On Tue, Apr 7, 2009 at 11:52 PM, Tony Li <[email protected]> wrote:
> As the conversation has died down, I'm going to guess that we've converged.
>  The consensus check can be found here: http://doodle.com/9sybb8dmk5phvp99

Tony,

Looks like I get to play lonely voice of dissent: these definitions
don't provide me with a useful language to talk about routing. With
these definitions adopted, I would need to strip the words from most
of the documents I've produced as they simply don't fit.

In addition, the definition of address is demonstrably wrong.


> address    An address is a name that is used as both an interface
>           locator and an endpoint identifier.

Counterexample #1: RFC1918 IP addresses used inside a LAN consisting
of a personal computer, an ethernet hub and a NAT device. The IP
address in this example contains no locator semantics whatsoever, at
least not as you've defined a locator.

Counterexample #2: The IP address of an anycasted DNS service. The IP
address neither identifies an endpoint nor locates a particular
interface's points of attachment to the network. Instead it identifies
a service and specifies multiple locations via which that service can
be obtained.

Regards,
Bill Herrin


-- 
William D. Herrin ................ [email protected]  [email protected]
3005 Crane Dr. ...................... Web: <http://bill.herrin.us/>
Falls Church, VA 22042-3004
_______________________________________________
rrg mailing list
[email protected]
http://www.irtf.org/mailman/listinfo/rrg

Reply via email to