In general, any time you have a set of values that can change over time,
and there is a reason for the community to know the currently-valid set of
values, a registry is a Good Thing.
One example of when a registry is NOT needed: error codes in
draft-ietf-rohc-sigcomp-nack, where the author
Date:Mon, 20 Dec 2004 14:28:52 +0100
From:Harald Tveit Alvestrand [EMAIL PROTECTED]
Message-ID: [EMAIL PROTECTED]
| In general, any time you have a set of values that can change over time,
| and there is a reason for the community to know the currently-valid set
There is the beginning of yet another debate on the SIMPLE mailing list
about do we need an IANA registry from some fields in a specific draft.
There does not seem to be a clear opinion on when to create a registry or
not and past activities are not 100% consistent. I'm sure this has come up
many
If there is something in an RFC that clearly is a parameter, then it
is not a bad idea to have an IANA registry. By creating a registry, you
can yourself set the requirements for assignment. For example, is an
RFC needed? Is IESG approval needed? Is just review on a mailing list
(and an
