> I am worried that the pretty new term "Locator" is going to be
> bent, too
It already has been.
The original definition of 'locator' (RFC-1992) was of a structured
location-sensitive name for an interface which _did not necessarily appear
in packets_ (since people kept assuming an 'address' had to appear in all
packets, which was not true in that design).
No big deal... "The street finds its own uses for things."
Noel
________________
Though I would really hope that people obstain from abusing this term.
Promoting the TARA-locator I am not afraid that any other type of locator would
surpass its aggregatability.
Indeed: Each geopatch can clearly recognize the 8 surrounding geopatches, and
each such cluster of geopatches can clearly recognize an even larger geopatch
cluster with itself in the middle thereof. This is a different league compared
with country-code based aggregation for instance.
But:
The primary objective is not aggregatability. Instead the primary goal must be
that a locator is ROUTABLE !!!
This is why I insist that a locator of a node must denote the location of that
node. If this is not acceptable, then I suggest to find another name instead of
"locator" that is more suitable for the intended use.
Heiner
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg
