Here is a very rough strawman idea just posted to the CIDRD-WG. ...
Autonomous System Aggregation Protocol
Without commenting on the technical details (I've always held that mapping
from IPv4 addresses to something else is one way to provide routing that
scales, as well as "portable" IPv4 "addresses"), let me just point out that
this scheme requires a certain amount of protocol development and
documentation, as well as implementation and deployment. As well all have
seen, the Internet seems to have a hard time doing that in less than three
years, flat out.
Also, the CIDR RFC (RFC-1519, 1993, three years ago) explicitly says:
the Internet ... is soon to face several serious scaling problems.
These include: ... Growth of routing tables in Internet routers
... It has become clear that the first two of these problems are
likely to become critical within the next one to three years. This
memo attempts to deal with these problems by proposing a mechanism to
slow the growth of the routing table
In other words, the growth of the routing tables was seen as a problem even
then, and CIDR (and "supernetting") was the solution that was proposed, in
a timely fashion, to that problem. RFC-1519 goes on to note:
The proposed solution is to topologically allocate future IP address
assignment ... this plan neither requires nor assumes that already
assigned addresses will be reassigned ... routing technology will be
capable of dealing with the current routing table size and with some
reasonably small rate of growth.
If this was an unacceptable solution, the time to say so, and make alternate
plans, was back then. We've put a lot of code in place assuming that this was
going to be the way we solved this problem, and it's very late in the game
to say "oops, not the right approach".
Noel