Re: Workshop notes and next steps

On Feb 21, 2011, at 7:01 PM, Glenn Adams wrote:

Let's just say I find the claim "if DASH is not RF then it will not be successful, since there are RF alternatives... that noone is ever going to make any money out of royalties on this thing" to be both naive and idealistic.

Well, it's interesting that you think there is a business case in licensing DASH IPR for those that have it. I think it very unlikely that Netflix would ever pay royalties for something like that which we know we can easily work around. I don't think that's naive or idealistic. And it applies to all companies that might use this technology. That doesn't mean noone will try and, to be clear, I cannot speak for other companies and am not making firm predictions. I just think it is well within the realms of possibility that the companies that have invested in this will choose to take the benefits of a widely adopted standard that they were all mainly aiming for in preference to chasing a non-existent business case for royalties.

You also have to understand the technology involved: DASH defines *formats* for media and manifest files and does not define any client procedures. It is difficult to see what could possibly be *essential* IPR in such a format at all, and more difficult to see what there could possibly be that cannot be worked around or isn't essentially prior art, especially given the already stated RF positions of companies that have pioneered this technology. Again, I'm not making definitive statements and the situation is unlikely to be crystal clear soon, but this is not a video codec we are talking about here.

3G-DASH was not developed with the express intent of being RF, any more than MPEG-2, MPEG-4, etc. were.

If, however, you can persuade all of the participants in the 3G-DASH development process, through whatever means, to retroactively agree to RF terms (of a sort that satisfies, e.g., the current W3C patent policy), then go ahead. I will remain skeptical however, and will question whether using the W3C as a vehicle for that purpose is appropriate, particularly if there is no definite plan for a W3C specification to normatively reference the work.


I hope that we would all like to see a common adaptive streaming solution supported in HTML5 products. Without that HTML5 can never displace Flash (IMHO). I doubt there could be a "definite plan for a W3C specification to normatively reference the work" in advance of some more information on the IP issue, so putting that first would create a circular dependency.

...Mark


G.

On Mon, Feb 21, 2011 at 7:23 PM, Mark Watson <watsonm@netflix.com<mailto:watsonm@netflix.com>> wrote:
Glenn,

The 3GPP and MPEG patent policies require organizations to declare essential IPR and indicate the terms upon which they are prepared to license it, either RF (aka "Option 1" in MPEG), RAND or not at all. There is no up-front assumption what the terms will be.

It's clear to me, and based on my conversations to many others involved as well, that if DASH is not RF then it will not be successful, since there are RF alternatives. So everyone should be clear that noone is ever going to make any money out of royalties on this thing.

Of course it's possible that some company will choose to kill the technology they have just spent 2+ years creating, but that is by no means a certainty. It's even likely that some organizations who contributed peripheral technologies will try to claim RAND terms for them, and then we would unfortunately be into the "RF profile" game.

Having said that, the message above needs to be communicated clearly within companies and for that people need some external evidence that this is indeed the case (that DASH will not see widespead adoption if it is not RF). And that is the point of this thread and the letters to 3GPP and MPEG members.

It's not necessarily a goal to have DASH written into the HTML specification, although there are extensions to HTML5 to enable adaptive streaming whose design might be informed by the need to work with DASH specifically. The goal is to have a single adaptive streaming standard that a range of communities can rally around - where and how things might be documented we can discuss.

...Mark


On Feb 21, 2011, at 2:51 PM, Glenn Adams wrote:

3GP-DASH was developed in an organization in a context where RF policy did not apply, but where RAND policy applied. It seems a little naive to think that one could retroactively ask for a RF policy to be applied. Any participant may just say NO if asked for a disclosure if none were required by the rules of the originating organization, and would be under no obligation even to respond to such a request.

You either develop a technology with RF terms or with RAND terms. You can't change the former into the latter merely by asking the participants for disclosures.

In any case, it is extremely unlikely that HTML5 would reference 3GP-DASH, either normatively or informatively. The HTML WG has already written off the idea of referencing any media type. So why would they not treat media streaming protocols known to be developed under non-RF terms any differently?

I just don't get the point of this thread. The outcome is already known.

Regards,
Glenn

On Mon, Feb 21, 2011 at 3:25 PM, Mark Watson <watsonm@netflix.com<mailto:watsonm@netflix.com>> wrote:
Kazuyuki, all,

In terms of continuing the discussion, the issues I had with the letter as drafted were as follows:

(1) RF "Profiles": at the workshop there was quite strong support (I thought) for asking for DASH to be Royalty Free and considering RF profiles only as a fallback. The issue I see is that as soon as you ask for an "RF profile" you admit defeat on the objective of having the whole thing RF. Secondly, "RF profile" implies a possibly lengthy process of profile definition. Finally, does an "RF Profile" really meet the needs of the W3C membership ? What happens when some new feature not included in the RF profile is needed in W3C - do we have to go through this process again ?

(2) I see the letter is addressed to "3GPP members" but I think we should be more explicit in the text that we understand that the request is directed at individual 3GPP/MPEG member companies.

(3) It is important to be precise about what is meant by "royalty free" - people are trying to get some traction with lawyers within their companies with this and lawyers tend not to answer unclear questions (well, except with more questions). In the case if MPEG, I think we are asking for member companies to make IPR declarations as soon as possible and encouraging them to indicate "Option 1" terms or something equivalent.

