shae

I got to meet Jef Raskin in the early 2000s when he gave a talk in Sweden.

He was a fun person to chat with, lots of great ideas and the drive to work out the details.

Curse Tablets are early LLM skills?

After decades of dealing with Elsevier, Springer-Verlag and the rest; I hope they all go out of business.

I use https://garnix.io/ for all my Nix CI, works great.

I go to the library, farmer's markets, do group classes at the gym, and join groups that match my interests.

I'm the same way. I require people time, and work from home wears me down.

I saw this used to obfuscate spam yesterday! Yay?

I think the solution is to launder all research papers through LLMs so the papers are no longer copyrightable, and let the rich journal owners fight with the LLM owners.

For source code diffs where a tree sitter grammar exists, difftastic is the best choice by far. It's better than you think it is.

https://github.com/Wilfred/difftastic

No really, if you haven't tried it, it's better than you think it is.

https://www.scannedinavian.com/tools-built-on-tree-sitters-c...

(I know, already mentioned later in comments by leeoniya, still deserves a top level comment!)

I've read lots about content defined chunking and recently heard about monoidal hashing. I haven't tried it yet, but monoidal hashing reads like it would be all around better, does anyone know why or why not?

At the 2018(?) ICFP, I sat between John Wiegley and Conal Elliot. They talked about expressing and solving a programming problem in category theory, and then mapping the solution into whatever programming language their employer was using. From what they said, they were having great success producing efficient and effective solutions following this process.

I decided to look for other cases where this process worked.

I found several, but one off the top of my head is high dimensional analysis, where t-SNE was doing okay, and a group decided to start with CT and try to build something better, and produced UMAP, which is much better.

In short, this does work, and you can find much better solutions this way.

(random link https://stats.stackexchange.com/questions/402668/intuitive-e... )

I suspect the H pattern they mean is where the gear shift is on the steering column, not on the floor. I long ago owned a 1945 Dodge truck with that shifting setup.

When you finish the game you get the credits for who created it.

There's also a threejs reference in a hard to reach area.

I put my trust in someone on a case by case basis, unless they're going to cheat someone. Then I don't trust them.

What if we train them and they leave?

What if you don't and they stay?

The easiest and fastest way to get everything installed is ghcup https://www.haskell.org/ghcup/

As for being successful, there are several nice books, and several active forums. I've gotten good answers on the Libera IRC network #haskell channel, and on the Haskell matrix channel #haskell:matrix.org

If you want to get started without installing anything, there's the exercism track: https://exercism.org/tracks/haskell

I've heard good things about Brent Yorgey's Haskell course ( https://www.cis.upenn.edu/~cis1940/spring13/lectures.html ) but haven't tried it myself.

I'd say this many sensors in a wireless package can do many things.

My first project would be exercise monitoring. You can measure heart rate, blood oxygen levels, and body temperature in one place.

I would then add something fun like comparing beats per minute of the music to heart rate.

I like the creators' suggestion of using the shape of the ear canal for authentication, I bet that's harder to steal off a video recording than fingerprints or a retinal pattern.

I'd also use this for voiceless speech, as the creators suggest.

I could think of more, but I won't be able to afford the dev kit, sadly.

I'd love to buy this, but I can't afford 2500 euro for a personal fun project.

I have previously purchased and written[1] about the Tympan.org open source hardware hearing aid[2], but it's about $300 which I can reasonably purchase.

If this comes down in price, I'd love to build some neat things with this hardware.

[1] https://www.scannedinavian.com/open-source-hardware-hearing-... [2] https://shop.tympan.org/collections/all

This is super cool! I'd like to build this sort of thing myself, do you recommend any tutorials?

If Python object methods returned `self` by default instead of `None` you could do this in Python too!

This is my biggest complaint about Python.

I care about the edge between "this value fails, one value over succeeds". I wish shrinking were fast enough to tell me if there are multiple edges between those values.

This sounds like graph reduction as done by https://haflang.github.io/ and that flavor of special purpose CPU.

The downside of reducing a large graph is the need for high bandwidth low latency memory.

The upside is that tiny CPUs attached directly to the memory could do reduction (execution).

I have this laptop, with 192 GB of RAM and an ADA 4000 GPU. I haven't tried LLMs but it's an amazingly powerful laptop.

This sounds like a useful way to mix in statistics and get useful approximations. I'm reading one of the survey links and it's approximately eye opening.

I love doing research on subjects that excite me, but I found that joining a PhD program would mean I would only get to work on what my advisor found exciting.

If you get lucky and find a PI that is interested in exactly the same thing, you're good. You might get lucky enough to find someone who is interested in something near what you want. Most likely, you will be assigned the thing you do for your PhD.

I tried to help with the Linux kernel once in 2001 or so. I decided my calm was worth more than dealing with abrasive kernel devs.

I'd play this tower defense game where I have to organize my city to resist zombies.

I tried helping with the Linux kernel once, in 2001. I haven't since. I don't need abrasive people in my life.

My swedish books from the 1800s have ads inside.

I wish we could create shared open source for schools and libraries.

Denotational Semantics is the basic tool of software design.

Meaning, model the whole thing in math, then you know whether the implementation matches the design.