Jim, I think I can see how to write a library that would scale across any number of servers and I don't think it would be too difficult. I've got other itches to scratch before I'd get to it, but if people started asking for it (or someone wanted to sponsor the work) I might decide to do it sooner.
Jim Jim Powers wrote: > Re: not all web applications have to scale - true, but I haven't worked on > one of those since about 1998 (this is an analog to Rich Hickey's statement > about not having to build non-concurrent programs in the last N years), > including "admin-like" things. Further: once you build all those > non-continuation-based components to support the "real site" it simply makes > no sense to write the "admin portions" in a completely different style > (although you clearly can). Also, "scaling up" portions of your web site > tends to have an unwelcome "need it sooner than you are ready" property. > Clearly what would be nice would be a distributed runtime that transparently > handles this problem. No, not Erlang, which still requires explicit passing > of data to specific nodes, but something that can move code and data around > transparently. > > Re: javascript - kinda, maybe. Although you can use this kind of approach > with something JS/AJAXy it is not completly clear what are the advantages. > Personally, things seem to be moving to Comet and the server-side is holding > less state and acting more as a soup of stateless calls. > > Just sayin' > > On Nov 30, 2009 6:57 AM, "Joel Westerberg" <joel.westerb...@gmail.com> > wrote: > > > Not every web application has to scale. I think that continuation based > stuff rocks for adminstration interfaces. > > The main benefit with continuation based stuff, is that it's possible to > build something that is more application like, so that one can avoid > building wizards, and having to split up stuff into separate steps. Another > good approach for managing state is to build the application in javascript. > > On Mon, Nov 30, 2009 at 5:09 AM, Jim Powers <j...@casapowers.com> wrote: > > > > This does indeed look... > -- > > You received this message because you are subscribed to the Google Groups > "Clojure" group. To post t... -- You received this message because you are subscribed to the Google Groups "Clojure" group. To post to this group, send email to clojure@googlegroups.com Note that posts from new members are moderated - please be patient with your first post. To unsubscribe from this group, send email to clojure+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/clojure?hl=en
