Subject: Re: RFC: route(8) host/bits vs. net/bits
To: Christopher W. Richardson <cwr@nexthop.com>
From: Tom Ivar Helbekkmo <tih@eunetnorge.no>
List: tech-net
Date: 05/12/2005 19:46:59
cwr@nexthop.com (Christopher W. Richardson) writes:

> It may violate POLA, but it is expected if you've been around
> networking a while.  That address is by default a class B.

It seems unnatural to me to use network class semantics for addresses
that are specified using classless syntax.  10.1 is 10.0.0.1 class A,
but 10.1/16 is 10.1.0.0 with a 16 bit subnet mask.  At least that's
what feels right to me.  (Not that that cuts any ice, I know...)  :-)

...but even if I'm wrong, I'd suggest that it makes sense to have
output from applications *not* use shortcuts, to keep people from
misunderstanding -- and to make sure the output will always be parsed
as the same value if given as input.

-tih
-- 
Don't ascribe to stupidity what can be adequately explained by ignorance.