--On Wednesday, May 11, 2011 21:53 +0000 John Levine
<johnl(_at_)taugh(_dot_)com> wrote:
...
RFC 5322 has the definition of atext, so there's no dangling
reference. This is at most an editorial issue, where it might
have been a good idea to be more specific about what symbols
are imported from where, so those of us doing a grep for
"atext" would find it.
And, if we ever get around to doing 5321bis ("two maturity
levels" is, in some respects, an argument against doing so in
the absence of significant errors), explicit identification of
all production definitions are imported from other documents,
and which documents they are imported from, is already on the
list.
Importing those things from (now) 5322, rather than duplicating
them, was an explicit decision made during the DRUMS period.
One can, of course, debate whether that was the right decision.
But, it is, somewhat more specific identification of sources
than is provided by Section 4.1.2 would be useful to some people
and harmless to others (as long as we do it correctly). It was
actually suggested during the IETF Last Call on 5321; we decided
to not do it at that stage because we were concerned that doing
so at that late point in the process would introduce errors.
john