On 3/13/98 at 4:57 PM -0600, Ned Freed wrote:
It was my understanding that you objected to the requirement that agents
generating wrap=yes also be able to generate wrap=no, or at least questioned
the need for it. If that's not the case then we are basically in agreement
and can proceed with the wrap parameter and a requirement that it be possible
to generate wrap=no material.
I did question the need for the requirement, but I didn't object to it per
se. Here's my reasoning, though you can all skip down to the bottom of the
message if you just want the conclusion:
Right now, the MIME documents make no claim about what the number of
columns that you are required to keep your short lines too; the only thing
it says is that you're supposed to send them short enough such that "It
should not be necessary to add any line breaks to display "text/plain"
correctly" (2046). My claim all along (with Laurence, Steve, and others I
believe) is that this turns out to be an impossible requirement to meet
because senders can't tell how short to make the lines from this dictum. At
best, you can follow the rule, "Lots of people won't have trouble if you
keep them below 80 columns, but some will". For some class of clients, it's
going to be necessary to add line breaks in order for many text/plain
messages to "display correctly". Arbitrarily putting a dividing line at
clients who have 80 column displays and don't do any further wrapping is
fine, but let's be sure we understand that this is what's going on. And
this failure of 2046 is going to be present whether we do text/paragraph or
a wrap parameter.
If there is consensus that 2046 really meant "short lines, meaning less
than 80 columns", then this means we have to add a requirement to the wrap
parameter document saying "You MUST still be able to generate short lines,
meaning less than 80 columns, and wrap=no". That's fine. But understand
that text/paragraph is going to have to make the same clarification: "2046
says that you still have to generate text/plain and that means you still
have to be able to generate short lines, and that has always meant less
than 80 columns." If text/paragraph doesn't need to make that clarification
to 2046, then there's no need for the requirement statement in the wrap
document; either 2046 needs clarification on what it allows in text/plain
or it doesn't.
All that said, I think you have heard now from at least Steve and I that we
have no objection to the wrap=no requirement if consensus is that this
clarification of 2046 is necessary; we have all the support to do that in
Eudora now and we don't think it's any big burden on other implementations.
You have heard objections to text/paragraph due to its serious
interoperability problems. So I think we can proceed with the wrap
parameter as you stated and abandon text/paragraph.
pr
--
Pete Resnick <mailto:presnick(_at_)qualcomm(_dot_)com>
QUALCOMM Incorporated
Work: (217)337-6377 or (619)651-4478 / Fax: (217)337-1980