On Wed October 13 2004 04:29, Nick Ing-Simmons wrote:
Bruce Lilly <ietf-822(_at_)imc(_dot_)org> writes:
Evidently you fail to grasp fundamental logic, the nature of the
problem with RFC 2822, and the nature of software development.
If we are going to be patronising
No "patronising": simply pointing out (repeated) flaws
in Mr. Lindsey's "reasoning" (using the term loosely).
you have failed to grasp the
nature of e-mail communication in the real world.
The issue isn't limited to email.
In the real world we are lucky if mail tools even attempt
to conform to the spirit of an RFC let alone the letter.
Nevertheless we have specifications for how things
are supposed to work, to ensure that conforming
implementations interoperate (and as a side-effect
to be able to identify specifically what is wrong with
a non-conforming, non-interoperable implementation).
The fact that there are lousy implementations does
not itself indicate that the rules should be discarded;
the lousy implementations should be discarded if they
cannot be repaired.
On the other hand the consequences if using the wrong sort of compare
for these fields seem to be rather unimportant.
Beware the rule of unintended consequences; such
a poor choice is likely to cause a problem at the
most inopportune moment.
Doing case sensitive compare "seems to work well enough" even when it is a
domain name.
You are certainly entitled to your opinion, just as
a certain well-known headstrong, swaggering dumbass
was entitled to his opinion that dropping bombs "seems
to work well enough" when what was called for (literally)
was merely a little more diligence in determining
that certain alleged objects which were claimed to
have existed in a certain region in fact did not exist.
Others do not necessarily share that "seems to work
well enough" attitude. Still others may suffer the
(possibly) unintended consequences of such a choice.