--On Friday, October 09, 2015 02:33 +0000 l.wood@xxxxxxxxxxxx
wrote:
>> RFC 2152 may be informational, but it is normatively
>> referenced from a standards-track document.
>
> (which one?)
RFC 3501, the base IMAP4 Rev 1 document.
> well, that makes a mockery of the whole
> informational/standards-track distinction.
"Mockery" is a little strong, don't you think? While I
personally believe we are almost always better off reclassifying
or reissuing/updating informational or lower-maturity-level
documents that are needed to support newer standards track ones,
we have, IIR, pretty much always had mechanisms for making
exceptions and allowing so-called "downward references".
> The RFC Editor errata process is really just for determining
> (formally) if there is an error. After that, it can be pretty
> open; and community consensus is not formal.
You may think so but that process assigns proposed errata to a
category. The mechanism used to do that typically involves a
discussion with authors and/or a small number of expert and
then, for IETF Stream documents, a discussion within the iESG,
not the broader community. In recent years, when an error has
not been completely obvious and/or trivial or changes the
technical subject matter of the spec, the tendency has been to
assign the category "hold for future revision", which is
completely useless as far as informing the user about what to
do.
As has already been pointed out, it is clear that the text of
2152 is not as clear and precise as one might like. The
question is what should be done about it in terms of what is
required and/or recommended. The errata process, as now
constituted, is not good at answering questions like that.
best,
john