Re: [AVT] updated "Clearmode" (64kbit/s unrestricted) Internet-draft

[Magnus Westerlund <magnus.westerlund@era.ericsson.se>]

Hi,

I have some comments on this draft.

1. Chapter 2: First sentence seems strange it looks like it lacks 
context. "The profile .." refers to something which I don't know what it 
is.

2. Maybe a little more text on how data is put in the payload format. 
Especially around the number of samples you are allowed to put into each 
packet.

3. End of chapter 2: "The timestamp SHALL be set according [5]"
Shouldn't this be "The timestamp SHALL be set according to [5]."

4. Chapter 3: Optional parameters. I think "maxptime" is a valid one to 
use. The existing RFC with this parameter is RFC 3267 but the updated 
SDP draft also contains a definition of it.

5. Chapter 3 is missing rules on how you apply the mime parameters to 
SDP. Please see draft-ietf-avt-rtp-mime-06.txt for such a chapter.

6. Reference list in chapter 5. Shouldn't there be nice [] around the 
numbers?

7. Time to read the ID nits. You are lacking a couple of necessary 
sections, like Copyright statement and IPR statement. See 
http://www.ietf.org/ID-nits.html

8. When you don't transport ISDN 7 kHz voice in this and actually carry 
some other type of data. Isn't of interest to mention RTP relevant 
mechanisms to protect the stream against packet loss. There should be 
significant considerations of this problem.

9. Please look at the formatting also. There are certain parts that 
would look nicer if they are done as bullet list instead of one sentence 
paragraphs.

Best Regards

Magnus

Kreuter Ruediger wrote:

> FYI, 
> I have submitted an update of the CLEARMODE Internet draft. You find it at
> http://www.ietf.org/internet-drafts/draft-kreuter-avt-rtp-clearmode-01.txt
>
> The major update, due to comments according the support of Nx64kbit/s
> and similar standardization activities in ITU's SG 11, is the introduction of
> Nx64kbit/s support. Two remarks to this issue:
> 1) 1x64kbit/s is used in current VoIP products and supported by most DSP 
>    Providers. Therefore I would like to keep Nx64kbit/s with N>1 as an option.
> 2) We currently do not have a VoIP product supporting Nx64kbit/s, therefore 
>    I introduced this merely as a discussion input.
>    So if anybody has a working Nx64kbit/s solution I would like to get some 
>    comments on the packatization method in that case. If nobody has a
>    solution I will delete this feature from the next version of this draft.
>
> Besides, the first draft initiated a controverse discussion, if this should be 
> "audio/CLEARMODE" or e.g. "application/CLEARMODE".
> IMHO none of the existing IANA content types really fits for CLEARMODE. 
> Since 1x64kbit/s is also used for 7kHz voice (despite G.722) I kept "audio"
> in this draft. 
> Regards
>   Ruediger
> _______________________________________________
> Audio/Video Transport Working Group
> avt@ietf.org
> https://www1.ietf.org/mailman/listinfo/avt
>
>  
--

Magnus Westerlund 
Multimedia Technologies, Ericsson Research ERA/TVA/A
----------------------------------------------------------------------
Ericsson AB                | Phone +46 8 4048287
Torshamsgatan 23           | Fax   +46 8 7575550
S-164 80 Stockholm, Sweden | mailto: magnus.westerlund@era.ericsson.se



_______________________________________________
Audio/Video Transport Working Group
avt@ietf.org
https://www1.ietf.org/mailman/listinfo/avt