Iljitsch,
(Most of) the vendors set a very bad precedent by making this
address space unusable because they couldn't find a documented
intended use. It's important to reverse this as much as possible by
saying that was the wrong thing to do lest we end up with IPv6 code
that won't accept anything outside 2000::/3.
I think the IPv6 specs are very clear on the unicast and non-unicast
parts of the address space and include specific instructions to this
effect. So I think the scenario you cite is unlikely.
Unfortunately, even with CIDR this wasn't made clear with IPv4. The
Class E (aka 240/4) space wasn't ever defined. The biggest issue I
see with 240/4 we don't know for sure what works and what doesn't.
It's hard to make a decision without that data.
Maybe at some point someone will find a use for 240/4, but I
wouldn't count on that.
The only scenario that makes sense to me to use 240/4 as non-
reserved address space is if it's use can help move us to IPv6
(e.g., Plan A).
By that logic, shouldn't we stop all IPv4-related work?
Yes.
Bob
_______________________________________________
Int-area mailing list
[email protected]
https://www1.ietf.org/mailman/listinfo/int-area
