On 20-apr-04, at 14:26, Geoff Huston wrote:
You appear to be saying that using 224.0.0.0/4 in contexts such as used
by the current RFC 1918 space does not sacrifice 'usable' unicast
address space. Surely, however, if these addresses are 'useable' in
1918
contexts they would be equally useable in a global context?
I don't think this is the case. If an organization were to adopt
addresses from 224.0.0.0/4 for use in an RFC 1918-like fashion, they
would only have to make sure that all their internal systems are able
to treat class E addresses as regular unicast addresses. However, class
E addresses would only be globeally useable if and when all systems
connected to the net throughout the globe are able to use them as
regular unicast addresses.
I personally do not see any value in using this address block up in a
1918 role.
I tend to agree, not having heard the case for additional private space.
My reasoning for this view is that life would be getting
mildly interesting if we ever reach the point when we are forced to use
this space, but life could be interesting a little sooner if we act now
to burn up this space for use in private / closed network contexts
I'm not sure whether you are trying to optimize for an interesting or a
less interesting life...
and
thereby make it completely unavailable to the public network. i.e. I
hope that collectively we, for some suitably large value of "we", will
manage this evolving situation so that the public Internet _never_
needs
to clamber into using the Class E space
Why?
The addresses are there. Leaving them unused while address assignment
policies grow more strict over time doesn't make much sense.
for me, it
is some small comfort to know that if we do reach the point we we've
managed to assign, allocate and otherwise distribute these 219.98 /8
addresses, there is something left in the address cupboard marked 'E"
that conceivably is there to use if we really _have_ to.
My comfort level would be much higher if by the time that we need the
extra address space, we have a fighting chance of actually being able
to use it. So I think it would be a good idea to make it very clear
that implementations must, in the absence of more specific information,
regard class E space as regular unicast space, the same way the IPv6
addressing RFCs spell this out for IPv6 address space that hasn't
received a specific purpose yet. If we do this now we have ten years or
more to clean up implementations.
_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www1.ietf.org/mailman/listinfo/ietf