Rain Man musings ... A splatter of even 25 toothpicks is likely to contain ambiguities, where, because of overlaps, the image is insufficient to distinguish each toothpick. I've been skeptical of Sach's "prime twins" ever since reading about them. With concerted effort, you might be able to memorize a coded table of six-digit primes. (Coding as 100K hex digits.) Recognizing 10-digit primes accurately is very hard to credit. The twins could fake it by memorizing a hundred 10-digit primes. More interestingly-- the film-maker may have thought his audience would recognize that 246 is exactly 4 short of 250. [fifty year old memory below - details unreliable] I recall reading a WSJ article ~1970 (google can't find it -- have they digitized that far back?) about a counting fad that started in New York. Someone purchased a box of 100 paper clips. When they opened the box, it seemed short, so they counted and found ~70. Their story went viral, and sparked a fad of counting: sheets in a ream of paper, etc. A surprising number of "short count" cases turned up, enough to suggest that some manufacturers were deliberately cheating. The NY attorney general became involved. It was decided that (true) average counts were allowable if the deviation wasn't too big. Some makers routinely overfilled by 5-10%. Rich ---- Quoting Hans Havermann <gladhobo@bell.net>:
MG: "I have no idea how credible any of this is, but I'm just not sure that the particular scene you cite should have (instantly :-) ruined the credibility of the character."
It's in my skeptical nature. I don't even believe that a box of 250 toothpicks will necessarily contain 250 toothpicks. That at least would be an easy one to demonstrate or counterexample. The matter of a hearsay human ability without the benefit of a rigorous laboratory test, well, not so much.
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com https://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun