<snip>
:> :> why not just use some separate subsystem for caching, eq. a squid :> in httpd accelerator mode?
<snip> This would only allow you to cache on web page boundaries not on object boundaries. I'd like to be able to write a DTML Method: <dtml-var standard_html_header> <dtml-var always_dynamically_rendered_object> <dtml-var possibly_cached_object> <dtml-var standard_html_footer> Moreover, I want the cached thing to be aware of acquisition, etc.. I think this necessitates its (rather close) integration with Zope. OTOH, am I in the minority of people that thinks cacheing on web page boundaries (i.e., REQUEST boundaries) is interesting but not compelling? A perfect case in point is the zope beta site. I always want to display the users login (which would be computer on ever hit) and say, "Welcome rob" However, I'd also like to cache the a rendered <dtml-var news_items> object. --Rob
On Tue, 14 Sep 1999, Rob Page wrote: :<snip> :> :> :> :> why not just use some separate subsystem for caching, eq. a squid :> :> in httpd accelerator mode? :<snip> : :This would only allow you to cache on web page boundaries not on object :boundaries. true. setting the "modified" time of the web page according to the youngest modify-time of the constituting object could be some kind of compromise... :Moreover, I want the cached thing to be aware of acquisition, etc.. I :think this necessitates its (rather close) integration with Zope. OTOH, :am I in the minority of people that thinks cacheing on web page :boundaries (i.e., REQUEST boundaries) is interesting but not compelling? this also could be done with setting modify times appropriately, if an object which lends itself to acquisition is changed, set the modified-dates of the objects which acquire from that object. dont get me wrong, i think that cacheing within zope is a good thing... but additionally, i think it would be a good idea to take care of the caches outside of zope, ie. the proxies around the world... (did i mention that http://zope.org is not very cacheable ;)) peter. ps.: please forgive my (bad) english _________________________________________________ peter sabaini, mailto: c c c p@oeh.tu-graz.ac.at, http://oeh.tu-graz.ac.at/~cccp/ - - - - - - - - - - - - - - - - - - - - - - - - - Any sufficiently perverted technology is indistinguishable from Perl. (ca. a.c. clarke) -------------------------------------------------
participants (2)
-
Peter Sabaini -
Rob Page