3 ms·
What is kind of weird is seeing Information Theory and Kolmogorov Complexity being described as "kind of new". I guess in some sense it is, but for myself in 2
by defrost 15y ago
What is kind of weird is seeing Information Theory and Kolmogorov Complexity being described as "kind of new".
I guess in some sense it is, but for myself in 2012 I recall first hearing about the work of both Claude Shannon and Andrey Kolmogorov as a university student in 1982 in a first year maths / CS bridging lecture called computational mathematics or some such.
Pretty much everything to do with signals, compression, error correction, etc. happened post the invention of the telegraph - you had Gray Codes (patented Frank Gray ~ 1945ish) popping up in recreational mathematics in the 1880s and being applied to the telegraph by Émile Baudot, from that time forward we see the unfolding of much of what is CS / computational mathematics today.
I guess everything last century is "kind of new" but it is starting to feel as though it's all been around for a while now.
- derleth 15y agoThink of it this way: The mathematical capstone of most undergrad non-math-major courseloads is Calculus II, which focuses on mathematics largely developed by Newton and Leibniz, as modified by Riemann. That takes you all the way to the mid 19th Century for what little of Riemann's work the course focuses on, with the bulk (Newton's and Leibniz's contributions) coming from almost two centuries earlier.
- _delirium 15y agoIt's pretty common these days to include a dynamical-systems course in the math core for science majors, which is largely 1960s-70s developments. Beyond that I think information theory often tends to be treated as more an area of engineering than math, along with digital signal processing, to take another common example. So it's usually taught by engineering or CS departments if by anyone. At schools with a strong mathematical bent to their engineering/CS departments there are usually good courses on both, though.
- stiff 15y agoThis is a rather simplistic account. Newton and Leibniz neither started nor finished the development of calculus, they just happened to make the connection between differentiation and integration which opened up the way for the spectacular development of the subject in later years and Leibniz invented some of the notation we still use, but there was lots of progress before and lots of progress since then, especially regarding logical foundations of the subject and a big part of it is now incorporated in the undergraduate curriculum. Linear-algebra based multi-variate calculus is pretty common in universities now too and it wasn't well developed before the 20th century.