Jim Choate wrote:
Forwarded message:
Date: Tue, 15 Dec 1998 17:50:56 -0500 From: Robert Hettinga <rah@shipwright.com> Subject: DARPA's Active Network page
I saw an article on this in InfoWorld. NAI/TIS and DARPA talking about putting self-switching packets into the network. I don't see this as too far off from packets purchasing their passage across the network with micromoney of some kind.
How does the packet know what is a good price or a bad price as it makes each hop? Whose processor is it using to run its program to evaluate the costs? Who pays for that processor cost on a node that isn't selected? Who has access to the audit trail? How does a reliable copy get back to the originator? How does it do this in an anonymous network?
Joy, just what we needed, a twist on the old source-routing packets. A bad idea within a bad idea. :( Gee, how many of your routers allow source routed packets these days? To add to the above question, how big will the over head of the packets be? How will the routers know that the packets aren't forged and that they contain real micromoney and not already spent micromoney (possibly for the purposes of denial of service attacks or other hacking attempts)? How will they do this in real time and not incur huge propagation delays? IMHO, This has the words "Stupid Idea" written all over it. -- ---------------------------- Kaos Keraunos Kybernetos ------------------- + ^ + Sunder "The real aim of current policy is to /|\ \|/ sunder@sunder.net ensure the continued effectiveness of /\|/\ <--*--> ALLOW FREE EXPORT OF US information warfare assets against \/|\/ /|\ STRONG CRYPTOGRAPHY! individuals,businesses and governments \|/ + v + PROTEST WASSENAAR!!! in Europe and elsewhere" -- Ross Anderson ------------------------------ http://www.sunder.net ---------------------- RESTRICTED DATA - This material contains RESTRICTED DATA as defined in the Atomic Energy Act of 1954. Unauthorized disclosure subject to administrative and criminal sanctions. NOFORN ORCON WNINTEL SIOP-ESI CNWDI