ietf-822
[Top] [All Lists]

Re: I-D Recommendations for Automatic Responses to Electronic Mail

2002-07-02 01:46:07

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

On Thursday 06 June 2002 22:15, Russ Allbery wrote:
Keith Moore <moore(_at_)cs(_dot_)utk(_dot_)edu> writes:
Also, the SHOULD use "in-reply-to" and MAY use "References" is
silly. In-Reply-To fields are enormously varied and hard to parse;
"References" fields are much better.  We should definitely
encourage people to support References over In-Reply-To.

Seems like in-reply-to is more widely supported, though.  I'm
curious to hear what other people think about this.  It wouldn't
bother me too much to make both of them SHOULD.

References is definitely more widely supported than In-Reply-To for
actual practical purposes.  Many, if not most, In-Reply-To headers
contain so much unstructured, extraneous information that they're
essentially only informational to humans and cannot be effectively
used by software. References has never had this problem, and is
therefore clearly superior technically.

I wholeheartedly agree that we should be recommending References over
In-Reply-To at every opportunity.

The "problem" is that some mailers (e.g. KMail) depend on in-reply-to to 
thread messages. Yes, kmail should use References, too, but...
One of the problems is that the IMAP fetch envelope command only returns 
in-reply-to, not references:

RFC2060:
                     The fields of the envelope structure are in the
                     following order: date, subject, from, sender,
                     reply-to, to, cc, bcc, in-reply-to, and message-id.
                     The date, subject, in-reply-to, and message-id
                     fields are strings.  The from, sender, reply-to,
                     to, cc, and bcc fields are parenthesized lists of
                     address structures.

And fetching the envelope structure (or ALL) is so much faster than even 
(ALL BODY[HEADER.FIELDS (REFERENCES)])

So it should be "SHOULD" for both "references" and "in-reply-to".

BTW: The only misuse of In-Reply-To I've seen is from early versions of 
mutt 1.2.x or 1.3.x.

Marc

- -- 
Marc Mutz <mutz(_at_)kde(_dot_)org>
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.0.7 (GNU/Linux)

iD8DBQE9IWg73oWD+L2/6DgRAo8AAKCgjJzSqwW8Tq1Ux4lUhbYDwpIOrACdGuyl
XyNBdy9ZRAhcRGT3wXSdfUs=
=3ymO
-----END PGP SIGNATURE-----


<Prev in Thread] Current Thread [Next in Thread>