'dice' or 'approximation', does it matter?

I've been meaning to do some new digging on Einstein's enigmatic complaint. In a recent program on Channel 13 (I think, but I can't locate it now) a recognized physicist portrayed Einstein as unable to accept uncertainty in nature, and that view seems to be becoming one of the prevalent understandings of the issue (see Wiki link below). On the face of it, since Einstein was a founder of statistical physics, it seems unlikely. "God doesn't roll dice", is about something else. One of the things I finally found today to expose the deeper issue was Niels Bohr's long, polite, emphatic last-word on the subject (Bohr 1949). Bohr says that what Einstein objected to in QM was the elimination of causality and continuity.

"Yet, a certain difference in attitude and outlook remained, since, with his mastery for coordinating apparently contrasting experience without abandoning continuity and causality, Einstein was perhaps more reluctant to renounce such ideals than someone for whom renunciation in this respect appeared to be the only way open to proceed with the immediate task."

Curiously, the violations of theory or nature expected by both sides in this long debate don't seem to have turned up in the many decades of argument and experiment. QM works fine, so apparently the bizarre way in which QM treats physical events as occurring without taking any time or involving any process, i.e. abstractly following rules in the complete absence of any means for doing so, doesn't matter. Both Einstein's (impossible) and Bohr's (necessary) views on the matter seem to have been simply wrong.

I guess my preference is the conservative approach. If it doesn't matter whether the disconnects of nature expressed by our best tool are physical or informational, there's no need to argue about it (i.e. within the 'shut up and calculate' school of thinking). The matter is far from settled, I realize, since provocative proofs like those of Bell's hypothesis seem to support the idea that QM's weirdness is physically real. That real weirdness still appears to be entirely contained, and to not violate causality and continuity anywhere other that within QM, however.

Where I think it may ultimately matter is in encouraging the idea generally that nature functions as a set of abstract rules without processes, rather than through incompletely understood physical processes which our rules approximate. I think whether you interpret nature is physical or informational on a macro scale probably matters a lot. The two models at least appear functionally different and need to be looked at.

The central problem I see with interpreting physical events as a function of rules is that those rules need to either refer to definable things, or to have a player. I don't think either of those is demonstrable as a generality, and the opposite is much more the usual appearance of the problem.

Is there anywhere it would really matter, one way or another?

------------- Niels Bohr 1949 Discussion with Einstein on epistemological problems in atomic physics. http://minerva.tau.ac.il/physics/bsc/3/3144/bohr.pdf

------------- Wiki link - http://en.wikipedia.org/wiki/Albert_Einstein "Einstein never rejected probabilistic techniques and thinking, in and of themselves. Einstein himself was a great statistician, [19] using statistical analysis in his works on Brownian motion and photoelectricity and in papers published before 1905; Einstein had even discovered Gibbs ensembles. According to the majority of physicists, however, he believed that indeterminism constituted a criteria for strong objection to a physical theory."