Wouldn't the Content-MD5 HTTP header help alleviate this?
On Tue, 14 Jan 1997, James R Grinter wrote:
> On Tue 14 Jan, 1997, Richard Hall <hall@charon.ns.utk.edu> wrote:
> >and it is configured so as to exclusively and optimally serve this
> >purpose. This machine is basically mirroring netscape.com.
>
> that could bring up a whole host of copyright issues and hoards of
> lawyers beating down your door.
>
> >3. The "mirror squid" attaches an accurate expires meta-tag to pages it
> >serves so that they can be reliably cached locally, thus minimizing
> >traffic to the "mirror". All objects would expire at the time when the
> >prefetch program runs.
>
> you still have to deal with people clicking 'reload' or otherwise
> forcing a re-fetch; and is saving time on one single transfer (the
> pre-fetch versus the first client to request the file) really worth
> it?
>
> Actually netscape.com reminds me of another problem: you can fetch
> documents from their servers using a variety of names:
> home.netscape.com (preferred?), www.netscape.com, www{n}.netscape.com,
> home.mcom.com (old browsers), www.mcom.com, etc... all affecting
> the efficiency of a cache. (They're not the only guilty party, though.)
>
> Some sort of host-aliasing directive (One could take it a step
> further and implement URL aliasing, but then you might as well use
> the redirector, as others are doing) could possibly be simple to
> put inside the core Squid code?
>
> James.
>
Received on Tue Jan 14 1997 - 17:07:43 MST
This archive was generated by hypermail pre-2.1.9 : Tue Dec 09 2003 - 16:34:04 MST