ietf-822
[Top] [All Lists]

RE: text/paragraph or wrap=yes/no

1998-03-13 14:50:45
At 1:58 PM -0600 3/13/98, Dave Crocker wrote:
Further, messing with text/plain by adding a parameter scares me a bit.
Text/plain is so fundamental to MIME's deployment that I believe we should
not mess with it

I agree with you that it shouldn't have been messed with.  But Ned already
messed with it in 2046 by adding the "displayable without wrapping"
requirement.  Call it a clarification, say it was what you really intended
all along, fine.  It still remains that formally compliant implementations
became formally non-compliant when he did that.  I'm no longer asking for
this to be rolled back, I'm just pointing out that text/plain has already
been messed with.

even if the argument in favor of a text/plain parameter
were strong, which it isn't.

A large part of the user community would be significantly inconvenienced by
text/paragraph.  I suspect least as large a part as are currently being
bothered by paragraph text in text/plain (go look at the list of mailers
failing to display text/* easily, and count their users).

I understand the desire to punish these non-compliant implementations.
Unfortunately, the old saw "this is going to hurt me more than it hurts
you" comes into play.  They're too big for me to punish.

So I will not at this point in time generate text/paragraph.  However, I
will continue to send unwrapped text when users instruct me to.  I would be
happy to label this with a parameter in either content-type or
content-disposition or anywhere else you please (how about message-id?
Received?  :-)).  But I will not send it as text/paragraph until there is
better compliance with the text/* rule (at which point I have no objection
to it at all).

The interoperability problem is in the near term better solved by a
parameter, not text/paragraph.  The only problems I've heard with the
parameter are fairly vague procedural misgivings, and a wish that nobody
would send paragraph text.  I understand the former, but I think it's
outweighed by the practical disadvantage of text/paragraph, at least for
now.  As for the latter, I just don't agree.

<Prev in Thread] Current Thread [Next in Thread>