[dccp] Re: 3448-bis - datalimited to app-limited?

[Sally Floyd <sallyfloyd at mac.com>]

Arjuna -

>  Just to avoid ambiguity, in 3448-bis, would it be better to change 
> data-limited to application-limited? This would conforms to RFC2861.

I actually like the term "data-limited" better.

RFC 2861 uses the term "application-limited" to contrast with
"network-limited".

RFC3448bis discuss two kinds of application-limited conditions.

The first is "idle" (e.g., in Section 4.4, considering whether
the sender has been idle since the nofeedback timer was sent.

The second is data-limited (e.g., in step (4) of Section 4.3,
considering whether the interval covered by a feedback packet was
a data-limited interval.  (It is not possible for the interval
covered by a feedback packet to have been an idle period.)
And the draft specifically defines the term "data-limited".

The text and the tables in Appendix C use the fact
that "idle" and "data-limited" are different things.

So I will add a note clarifying why we are using "data-limited"
instead of "application-limited".  And then, to be consistent,
I might change the faster restart draft back to "data-limited"
instead of "application-limited".

OK?

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