IMHO, if they purchased the books etc, it's fair because, for me, it's like a biological neural network (i.e human) reading and watching the contents and memorizing them.

CMIIW

A bit off topic about the word "lahar", we Indonesians use that word to mean "lava", and we use "lahar dingin" (lit: cold lava) to mean volcanic mudflow.

and because it is comparative ratio (cm / cm), it works in any length metric. CMIIW

pleasantly surprised knowing that js of ocaml (that Bonsai used) is still active https://github.com/ocsigen/js_of_ocaml

from the site

> For large point sets, Delaunay32 is over 10× faster than delaunator-cpp and around 4× faster than Fade2D.

For one million unconstrained points on the reference Apple M1 system, eight-thread Delaunay32 completes in roughly 50 ms.

Using that as the 1.0× baseline: • Delaunay32, one thread: ~2.8× runtime. • Fade2D, automatic threads: ~4.5×. • Fade2D, one thread: ~6×. • delaunator-cpp: ~11×. • Triangle: ~11×.

I believe they mean "the windows version does not need WSL layer", as stated right after that

I mean sure? But WSL isn't a problem for me on Mac, so its presence makes me feel like the product is Windows-specific.

Well yeah, AI loves to focus on things that sound important but aren’t actually that important. Like “unapologetically native.” Um, no one is shaming native apps. There’s no actual drama here. But of course the AI doesn’t actually think, it’s just adding text that looks shiny and important and catchy

lately I saw a post [1] about Doorman Fallacy [2]. I strongly believe this is another example of that.

[1] https://rozumem.xyz/posts/17

[2] https://en.wikipedia.org/wiki/Doorman_fallacy

is that supposed to be https://tristonarmstrong.com/blog/custom-rust-errors ?

I got lucky I saw and screenshot when this had 222 points with 55 comments.

<http://ibb.co.com/1Y7QFB8N>

Oh, yea, I know of it. I just found out about this new project and want to share it here.