The problem is: what is a page?
Maybe I should have used 'object' instead, because in such a setup everything involved would be cached, including graphics, animations, audio, video etc. From one or more dynamic sources; Zope could be one of them.
For instance, if you got to http://www.zope.org/, you'll get a different page than me. Why? Because my page is personalized. However, the HTTP header Etags could help by giving each version of a page a unique id.
The win these days is in smarter caching which is more finely-grained than at the page level.
I agree. What I would be interested in achieving is a 'layered' caching structure, where a webpage can be assembled from multiple sources, each with it's own 'lifespan'. The graphics stay in the cache, the content changes until it is archived and stays more or less the same over long periods of time It allows me to use a distributed, load balancing network of data sources with a certain amount of redundancy and just the level of 'on the fly' I want. I am a long time Frontier 5 user, so I am very accustomed to easy ways to manage code, templates and content in an ODB and just render the final page whenever necessary.
I _think_ that Squid has a protocol like this to allow caches to send messages to each other.
Interesting. I'll dig a little deeper... Cya Jonathan -- UR Communications - Solutions for a wired world Who, what & where @ http://www.ur.nl/