[11127] in Commercialization & Privatization of the Internet
No subject found in mail header
daemon@ATHENA.MIT.EDU (com-priv-forw@lists.psi.com)
Tue Mar 22 06:34:05 1994
Date: Tue, 22 Mar 94 01:37:26 EST
From: com-priv-forw@lists.psi.com
Apparently-To: com-priv02-out
PS: Hey, Scott, thinking more about it, we could think about hijacking
the vBNS as a new sandbox. The notion of it being procurable
operational infrastructure is baloney anyway, I think, as we are really
talking about collaborative research and development efforts for high
demanding applications, rather than the aggregation of 10000 flows or
so per timestep that we see today (I can point you to some flows paper
we wrote a while back that describes a little of the current situation,
if you like). We really need to tie application and networking
requirements together, perhaps even with the multicasting stuff that
you are so interested in. We can then think about playing with
sloshing bandwidth around for single or few applications (similar to,
but more broadly applied to multiple usages, than what we do on the
gigabit testbeds), perhaps by means of SVCs or so (if the vBNS
agreement allows for that (which I don't know)). Or at times perhaps
even go as far as dedicating the whole vBNS network to specific
experiments. Lotsa research opportunities.
Hans-Werner
PS: Sorry that them apparently little tit for tat games between phone
companies winning and protesting each other's agency awards are
jeopardizing progress. Especially since the current overall sandbox is
big enough that someone's progress and market share is probably more
gated by their efforts and quality of deliverables than some petty
award from some federal agency for a small handful of ATM connections.
Kind of amusing seeing people argue almost in one sentence that the
gummint investment into the current infrastructure is negligable (even
saw something the other day that almost questioned the relevance of the
Fed's efforts for all along), and in the next sentence rant and rave
about them Feds jeopardizing their market share. When, for that matter,
NSF for example is really trying hard not to "own" operational
infrastructure, and largely just helping right now that the system is
not falling apart.
