Re: [rtcweb] H.264's high-low play (Was: H.264 IPR disclosures (or persistent lack thereof))

[Basil Mohamed Gohar <[e-mail]>]

On 12/14/2013 04:29 PM, Eric Rescorla wrote:
> On Sat, Dec 14, 2013 at 1:16 PM, cowwoc <[e-mail]> wrote:
>> On 14/12/2013 3:00 PM, Silvia Pfeiffer wrote:
>>>
>>> While I'm all for vp8, this argument is a far shot. I actually think that
>>> the longer we wait, the easier it will be to pick vp8.
>>>
>>> Did you notice that the only objection to choosing vp8 that returns in the
>>> survey is the Nokia ipr statement? There is no mention any more of lack of
>>> hw support? When Google makes the binary offer that Cisco made for h264,
>>> that goes away. I wonder what objections will remain??
>>>
>>> Time is actually on the side of royalty free in this case faict.
>>>
>>
>> I think there is a bit of a fallacy at play here. H.264's installed base in
>> the context of WebRTC is exactly zero. Why? Because there is no meaningful
>> encoder support on any platform.
> 
> "Installed base" here also means the very large number of
> existing devices which speak SIP and which have H.264-only.
> 
> With that said, there are also a large number of devices with
> built-in H.264 encoders.
> 
> 
>> IPR-issues aside, does anyone honestly think that we'd be better off with a
>> video codec monopoly? I'd much rather see VP8 and H.264 compete on equal
>> footing. I believe that one of Mozilla's primary goals for entering the
>> mobile market was to create competition in that space, which is why I hope
>> that they will push for competition in this space as well.
> 
> I'm not sure what you think "Mozilla" should do other than what it's doing,
> which is to support both codecs.
> 
> -Ekr

Ekr,

Can you provide something that points to this:

> "Installed base" here also means the very large number of
> existing devices which speak SIP and which have H.264-only.

as a fact?  I'm particularly interested in the "very large number" part
more than the H.264-only part.  How many SIP devices are really out
there that have any a passing interest or chance at being an interop
target for rtcweb?

The reason I ask is because amongst the most pressing strengths of H.264
as MTI for rtcweb is interop, but interop is not a primary goal of
rtcweb.  Introducing the burden of H.264 into the spec for new devices,
technologies, and methods should not be a factor if the true relevance
is minimal or, in fact, small enough to be considered non-existent.

-- 
Libre Video
http://librevideo.org