From: Ralph Babel (rbabel@babylon.pfm-mainz.de)
Date: Wed Apr 18 2001 - 15:33:16 CDT
Russ Allbery wrote:
> the user can arbitrarily edit the Subject header.
> Including removing Re: and inserting Sv:. And there's no
> way for the protocol to distinguish between user behavior
> and software behavior. That makes the MUST NOT pretty hard
> to justify, as much as I hate that bogus mistake too.
Right, and this applies to many
other items in "our" draft as well.
For instance, there's no way to tell whether an article that
does not contain a "References:" header and the "Subject:"
header of which starts with "Re: " was generated this way
deliberately by a user starting a new thread or by broken
software that discards all references.
Similarly, users are free to introduce their silly
quotations etc. at the end of their articles with anything
they please (or not at all), so - from a protocol point of
view - it's pointless to insist ("MUST") on the signature
delimiter being dash-dash-space; if no space follows the
"--", it's simply not a signature. Tough, but not at all
a protocol issue, so RFC 2119 keywords MUST NOT be used.
What's so difficult about simply _describing_ a convention
("a signature starts with '-- '") instead of insisting on an
unenforceable rule ("a signature MUST start with '-- '")?
Section 4.3.2 ... *sigh* ... don't get me started ...
"Give me a Standard large enough and a Committee to
discuss it, and I will prevent the Earth from moving."