transition to multi6 [was: Fwd: Minutes / Notes]

[Erik Nordmark <Erik.Nordmark@sun.com>]

> Actually, this leads me to think. If interoperation with unmodified hosts is
> important, then all this discussion about alternative namespaces to identify
> endpoints is moot. At that point, there are precisely two mechanisms
> available:

There is more flexibility if we e.g. introduce a new record type in the DNS
for the nodes that support the new mechanism (call it
an "ID", "WHO" or "WHICH" record if you'd like).
A multi6-aware node will look for such record and, if found, it can do the new
song and dance to get id/loc separation. If not found it falls back to
looking for AAAA.

  Erik