On Fri, 30 Jul 2004 11:15:25 -0500, wayne wrote
In <009701c4764d$e56d4740$c9006bd8(_at_)jdk> "John Keown"
<jdk(_at_)nni(_dot_)com> writes:
Meng could you please chime in as to whether this is an acceptable form for
a spf record.
I'm not Meng nor Mark, but would be shocked if either of them thought
that 1.2.3.4/24 and 1.2.3.0/24 should be treated differently. If
either of them did, I would argue with them that they are wrong.
I think every piece of software I've used that handled CIDR notation has
treated both cases the same. In fact I've used both in the past.
Since it's a mask (a bitwise-AND operation), I don't see why, say,
192.168.1.0/24 should be treated any differently than 192.168.1.1/24.
I can't think of any situation where it would be adventageous or meaningful to
treat them differently. If John can come up with a situation where treating
them differently makes sense, then I'll concede he has a point.
Zeroing out all the masked-off bits is a good convention, and makes the
notation prettier and easier to read in most situations, but I don't see why
it should be required. It's not the job of software to enforce pretty-printing.