Mozilla Expires: 'correctness' checks?

As part of the cryptozilla effort which DSTC hosts, I decided to actually
look at the netscape mozilla code instead of just managing its hostbox.

The code for Expires: and IMS is not as trivial as I'd expected. Some of
the comments look very like a netscape browser is *always* going to check
a URL, even if a valid Expires: header is in-hand.

Obviously, this would have an impact on the 'minicall' load going through
a cache. (is this class of traffic what some IETF lists call mice traffic?)

Myself, I think that if a \*.{gif,jpg,png} instance has a valid Expires:
header, then it should be trusted over any other IMS policy which makes one
want to go and look for it again. \*.html, I can understand people being
more skeptical on the age hints.

Wouldn't this tend to make the browser also appear 'zippier' if the image
decor on a website was self-consistent and re-used?
  
Maybe a real code expert can comment on the implications of the netscape
methodology in this area? Its not hard to find :-)

        find . -type f -print | xargs grep -i expire

and a bit of removal of the DNS and internal threads expiry primitives...

-George

--
George Michaelson         |  DSTC Pty Ltd
Email: ggm@dstc.edu.au    |  University of Qld 4072
Phone: +61 7 3365 4310    |  Australia
  Fax: +61 7 3365 4311    |  http://www.dstc.edu.au