Comments by "SterileNeutrino" (@SterileNeutrino) on "The Math Sorcerer" channel.

  1. I just have "Mathematical Methods of Classical Mechanics" by Vladimir Igorevich Arnold, originally by Nauka, Moscow, 1974, but now available in 2nd edition at Springer, No 60 of the Graduate Texts in Mathemtics series.
    3
  2. I think I had it on my shelf, but I decided to make space (I checked, it's not there). It was like programming the Babbage Analytical Engine. It was old even compared to the old stuff I have around. Young ones use C (rather outdated, fickle) and Python (should never have left the 1st semester classroom, still better than bash) and stacks of books on the myriad of tools we have around these days. Out of common clay we make data processing.
    2
  3. Well, you need optimization in all manner of low-dimensional problems, including getting rockets on target, so it's certainly still useful. High-dimensional and extremely high-dimensional problems are of course not really amenable to such approaches.
    2
  4.  @ДмитрийВасильев-я6б  Not at all. Planning is scheduling or examing your tree of future possibilities before it transforms into the certainities of the past. The schedule may go haywire. Then you plan again. "Enforcement" is really an orthogonal problem, they may not be anything to enforce (e.g. how quickly your fuel tank empties as head for the Moon surface)
    2
  5. There are probably good online course in all kinds of subjects of interest (Coursera and MIT certainly has those). Interaction is key, in old times we wrote down the exercises which one should still do but these days one has the interactive systems and the markup-driven "notebooks". The problem is time... learning and solving even the simplest things takes a lot of time
    1
  6. Felleisen's books on Scheme ("The Little and the Seasoned Schemer") is a pretty good intro to programming in a (sadly untyped) functional language. After that, followup with Prolog which will yield a number of aha experiences (There is "The Reasoned Schemer" as an intro into logic programming with Scheme/Racket syntax which is a bit weird but okay too, and there is "The Little Prover" (theorem-proving) and "The Little Typer" (dependently typed languages, but I would rather use Idris for that), so apparently computer science is growing up slowy and moving out of the "let's program anything it's gonna be fine" stage. A functional language should be compelling to a mathematician and not everyone wants to go full Haskell or worry about the compressed crud that is "ECMAScript". Avoid Python, it's highly regarded (I don't know why it even exists except that everyone uses it because everyone uses it), and you don't want to do system programming in that language for the PDP-11, "C" or its overly baroque "C++" by the guy who didn't quite get object orientation.
    1
  7. Lockhart's Lament on Maff Education (Math being called racist and colonialist in decaying empires like the US to add insult to injury): "The first thing to understand is that mathematics is an art. The difference between math and the other arts, such as music and painting, is that our culture does not recognize it as such "Other math courses may hide the beautiful bird, or put it in a cage, but in geometry class, it is openly and cruelly tortured."
    1