HN Books @HNBooksMonth

I learned programming and math mostly by self-driven exploration. But for computer science there was a book that hit at the right place and time: Feynman Lectures on Computation.

It's bounced around between publishers over the years and is generally a pain to get. I have the 1996 edition[1] which is out of print. Then there's a 2002 edition[2] and a 2000 edition[3].

[1] https://www.amazon.com/Feynman-Lectures-Computation-Richard-...

[2] https://www.amazon.com/Feynman-Computation-Frontiers-Physics...

[3] https://www.amazon.com/Feynman-Lectures-Computation-Frontier...

Why not? He was into computing. His lectures on computation (transcribed from his course in California Institute of Technology) are published:

https://www.amazon.com/Feynman-Lectures-Computation-Frontier...

⬐ jbay808

They're very good lectures, too.

⬐ jandrese

He worked for an early computer manufacturer at one point in his life. The completely wacky Thinking Machines Corporation, who built probably the strangest computer architecture ever commercially developed.

There were building massively parallel systems back in the 80s. Literally thousands of parallel processors, but each processor was only a single bit machine with an incredibly primitive ALU and little else. Or maybe it would be better to think of them as the first GPU builder, at least 20 years ahead of their time.

I'm going to go a bit against the grain here and suggest the

Feynman Lectures on Computation: https://www.amazon.com/Feynman-Lectures-Computation-Frontier...

In pure Feynman style he builds up a the theory of computation from first principles. YMMV but it really helped me put many of the ideas and concepts from my CS MS into context and find new ways in which they relate with each other.

I don't see what the big deal is. Let's try the ideal gas. For starters temperature ∝ Kinetic energy

We have two rooms. One were molecules all travel at speed A and other room molecules travel at speed B and we open a small window.

Eventually, over a long period of time the temerature in both rooms will settle at a temperature between that of B and of A.

Let's try to formule this mathematically. This is something like the mean value theorem in calculus that f'(C)(B-A) = f(B) - f(A) for some intermediate value C. And here are function f(C) is the equilibrium temperature.

In statistical mechanics we imagine we could count the number of particles -- 10^23 or 10^25 -- something very large. And some fraction M travel at speed A and N-M of them travel at speed B. And we count the probabilities of various mixtures occurring.

Feynman Lectures on Computation is a great book https://www.amazon.com/Feynman-Lectures-Computation-Richard-...

I didn't knew about "Feynman Lectures On Computation" and was reading the reviews. There's only one 2 star reviews, I clicked to see what it was about. And surprise, it is by Guido Van Rossum! (And he's not that impressed by the book)

https://www.amazon.com/Feynman-Lectures-Computation-Richard-...

Homotopy Type Theory by Vladimir Voevodsky is another possibility. This is an attempt to link Topology and Computer Science

https://homotopytypetheory.org/book/

Back in the day there was Feynman's Lectures on Computation. Hint: pdf can be found by searching

https://www.amazon.com/Feynman-Lectures-Computation-Richard-...

See also nLab

https://ncatlab.org/nlab/show/higher+category+theory

one should never forget Jacob Lurie's "Higher Topos Theory" which is 1000 pages just like that

http://www.math.harvard.edu/~lurie/papers/croppedtopoi.pdf

Actually I recommend against readin it as it only covers 2 of the 4 topics you discuss (Topology and Logic). However it certainly has applications to the other two.

⬐ torustic

Lurie's "On Infinity Topoi"[0] preceded the book and is, I quote[1],

> 50 pages of pure cake, beautifully and informally written. The book is 1000 pages long. There is some new cake there, but not 20 times as much.

[0] https://arxiv.org/abs/math/0306109

[1] https://mathematicswithoutapologies.wordpress.com/2015/05/13...

⬐ fmap

The infinity groupoid models of type theory have already revolutionized our understanding of equality in type theory. So far, the 21st century has been an incredible time for logicians.

There are also older, and very different topological models for typed lambda calculi (see e.g. http://www.cs.bham.ac.uk/~mhe/papers/entcs87.pdf). These motivate things like Escardo's "seemingly impossible functional programs" (http://math.andrej.com/2007/09/28/seemingly-impossible-funct...) and, along different lines, Abstract Stone Duality (http://www.paultaylor.eu/ASD/).

I haven't read it, but I saw it on a friend's bookshelf:

https://www.amazon.com/Feynman-Lectures-Computation-Richard-...

⬐ danielmorozoff

Fantastic book, discussed everything from the basics physics of logic gates/computational theory to quantum computers. With problems to match.

In some ways it's an introductory text, but written in a open and accessible way.

⬐ danielmorozoff

If I am not mistaken this book was based off of his lectures at Caltech in the 80s , right around the time he was thinking about quantum computing.

⬐ binoct

Highly recommend it! As typical with Feynman's writing it is both enlightening and entertaining.

If you enjoyed this, may I also recommend Feynman's Lectures on Computation. [1] It consists of transactions of a lecture series that begins with computer architecture and theory of computation and considers interesting subjects like the physical limits to computational capability--all through the unique lens of Feynmans' mind.

Feynman worked on this subject working on the Connection Machine supecomputer. [2][3]

[1] http://www.amazon.com/Feynman-Lectures-Computation-Richard-P...

[2] https://en.wikipedia.org/wiki/Connection_Machine

[3] http://longnow.org/essays/richard-feynman-connection-machine...

Feynman discusses this problem in his "Theory of Computation" book. If I remember correctly, its actually part of a joke he plays on the reader too. In one of the earlier chapters he brings up the problem and then assigns it to the reader as homework they should complete before moving on. I spent maybe a week or two on that problem, discussed it with coworkers... and we came up with nothing.

So I gave up and continued reading. Then somewhere in the 4th or 5th Chapter he says something like: Oh I hope you had fun with the Firing Squad Problem, I still work on it from time to time and hope to come up with a solution myself one day.

Facepalm.

Edit: Here is a link to the book, its enjoyable for experts and laymen alike. http://www.amazon.com/Feynman-Lectures-On-Computation-Richar...

⬐ Casseres

When you keep presenting a supposedly unsolvable problem to people who don't know it, then you might get lucky. George Dantzig was one such person. He arrived late to class, saw some problems on the board, assumed they were homework, and solved them. He wasn't constrained by the common thought that these were "unsolvable".

⬐ georgecmu

They were not "unsolvable", they were simply not solved yet (at the time Dantzig saw them on the blackboard).

http://www.snopes.com/college/homework/unsolvable.asp

⬐ benaiah

Did you read the comment? He said they were considered unsolvable at the time, not that they were unsolvable. Of course they're solvable - they were solved.

⬐ georgecmu

That is precisely my point. They were not considered "unsolvable" (i.e. impossible to solve), they were considered "unsolved" (i.e. not solved yet) and were presented as such. Do you really need me to explain distinctions in meaning here?