> From: Scott Brim <[email protected]>
> But where's the difference? ... The more structure there is in the
> name, the less work routing has to do. On one end of the spectrum you
> have a strict source route (or a circuit or a PIP "address").
Go back to Saltzer - or to Shoch, for that matter. There's a _fundamental_
difference between the place you're trying to end up at, and the _path_ you
use to get there.
> On the other end you have MACs. It's all routing, it's all forwarding,
> ... but there is a continuum of how much structure there is and how
> much work you have to do.
Look, step back far enough and _all_ this stuff is just 'stuff used to move
bits around'. [Sarcasm=on] So why don't we throw away _all_ the terminology,
and just call them all 'sumbas'? (Including TCP and HTTP, they have the same
high-level point - if your viewpoint is high-level enough.) [Sarcasm=off]
We're trying to draw distinctions between things that _have significant
fundamental differences_, one that have real consequences in the engineering -
like the one above, between _where_ you want to end up, and the _path_ you use
to get there.
> it's all "locators" (meaning things used by forwarding)
To most of us, "locator" means 'a _structured_ name for a place in the
topology'.
And, in a very fine irony indeed, we've circled around back to August 1993,
because the _original_ definition of "locator" did _not_ include use "by
[the] forwarding": the term 'locator' was defined _precisely_ to have a term
that meant 'structured name for a place in the topology that is _not_ used by
the forwarding'.
I feel like I'm an extra in "Groundhog Day".
Noel
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg
