W3C home > Mailing lists > Public > www-talk@w3.org > May to June 1999

Re: Tracking hits without cache-busting

From: Koen Holtman <koen@win.tue.nl>
Date: Tue, 29 Jun 1999 13:37:54 +0200 (MET DST)
Message-Id: <199906291137.NAA05929@wsooti08.win.tue.nl>
To: liberte@w3.org (Daniel LaLiberte)
Cc: mcmanus@appliedtheory.com, danielh@econ.ag.gov, www-talk@w3.org
Daniel LaLiberte:
[...hit metering..]
>One workaround might be to send out all documents with the following header:
>  Cache-Control: must-revalidate, maxage=0

Picking a nit: it is max-age=0, not maxage=0.

>Then compliant clients should always make If-modified-since requests to
>the origin server.   I'm using this at hypernews.org, but not for hit
>counting - just because any message page might have a new reply.

Note that 'must-revalidate' is the strongest HTTP/1.1 revalidation
directive you can use on a proxy.  It is meant to be used primarily if
a failure to revalidate (for example because the link from the cache
to the server is down) gives incorrect application behavior.  A
friendlier way of cache-busting to do hit counting would be to use
max-age=0 only, or s-maxage=0 (see rfc2616).  And as always, leave out
a the Last-Modified header or add an Expires to deal with legacy
HTTP/1.0 caches.

>But IE 5 might have a bug regarding this must-revalidate feature: it
>doesn't redisplay a page when you revisit it from your history with Back
>and Forward keys until you hit enter in the location field, or something
>like that.

That is not a bug in IE5, that is the exact behavior required by the
standard.  Back is supposed to show the old page you have seen before,
not fetch a new copy.  See for example section 13.13 of rfc2616.

>Daniel LaLiberte

Received on Tuesday, 29 June 1999 07:38:01 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:33:01 UTC