Re: [AVTCORE] Coments on draft-ietf-avtcore-rfc5285-bis-05

[Roni Even <[e-mail]>]

Dale,
If extmap-allow-mixed is declared it means support for both one-byte and two-byte. Without it  there is no way to signal support for just one-byte (or declare support for two-byte) in RFC5285 . It can be assumed based when the  IDs used in the offer are with IDs 1-14, still it  does not say that in this case only 1-byte are supported. Since this is the text in RFC5285, I am not sure if we can change without agreement from the WG.
Roni

> -----Original Message-----
> From: Dale R. Worley [mailto:[e-mail]]
> Sent: יום ב 13 פברואר 2017 20:36
> To: Roni Even
> Cc: [e-mail]; [e-mail]
> Subject: Re: Coments on draft-ietf-avtcore-rfc5285-bis-05
> 
> Thanks for looking at my comments and doing what you can with them.  I
> suppose overall, I wish that 5285 was a lot clearer, and in particular, that it put
> in one place all the rules that relate offers and answers.
> But that would require a complete editorial overhaul.
> 
> However, there is are a couple of point that could be addressed.  In section
> 4.1.2 paragraph 2, it says
> 
>    Since it is expected that (a) the number of extensions in
>    any given RTP session is small and (b) the extensions themselves are
>    small, the one-byte header form is preferred and MUST be supported by
>    all receivers.
> 
> This says that all receivers must support the one-byte format.  But I can't find
> a similar statement that all receivers must support the two-byte format, so it
> seems that support for that format is optional.
> And if that's true, it means that a device that can receive only the one-byte
> format needs a way to force the sender to only send the one-byte format.
> 
> Another point is:
> 
> > > 4.2.  One-Byte Header
> > >
> > > The ID value 15 is reserved.  But there is also the invalid case
> > > where the ID field is 0 and the len field is non-zero.  That case
> > > should probably be included in the same exception processing as for
> > > ID value 15.
> >
> > > [Roni Even] It says that ID=0 MUST not be used value =0 is padding
> > > (both ID and len field) so you are asking for text about what to do
> > > if there is  header with 0 in the ID part but some value in the len
> > > part. I assume that good policy will be the same as for ID 15 but
> > > any error behavior can be expected (ignore all header extension
> > > ,treat like padding ...) since this case is not discussed in
> > > RFC5285.
> 
> My thinking is that if the draft is careful to specify the processing of the
> invalid case of ID=15 in the one-byte format, then it would similarly make
> sense for the draft to specify the processing of the invalid case ID=0,len>0.
> 
> Dale