[Iccrg] update to cc-rfcs draft

[Sally Floyd]

Wesley -

>  Michael and I feel
> that this summary is fairly complete at this point, and would like to
> wrap it up in the near future, so we'd appreciate some additional 
> reviews
> from the group.

I read it over, and it seems good to me.

Some fairly minor feedback is below.


In the description of RFC 2481, you might want to add that it is
obsoleted by RFC 3168 (even though that is also mentioned
two paragraphs down in the discussion of RFC 3168).


In the description of RFC 3540:
The semantics of the bit combination in question are specified in 
[RFC3168].
 From [RFC3168]:  "The use of two ECT codepoints essentially gives a
one-bit ECN nonce in packet headers."

So the first sentence would be more correct as follows:

"This document specifies a nonce mechanism that uses an ECN
bit combination that is not used in [RFC2481], but that is
specified in [RFC3168] to allow a one-bit ECN nonce.
This nonce mechanism includes the definition of a Nonce Sum ..."


Also in the description of RFC 3540:
"there is currently some controversy as to whether the
mechanism is really effective".
I am aware of the controversy of whether the nonce is the
best use of this codepoint, but I am not aware of any controversy
about whether the nonce mechanism is really effective at preventing
the receiver from lying about receiving packets undropped and
unmarked.


Typos:
Section 3:  "may discussed" -> "may be discussed".

- Sally
http://www.icir.org/floyd/