> >(i.e. if you're not very smart, you can't tell the difference
> >between 13.0.0.0 referring to the class A network 13/8 and
> >13.0.0.0 referring to the subnet 13.0.0/22.)
>
> that's really lame.  it sort of "implies" that i *can* tell the
> difference between
>
>   13.1.0.0/16
>   13.1.0.0/20
>   13.1.0.0/24
>
> simply by looking.  after all, those don't have "all-zeroes" subnet
> mask.

Uh...  The IOS thingie can be understood if you think of what
plain old RIP or IGRP would do with those "subnets".  Those
routing protocols do not carry any explicit network masks, so
have no way to distinguish between subnet zero and the network
route as such.

Of course, if you're doing variable-length subnetting you have
long ago converted to a routing protocol which carries explicie
prefix length information in the routing protocol updates, and in
that case, doing "ip subnet-zero" makes sense.

Oh, BTW, I don't find it particularly difficult to make sense out
of IOS, Bill, at least not in this case...


Regards,

- H=E5vard