At 9:34 PM +0200 6/23/06, Julian Reschke wrote:
Related to this, I'd still like to find out the background history
on why this is on the standards track. It was claimed that
extensions to standards track documents need to be on the standards
track as well, but that doesn't seem to be correct, as far as I can
tell.
It is not correct that they *need* to be, but there is no reason for
them not to be. In general:
- If a protocol is worked on in an open context, standards track is preferred
- If a protocol is worked on in private and the author doesn't want
to take changes from others, Informational is most appropriate
- If a protocol is really an experiment that may not even get
implemented, Experimental is most appropriate
It certainly doesn't make sense (because it would prohibit
experimental extensions to standards track protocols), nor does it
reflect past decisions (for instance, see RFC2774 and RFC4437).
It is not clear to me how people can so consistently misread RFC
2026, which is the current definition for when something should be an
Experimental RFC. Both examples you give are poster children for the
Inappropriately Labelled RFC Foundation.
(sorry for getting off-topic)
Not off-topic at all, given that we will probably have more
extensions in the future.
--Paul Hoffman, Director
--Internet Mail Consortium
