Spin polarization and the Einstein--Podolsky--Rosen paradox in the Monte Carlo event records

In the future high energy physics experiments, the question of properly matching the phenomenological programs that describe different parts of the physics processes (such as hard scattering, hadronization, decay of resonances, detector response, etc.) is very important. In the past, FORTAN common blocks filled with lists of objects (particles, strings, clusters, etc.) of defined properties, origins and descendants were in use. Similar structures are now envisaged, for future programs, to be written in languages such as C++ or Java. From the physics point of view such an approach is not correct, since this kind of data structures impose certain approximations on the physics content. In the present paper, we will explore their limits, using examples from the physics of W's, tau's and the Higgs boson, still to be discovered.