This is child abuse. What sort of monster uses vim for s-expression based languages? Everyone knows Emacs is the only sane way to work with them.

I mostly learned with pen and paper and tracing in my mind. We were 18 though

Pen and paper? Excuse me, but real programmers use butterflies.

This was just standard procedure at university in 2002. Nothing special. It was me and 200 more people. Pretty sure all the exercises were lifted or inspired by The Little Schemer even thought it took me years to learn about that book.

What's this better stuff?

Shrek for example.

I thought you were going to say skibidi toilets.

Python has so many ways of doing things it's a distraction, doubly so when you add all the special sauce for tensor manipulation.

Scheme by comparison has only one way of doing things and gets out of your way.

Fortran is about numerical programming without having to deal with explicit addresses. Symbolic programming is something like lisp.

Llms do not manipulate symbols according to rules, they predict tokens, or arbitrary glyphs in human parlance, based on statistical rules.

Fortran internally transforms expressions to assembly using symbolic computation. a+0->a, a+b->b+a.

LLMs are very much supported by symbolic computation. These "arbitrary glyphs" are computed by symbolic computation. The output of LLMs is constrained by program language grammar in code generation tasks. LLMs are often run with beam search, which is symbolic computation.

Shake your head and move on.

It's not like we haven't seen closed source applications become hostile to their users before. And it's not like we didn't warn people about it.

It's that no one knows if a system works.

Since you have the data could you show how far you can see in every direction rather than the longest direction?

Yes, but, there's one small problem, storing this extra data for the whole world starts to run into the petabytes!

what if you put a limit on it? like only a dozen views per point, and only views that are at least 100km. or only do this for peaks or points of interest. or some combination like more views for higher peaks, less for lower ones.

Ohh yes, nice ideas.

i don't know if that makes sense but it could be interesting to split the views of any point into multiple segments, and then for each segment find the largest distance. next between two of these largest views find the shortest view. if the ends of each view are connected by lines the result would be a zigzag circle around the starting point that gives you a rough idea of the visible area.

most points would be on a slope so they would just have a half circle, only peaks themselves would have an all around view.

This would be incredible - please add this!

I would usually keep quiet and ignore post like this because they are extreme timesinks, but to give a perfect example: Waco. It happened under the other side. The official story is bunk and half the country that is supposedly anti-authoritarian cheered it on.

I don't care to have the conversation or change anyone's mind but your post is the perfect example why people in the middle disengaged from the loud minority that takes over online spaces.

Hi, I consider myself a moderate. Feel free to ask me any questions. Happy to knock your silly strawman down. :)

I like to call it Canadian girlfriend coding.

My favourite part of a Canticle for Leibowitz is the manual auto regressive model the old monk is using to recover damaged books. I remember reading the gpt2 paper and thinking hang on...

I just re-read Canticle after decades. It's startlingly well written, at least the first two parts. Head and shoulders above most 50s pulp. A masterpiece really. It was assigned reading in my high school. I wonder if anyone under 40 reads it anymore