> Ted Lemon <mailto:[email protected]>
> Thursday, October 23, 2014 11:16 AM
>
> Right, 'cuz there's nothing at all difficult about getting ICMP to
> work... :)

understood. but that's part of what makes this a good solution. systems
need to learn to live with hosts whose names they cannot guess. icmp
even when it gets through can be spoofed either by the remote system or
any intermediate system. that exactly matches the confidence and
dependency levels that hostname-depending programs should have.

internet service and internet access should be thought of differently.

william simpson's icmp idea came with the observation that only a host
can reliably report its own name, and only while it's up, because it
might have a new name from time to time.

the impedance mismatch between the understood and the possible in
today's PTR system led to the creation of <http://enemieslist.com/>
which allows those of us who want to know when a PTR was
machine-generated and that the host's actual name is "amnesia" and
should not be able to do the things that real hosts which actually have
names should do, can undo all the $GENERATE complexity that internet
access providers use to "jump through the hoops" of pretending that
their internet access IP addresses actually have host names, when in
reality, they don't and oughtn't.

-- 
Paul Vixie
_______________________________________________
DNSOP mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/dnsop

Reply via email to