Comments by "" (@TheDavidlloydjones) on "3Blue1Brown"
channel.
-
7
-
4
-
4
-
2
-
2
-
2
-
2
-
2
-
2
-
At 2:00, a line is one dimensional, a square is two-dimensional? Agreed. But then "The space we live in is three-dimensional, and so on..."? Hunh?
People are always saying "You can go up and down and front and back, and sideways, so that's three." But I can go sideways in a whole bunch of different directions. Left and right, and east and west are all directios. Down is a pointer in my local gravity field, and up is a wavy moving line which varies with the phases of the local Moon, mostly.
We face great puzzles (and huge wastes of money) in academic physics, and it seems to me this is largely because we don't talk sense to ourselves in the most elementary ways. Keeping mathematical degrees of freedom from getting confused with physical dimensions, and physical dimensions and directions from getting confused with each other might be a start, wouldn't it?
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
1
-
At 6:45: Kolmogorov was very much not a 19th-century mathematician. He lived 1903-98 and his career spans, exemplifies, and affected both modern hyper-industrialism and the triumphs and disasters of the Soviet experiment and its failure. One of the key events of the 20th century.
There are many good Kolmogorov stories around, but my favourite is the one, perhaps apocryphal, about steel industries. At one point in the 1940s, when he was working on linear programming, he conjured up a model for a steel industry. Some Russian bureaucrat is supposed to have looked at it admiringly, with an air of Yess, Kamerade, ve vill adopt zis right away, but asked "But what is this vector over here."
"Those, Comrade Director, are the prices," said Kolmogorov.
1
-
1
-
1
-
That horrible median score on the Putnam? No big deal. The median number of errors found in a Microsoft operating system by its pre-release reviewers is zero.
The top few reviewers, however, tend to turn in numbers like 160, 90, 68, 55,... lots of twos, lots of ones, and then about 200,000 people who just wandered in clueless. They're zeroes -- and that's the number of improvements they come up with, too.
Hmmm. How about a video on Zipf's Law, power distributions, and maybe a few of the other fun ones. All the economists out there deserve to find out more about heteroscedasticity, for instance, doncha think?
1
-
1
-
1
-
Uncomfortable moment: when you try to tile all of infinite space by making a spiral of little squares each of which contains a Hilbert Curve, haven't you just fallen into the problem that the snake curve had, back earlier in the video? Haven't you made a jumping, discontinuous, mess out of all those pleasantly well-proven-out little Hilbert-curves squares?
It would be nice if somebody could post a reassuring No, 't'ain't so. Alternately, is there a better way of tiling-out the plane with a Hilbert Curve of little Hilbert-curve-squares instead of that dumb* spiral?
I think the answer is yes to the first, it's a mess, but then yes to the second (whew!), it can be made to work, but I'm not sure.
_____________________________________
* For once "dumb" as stupid is the same thing as dumb having lost the power of sound.
1
-
1
-
Grant,
Damn but your expressive little blue Pi's are intelligently done!
I think maybe you should be in the running for a Pulitzer in cartooning for them!
I also think your Einstein quote, to the effect that the cultic "math and reality are identical" is ver-ree apropos. When I first read Eugene Wigner's paper, https://en.wikipedia.org/wiki/The_Unreasonable_Effectiveness_of_Mathematics_in_the_Natural_Sciences, I thought it must be a bad translation from the Hungarian.
It turns out he was serious. Ugh! There are a whole slew of things wrong with it, among other things that for every mathematical construct suggesting a reality there a bunch of other ones with different implications. I think the Big One, though, is that most of the mathematics Wigner celebrates takes the form of differential equations, and these are hypotheses, not conclusions. Even their solutions are mere suggestions, not statements about reality.
The Wiki entry on Wigner's paper does him a great favour, suggesting a tentativeness that I don't see in the original; to my mind, the paper is emphatically wrong, and your Einstein quote gets it right.
1
-
Once you've figgered out the percentages, 91.7, 94.2 and 92,6, the question arises, is 1.6%, the difference between the two big ones, even a difference? My rule, purely emotional I believe, is yes since 48 and 186 are "big numbers," a big number being completely arbitrarily defined as anything over about 17 and "about" being an undefined element, like a point or a line. FWIW, the long run feels like a time period where you can feel the difference between 1% and 3% compounded annually, so maybe five years.
Alternatives welcome. Anybody got any real science on these? How can 17 be big while a mere five is long? I dunno.
1
-
1