Gorry -
I understand that RFC3448 will be obsoleted by publication of this
documents. Implementors of TFRC need to carefully consider the new
draft within the last-call!
As for RFC4242 (CCID-3), the language in the RFC is clear:
3.1. Relationship with TFRC
The congestion control mechanisms described here follow the TFRC
mechanism standardized by the IETF [RFC3448]. Conforming CCID 3
implementations MAY track updates to the TCP throughput equation
directly, as updates are standardized in the IETF, rather than wait
for revisions of this document. However, conforming implementations
SHOULD wait for explicit updates to CCID 3 before implementing other
changes to TFRC congestion control.
My interpretation of this is that the replacement for RFC 3448 does
NOT change the throughput equation, but it does modify TFRC
congestion control. There is therefore no implicit update to RFC4342
- this would still depend on RFC3448.
However, I know several of the contributors to the current work have
been evaluating this in the context of a modified RFC 4342 (in fact
this includes many features already in RFC 4342). Should/could this
update CCID-3? Would this invalidate the recently stable releases of
CCID-3? Should it define CCID-5?
My suggestion would be that we add a sentence to rfc3448bis saying
that "CCID-3 and CCID-4 implementations MAY use rfc3448bis instead
of RFC 3448." Or "SHOULD" - either one would seem reasonable to me.
If it says "MAY" instead of "SHOULD", then there are probably some
parts of rfc3448bis that fix corner cases that should be separately
labeled "SHOULD" for CCID-3. But if a new document is required,
someone other than me would have to volunteer to take that on - I
am semi-retired these days, working less than half time, and am not
going to be taking on any new commitments.
The *intention* is that there is nothing in rfc3448bis that requires
senders and receivers to be coordinated as both following rfc3448,
or both following rfc3448bis. In particular, Section 3.2 on Packet
Contents has essentially not changed. (And DCCP and CCID 3 add
their own requirements for packet contents that are independent
from RFC 3448 or rfc3448bis.)
This may be a good topic to discuss on this list and to raise at the
WG meeting.
That sounds good to me.
- Sally
http://www.icir.org/floyd/
