Re: Unhappy with the complexity of pop-imap-downgrade

>> 1.  Since downgrades aren't reversible, there isn't really an interop
>> reason to specify the details.  Say that for headers that contain UTF8
>> text, do some combination of MIME encoding where possible (most
>> headers), redacting parts of them (address headers), and throwing them
>> away (trace headers).
>
> Current draft.
> Tried not to break current MUAs. (Except allowing empty from:).

Agreed, but we could strip out a great deal of the detail and just say 
here are the three techniques you can use to fix headers, use whichever 
ones you want.  Since the transformations aren't reversible and the user 
can't reply to a message with an EAI return address, the details don't 
really matter.

>> 2.  To deliver an EAI message to a non-EAI MUA, wrap it as a
>> message/global, copy whatever headers you can into the wrapper
>> message, and it's the MUA's problem.
>
> Do you think that end user understands what happend?

I don't know, but I don't see any reason to think that it's more or less 
baffling than a message full of Downgraded-this and redacted that headers. 
The IETF has never been very good at user interface design.

>> 4.  For the rest of EAI, we made the decision that the EAI and legacy
>> mail streams are parallel but separate, if you want to send an EAI
>> message and there isn't an EAI path to the recipient, you lose.  Why
>> should this be any different?  If the user doesn't have an EAI MUA, he
>> can't pick up mail from an EAI POP or IMAP server.
>
> Any MUA can access EAI POP servers which offer EAI messages for the MUA.
>
> Port number and protocols are the same.

We could simplify section 3.1 of RFC5721bis and section 4 or RFC5738bis, 
so that if the client hasn't said UTF8, requests to retrieve EAI messages 
fail.  They currently offer a choice between fail and downgrade, just take 
out the downgrade choice.

Regards,
John Levine, johnl <at> taugh.com, Taughannock Networks, Trumansburg NY
"I dropped the toothpaste", said Tom, crestfallenly.