[8905] in www-talk@info.cern.ch
Re: Session-ID proposal
daemon@ATHENA.MIT.EDU (Jim Seidman)
Mon Aug 14 10:43:11 1995
Date: Mon, 14 Aug 95 09:40:35 -0500
To: dmk@allegra.att.com (Dave Kristol)
From: Jim Seidman <jim@spyglass.com>
Cc: www-talk@www10.w3.org, http-wg@cuckoo.hpl.hp.com
Resent-From: www-talk@w3.org
At 05:48 PM 8/10/95 EDT, Dave Kristol writes:
>I have assumed (erroneously?) that a caching proxy must send
>conditional GETs to the origin server. If so, there's already the cost
>of a connection. The State-Info (previously "Session-ID") can ride the
>request almost for free.
This isn't necessarily true. If the server sends "Expires" headers, then
the proxy can pretty safely avoid sending if-modified-since requests, for
example.
Even without "Expires" headers, some caching proxies just use simple
algorithms to avoid doing a GET on each request, e.g. only re-requesting
text/* files if they're at least an hour old. This applies not only to
proxies, but also to clients with persistent caches such, if I have a
document still cached on my hard-drive from an access yesterday.
Given these considerations, and the slowly increasing use of "Expires"
headers, State-Info could be expensive indeed.
--
Jim Seidman, Senior Software Engineer
Spyglass Inc., 1230 E. Diehl Road, Naperville IL 60563
