On Apr 14, 2013 9:46 AM, "Tristan Slominski" <[email protected]> wrote: > > A mechanic is a poor example because frame of reference is almost irrelevant in Newtonian view of physics.
The vast majority of information processing technologies allow you to place, with fair precision, every bit in the aether at any given instant. The so-called "Newtonian" view will serve more precisely and accurately than dubious metaphors to light cones. > > Obvious things in Newtonian view become very wrong in Einsteinian take on physics once we get into extremely large masses or extremely fast speeds. The Einsteinian view only becomes significant if the actual message endpoints (e.g. my computer and yours) are moving at relativistic speeds. If you're merely a large light-distance, that's predictable latency. Within computers, or a wired network, the Einsteinian view is irrelevant. And, even in relativistic systems, we can predict much. > > When an actor sends messages, there is an information "light cone" that spreads from that actor to whatever actors it will reach. Actors model has so many inconsistencies with a light cone metaphor that the comparison misleads more than clarifies. Light, unlike actors messages, travels at a very predictable speed. Objects traveling through the light cone, even at relativistic speeds, will observe a predictable subset of it. Messages, unlike light, have a named destination and are delivered again on error. Messages have unpredictable delays that are in no way essential according to any physics we know. Actors model has much more in common with a Newtonian post office than an Einsteinian light cone. > > Within an actor system, after a creation event, an actor is limited to knowing the world through messages it receives. This seems to me to be a purely empirical knowledge (i.e. coming only from sensory experience). Yeah, that's a good property to have. Mostly. Assuming you never need to debug or extend the actor system. Hmm. Well, whether or not this property is good, it's easily achieved in many non-actors models, so it is a weak selling point. In any case, if you want to be "close to our universe", then starting with a relativistic model is "doing it wrong" because it won't adequately explain behavior at small scales. Seek instead a model for the small scales where relativistic observations are an *emergent property* at large scales. Reversible cellular automata are much more promising for your stated goal.
_______________________________________________ fonc mailing list [email protected] http://vpri.org/mailman/listinfo/fonc
