Re: [re-ECN] RE-ECN without ECN.

[Matt Mathis <[e-mail]>]

On Mon, Oct 5, 2009 at 2:51 PM, Fred Baker <[e-mail]> wrote:
>
> On Oct 2, 2009, at 3:42 PM, Matt Mathis wrote:
>
>>  My bet that it
>> is a lot easier to add the duplicate detection to the ingress router,
>> since the owner is motivated to do so.
>
> is the presence of encrypted or tunneled data streams important here?

Clearly, any encryption prevents peeking at sequence numbers and
generating pseudo re-feedback.  However, for RE-ECN, any non-ECN
receiver or non-re-ECN sender prevents re-feedback....   On day 1 loss
based re-feedback would be way ahead, even if it doesn't work
everywhere.

My hope that these two schemes might be unified.  One then becomes a
migration (bootstrapping?) strategy for the other.  I am also finding
that imagining a dual implementation (both loss based and ECN based
re-feedback) leads to a whole lot of clarity on many side issues.
Its got a much better chance of being deployable, because the
partially deployed states provide at least some of the function of
full deployment.

Thanks,
--MM--
-------------------------------------------
Matt Mathis      http://staff.psc.edu/mathis
Work:412.268.3319   Home/Cell:412.654.7529
-------------------------------------------
Evil is defined by mortals who think they know
"The Truth" and use force to apply it to others.