Valdis(_dot_)Kletnieks(_at_)vt(_dot_)edu wrote on 21.03.00 in
<200003210834(_dot_)e2L8YaC22436(_at_)black-ice(_dot_)cc(_dot_)vt(_dot_)edu>:
On Mon, 20 Mar 2000 17:59:13 EST, "Simon St.Laurent" said:
It's a plausible risk only if HTTP 1.1 or other protocols are explicitly
reopened and modified. Otherwise, it's pretty easy to point to the spec
and say "No. That's forbidden." It might be a good idea to present these
If you haven't been in the industry long enough that it takes you more
than 30 seconds to think of at least 5 vendors(*) of software who aren't
dissuaded from doing something just because the standard says explicitly
not to do that, you haven't been around enough to see *seriously* broken
software.
Well, in case someone wants to expand the Accept: mechanism to allow this
kind of matching, have a look at the wildmat patterns used, for example,
in NNTP. It's already in a RFC, free source code is available, and it's
compatible with the existing spec.
MfG Kai