On Tue, 2002-04-30 at 20:04, Paul Prescod wrote:
>  Amelia A Lewis wrote:
>  > You *are* aware that HTTP is "not a MIME-compliant protocol"?  Section
>  > 19.4.1, first sentence.  Although it's possible to do MIME-like (to use
>  > the phrase from the RFC) stuff with it, it isn't the same.
>  >=20
>  > As long as the processors are packing and unpacking if they move betwee=
n
>  > protocols, it isn't that much of an issue.  But it isn't ... exactly ..=
.
>  > MIME.
> =20
>  I should have known to be more careful on xml-dev. ;)
> =20
>  AFAIK, all of the relevant strengths and weaknesses of MIME survive the
>  slight differences unaffected.

Depends on your use case.

In single-protocol usage it's fine.  Slight difference in construction
or decode.

In multi-protocol environments, you're usually fine, unless you
encounter the famous edge cases.  Most likely for http entry, mime
exit.  That is, since http is more-or-less a sort-of qualified superset
of mime, you can lose information crossing the boundary from http to
mime transport, unless you're aware of the differences.

Don't mean to be pedantic; I've been writing MIME code for the past
month.  There is a set of interop problems associated with Soap with
Attachments (or another protocol relying on MIME and HTTP with an
expectation that data will cross those boundaries) that I haven't seen
addressed, as yet.

Amy!
--=20
Amelia A. Lewis       amyzing@[...].com      alicorn@[...].com
Never imagine yourself not to be otherwise than what it might appear to
others that what you were or might have been was not otherwise than what
you had been would have appeared to them to be otherwise.
                -- The Duchess [Lewis Carroll]