Magos Logis Logan @jmw150@qoto.org

Favorite Theorem: Fermat's Last Theorem

There is just so much culture and memes behind that 300 year old problem. Also it was solved with algebraic geometry. Algebraic geometry is pretty neat.

Sigh. There is programming language research and new research language systems with provably unique logics...

And then there are the corporate sponsored propagandized common languages. They feel like everyone is hyper-political about which color of paint to use. It is not strong sapir-whorf. These are not actually DSLs. Art is what they are, mostly.

I really like reading Hartshorne's Algebraic Geometry honestly.

Took 4 hours just to get through 2 pages. But there is a lot of good content there. I do not get why everybody is so afraid of starting algebraic geometry with this book.

>> People understanding the math on my shirt
>> Instant dopamine boost

Compiler theory and machine learning have a lot of potential and connections with each other.

https://arxiv.org/pdf/1803.07244.pdf

by and for mathematicians
https://msp.org/

Man it feels good to be a paid scientist!

Seriously though... We finally get a job that pays well, that is not management, or requires near a decade of education. And you want to automate it...

https://sampl.cs.washington.edu/tvmconf/slides/Justin-Gottschlich-Machine-Programming.pdf

I specifically asked IT for a Linux dell workstation. They gave me Windows education edition, with a bunch of CISCO backdoors on it, and locked the bios.

Gee.. Thanks for the expensive paperweight.

Symbolic regression is a thing!? Data making programs and math has been around awhile apparently.

https://arxiv.org/abs/1905.11481

Very fun. Computational science has all of these neat things that you can do. Like discover new mathematics.

https://arxiv.org/abs/1802.09436

Neural network analysis is so underrated. Reality is so manifold-y.

https://arxiv.org/abs/2004.06093

Friendship ended with proofs. Now programs is my best friend.

"Your programming language choice is your social class" without actually saying it.

https://www.joelonsoftware.com/2005/12/29/the-perils-of-javaschools-2/

As a computer scientist, I dislike programming languages that tell me what to do.

So, Java, Haskell, and Rust have interesting ideas... But I do not like programming in them.

I also think idiomatic programming is a garbage idea. It is an evolved version of tabs-vs-spaces. Seriously, code how you want, and use a traspiler to get it in your company's standard form.

Scientific disciplines are like sports teams, and it is weird.

I have a theoretical physicist that does not believe in infinity. So he is a physics guy instead of a mathematician.

And I have a computer science guy that says all math is constructed. So he is a computer scientist instead of a mathematician.

The search for truth is really just the search for narrative.

It has its own assembly language now.

https://youtu.be/FDiapbD0Xfg

Coffee definetly helps.

Prefinals week here. Hang onto your butts.