ietf-822
[Top] [All Lists]

Re: Metamail & unquoted dots.

1992-07-08 10:54:05
John hits the nail smack on the head.

On 08 Jul 1992 09:55:26 -0400 (EDT), John C Klensin wrote:
Ned and others have explained the issue in terms of people being told by
their bosses "I don't care what the standard is, fix the code to accept
this or I'll find someone who will".  Those are *real* persuasive
arguments; they don't yield to principles of purity, etc.

I *still* get pressure from the few remaining TOPS-20 sites on the net to
`fix' the TOPS-20 mailer to treat bare line feeds as newlines, for the benefit
of broken sendmail.cf files.  ;-(

If one cannot get an agreement from every significant MIME-receiver, now
and in the future, to reject [unquoted periods]

I don't think it will happen.  Thanks to Marshall's opposition, I gave up.  I
now use a dot-omitted tspecials for parsing, albeit still using the dot
version for sending.  I could not justify a hard line on MIME when I have
similar garbage for dots in phrases in the name of interoperability.

You'll have a very hard time persuading me to change it back.

Consequently, I think we are obligated to either
  -- Come up with, and document, a Real Good Reason while dots have to
be on the list, a reason that is persuasive enough that no one really
tries this (or is quick to admit that they have a bug if they do).

No such reason exists.  There isn't any reason for tspecials to exist if dot
is included.  The BNF could be:
        token := 1*<any CHAR except SPACE, CTLs, specials, "/", "=", or "?">
I invented tspecials for the explicit purpose of *not* including dot.  I don't
know why dot was put in, or who did it.

  -- Change the rule.

As far as I'm concerned, the rule is already changed to `quote your dots going
out, but allow unquoted dots coming in' no matter what the MIME document says.
If the MIME document reflects reason and allows unquoted dots, it only means
that maybe a couple of years from now I may feel bold enough to stop quoting
dots on outgoing messages.  I have to think of my installed base, much of
which still has the older version that enforced the restriction.

-- Mark --

PS: here's the latest violation of MIME that's appearing, from several
different sources:
        Content-Type: TEXT/; charset=US-ASCII


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