Jeff Bone wrote:
>
> Paul Prescod wrote:
>
> > Er, who said you can't cache things in query strings?
>
> I assume you mean things with query strings. My info is a bit dated, but IIRC
> (from proxy work circa '94-'95) it used to be considered bad form for various
> reasons for proxy caches to cache "objects" (representations of resources)
> identified with URI that had query strings.

Yes, it was bad form for HTTP 1.0, but not HTTP 1.1:

"We note one exception to this rule: since some applications have
traditionally used GETs and HEADs with query URLs (those containing a
"?" in the rel_path part) to perform operations with significant side
effects, caches MUST NOT treat responses to such URIs as fresh unless
the server provides an explicit expiration time. This specifically means
that responses from HTTP/1.0 servers for such URIs SHOULD NOT be taken
from a cache"

> ... Maybe this isn't the case anymore;
> indeed, if the entire piece is considered as an opaque whole, on further
> consideration I'm not sure why this would be problematic. OTOH, its a bit
> semantically muddled; does the query string identify the resource or not?
I.e.,
> do the following refer to the same _resource_ or not?
>
> http://foo.com/some/resource
> http://foo.com/some/resource?p1=foo&p2=bar

No. I have seen nothing to indicate that they refer to the same
resource. URIs address unique identifiers unless some spec says
otherwise. HTTP doesn't say anything that would indicate to me that
those are the same resource:

* http://www.w3.org/Protocols/rfc2616/rfc2616-sec3.html#sec3.2.2

Paul Prescod