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

Re: making progress on State-Info

From: Sankar Virdhagriswaran, Crystaliz Inc. <sankar@fcrao1.phast.umass.edu>
Date: Sat, 09 Dec 1995 18:37:31 +0000
To: Koen Holtman <koen@win.tue.nl>
Cc: "Sankar Virdhagriswaran, Crystaliz Inc." <sankar@fcrao1.phast.umass.edu>, koen@win.tue.nl, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Message-Id: <9512092311.AA18286@fcrao1.phast.umass.edu>

>I think there is some confusion here.  The discussion was on whether
>the state-info mechanism proposed by Dave Kristol should be extended
>with a `path=' attribute as in Netscape cookies.  Someone suggested
>that this might allow for more session state to be kept without
>resorting to server side databases, if I remember correctly.  I
>remarked that I failed to see how that `path=' would allow this.
>

Sorry about the confusion. I agree with what Koen has to say. Also, servers
(or atleast CGI bin scripts) may choose to keep state. For certain class of
applications that is the right thing to do. The current proposal, as I see
it does not preclude it. In fact, Dave in his section on implementation,
attempts to bring up the issue and does not explicitly forbid it. He does
the mention the problems of some CGI bin script maintaining state between
sessions. Well, such is life in the state oriented, transaction world. They
all eventually become hairy beasts to maintain.

 
Sankar Virdhagriswaran                         Phone: (508) 287 4511
Crystaliz Inc.                                 Fax:   (508) 287 4512
Received on Saturday, 9 December 1995 15:27:53 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:37 EDT