(6bone) IETF-53 meeting (was (ngtrans) final ngtrans agenda for IETF-53 in Minneapolis)

[Antonio Querubin]

On Wed, 13 Mar 2002, Michel Py wrote:

> > Bill Manning wrote:
> > I am saddened by the thought that we move away from "where
> > everyone peers nicely with others" to something else.  Why
> > is this useful or desirable?
>
> It's not useful neither desirable. The point I will be trying to make is that it will happen at some point (don't get me wrong, I don't like it either, but the business model where everyone provides IPv6 transit for free is not going to last forever). In fact, some of the questions I plan to ask the floor for the purpose of triggering thinking about them are (roughly):
>
> 1. Is the evolution of the IPv6 backbone (pushed by market forces) to a
> v4-like tiered system unavoidable?

The evolution due to market forces is steered by the cost of bandwidth.
Currently the cost of providing bandwidth for IPv6 for most is relatively
low compared to IPv4 because the amount of traffic is low.  So few mind
providing transit currently.  When the IPv6 traffic increases to a
significant percentage of the total, anyone paying for bandwidth isn't
gonna care whether it's IPv6 or IPv4 bandwidth.  They'll begin applying
the same policies resulting in the same tiered system.

> 2. Is it the role of the 6bone or the IETF to think about it and
> possibly recommend something about it.

Depends on whether the IETF should be getting involved in adjusting
economic forces or finding ways of making the cost of bandwidth
negligible.  The latter in particular seems more like a basic technology
problem than an Internet architecture problem.