[e2e] New approach to diffserv...

[Jon Crowcroft]

In message <20020616104540.AB62670B at ab.use.net>, Sean Doran typed:

 >>| lets take the "middle" out of middleboxes and
 >>| justify the end2end means.
 
 >>Nice slogan, but insufficient marketing to make the argument 
 >>completely compelling.

you kind of ignored my concrete examples. 
again (for example, but i have more if you keep reisting:-):
1/ middleboxes are not in the middle because by definitoin they cannot
perform at warp factor 10 core speed  necessary. if they could we'd
just have a net matde of web proxy caches and smtp relays.

this means that i have an incentive to buy e2e solutiosn as soon as i
need hi speed access...which is a LOT Of people

2/ middleboxes dont solve any real security problems, they introduce
more. at some stage someone is gonna sue a middelebox provider 
who claims some security (or performance) enhancement and demosntrably
(catostrophically)_ fails to produce it - that will create some mighty
incentives too...

 >>A trio of counter slogans: 
 
 >>	show me the money

well, microsoft appreciate the idea of billions of $ margins instead
of 10s of 1000$  margins - who would you be today? :-)

 >>	demonstrate the greater utility for the greater number of people
IP does this (conter to your stat mux point)
 
 >>	illustrate how imposing the model will eventually lead to the
 >>	population at large accepting its moral superiority
imposition? no-one said much about imposition.....we're just debating
a point - the only imposition i find is ports being arbitrarilty
blocked by annoying ISPs:-)

 >>which should cover a good chunk of political philosophy beyond
 >>the obvious "let the market decide".
 
not quite - 
 >>I think an interesting area of research would be in analysing
 >>*why* end2end models for various real-world problems have *failed*
 >>to achieve the uptake enjoyed by non-end2end ("middle") solutions.
 >>No, interesting is too mild a word, given what one could learn about
 >>the mistakes (from an end2end-interest perspective) to date.
I compeltely agree

a good read in this area (but you have to go to source papers to get
the meat) is
Linked
by
Albert-Laszlo Barabasi
Perseus Publishing
ISBN 0 7382 0667 9

A fun read if you want a rant, but some legal framework too, is
Lawrence Lessig's stuff...


by the way, here's a gedanken experiment:
if middleboxes work well at the IP level, they ought to be a good idea
at the HTTP/URL level. install, and
now try go build a search engine
 
 >>

 cheers

   jon