[e2e] Is a non-TCP solution dead?

Bob Braden braden at ISI.EDU
Mon Apr 21 09:12:08 PDT 2003


  *> TCP definately has its problems.  And, one can make the case for CC
  *> being in the network layer (or transport or application).

!!??

Mark,

We *did* do congestion control at the [inter]network layer, it's called
Source Quench, and it was not a very good idea.  We could probably do
better at CC control at the network layer... e.g., we could use X.25.
That is apparently what Alex wants us to do.  The architectural
arguments in favor of connectionless networking will no doubt never
be resolved; some people love it, some people hate it.

This discussion has been bordering on noise.  Let's stop the vague
innuendo and specious complaints about implementation errors and
discuss something fruitful.

Bob Braden




More information about the end2end-interest mailing list