Re: Querystring vs. Squid Cacheserver

Written by you 22:03 09-03-99 +0100,Henrik Nordstrom
>> The issue is whether or not the script gives the same output given
>> the same input at a later point in time.
>
>If you have this view, then perhaps you should not cache anything.
>Someone may update their document.

I think we agree. My point was that I have trouble enough dealing with
cache-busters of the kind that sends bad headers for objects that else way
could be cached (ie MSIIS with their "cache-control: private" on
everything). I see a point in caching an object if it is somewhat likely to
get requested again within the objects life span in my cache.

As the case was I was referring to Andres' citymap which displays 4 images
depending on input parameters as city, street and house number in germany.
I must admit I havent checked in my logs but I find it highly unlikely that
the same search would be performed twice within eg a week (the house number
taken into consideration).

If my cache fills up with never again requested objects, then I would
consider it cache-busting too because the headers lead me to belive that
the object is cashable even though it makes no sense.

In the actual case I would have to cache http://someurl?a=1&b=1 and
http://someurl?b=1&a=1 and so forth. The script produced identical output
but the parameters got shuffled once in a while. My point was make it even
more cache friendly - then it will make sense (eg refer to the map by the
id it must have in the database and make a cross an another layer, then I
can cache it, but when added a cross with GD or something it suddenly gets
less interesting to cache).

Regards

-- 
Ole Møller olm@cybercity.dk, Sysadm CyberCity Internet
Win 95 -- The only OS with the Year 2000 bug built into its name!