Re: [rtcweb] Counting NOs (Re: Straw Poll on Nokia mincing)

[Mohammed Raad <[e-mail]>]

On Sat, Dec 21, 2013 at 4:17 AM, David Singer <[e-mail]> wrote:

> Responding to various people
>
[MR] OK, very convenient.

>
> >>> What members of this WG have been saying, and writing, is that they
> want interoperability. OK, so that can be achieved with at least one of the
> options in the straw poll that VP8 proponents have indicated is acceptable.
> It would seem logical and reasonable to expect AVC proponents to take up
> that option also.
>
> I am unclear as to which option you are talking about.
>
[MR] I was mostly thinking of option 3. (entities must implement both H.264
and VP8).

>
> >>> I think the biggest problem is that people only want to maintain code
> for one codec.
>
> I think it’s a preference, but not a total bar.  I think this is true of
> any organization: you want to use engineer-hours wisely.
>
> >>>  If you already have H.264 support implemented, you will
> >>> oppose every other codec, no matter how big the advantages.
> >>> Therefore, a compromise is not possible with such a position.
>
> Let’s see, I have suggested we look more closely at H.263, H.261, Motion
> JPEG, to name only 3.  I also have been trying to get an RF status for
> H.264 baseline (in several ways).  Cisco has tried to soften the practical
> position as well.  I don’t think it’s fair to describe this as a
> dog-in-manger attitude.
>
> Regrettably, there is an (unconditional) type-2 (royalty-bearing)
> declarations for H.264 baseline from a company purporting to support RF
> coding, as well. How helpful is that? It hardly swings the tide in a
> helpful direction.
>
> >>> > It has been pointed out that the MTI should satisfy options a) or b)
> on the
> >>> > ietf declaration form. As matters stand AVC/h264 does not satisfy
> that, this
> >>> > has been made very clear through the ISO process.
>
> But at least it satisfies (c).

[MR] you had written in a previous post that the WG is looking for options
(a) and (b), and thus the reference to that comment here - very unsporting
to simply shift the goal posts.

>


> >>> On the other hand, VP8 has
> >>> > so far withstood court challenges, as you point out
>
> I am unaware of a result from Germany still (it was expected in JUNE,
> wasn’t it?). And I believe that we have been told that those cases only
> involve a subset of what what was declared.
> And the formal declaration stands, something which H.264 at least does not
> have (in any body). Generally standards bodies — and especially the
> engineering sides — have to go by the declarations.  Occasionally (e.g. at
> the W3C) a formal process is run to analyze and handle them.
>
[MR] H.264 has this type of thing
http://news.priorsmart.com/fastvdo-v-apple-l4q2/  , which is not the same
as formal patent declarations, I agree.

>
> Oblig IANAL.
>
[MR] We know.

>
> >>> - something which AVC
> >>> > could not BTW
>
> You’re saying you are aware of IPR declarations against H.264 that are
> unlicensable and would stand up in court?  Pray tell.
>
[MR] not sure how the previously linked case has concluded, but the judges
comments (available at
http://www.plainsite.org/flashlight/case.html?id=1724130  along with other
details) seem to indicate a number of agreements with what was being
claimed. I am also not a lawyer, but I would conclude that this is an
example of one case where there are still question marks about how easy it
is to get a license that covers all of the different parts of AVC/h.264. I
suspect this was concluded by some kind of license agreement, although I
have not been able to find any publicly available information on this.

>
> >>> > It would seem very clear that the codec that is closer to the
> royalty free
> >>> > ideal situation is VP8.
>
> VP8 is formally *unlicensable* at the IETF.  Ignoring this does not
> increase credibility of your arguments.
>
[MR] You see, that nice long list of patents you so avidly keep on going
back to is yet to be shown to be relevant to anything related to VP8 (you
mention that a few lines before this). The "formal declaration" argument
would hold some weight if the company making that declaration was not in
court getting negative results in its attempt to enforce what is in that
declaration. Copying and pasting patent numbers into an online form does
not mean those patents are relevant. But I do believe that I have found a
public statement from Nokia saying that they will not license patents that
read on VP8 at
http://www.fosspatents.com/2013/03/nokia-comments-on-vp8-patent.html which
is interesting given that they have just confirmed the licensing of their
patents to MS, so you would expect to have at least one case where the
claimed to be relevant patents are licensed.

[MR] going back to my "convenient" comment at the top - cutting different
points from different posts by different people really does build an easier
argument to deal with. Since you are concerned about credible arguments, I
would suggest not to do that.

>
>
> David Singer
> Multimedia and Software Standards, Apple Inc.
>
> _______________________________________________
> rtcweb mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/rtcweb
>



-- 
Mohammed Raad, PhD.
Partner
RAADTECH CONSULTING
P.O. Box 113
Warrawong
NSW 2502 Australia
Phone: [phone]
Email: [e-mail]