(4) It would be good to name a W3C legal contact in the letter - there may be companies who would like to discuss with W3C exactly what kind of Royalty-Free promises might be acceptable under the W3C policy.

Thanks!

...Mark



On Feb 19, 2011, at 6:21 AM, Kazuyuki Ashimura wrote:

> Hi Mark,
>
> It was very nice to meet you in Berlin during the workshop, and we are
> very sorry to make you confused and surprised.  Please let us add some
> clarification about the situation.
>
> Actually, we originally thought we should have put the summary and the
> minutes of the workshop together first and clarify what was discussed
> during the workshop so that we can start discussion about what we should
> do for the next steps.  We thought so, because we should get consensus
> within the whole Interest Group if we really send out an email as an
> official message from the group to a specific external group like 3GPP.
>
> However, unfortunately, there was a timing issue that there would be a
> 3GPP meeting planned this week from Thursday, Feb. 17th, and we needed
> to send some initial message to start discussion with them right away.
>
> As you know, there was a strong request for an RF profile of DASH for
> the Web supported by many of the Berlin workshop attendees.  So we (=4
> IG co-Chairs and 2 W3C staff contacts) talked with each other and the
> conclusion was that it would make sense for us to send a simple initial
> message to 3GPP, like:
> [[
> We in the W3C Web and TV IG need an RF profile of DASH for the Web.
> So we request to work together with you to prioritize the profiles
> that should be RF.
> ]]
>
> We thought a short statement like the above should be comparatively
> less controversial and have broad support by the other IG participants.
>
> We are very sorry we could not explain the difficult situation and why
> we were in rush beforehand.  We think we should ask all the IG
> participants, esp. official W3C Member participants (including you),
> for opinions about how to continue the collaboration with other
> standardization organizations.  So we would like to continue the
> discussion on this list.
>
> Regards,
>
> Kazuyuki Ashimura
> for the Web and TV Interest Group Chairs and staff contacts :)
>
>
> On 02/17/2011 01:23 AM, Mark Watson wrote:
>> Francois,
>>
>> That's a surprise - I thought the letter was to come from the IG, not the co-chairs. It was mentioned several times in the meeting that it would come from the IG, so I saunas we would get to review and agree it first. (Kaz did not allude to you sending it, just to you preparing it).
>>
>> There are several areas where the letter does not align with what was discussed in the workshop.
>>
>> I hope we can discuss the letter to MPEG and send it from the whole IG.
>>
>> I think we should do that asap. As I understand there are internal processes within some companies that will benefit from this letter, so we should not wait on MPEG meeting dates.
>>
>> ...Mark
>>
>>
>>
>> Sent from my iPhone
>>
>> On Feb 15, 2011, at 11:47 PM, "Francois Daoust"<fd@w3.org<mailto:fd@w3.org>>  wrote:
>>
>>> Hi Mark,
>>>
>>> On 02/14/2011 08:46 PM, Mark Watson wrote:
>>>> Also, what are the logistics for the next steps in the IG ? Particularly the letter from the IG to MPEG and 3GPP. Is someone working on a draft ?
>>>
>>> As Kaz alluded to, we've prepared and sent a short letter to 3GPP members as discussed during the workhop, copied below. There is more time for MPEG as their next meeting is in March, IIRC, so this second letter may (and should!) be addressed by the IG.
>>>
>>>
>>> -----
>>>
>>> Dear 3GPP Members,
>>>
>>> The W3C has recently launched a Web and TV Interest Group, set to identify requirements and potential solutions to ensure that the Web will function well with TV.
>>>
>>> During the second W3C Web and TV workshop, held in Berlin on 8-9 February 2011, it came to our attention that many participants of the workshop are interested in adopting DASH as a possible adaptive streaming mechanism on the Web. As a result, the organizing chairs of the workshop have asked the co-chairs of the Interest Group on Web and TV to convey this interest from workshop participants to 3GPP members, and to inquire about the licensing status of DASH. W3C has a strict royalty-free patent policy on the technologies that get adopted as core Web technologies.
>>>
>>> We would very much appreciate if you could inform us of the current status of the specification of DASH and the possibility to work together on a royalty free profile of the specification for potential integration of DASH as adaptive streaming mechanism for audio and video in HTML.
>>>
>>> 3GPP members willing to discuss the topic with participants of the Web and TV Interest Group may use the public public-web-and-tv@w3.org<mailto:public-web-and-tv@w3.org> mailing-list, whose archives are publicly available at:
>>>  http://lists.w3.org/Archives/Public/public-web-and-tv/
>>>
>>> Companies and individuals willing to know more about W3C and W3C process may get in touch with Francois Daoust<fd@w3.org<mailto:fd@w3.org>>  and Kazuyuki Ashimura<kaz@w3.org<mailto:kaz@w3.org>>, W3C staff contacts for the Web and TV Interest Group.
>>>
>>> Best regards,
>>>
>>> Francois Daoust,
>>> On behalf of the W3C Web and TV Interest Group chairs.
>>> -----
>>>
>>
>>
>
> --
> Kaz Ashimura <ashimura@w3.org<mailto:ashimura@w3.org>> http://www.w3.org/People/all#ashimura
> Tel:                                                 +81 466 49 1170
>