Re: radical solutions

[Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>]

Thomas;

> > I don't see GSE as being particularly different from Guruprasad's
> > architecture, although the details of MO's proposal may be fine-tuned
> > (e.g., there's clever things we can do in the lower 8 bytes to make
> > any mapping between them and the top 8 bytes easier, and which may
> > eliminate use of the few bits robbed for local-to-the-site routing),
> 
> There are some interesting details buried in here.

GSE is a paricularly poor design variation of 8+8 proposal.

> If you want to map
> from the lower 8 bytes into something that can be used in the upper 8
> bytes for routing, one likely needs a heirarchy in the lower bytes
> (e.g., ala DNS delegations) so you can do the mapping in a scalable
> way.

That's not a point of GSE.

> Once you impose heirarchy, you lose the ability to just use MAC
> addresses (aka EUI-64s) -- they only have two levels of heirarchy and
> the heirarchy doesn't match what the delegation structure probably
> needs to be (i.e., they reflect the manufacturer not the device
> owner). As a consequence, one loses stateless addrconf and one still
> needs a way of assigning the lower bytes (i.e., identifier) to
> devices.

Stateless autoconf is just another hype that nothing is lost.

						Masataka Ohta