(07-24-2016, 04:36 PM)Thomas Ray Wrote: Richard Gill's criticism is refuted. http://arxiv.org/pdf/quant-ph/0212085.pdf

[...]

That paper is not a refutation but a confirmation of the criticism of myself and others. See page 4, point (vii), and the further discussion on page 5. The Hess and Philipp model is not a local hidden variables model.

As Hess and Philipp themselves admit, the distribution of the hidden variable in Alice's measurement station at any time has to depend on Bob's setting at the same time.

Technically, the problem with the model is hidden in formulas (28) and (29). For each given m, the integral of sigma_a tau_b kappa over v depends on a and b. There is no way to use this representation to simulate the hidden variables (u, w, m) needed to generate Alice's outcome without knowing Bob's setting as well as Alice's. How do I know this? Because the model reproduces the singlet correlations and hence violates the CHSH inequality.