Re: [rtcweb] Agenda time request for draft-marjou-rtcweb-audio-codecs-for-interop-01

[Tim Panton <[e-mail]>]

On 13 Mar 2013, at 19:45, Jean-Marc Valin wrote:

> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
> 
> On 03/13/2013 09:14 AM, Xavier Marjou wrote:
>> - In order to reach a compromise, we would like to add some text in
>> the WG draft draft-ietf-rtcweb-audio providing incentives for the
>> browser to use these three codecs: make them mandatory to implement
>> when there is no cost impact on the browser (e.g. if codec already
>> installed, paid by the device vendor...).
> 
> I think this is the main faulty assumption here. "Royalties already
> paid" does not imply "free". There's a real cost here because
> supporting AMR, AMR-WB, G.722 and any other codecs added to that list
> means that:
> 1) Someone has to write all the code for actually using these codecs
> in their RTP stack.
> 2) There is no standard interface for accessing these codecs, so a
> browser vendor would have to write code for each of these (often
> undocumented) interface and *test* it on every device.
> 3) This code needs to be maintained and fixed for security
> vulnerabilities.
> 4) If there's any issue (e.g. buffer overflow) in the platform
> implementation of the codec, there's nothing the browser vendor can do
> about it.
> 
> Oh, and that cost would be paid by pretty much all browser vendors
> because they're all running on at least one platform that supports
> each of these codecs.

And that assumes that the codec license signed by the hardware maker permits the use of these 'no-cost' in all the possible ways
the javascript user wants. E.g. is it limited in number of channels ? Is the license symmetrical wrt encoding and decoding ?
Is it available to 3rd party browser makers (eg chrome on blackberry? Or to products derived from the native browser browsers
or that embed browser objects (like cordova/phonegap) ?

T.