Dave Aronson (davearonson@bigfoot.com)
Wed, 14 Mar 2001 18:08:31 -0500
"Eric Burger" <eburger@snowshore.com> wrote:
> We rejected making critical content a parameter to Content-Disposition,
> since people felt that would bend the semantics of Content-Disposition
> a bit too much.
And it bends those of Content-TYPE *less*? Methinks I missed this
discussion; there seems to be nothing relevant in the past several months
of VPIM WG mailing list archive. (Found the archives, thanks, Glenn!) If
you mean in the draft itself, I don't see much logic there beyond that it
"overloads the meaning of the entity". I'll admit I'm not intimately
familiar with the original intent of that entity, but it seems to me that
criticality is indeed an instruction on disposition. One obvious case is
that if something with criticality of CRITICAL can't be rendered, DISPOSE of
the message. B-) Sorry if I'm beating a dead horse, but it was dead when I
got here, honest, officer!
> The argument for making critical content a parameter to Content-Type
> is that all IMAP servers return Content-Type parameters in the
> BODY/BODYSTRUCTURE methods. They would not have to implement a
> BODYSTRUCTURE extension to return Critical-Content.
"Because it was there" may have worked for Hillary (Sir Edmund or Clinton,
take yer pick), but I don't think that's quite the sort of deep technical
analysis usually expected of the IETF. B-)
> The argument against making critical content a parameter to Content-Type
> is that it goes directly against two paragraphs in RFC 2045:
>
> There are NO globally-meaningful parameters that apply to all media
> types. Truly global mechanisms are best addressed, in the MIME
> model, by the definition of additional Content-* header fields.
>
> a CRITICALITY parameter is a globally-meaningful parameter.
I'm not sure I quite grok what you mean here. It's not
"globally-meaningful" in the sense of applying to the entire message; it is
only for one piece. (Sounds like it's probably got some locally-specific
meaning explained in the RFC; I'll go check it later, as I'm writing this
offline.) However, the other argument against it:
> A Content-Type parameter is supposed to be for
> modifiers to the Content-Type only:
is what knocks it out for me. The criticality has no relation to what type
of thing is actually inside -- especially since there could be parts of the
exact same type and different criticality, such as Voice-Message versus
Spoken-Receipient. (Sure, you care if the message itself gets through, but
do you really want to communicate with anyone who has to be reminded who he
is? B->) Therefore, at least IMHO, it should not be in Content-Type.
> Do we really care?
Enough to argue about it -- but that doesn't say much! B-)
> I'd like to hear from IMAP server vendors.
Well, seeing as I'm currently Between Positions (see sig) and might possibly
get hired by an IMAP server vendor....
-- Dave Aronson, Sysop of AirNSun free public Fidonet BBS @ +1-703-319-0714 The above opinions are MINE, ALL MINE, but for rent at reasonable rates. I'VE BEEN LAID OFF AGAIN; see http://djajobhunt2001.tripod.com for info.
This archive was generated by hypermail 2.0b3 on Thu Mar 15 2001 - 01:04:33 IST