steve/brian/brian,
I read the spec wearing my ietf spec reveiwer and architecture hat. I
like the clarification on zones. It is all clear to me. I have not one
issue with it at this point. Clearly as I start to envision what it
means to the IP and transport layer in our implementation and the API
suite other questions may arise. But thats step 2.
Also my proposed fix for the multisited app server would not interfere
at all with your spec here.
Very roughly. What I am thinking about is when the address is passed to
the app and its a sitelocal a unique identifier would be embedded in the
sitelocal address which would make it unique across multiple sites or in
cases where two sites merged or whatever. My intention is NOT to
prevent using the zone/scope semantics but make them transparent to
application servers. I found many holes and have only plugged half of
them thus far but I think they can all be plugged. The unique
identifier would never be passed on the wire. There may be one
exception and that is if Erik's draft is to keep moving he may be able
to use the way I want to do this with ND and then they would exist
within the context of a sitelocal ND msg.
Thanks for doing this work we needed it bad.
regards,
/jim
--------------------------------------------------------------------
IETF IPng Working Group Mailing List
IPng Home Page: http://playground.sun.com/ipng
FTP archive: ftp://playground.sun.com/pub/ipng
Direct all administrative requests to [EMAIL PROTECTED]
--------------------------------------------------------------------
