I have reviewed <draft-ietf-ipv6-scoping-arch-00.txt> and have the
following two comments:
a) I think it would be helpful if the document would refer to
<draft-ietf-ipv6-deprecate-site-local-01.txt> in places where it
explains why it does not cover site-local unicast addresses.
b) I have a more serious issue with the default zone. The text says in
several places that the zone index zero might be used to indicate
the default zone but it explicitely allows to use other values. In
section six, it says:
Similarly, an implementation may choose an index value other
than zero to represent the default zone.
I am not sure why this is helpful. Is there a particular reason why
we can not just say that the default zone is indicated by a zone
index which MUST (or SHOULD if we have to compromise) be zero?
Note that I am editing some textual conventions for MIBs where the
language in place right now is much clearer that zero means default
zone. This means that an implementation which chooses a different
value to denote the default would have to translate this value for
the MIB interface, which really seems awkward. Hence, I would
prefer if the scoping architecture document could be more precise
here.
/js
--
Juergen Schoenwaelder International University Bremen
<http://www.eecs.iu-bremen.de/> P.O. Box 750 561, 28725 Bremen, Germany
--------------------------------------------------------------------
IETF IPv6 working group mailing list
[EMAIL PROTECTED]
Administrative Requests: https://www1.ietf.org/mailman/listinfo/ipv6
--------------------------------------------------------------------
