Re: New Internet Draft on automatic (end-user) tunneling for SSM

[Jon Crowcroft <J.Crowcroft@cs.ucl.ac.uk>]

In message <3A9E5A41.813E3D0E@21rst-century.com>, Marshall Eubanks typed:

 >>I still do not see how auto tunnels will scale. If my tunnel relay's will only accept a fan-out of 2,
 >>then to serve 2048 people will require 10 generations of tunnels (and about 1000  
 >>tunnel servers). 


assume that we have the same limit roughly as with the old mbone
mrouted tunnel overlay - was fanout of 6 as far as i recall...
its still, as yo usay, a problem - in fact, the exact place you want
this (at an ACCESS ISPs dslams etc) is where you wont get it...

 cheers

   jon

i am a big fan of quantium computing - quantum computing allows n bits
to store up to 2^n values _simultaneously_ :- with quantum computing,
IP multicast becomes much simpler, and ipv6 becomes obsolete...
you can simply address all the end systems in a group using 
all their seperate IP unicast addresses, but it takes no more bits 
than a current IP header.

of course, you need a $1M MRI/interferometer in all your routers to
sort out the packet entanglement, but hey, thats today's prices;
when quantum internet addressing and routing meets optical switching 
the world will be a safer place, probably. a neat side effect of
quantum ip (quip, TM) is that we can employ schroedinger's approach to
mesage payload encoding too, which means that the fbi's carnivore
systems won't be able to decode what we've sent (well they will, but
they will have _chosen_ what to observe by observing it, so it can't
be used as evidence in court about what we did send) - feynman said it
all of course.