Re: [EAI] Request for advancement of draft-ietf-eai-{utf8headers, smtpext} to Experimental

["Frank Ellermann" <nobody at xyzzy.claranet.de>]

[mailed separately]

Harald Alvestrand wrote:

> Charles Lindsey and Frank Ellermann have argued strongly
> against a specific point in the specification: The
> transmission of EAI messages in a subtype of "message".

Just in case if it is unclear:  These objections are to
some degree orthogonal.  Charles argues that EAI cannot
overrule an "EXPRESSLY FORBIDDEN" in MIME, while I came
to the conclusion that this is actually an error in MIME,
compare <http://article.gmane.org/gmane.ietf.ima/2150>.

I argue that EAI has no business to update MIME at all,
EAI could solve Charles' issue with an erratum.  I like
message/global, but think its definition is flawed, if
there is no UTF-8 in the header (singular) the message
is implicitly a message/rfc822, not a message/global.

The EAI definition uses a convoluted MIME part parsing
checking headers (plural) of the message and MIME parts,
for the purpose of updating MIME allowing UTF-8 also in
MIME part headers.  IMO it's incompatible with MIME 1.0
and also no "KISS" solution.

> One minor ABNF error has been identified, and will be
> corrected with other Last Call comments.

I hope that covers the erroneous 1*FWS instead of FWS.

 Frank

_______________________________________________
IMA mailing list
IMA at ietf.org
http://www.ietf.org/mailman/listinfo/ima