Re: [EAI] POLL: on the "what MIME type" question

["Charles Lindsey" <chl at clerew.man.ac.uk>]

On Fri, 22 Jun 2007 07:15:49 +0100, Harald Alvestrand  
<harald at alvestrand.no> wrote:

> This poll asks questions about the carriage of messages that make use of
> the extensions described in draft-ietf-eai-utf8hdrs, when they are part  
> of another message. In the text below, we call them UTF8SMTP messages,  
> without prejudice to what we end up naming them.

I must protest most strongly that we are being asked to choose netween two  
alternatives of which one has currently no clearly agreed technical  
description, making it impossible to make an informed choice between them.


> QUESTION 1: Should there be a new MIME type for UTF8SMTP messages when
> carried inside messages in an UTF8SMTP environment?
>
> A: No, message/rfc822 can be used.
> B: Yes.
> C: I have another opinion, which is....

A.

The only argument presented against A is that such messages might leak  
into the existing environment without being properly downgraded first.  
Leaks happen because standards have been accidentally violated, usually by  
unusual hand-written scripts which have not been implemented with the care  
expected when writing agents intended for widespread use.

The question that arises is whether introducing a message/utf8smtp will  
reduce the number of such 'accidents', bearing in mind that current agents  
and the expectations of script writers are unlikely to be aware of the new  
message type. For sure, the number of leaks will indeed be reduced, but  
will the proportion of leaks reduced be significant in comparison with the  
number that remain? Sufficiently so to justify the introduction of the new  
type?

> QUESTION 2: Should the same MIME type be used to carry UTF8SMTP messages  
> in
> a SMTP environment (without the UTF8SMTP extension, and without requiring
> the 8BITMIME extension)?
> This implies that it's possible to encode the message.
>
> A: Yes.
> B: No, that should not be allowed.
> C: I have another opinion, which is....

C. It depends on how well such messages will be propagated intact within  
the present network. This is no problem if the answer to Q3 is 'B', but  
otherwise it appears to be the case that NO current MTA will downgrade  
them from 8bit to Q-P or to Base64 (I have repeatedly asked for a SINGLE  
example of a current MTA that would do so, but none has been forthcoming,  
whereas several current MTAs with significant current usage that will not  
do so have been reported - notably sendmail).

If it were already the case that many/most MTAs already handled this case,  
then the implied violation of that EXPRESSLY FORBIDDEN in RFC 2045 could  
perhaps be overlooked; but apparently that is NOT the case.

Therefore, the only possible answer to this question is 'A' if the answer  
to Q1 is 'A', and 'B' if it is not.

Alternatively we could state explicitly that this new message type was not  
expected to propagate through non-8BITMIME environments (as seems to be  
suggested in the latest DSN draft); but that is outwith the scope of the  
question as asked.

> QUESTION 3: Should the new type be a subtype of Message/?
>
> A: Yes
> B: No, it should be application/something
> C: I have another opinion, which is....

C. 'A' if the answer to Q2 is 'B', otherwise 'B'.

But 'B' is ugly and counter-intuitive. And indeed the ugliness of 'B' is  
the best argument for not anwering 'A' to Q2.

> QUESTION 4: What should the new MIME type be called?
>
> A: Message/utf8smtp
> B: Message/i18n
> C: Message/international
> D: Message/global
> E: Message/rfcxxxx (to be assigned)
> F: Message/rfc822 (consistent with an "A" answer on question 1)
> G: Message/utf8
> H: Message/rfc822u
> I: Message/mail
> J: Message/i18n-email
> K: Message/eai
> L: Message/smtp
> M: Message/smtp8
> N: Message/utf8eai
> O: Message/utf8-headers
> P: Message/utf8-email
> Q: Message/ima
> R: Message/intl-email
> S: I have another opinion, which is....

Presumably if Q1 is anwered with 'A', then this Q4 is redundant.

Otherwise, it needs to include the word 'mail' plus either 'utf-8' or  
'international' (or variations on them such as 'email' - even 'smtp' - and  
'i18n').

So that restricts the choice to one of A, J, O, R, of which I prefer O  
(except that has been pre-empted by the new DSN draft - but that is  
fixable, and should have been a text type anyway).

--
Charles H. Lindsey ---------At Home, doing my own thing------------------------
Tel: +44 161 436 6131 ;                      
   Web: http://www.cs.man.ac.uk/~chl
Email: chl at clerew.man.ac.uk      Snail: 5 Clerewood Ave, CHEADLE, SK8 3JU, U.K.
PGP: 2C15F1A9      Fingerprint: 73 6D C2 51 93 A0 01 E7 65 E8 64 7E 14 A4 AB A5


_______________________________________________
IMA mailing list
IMA at ietf.org
https://www1.ietf.org/mailman/listinfo/ima