[e2e] What should e2e protocols know about lower layers?

[Michael B Greenwald]

   Wed, 17 Oct 2001 14:55:10 +0200
   Marcel Waldvogel <mwl at zurich.ibm.com>

   I agree that TCP in the congestion avoidance phase will take forever to 
   absorb the remaining bandwidth. But I also believe that ... [reasonable
   arguments deleted]

I agree that the arguments you make in this note are more convincing, but I
think you misunderstood my note.  I was not advocating turning off TCP
congestion control myself, and I understand the arguments against doing so.
I was only answering your question ("Are these not the problems, or do
these solutions not solve them appropriately?") as to why the specific
arguments you mentioned in your earlier note (e.g. "slow start will take no
time" & "simply remembering the [previous] send window") would be unlikely
to convince those people who do want to turn off TCP congestion control in
so-called local situations.