Re: Auto-tunnel Rant

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

In message <7650.985272262@dstc.edu.au>, George Michaelson typed:

 >>I believe large amounts of edge network will *never* be multicast enabled
 >>because the clue density drops as you approach the edge.

i would ameliorate this comment - as edge nets have lans (e.g. dsl
access via ether not usb) edge sites might well have multicast but be
one hop removed from the core, but with no applciatio nlevel box
between them and coree multicast capable nets.
 
 >>So without this feature, end-users in 2 or 3 subtier networks will be
 >>unable to link back to core multicast meshes.

 >>Surely your arguments apply just as equally to the proliferation of
 >>auto-discovery of unicast route/path methods embedded in the applications?
 >>
 >>And, while s/w developers may have less motivation, content *hosters* have
 >>to scale backend platforms for large amounts of parallel feed knowing that
 >>they could reduce their copy cycle to a fraction if they had effective fanout.
 >>
 >>Watching two people in the same cubicle connect to the same service and
 >>listen to the same source on a 2 second lag from each other tends to
 >>re-inforce a view there is a role for a simple lightweight tunnel method.
 >>
 >>You might as well rant against SSH enabled VPN and Cisco GRE at the same
 >>time.
 >>
 >>cheers
 >>	-George

 cheers

   jon