[11566] in www-talk@info.cern.ch
obeying the protocol
daemon@ATHENA.MIT.EDU (William F. Hammond)
Thu Apr 6 12:40:16 2000
From: "William F. Hammond" <hammond@csc.albany.edu>
Date: Thu, 6 Apr 2000 12:32:09 -0400 (EDT)
Message-Id: <200004061632.MAA09832@hilbert.math.albany.edu>
To: www-talk@w3.org
Cc: onank@kanald.com.tr
Resent-From: www-talk@w3.org
Usal Onan KARAGOZOGLU <onank@kanald.com.tr> writes to www-talk@w3.org
on Thu, 6 Apr 2000 16:10:08 +0300:
> What does the extension *.stm stand for? And what does it do?
Suffixes have NO MEANING across the network. Most end-user clients
(i.e., browsers) offer a page-information function so that a user can
determine the content-type of an item served through http.
That topic is, in fact, not appropriate for this list.
But perhaps it *is* appropriate for me, with slight change of topic,
to complain about a user client (which I happened to find on a piece
of hardware purchased new three months ago) disregarding the http
content-type header in favor of what it perceived as a "suffix" in a
url.
Misbehavior is compounded into absurdity by the nature of this
particular inferred suffix, which is ".sgml" served with content-type
"text/plain", and, which, apparently, this client thinks is something
it knows how to render as other than text/plain, with the result that
much of the text/plain content is lost.
I tolerated this when I saw it in 1996, but it has been wrong since
the dawn of HTTP/1.0, and I find it outrageous now.
Beyond that I believe that there is a tacit understanding that no
general purpose user client will ever get any closer to internal
handling of "text/sgml" than for internal handling of "text/xml".
Isn't that exactly why XML is sponsored by W3C?
It should never be assumed that "text/sgml" is "text/xml".
Absent user (or site) client configuration for an external
application, an item served as "text/sgml" should be reported as not
renderable and the user offered a "download".
And let's hope that user clients in general avoid being territorial
about "text/xml" by always honoring a user (or site) configuration of
an external application for that content-type.
William F. Hammond Dept. of Mathematics & Statistics
518-442-4625 The University at Albany
hammond@math.albany.edu Albany, NY 12222 (U.S.A.)
http://www.albany.edu/~hammond/ Dept. FAX: 518-442-4731
Never trust an SGML/XML vendor whose web page is not valid HTML.
And always support affirmative action on behalf of the finite places.
