Brown, John Mickey wrote:
Nice write up. It seems to convey the same topics I thought of during the
conference. I told David Ungar that I agreed mostly with his premise because at
a macro scale (distributed computing), we've been dealing with this issue of
scaling and data staleness for decades. I've often asked the business the
tolerance of receiving correct timely data during requirements. Pure
synchronized solutions in the enterprise are extremely costly.
Although the application may be different and have different constraints at the
micro level, there's opportunity for the two camps (micro and enterprise) to
learn practices, patterns, and innovation from each other that can impact both
fields.
I'm particularly interested in how to apply the Actor language concepts into
enterprise business applications to allow it to be more mainstream (perhaps in
the ESB and Fabric space)
Glad to find someone else that recognized the same parallels (pardon the pun).
FYI: Carl Hewitt, the guy who coined the actor concept, has been doing
some interesting writing along similar lines, c.f.,
http://knol.google.com/k/carl-hewitt/indeterminacy-in-computation/pcxtp4rx7g1t/32#
There seems to be a growing school of thought that large, distributed
systems are always inconsistent, and we have to find new metaphors and
design patterns that take this into account. A lot of the work in the
NoSQL space, around MVCC (multi-version concurrency control) and
"eventual consistency" as embodied in, for example, CouchDB, or for that
matter distributed CVS systems like Git, are steps in this direction.
Erlang's implementation of the Actor model is a particularly interesting
platform in this regard.
Miles Fidelman
--
In theory, there is no difference between theory and practice.
In practice, there is. .... Yogi Berra
_______________________________________________
fonc mailing list
[email protected]
http://vpri.org/mailman/listinfo/fonc
