Re: Reserved addresses in the SSM range

[Colin Perkins <csp@east.isi.edu>]

--> "Dr. Carsten Bormann" writes:
>> For example, we could
>> modify RTCP such that responses to RTP streams sent to an SSM address
>> would be directed to the sender instead of the group.
>
>Just remember that the RTCP receiver reporting algorithms depend on seeing
>every other report to gauge the group size and adjust the reporting
>bandwidth appropriately (see section 6.2 of RFC1889 or, better,
>draft-ietf-avt-rtp-new-08.txt).
>RTCP currently has no way to reflect this information back from the sender
>(which is the only one who will have it if RTCP is unicast).
>For NORM (NAK oriented reliable multicast, rmt working group), we will
>probably put a sender-based group size estimation scheme into the protocol
>to make this work.

We did figure out a way of making this work during a side discussion at the
Oslo IETF, but I don't think it ever got written down..  Maybe it's time to
resurrect those ideas?

Colin