W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1995

Re: Expires, Last-Modified, Pragma: no-cache etc.

From: Albert Lunde <Albert-Lunde@nwu.edu>
Date: Fri, 18 Aug 1995 08:42:12 -0500 (CDT)
Message-Id: <199508181342.AA278753332@casbah.acns.nwu.edu>
To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
> This somewhat changes the scope of Pragma, which is no longer restricted to
> proxies in the new paragraph.
> 
> I think the name "Pragma" confuses a number of people (cf Lou's initial
> interpretation), and it's not clear to all whether a Pragma header field is
> destined for proxies only, or for proxies and user agents. Based on the fact
> that not so many WWW clients and servers already implement Pragma, I propose
> that we replace the name "Pragma" with the name "Proxy". Maybe in HTTP/1.1
> specs ? It would be crystal clear that 'Proxy: no-cache' is not destined for
> user agents.

When I say the "Pragma" header orginially, my first throught was #pragma
in ANSI C: an escape hatch to provide implementation-specific information.

Now, this was added originally for the benifit of the CERN proxy server,
and the language of the specs reflect this, but I'm not sure we need
to stick with this if we can be clear what the scope of particular
Pragmas is... the nature of this sort of thing is that implementations
will (and usually should) ignore/discard unknown pragmas.

I _would_ like us to use some kind of current-practice or best-current-practice
interpretation of Pragma: no-cache and If-Modified-Since: rather
than redefining them in some more complicated/tricky way: thus I
tend to favor the suggestions that add new headers/pragmas to offer
new semantics.

(Should any of this discussion go to www-proxy or is that a dead list?)



-- 
    Albert Lunde                      Albert-Lunde@nwu.edu
Received on Friday, 18 August 1995 06:43:36 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:25 EDT