Moving Event Classes to Iana (was: other stuff)

["Sharon Chisholm" <schishol@nortel.com>]

hi

Just to confirm. We are both saying that Notifications are defined in
the data model not IANA, but you are proposing an alternative way to
specify the event classes? Currently they are an enumerated list in the
specification. You are suggesting instead to create an IANA registry?
You are suggesting an initial value for this registry to be a much
smaller subset of event classes then is defined in the current draft.
You are also suggesting to make it URI-based.

I think we can specify a reasonably complete list, so IANA seems a bit
overkill, but as I've mentioned, I have a non-functioning crystal ball,
so I'm open to the idea.

I'm not sure we actually want to make these URI-based instead of an
enumerated list. It is the old SNMP rowPointer versus enumerated list
argument. I always hated rowPointers since there was really no way to
validate them or ensure that there were not 234234234 ways to say the
same thing.

I also worry that we would start out way too conservatively with the
list of event classes or that we might try to block additions to the
list unnecessarily. But I guess we can discuss what the list is and what
IANA's method of working will be and take it from there. 

Note that the content portion of this specification (aside from event
class) has always just been a non-normative appendix. It was added to
provide context to the operations. If there is consensus to remove it,
then that isn't a big deal. 

Sharon

--
to unsubscribe send a message to netconf-request@ops.ietf.org with
the word 'unsubscribe' in a single line as the message text body.
archive: <http://ops.ietf.org/lists/netconf/>