hardmath123

I believe the heat of a wildfire is literally the trigger for the cone to open up and release the redwood's seeds! (I don't know the exact mechanism for this, though.)

I implemented such a version in Scratch back in 2016! https://scratch.mit.edu/projects/117304786/

Here’s a gif of that if the project player doesn’t work for you: https://github.com/kach/gifblocks/blob/master/gallery/fourie...

The N programming language (1990?) is spookily reminiscent of PyTorch!

> Expecting a wide use of neural network algorithms in the near future, our objective is to get a complete software development environment for programming and testing new applications. We intend to produce a high level language for neural network specification, as a part of such an environment.

> The language we propose is characterized by a high degree of modularity, based on parameterizable data structures, with functionalities in the form of update methods attached to them. Composition rules of structures and methods enable to build, step by step, more complex structures from smaller ones previously defined. Objects are viewed as autonomous modules which are linked through plugs for communications. We particularly cared for the parallelization of methods running concurrently on different objects of the network. The syntax is largely related to those of the C and C++ languages.

https://link.springer.com/chapter/10.1007/978-3-642-76153-9_...

I came across this problem when I wrote a small program for a humanities professor to help him draw some diagrams. He didn't want a web app ("it'll disappear once you graduate!") so I wrote him a mac app and emailed it... it was such an adventure getting him to ignore the security warnings to run it...

This is a great question! I don’t actually know the precise condition for box-diagram-ability of a DAG, but I’ve spent a long time thinking about it; it’s partly why I wrote that program. It’s not a deterministic algorithm, it’s powered by a SAT solver which may return “UNSAT” indicating failure. If anyone has a better answer I would be _very_ interested in reading about it!

A couple years ago I wrote a program to make dependency diagrams just like this! https://github.com/kach/tower-of-power

Here's `pdf-redact-tools` from Homebrew: https://github.com/kach/tower-of-power/blob/master/gallery/p...

Here's the CS core at Stanford (dependency = prereq): https://github.com/kach/tower-of-power/blob/master/gallery/c...

See also: "learning" Conway's Game of Life configurations by gradient descent.

https://hardmath123.github.io/conways-gradient.html

A while back I stitched together a "hyperlapse" of Stanford's Hoover Tower using lots of Flickr-scraped images. Everything was aligned using "classical" CV tricks and I was really happy with the results. I wonder how NeRF-w would fare on this data?

https://github.com/kach/hootow-hyperlapse

Stanford has an excellent physics course, "Back-of-the-Envelope Physics," based on Mahajan's work. On the first day we worked out how much to feed a baby every day assuming they are spherical heat emitters of radius 1 meter... (also, you got points off the problem sets if your answers were _too_ precise).

Cool project! Yes, it would be really surprising if this "worked," because SHA256 is meant to be very difficult to reverse. But it's interesting that the failure mode is gradient vanishing as opposed to, say, getting trapped in a local optimum.

Once, when I was taking a music theory class, I tried to use the Z3 constraint solver to automatically harmonize melodies… it worked surprisingly well! https://github.com/kach/recreational-rosette/tree/master/mus...

Comfortably Numbered (http://hardmath123.github.io) is about how you can learn so much about the world just by thinking deeply about the simplest things — shadows, trees, orchards, laundry, fish. The lesson is to keep your eyes open.

More on this, for the curious: https://codewords.recurse.com/issues/three/algebra-and-calcu...

Ah, we used Morin's excellent book in my undergrad physics class. =)

Can folks at Stanford come by and watch live? (This is the first I've heard of the Stanford Chess Club…)

Oh, wow, beautiful. I wonder if there's a way to rephrase this as a diophantine approximation problem? If one ray must be axis-aligned then of course it's just the convergents for √3, but with two rays it's a little trickier.

> Contactless student ID cards. With contactless student ID cards in Wallet, you can use your iPhone all over campus to access places like your dorm, the library, and events; or pay for laundry, snacks, and dinners.

This sounds amazing.

They are the two "Blums" in "Blum Blum Shub."

Frink is like 80% of the reason I passed first-year physics. It's like a typechecker… for science!

Hehe, I worked on one of these last year when I took a music theory class! You might enjoy this: https://github.com/kach/recreational-rosette/tree/master/mus...

Yes! SAT solvers are an amazing secret weapon for solving tricky problems. I've been collecting some fun examples of solver-aided programming here: https://github.com/kach/recreational-rosette (also, not in that repo, but still interesting: https://github.com/kach/tower-of-power).

I took a physics class taught like that last year, and my experience was the opposite. Because the teaching team expected that students had read and understood the material before class (hint: very few people read it, even fewer understood it), the TAs and professors went waaay faster than they should have, and the class was one of the most stressful experiences I've had at college. Nobody liked it.

The last few pages have a nice general history of communications. Gives some great perspective.

Neat! It would be cool to port this to one of the node option-parsing libraries, perhaps launching a small Electron-ish webapp for the "gooey."

I'm a big fan of Idyll, especially how easy it is to get started with. Here's a screenshot of something I made not more than 20 minutes after first seeing it [1].

Also, can I put in a plug for nearley [2], the parsing library that Idyll uses? The Idyll parser is here [3].

[1] https://imgur.com/a/5I143

[2] http://nearley.js.org

[3] https://github.com/idyll-lang/idyll/blob/master/packages/idy...

(To be clear, it shells out to an existing SAT solver. If I understand correctly, the 1500 lines of C implement the "modulo theory" part of "satisfiable modulo theory" — that is, they "compile" things like bitvector addition into pure boolean circuits.)

Hah, I'm in the same but opposite situation: I, too, am working on a Coq-like proof assistant, but I understand the type theory far more than the topology needed for HoTT.

Do you have any suggestions for simple introductions to HoTT, especially for someone without the topology background?

One attempt was Hummingbird. http://www.hummingbirdnotation.com

Yeah, I'm sure things are better. And your typical developer will probably be just fine (I mean, we know they are because Python is so popular!).

But for a beginner who is just trying to figure out how to share his or her code, it's still very confusing. And yes, part of that is definitely the fact that all of the mistakes of the past and their messy workarounds still show up in Google searches, muddying the waters; this obviously gives npm an edge. But even so, I feel that npm just does a better job when it comes to being learnable: people pick it up right away with very little trouble, and you can have your first "public-facing" module up on day one with very little effort. Indeed, in a way, the fact that it's so easy is what's giving it the most grief right now!

So there are definitely lessons to be learned there, for everyone.

A while back, I wrote some of my thoughts on Python packaging. I got excited seeing this title, but reading the article, I remain unconvinced. The whole thing is still way too confusing, and it's still not nearly as intuitive (or teachable) as "npm install ___".