Hi Keith -
At 12:56 9/16/2003, Keith Moore wrote:
> An interesting point, but probably one better made to ICANN. Try
> icann(_at_)icann(_dot_)org(_dot_) Since its not a standards or protocol
issue its
> probably not an IETF issue.
I disagree that this is not a protocol issue, as it certainly affects
operation of protocols. There are probably issues for the DNS protocols
also - if nothing else it may be that some claification of the
specification is needed.
I agree that there may be protocol issues, but the email specifically went:
I think that managers of the TLD's should not be allow to use wild card
RR's in this manner.
The remainder of the email was in the same vein. That's an operational or
policy issue best addressed elsewhere - and when resolved we should take a
whack at the protocol implications. E.g. how do we get the protocol to
enforce the policy or does the policy break the robustness principle, etc..
It can be quite reasonable to make wildcard assertions about RRs that
are all within the same administrative domain, but arguably this
condition is not met for the COM or NET zones.
Agreed - but again, unless it breaks the protocol or has an adverse impact
on robustness, (and not just some number of bottom lines) its probably
better to resolve the policy issue before putting fingers on the protocol.
If someone wants to bang on DNS absent policy discussions - this may be the
right place to start, with a very quick trip to namedroppers once the
relevance was established.
My $.02...
Mike