Re: [rtcweb] Few comments on draft-ietf-rtcweb-audio-07

[Jean-Marc Valin <[e-mail]>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hi Cullen, Magnus, Colin,

Thanks for the review. I just updated the draft. As some of you
pointed out, I changed all instances of "client" to "endpoint" to
match the terminology. See below for more:

On 16/04/15 02:28 PM, Cullen Jennings (fluffy) wrote:
> 2) Need to tweak Sec 2 to include NOT RECOMMENDED:

Done.

> 3) To avoid a sec considerations comment, I’d probably bolster
> them by adding something like:
> 
> For security considerations regarding the codes themselves please 
> refer their specifications.  Likewise, consult the RTP base 
> specification for security RTP-based security considerations.

Added that text, along with references.

> 4) The last sentence in the sec cons is likely to attract flies.
> Can we point to the two security drafts instead?

Added many security references. I didn't delete the sentence itself,
but I can do that too if you think it should go.

> 5) I’d probably the refer to the opus rtp payload draft from -03
> to -08 (it’s in IESG eval - yippie) and the reference to the audio
>  codecs for interop from -00 to -01.

Done.

> 6) The draft has "Because of this, clients MAY increase the gain 
> before playback." To make consistent sound levels across browsers,
> I wonder if this should continue to say something like - "many
> browsers increase it by X db" and insert whatever value of X that
> chrome and FF think they should be using. I don't feel strongly
> about adding this one way or the other but can the authors give it
> a little thought and see if they think it is needed or not.

I don't know that browsers are doing that yet and I'm not exactly sure
what would be a good value there. I suspect it's more of a W3C discussio
n.

On 20/04/15 03:40 PM, Colin Perkins wrote:
> - Section 3 says "Clients MAY use the offer/answer mechanism to 
> signal a preference for a particular mode or ptime". Is this just
> of Opus, or in general? It might be worthwhile saying something
> explicit about acceptable ptime values in general.

There was some discussion on acceptable ptime values in the past, but
there was no consensus achieved.

> - Security considerations ought to explicitly point to the
> security considerations of RFCs 3389, 4733, 6716, and 
> draft-ietf-payload-rtp-opus. It should possibly also point to the 
> draft-ietf-rtcweb-security-arch-11, and the security
> considerations for RTP use in WebRTC in
> draft-ietf-rtcweb-rtp-usage-22

Done.

Cheers,

	Jean-Marc
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJVQqhNAAoJEJ6/8sItn9q9bP4H/2BwbcFqUo8V0BuG/mRWdzxv
W5pAhsphO+4hxSV2c6bGABRR3KRkjqHklRpQHebSQ/yAhQEWRWWZXoSeX/sqfMqD
RODCF56ZLMIfPtep3ss7eLaRN3qjMfJMbyP7nvTzYqZujbcnlQzw11pJSywMKVqh
Yg0/Fa9X+C5ymWRE3Y9v2ITCoVDwd5A4MnHO3FY+3gotmsT/6+RRgccbXGyqUp2t
OLMknp3lA3ofidZAWkQ1lMv9D/GrLxCqpya2w/4LGZFCK++q0YBvfxlyXjoL7N9l
CKtTreovnhJwULcyp5RVEr3G8BJ0YniOGfkjHuslfAg/L45jVmP/pMxorR7s1LU=
=Afic
-----END PGP SIGNATURE-----