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

Re: Idempotent

From: Henrik Frystyk Nielsen <frystyk@w3.org>
Date: Thu, 31 Aug 1995 10:08:53 -0400
Message-Id: <9508311408.AA10389@www20>
To: Harald.T.Alvestrand@uninett.no
Cc: Paul Leach <paulle@microsoft.com>, http-wg-request%cuckoo.hpl.hp.com@hplb.hpl.hp.com, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com 

> Why not use "cacheable" instead of "idempotent"?

No, that doesn't work - "cacheable" refers to the object, "idempotent" refers 
to the method. I would rather say that an idempotent method does not change 
the topology of the Web.

The reason for using "topology" and not "state" is that in some cases, 
idempotent requests can change the state of the web - especially if log files 
are considered as a part of the state of the Web, or for example if a document 
can be accessed 5 times.

-- 

Henrik Frystyk Nielsen, <frystyk@w3.org>
World-Wide Web Consortium, MIT/LCS NE43-356
545 Technology Square, Cambridge MA 02139, USA
Received on Thursday, 31 August 1995 07:19:59 EDT

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:31:27 EDT