Re: [EAI] RE: I-D ACTION:draft-duerst-mailto-bis-03.txt

At 07:35 06/10/28, Frank Ellermann wrote:
>John C Klensin wrote:
>
>> updating mailto when appropriate is an explicit work item for the WG.
>
>Maybe the charter needs an update then,

Yes indeed. I was explicitly raising the question of whether
a mailto: update should be part of the charter or not when
chartering was discussed, and I seem to remember that the
answers were mostly along the lines of "we may need coordination,
but don't make this an item of the WG". I have just reread the
WG charter (at http://www.ietf.org/html.charters/eai-charter.html),
and I have not found any trace or hint of mailto: in there.

Digging a bit, I found the following mail that provides some
additional pointers to the discussion before chartering:
http://www1.ietf.org/mail-archive/web/ima/current/msg00140.html.

>and the mailto draft could be renamed to draft-ietf-eai-mailto-bis.

I would be happy to do this (I can't speak about my coauthors),
but I don't think it's necessary, and I think first there should
be a clear expression of preference from a substantial part of
the WG that we indeed want to change the character in this direction.

>For the 2368bis part trying to
>get in line with 3986 incl. fixes for several 2368 issues it might be
>better to ignore UTF-8 for the moment.  Where <pct-encoded> is allowed
>adding UTF-8 later is no problem.

Frank, here I have to disagree with you quite strongly.
Allowing arbitrary byte salad first and then later trying
to limit that to UTF-8 is a very bad idea. It has happened
by accident a few times, but it would be a big failure to
do that deliberately. Also, it would totally leave implementers
in the dark about what to do with those octets. Take for example
   mailto:abc@example.com;body=abc%ab%cd%efdef
How is an MUA supposed to send this body? What's the charset?

>> If you submit this for standardization now, and the IESG follows 
>> precedent, the document will be referred to the EAI WG for evaluation.

Yes, that's why I'm sending it here, for pre-evaluation.

>There are several issues in the mailto draft even without mentioning
>UTF-8, I18N, or IRIs.  It needs a new clean decrufted 3986-compatible
>version as a solid base before adding new features.

Frank, if you send me a list of things that need to be fixed so
that mailto: gets 3986-compatible, I'll be glad to do all these
fixes. Or do you think they already have been made in
draft-duerst-mailto-bis-03.txt? Then please say so, thanks.

>It's too complex to get this right in one giant jump.

I completely disgaree. There are already some mailers, e.g. Opera
mail, that just do the right thing for %-encoding and UTF-8.
There are others that are waiting for directions. A spec
that allows to send email, but not a word besides US-ASCII
(in the body, where this is possible since ages) is not something
the IETF should approve in this day and age.

Regards,    Martin.

#-#-#  Martin J. Du"rst, Assoc. Professor, Aoyama Gakuin University
#-#-#  http://www.sw.it.aoyama.ac.jp       mailto:duerst@it.aoyama.ac.jp     

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