What I mean is that merely separating the process by which you choose each measurement setting and the entangling event does not actually close even this restricted version of the loophole. You need a further assumption that this choosing process produces a distribution of outcomes which are uncorre No worries at all! I think I pressed the wrong button :-) And it is good to see that my answer was not totally off! Okay, I don’t disagree. You seem to be implying that making this split is artificial? Why should this be the case for setting dependence (freedom of choice) but not for outcome dependence? I put in a feature request for sorting comments by date. https://github.com/scirate/scirate/issues/328 I had answered in a similar fashion (a second answer after a first answer), but also my comment is floating around :-) Note: Because I hit the wrong button, there’s a stray comment (by me) floating around that answers this question. It’s just not in this thread… Good point about sorting …but it wouldn’t have been an issue if I’d used the site correctly! I intended my comment to be a reply to Tom’s, but (due to incompetence) it ended up standalone instead. Oops. Anyway, SciRate does keep threads together, which should go some way to helping with organiz That would indeed be helpful – you should definitely email the developers about this! Not sure if they read every comment ,-) I like this use of SciRate as a ‘reading group.’ It’s a collaborative and friendly use of the comments. Perhaps SciRate can add a feature to ‘sort by date’ rather than by rating/points, which would be useful for such discussions? Source.