From: John Stanley (stanley@peak.org)
Date: Fri Dec 13 2002 - 11:09:24 CST
Charles Lindsey (chl@clw.cs.man.ac.uk):
>Remember that this issue started when someone posted with
>Followup-To: Poster in one of the uk.* groups, and there were reports of
>at least two systems that treated it case sensitively, and therefore
>failed to do the followup.
And then:
>Note also that my revised text says that agents MAY be liberal and treat
>it as case insensitive (no harm can arise, because single-component
>newsgroup-names are forbidden anyway).
I guess rejecting an article because the user's agent has treated the
Poster value as case-insensitive (as this draft now permits it to do)
isn't what would be classified as harm, huh?
Henry Spencer (henry@spsystems.net):
>Backward compatibility is an entirely sufficient reason, given that there
>is no benefit from making it case-insensitive.
I guess not rejecting articles where an agent has treated the value as
case-insensitive isn't a benefit, huh?
Matt Curtin (cmcurtin@interhack.net):
>My case here is based on the value of backward compatibility, and the
>supposition that USEFOR is a standards-track document (documenting
>existing behavior).
Yes, that's right. Existing behaviour is apparently that this value is
case-insensitive for the users. Otherwise there would have been no report
that the value "Poster" was being rejected. In other words, both natures
are "existing practice". Which of the two is a) more permissive and b)
less damaging? Case-insensitive. An agent that treates the entry as
case-insensitive will never reject the entry, even if the generating agent
is broken and treats it case-sensitively.
> ... we need to rate backward compatibility and documentation of
> what is currently happening high priorities.
Yep. Since it is more compatible to treat it as insensitive, let's do
that.
Henry Spencer (henry@spsystems.net):
>Why? The fact that users may be entering the wrong thing by hand ...
Why should it be the wrong thing? The only reason that it is the wrong
thing is because the wrong person is arguing that it should be the right
thing. There is no technical reason for it to be the wrong thing. If there
is no technical reason, and there certainly is no other reason, why should
there be an artificial limit?