[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [RAM] Ramblings about "locator"
Hello;
On Jun 15, 2007, at 3:11 AM, Brian E Carpenter wrote:
On 2007-06-15 05:03, Ved Kafle wrote:
Marshall,
...
Maybe the essential point is that a locator can at least in
principle be mapped to topology and an identifier can't.
If in Ethernet what is essentially a random number (the MAC
address)
becomes
a locator then maybe the distinction will never be crisp.
Well, that horrible thought crossed my mind. I think I disagree
with Jarno's comment, because as soon as an Ethernet address
gets into a spanning tree it has become a locator (at level 2).
So, the distinction between an identifier and a locator is
dependent on
the avalaibility of routing/switching infrastructure. If networks
possess routing/switching insfrastructure based on a
identifier namespace, the identifier simultaneously functions as a
locator.
My original RAMble (sorry) ended thus:
Maybe the essential point is that a locator can at least in
principle be mapped to topology and an identifier can't.
I think that's correct and actually consistent with much that's been
said. A slightly different way to say it is that a locator is a
handle for a route.
I think that the missing piece here is _intent_.
A locator is _intended_ to be mapped to or congruent with typology.
An identifier is _intended_ to be unique and independent of typology.
Of course, once something is created, how it is actually used (or
misused) is no longer under control,
and it is safe to say, "if it can be, it will."
So, you can't be crisp about usage; the goal should be to be crisp
about intentions.
Brian
Regards
Marshall
_______________________________________________
RAM mailing list
RAM at iab.org
https://www1.ietf.org/mailman/listinfo/ram
_______________________________________________
RAM mailing list
RAM at iab.org
https://www1.ietf.org/mailman/listinfo/ram