Peering versus Transit

Barney Wolff wrote:

Is it really the case that people with routers at exchange points actually
consider a packet addressed to one of their own customers to be theft of
service? So far, I note, we haven't heard any position expressed by
any of the big folks, just by others outraged on their

This, of course, now includes me. However I think have a
different perspective than the other folks lumped in with
me in the last sentence...

Dumping traffic skews engineering projections for traffic
loads, which can have dramatically nasty effects.

A particular vendor's old-style FDDI interface has been
built, as with other cards, with the assumption that
traffic will be bidirectional. The buffer management
policy really is geared towards send a frame, receive a
frame, and anything else works poorly, particularly under
heavy load.

In one hypothetical case, consider a large provider that
is being dumped upon by another provider which is
outraged that the large provider consistently refuses to
peer with them. This could have the result of pushing
traffic inbound towards the larger provider over the
threshold at which the difficulties in buffer management
become highly noticeable, which leads to some 15% packet
loss at times for customers of the larger provider trying
to make use of that exchange point.

I would consider this hypothetical situation to be a
gross denial of service. Wouldn't you?

Moreover, even in the case where hardware with ancient
design criteria is not a key factor, having traffic arrive
unexpectedly will tend to reduce the accuracy and value of
other engineering predictions, and may lead to large
increases in infrastructure costs.

I would say that using a resource without permission that
has a real monetary cost to the owner of that resource is
a pretty canonical definition of "stealing".