Pete Resnick wrote:
- Remove NO-WS-CTL from ctext, qtext, utext, and dtext.
(That makes utext simply VCHAR.)
A kind of early "seasonal" gift, utext is simply VCHAR.
No more Escape-sequence-tricks in subjects, half of the
security considerations can go to the obs-section (yes,
I know, there were no security considerations about it)
- Create obs-ctext, obs-qtext, and obs-dtext which
contain NO-WS-CTL.
<obs-dtext> didn't exist in practice, IMO you can skip
this step.
(Question: Should they contain NUL as well?)
Not as part of the NO-WS-CTL cleanup, NUL was already
an obs-cenity in a few places, not a NO-WS-CTL. It's
in obs-qp (=> no new effect for 2822upd), obs-char,
(indirectly) in obs-text, and therefore (indirectly)
also in the existing <obs-utext>.
Keep NUL where it is if the "interoperabilty report"
shows that NUL is widely implemented and understood.
(But my old Netscape 3.x froze when it found a NUL)
Do we think this will recycle at Proposed?
Obsoleting cruft is allowed. I'm looking forward to
a justification why NUL is kept as an obs-cenity, it
ought to be "appeal-repelling" or something.
Frank