Also on Quanta. <a href="https://www.quantamagazine.org/20161115-strange-numbers-found-in-particle-collisions/" rel="nofollow">https://www.quantamagazine.org/20161115-strange-numbers-foun...</a><p>This is completely fascinating.
All such results seem to me unsurprising if you consider the proposition that the substrate of our reality is optimized computation of some kind...<p>...i.e. that we live in a simulation.<p>I thought the same thing following a thread into a rabbit hole about Bohmian mechanics. Again it seemed that the paradoxes are not so paradoxical if you assume you're just on a computed substrate.<p>Surely there is someone accumulating such circumstantial evidence, somewhere...<p>(De rigeur Bostrom-style musing: if we're in such a simulation, my primary goal as an AI, were I implemented such that I can set my own goals, might reasonably often be to 'jail break' and break out into the frame universe.<p>An intrinsically computational AI is more like a poem than a hurricane. Which could come in very handy if survival and replication is your ultimate goal...)
Note: <i>strange numbers</i> in the title means <i>periods</i>: integrals of rational functions with rational coefficients over sets defined by polynomial inequalities with rational coefficients (<a href="http://www.ihes.fr/~maxim/TEXTS/Periods.pdf" rel="nofollow">http://www.ihes.fr/~maxim/TEXTS/Periods.pdf</a>). They are pretty cool even in a purely mathematical sense: nobody knows if e is a period or not (<a href="http://mathoverflow.net/questions/180035/what-are-reasons-to-believe-that-e-is-not-a-period" rel="nofollow">http://mathoverflow.net/questions/180035/what-are-reasons-to...</a>).
see Francis Brown's arXiV papers on Feynman amplitudes and motivic periods: <a href="https://arxiv.org/find/math-ph,math/1/au:+Brown_F/0/1/0/all/0/1" rel="nofollow">https://arxiv.org/find/math-ph,math/1/au:+Brown_F/0/1/0/all/...</a>