Keith, I just don't think it works.
The hard reality is that there are lots of mail transport agents out
there that think they have license to mess with line breaks. If
everything gets converted to CRLF format before transport, I predict
that LOTS of existing transport agents will break, badly, because
they've suddenly started getting CRLF where they expected whatever the
local standard was.
This is the problem with confusing the model with what actually happens inside
a real mail system. It is completely possible to implement things so that
CRLF are _NEVER_ substituted into the contents of text messages. I know,
because this is precisely what my mail system does. Yet my system conforms
completely to Keith's model. It also conforms completely to the current
MIME draft.
I believe that not breaking existing transport has always been one of
the highest goals of MIME, which is why I object to this redefinition of
quoted-printable. The existing definition, in the current MIME
document, does not break transport; that's why I propose to keep it,
with the possible addition of explanatory prose of the kind I posted
earlier.
The current definitions are not wrong and I don't propose to change them.
The problem is that they are incomplete -- I think a model is needed to
make them complete. If an ancillary description of how to build an
implementation of this model on a UNIX system is also needed, so be it.
I hate to say it, but if we make the proposed changes to
quoted-printable, I'll be hard-pressed to consider the document as
standard-ready, because we don't have a CLUE as to how many existing
transport agents will be broken by the new definition. I'm not calling
it a show-stopper, but I do think it is the kind of thing that would
make us want to step back and wait a couple months before going back to
the IESG. -- Nathaniel
Since none of this affects the final results of how things get encoded in
any way, I fail to see how it could affect the standards-quality of MIME.
We're only talking about clarifications and explanatory text here. No changes
to how quoted-printable represents output are planned. I would be opposed to
such a thing myself, but I don't see it in Keith's model at all.
Incidentally, it seems that some folks view this discussion as an inpediment
to the progress of MIME to standards status. If this is in fact the case I
want to punt on this stuff NOW and get MIME out there! This material can
very easily be slipped in six months from now.
Ned