Justo wrote:gill1109 wrote:OK, so here is a new test of Fred's Mathematica code.
I set m = 1, and e = 100.
With settings (20, 130) I got outcomes (-1, 1).
With settings (20, 140) I got outcomes (1, 1).
Alice's setting stayed the same, Bob's setting changed, Bob's outcome did not change but Alice's outcome did change.
The hidden variable was the same in both cases.
https://www.wolframcloud.com/download/gill1109/Published/newCS-21-paper-Gill-test.nbThe notebook is a copy of Fred's notebook with a few lines altered at the beginning and the calculation of correlations replaced by simply printing the inputs and the outputs.
This proves that the outcomes are not computed according to the formulas written out in Fred and Joy's paper.
There has been some kind of spooky action at a distance.
Pardon me for stepping in but if that's true, there is a problem. That is the test of locality that Eugen's model does not pass. Eugen does not understand that no matter how you explain the result, it constitutes objective proof of nonlocality.
Dear Justo, thanks for stepping in! You are absolutely right.
I would also really like to know Michel’s opinion. He’s a great programmer.
A plain text (.txt) file is much easier to work from, than pdf. When you select text from the pdf, and copy and paste it into a programming environment, you discover all kinds of weird symbols and weird layout. Pdf files have been invented for display and conversion to paper while keeping typography fixed. They are essentially vector graphics files. Instructions for drawing smooth lines on paper down to the individual letters.
FrediFizzx wrote:@gill1109 Hmm... I guess I did that long explanation for nothing. Ok, let's review that and tell me where it is wrong.
https://www.wolframcloud.com/env/fredif ... 5-forum.nbIf you fail to do so, then the simulation is completely local which it is anyways.
Dear Fred, Your explanation is very helpful for the purposes of understanding how your code works, from an algorithmic point of view. Perhaps someone who is interested in your approach could convert it into concise pseudo code. At that point, I would enjoy trying my hand at writing a Python version. I think this would make various experiments a whole lot easier. For instance, does it help to replace Michel’s formulas with Pearl’s? John Reed’s Mathematica code is also very interesting. With the same random seed it generates identical streams of settings and outcomes as your Mathematica code does. It runs much faster, and it’s much more transparent. It seems to me to be a better implementation of the same algorithm and it lends itself well to a Python translation. If you and Joy can see your way to adopting it I think you’ll find your work easier to publish, and future readers will benefit.
No hat-eating by me. My goal posts did not move in 20 years. My challenge is a computing challenge, not a physics challenge. I think your notion of “local” differs from mine. I don’t mind if you don’t find my challenge interesting.
Joy has written that the purpose of his work is to challenge conventional understandings of metaphysical concepts (like “locality” and “realism”). I’m an amateur in that field.