Re: Last Call: draft-klensin-rfc2821bis (Simple Mail Transfer Protocol) to Draft Standard (4)

John C Klensin wrote:

 [2822 vs. 2822upd] 
> This is, of course, the sort of thing that can be changed in an
> RFC Editor note if, in fact, 2822 is ready and more or less
> concurrent.

Modifying the reference is simple.  Removing various NO-WS-CTL,
DEL, and a NUL in 2821bis itself, not inherited from 2822(upd),
for consistent usage with 2822upd (and net-utf8) isn't trivial.

Maybe the interoperabilty report will say "nobody uses it, and
some implementations don't support it".  It could also arrive
at the opposite conclusion, "most implementations survive it,
some even use it" (I'd be curious for *what*).  Or the likely
outcome "who knows, nobody mentioned NO-WS-CTL and NUL in such
obscure places as greetings or domain literals".

Obscure domain literals didn't work everywhere, and that's a
"simple" 2822upd <dcontent> case.  Where it's not yet clear
what 2822upd will do, move it to "obs" or to /dev/null.

If 2822upd moves it only to "obs", then 2821bis needs its own
syntax without NO-WS-CTL, because NO-WS-CTL in IPvFuture (or
rather the 2821bis name for IPvFuture) is already known to be
neither used nor interoperable.

For the other cases of DEL, NO-WS-CTL, and NUL we haven't seen
any tests so far.  For various 127 in the ABNF it's plausible
that they are mere typos, excluding CTL but allowing DEL makes
no sense (i.e. there would be a comment if that's intentional).

 Frank

_______________________________________________
Ietf mailing list
Ietf@ietf.org
https://www1.ietf.org/mailman/listinfo/ietf