On Sep 16, 8:29 pm, "Raoul Duke" <[EMAIL PROTECTED]> wrote:
> http://lambda-the-ultimate.org/node/2865#comment-42428
>
> maybe an approach which would be a not-very-big-diff-to-Clojure to get
> distributed computing.
I like the fault streams idea mentioned in the linked-to thesis, and
have been thinking about something similar for Clojure's agents.
As far as transparent distribution, I still don't believe in it.
My inclination is still towards loosely coupled, server mediated queue
abstractions (e.g. JMS, AMQP, XMPP), vs. strongly interconnected webs
of 'active objects'.
There are many architectural advantages to message queues - optional
message durability, the consumer being offline is not an error,
transparent proxying/multiplexing, heterogeneity, flexibility, and,
perhaps most important, moving workflow, resource management and other
policy decisions out of applications. They allow many difficult
distribution problems to be solved (once) by the message queue
subsystem and not by every individual application (over and over
again).
Rich
--~--~---------~--~----~------------~-------~--~----~
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
To unsubscribe from this group, send email to [EMAIL PROTECTED]
For more options, visit this group at
http://groups.google.com/group/clojure?hl=en
-~----------~----~----~----~------~----~------~--~---
