[Tsvwg] multiaddressing

In draft-crocker-mast-analysis-01 (now expired), the concept of
"multiaddressing" for solving multihoming and mobility problems is
explained.

The HIP groups and multi6 are focusing on multiaddressing solutions
built for "simple" transports that can only bind to a single address
with no means to update or extend that binding when it might be required
by mobility or multihoming.  Adding a wedge-layer to map dynamic locators
to fixed identifiers fixes these problems and allows the transport to go
untouched and remain simple.

A drawback of such wedge-layers, Mobile IP, and other systems that hide
mobility (or multihoming) from the transport is that current transport
mechanisms are not nearly "agile" enough to deal with end-to-end path
changes that can pop up instantly.  The transport is where all the path
estimates are stored (congestion control variables, RTTM, PMTUD, etc) and
making non-trivial path changes without some explicit signalling to the
transport can be dangerous.

If we're going to see multiaddressing wedge layers deployed, then it seems
like we should have a slightly richer interface between the wedge and the
transport so that reinitialization and reestimation of path state can happen
rapidly.

Architectures that don't include such wedges are what we were
considering in draft-eddy-tlmarch-00 (presented last week).  The idea of
rapidly fixing the transport's path estimation variables is roughly
visible in draft-swami-tcp-lmdr-04.  For architectures that *do* include
wedges, some similar form of signaling and responding to path changes is
still _required_.  Signaling has to occur both between the wedge and
transport locally and remotely, so that both half-connections can be
fixed.  Responses to these signals will have to be implemented
per-transport protocol.

It's not entirely clear to me how we can combine the strengths of a
multiaddressing wedge layer with the rapid-response of path property
estimates that we can have without the wedge in a transport layer
mobility scheme.  Thoughts?

-Wes