W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > May to August 1996

Re: New document on "Simple hit-metering for HTTP"

From: Jeffrey Mogul <mogul@pa.dec.com>
Date: Wed, 07 Aug 96 16:04:41 MDT
Message-Id: <9608072304.AA01934@acetes.pa.dec.com>
To: Paul Leach <paulle@microsoft.com>
Cc: "'http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com'" <http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com>
X-Mailing-List: <http-wg@cuckoo.hpl.hp.com> archive/latest/1237
    However, I think I could be convinced that there only needs to be
    one count per Etag. The idea was that even if there were only an
    English version of a page, having the server set "Vary:
    Accept-Language" would cause separate counting of hits for users
    with "fr" "de", etc, so that the potential for these languages
    could be gauged without creating pages for all of them. Being able
    to count them is, I think, a GOOD THING; but maybe one could create
    a separate Etag without having to create a separate physical copy
    of the page.  We'll think about it.

I don't believe this would be a good idea.  As far as I can tell, we
have no way to tell a cache "these two etags are for the same
entity-body".  That is, there is no possible way to avoid transmitting
the entire entity body twice if it is associated with two different
etags.  So if we were to adopt a "one count per Etag" requirement,
then this would force extra body transmissions between servers and
proxies ... just what we are trying to avoid!

I don't see that the Vary-based scheme we proposed is so difficult
to implement, and it seems to avoid unnecessary transmissions of
bulk data.

Received on Wednesday, 7 August 1996 16:11:07 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:40:17 UTC