> From: Keith Moore <moore@cs.utk.edu>
> I doubt that a significant percentage of users think in terms of
> "want[ing] identifiers for their machines that are independent of their
> location in the connectivity topology". You and I think that way
If they think "I want a NAT box so I can change ISP's", then I would say that
my formulation is indeed what they are thinking, although you're surely right
that they don't put it in those terms (just as the average person filling
their gas tank doesn't think "I'm loading in some energy stored in chemical
form, for later conversion to kinetic and potential energy").
>> Clearly, you can't have a single label which is both
>> location-independent (so it's provider independent) and
>> location-dependent (so that the routing works).
> It depends on what you mean by "routing".
The collection of mechanisms which establish paths, and cause packets to
travel along those paths.
> Users don't care about the mechanism for how those packets get there
> .. If the packets had to get redirected from a location-independent
> address to a location-dependent address, that would be fine with users
> as long as that operation happened quickly and reliably enough to suit
> the users' apps.
In which case you'd no longer have a "single label", but rather two, with a
mapping in between.
> but I don't think it's a matter of "clearly, you can't" do this.
I see no reason here to modify my statement.
Noel