[e2e] Technology marches forward at the expense of the net?

[Vasilis Friderikos]

I agree. The idea of using powerful coding schemes (beyond FEC) seems very
interesting. This can be seen as what is happening in a wireless link, i.e
you try to recover from bursty errors (fast fading) inside the MAC frame
(ARQ is afterwards), the difference is that now you use the same philosophy
but instead of recovering bits you recover packets (or cluster of
information). But I don't know what the consequences of that pseudo TCP
traffic will be on the dynamics of the queue in routers and other TCP flows
(from the point of view of comparing the throughput that TCP would achieve
along that specific path - this is the definition of TCP friendliness, isn't
it?).

regards,
Vasilis Friderikos