[e2e] Overly Overlay; Peer to peer is commonplace

[David P. Reed]

At 03:29 PM 12/31/2001 -0500, Craig Partridge wrote:

>In message <5.1.0.14.2.20011230071041.00a8a4c0 at kahuna.telstra.net>, Geoff 
>Husto
>n writes:
>
> >You are possibly working on the assumption that _someone_ knows, but they
> >are not telling the rest of us. The alternative explanation is that noone
> >knows and there is nothing to tell as yet!
>
> >From my perspective, what's interesting is that it looks like the
>mobile ad-hoc networking folks are getting ahead of the wire-line folks.
>
>They have to deal with rapidly shifting topologies, lots and lots of
>nodes, and tight limits on bandwidth they can use to pass around the
>routing info (and it has to work and be stable).

And meanwhile, the end-to-end protocol developers seem to be building into 
their models of the network various pernicious assumptions, such as:

  - that the path through the network involves a single series of links 
(i.e. no parallel routes are every used),

  - that the only sources of congestion are the flows from other users,

  - that round-trip times should be assumed to be relatively stable over 
short intervals,

  - that isochronous applications require isochronous links (like ATM) and 
isochronous router class-of-service,

  - that IP addresses should reflect topology because topology is stable, 
because putting in fiber requires backhoes,

  - that identity and authentication should be based in topology because 
network owners should act as identity brokers,

  - that packet checksums are only about pseudo-random noise sources, 
rather than adversaries,

  - that end-to-end encryption is not really that necessary, because 
tapping into routers and fibers is hard for script kiddies to do,

  etc.

Fortunately some of the ad-hoc mobile guys are not mired in the past or 
present, and might take up the challenge of thinking more clearly about 
end-to-end protocols that don't start to bind in these assumptions.




- David
--------------------------------------------
WWW Page: http://www.reed.com/dpr.html