On Jan 15, 11:34pm, D. J. Bernstein wrote:
}
} Bruce Lilly writes:
} > header field order can (and does) change, so there's no way of
} > guaranteeing that a header-encoding header-field would always precede
} > use of any specified encoding
}
} So what? Does your MUA also have trouble displaying From before Subject
} when the From line in the message comes after the Subject line?
As I recall the discussion at the time (circa RFC 1342), the issue was not
the ordering, but rather that each header field needed to be interpreted
in isolation, because assorted transports (and maybe even some MUAs) could
not be counted upon to pass through a header field with an unknown name,
or to pass it undamaged. (It was some while before senders really had any
confidence that the RFC 1341 [no, that's not a typo] headers would reach
all recipients, unless I'm misremembering.)
There actually was a proposal, coauthored, if memory serves, by Bob Smart and
myself, that defined an approach using multiple header fields.
The consensus at the time was that Keith's approach, which came out later, was
clearly superior, both for the reasons you gave and because it ended up being
quite a bit simpler. Again if memory serves, Bob and I agreed with that
concensus and withdraw our document from consideration.
Of secondary interest, but another reason for not using a separate field,
is that the encoding can represent multiple character sets and languages
in that same single header. (Obviously that's not so significant if a
"universal" character set is available, but language information is still
of interest even if it isn't absolutely essential.)
Our proposal at the time handled this to some extent, but at the cost
of considerable complexity.
...
It's one thing to accept a risk to your own data, but quite another to
standardize on something that imposes that risk on others, no matter how
unlikely you think it is that anything "really bad" will happen, and no
matter how desirable the outcome once all the breakage is "outgrown."
That, in my view, is why the proposal as it stands is not acceptable, and
is the issue that has to be addressed before an acceptable proposal is
possible.
Very nicely put. I agree completely.
Ned