Nick Drozd

Great writeup! I ended up following a few of the links and came across perhaps my favorite hand waving of all time for the validity of "The Spaghetti Code Conjecture"[1]. Basically: For N=0,1: the problem is trivaial/degenerate. For N=2: the Turing machine must go back and forth between states or else it'd just be stuck in one state forever, which wouldn't be very busy beaverish. For N=n+1: well we just proved it up till n, so it seems pretty unlikely that n+1 would be different, things don't really just suddenly change like that y'know? QED So I'm not too surprised that this conjecture is being called into question. [1]: https://nickdrozd.github.io/2021/01/26/spaghetti-code-conjecture.html

The OEIS is an amazing tool. I recently came across a sequence from an unusual source and then made the shocking discovery that the sequence already existed. It turned out to be a wild coincidence. Write-up here: https://nickdrozd.github.io/2021/04/12/math-fact-busy-beaver.html

To know whether you'd enjoy it (which is the only measure of “worth it” I can think of in relation to this book), you could start with one of the recent sections on combinatorial algorithms (Volume 4A and part of Volume 4B have been published so far): http://www.cs.utsa.edu/~wagner/knuth/ For example, 7.2.1.2 on generating all permutations (http://www.cs.utsa.edu/~wagner/knuth/fasc2b.pdf) has a lot of cool mathematics, some low-language programs, some alphametic puzzles, .... Or 7.1.4 on Binary Decision Diagrams (http://www.cs.utsa.edu/~wagner/knuth/fasc1b.pdf) is on a little-known (IMO) data-structure presented in Knuth's unique way, which can solve many counting problems. Stepping back a little, there are two ways you could read this book: • Try to engage with the mathematics, try each exercise for at least half an hour before giving up and looking at the solution, etc. (Doing it this way can definitely be a massive time investment.) If you'd like to do this, and find the mathematics difficult, you should read Concrete Mathematics which Knuth co-wrote with Ron Graham and Oren Patashnik, which is a highly enjoyable book in itself. • As mentioned in a recent post (https://nickdrozd.github.io/2019/05/17/knuth-check.html): > By the way, if you’ve ever thought about reading TAOCP, give it a try. A lot of people will tell you that it’s a reference work, and it’s not meant to be read straight through, but that isn’t true. The author has a clear point of view and a narrative…

this is a cool link, not an ELI5, but https://nickdrozd.github.io/2020/10/04/turing-machine-notation-and-nor…

Ha, the funny thing is, there are conjectures both ways! Some people think winners get noisier and noisier, the spaghetti code conjecture. Others actually think the winners get cleaner and cleaner, the clean code conjecture. Personally, I am slightly leaning towards clean code being more powerful. Chaos is hard to control. https://nickdrozd.github.io/2022/03/12/formal-theory-of-spaghetti-code…

I found this quite helpful: https://nickdrozd.github.io/2020/10/04/turing-machine-notation-and-nor… Despite a CS undergrad I don’t recall really learning any of these canonical representations of TMs before.

Counting the number of distinct TMs is not a simple task. The way you count it is the most broad (the count of all tables of values where each cell can have any (symbol, direction, state) combination). But this is a significant overcount of the bits needed to describe an arbitrary TM. for the BB(3, 4) case, there are only about 600B distinct TMs using the Tree Normal Form aka Brady's algorithm (https://nickdrozd.github.io/2022/01/14/bradys-algorithm.html) which comes out to <40 bits.

Related: https://nickdrozd.github.io/2020/08/13/beeping-busy-beavers.html and https://googology.fandom.com/wiki/Googology_Wiki

It turns out that a different post has a good answer! https://nickdrozd.github.io/2021/09/25/spaghetti-code-conjecture-false…

This should help: https://nickdrozd.github.io/2020/10/04/turing-machine-notation-and-nor…

He is still working on it and the latest check I know about was from a few months ago: https://nickdrozd.github.io/2019/05/17/knuth-check.html but I still agree with your first sentence.

The machines used in the article don't have to halt whereas the busy beaver only admits Turing machines that halt. For the purpose of this article the Turing machine either has to halt or has to enter some kind of degenerate state where it can still technically run forever but is in some sense stuck in a trivial infinite loop. The precise details of this degenerate state are explained in this article: https://nickdrozd.github.io/2021/01/14/halt-quasihalt-recur.html This degenerate state is known as quasihalting. The author points out that for the traditional busy beaver function, let's call it BB, then BB(4) = 107. For this modified busy beaver function, let's call it, QQ, QQ(4) ~= 33 million. The article goes into some detail why considering Turing machines that quasi halt instead of strictly halting provides some insight into some interesting questions, notably the spaghetti code conjecture.