From: John Stanley (stanley@peak.org)
Date: Wed Mar 31 2004 - 13:47:32 CST
Seth Breidbart (sethb@panix.com):
>"Re: " doesn't make pseudo-threading easier than no back reference.
>It does two things: (1) It specifies that the Subject wasn't changed
No, it does not.
>(yes, the user can "fool" this heuristic; tough. The user can break
>anything.)
The "user" isn't "fooling" anything. The "Re: " simply does not mean what
you claim it "specifies". The fact that the user can change the subject
while still keeping the "Re: " on the front proves that it does not mean
what you claim it does. At best, you can claim:
"If a subject starts with 'Re: ', and the poster uses a followup
agent that automatically inserts 'Re: ' when copying the subject
into the proto-article, and the user has not changed the subject,
then the subject hasn't changed."
Note that the entire statement depends on the last clause "and the user
has not changed the subject".
> (2) Specifying "Re: " is the strongest way to prevent the
>use of _other_ back references, which _do_ break things (from the
>viewpoint of someone reading Usenet).
And now we are back at the claim that the subject "Sv: something" actually
breaks USENET, which nobody has yet been able to prove. The only things
that break when presented with legal unstructured content in an
unstructured header are inherently broken and ought to be fixed, not
codified into a standard as normal.