Notices by ⛈️ Information ⛈️ (elucidating@mastodon.social), page 2

@mattly You gotta admit though, the coffee flavored milkshakes are amazing. in the top tier of milkshakes if we're being honest.A

@mattly One of the things I am delighted to realize is that categorization is obviously an internal feature of our brain we repurpose a lot, and it appears to be very fundamental to our biology and not an obscure or unique function. That obvious observation has wild implications.

I continue to do a survey of modern apologetics, moving on to what I think is a novel genre: political apologetics. This is what folks like Shapiro and Peterson do: justify why a hegemony around the Abrahamic God is absolutely necessary for society.

And I gotta say, Peterson really lays bare the bizarre "post-skepticism apologetics" genre. He reminds me of folks in the skeptical movement who pushed women out in 2010 and crashed the movement. Big words, obfuscated arguments.

Peterson is a mythologist who will openly tell you he does not believe in biblical literalism, but he'll do so in such a roundabout way, and demand an alleigance to a specifically Abrahamic higher power that will punish you. He goes so far as to argue atheists don't exist, except Stalin.

It's a wild, wild argument.

@mattly that said, I'll fight tooth and nail to let my daughter grow up in one of the few places in the entire United States where someone won't ask her, "No, no, where is your family from."

Even other parts of California are just casually racist to anyone even slightly off white, and near SF is a place where she's actually in the (slight) majority so this doesn't happen.

@mattly Well you see once you're here you're on a desperate treadmill not to leave because it's so insanely expensive.

For example it is interesting to think about if integers or invented or discovered.

But it is patently obvious and ridiculous to suggest anything other than an inventive origin for ohm, Maxwell or Einstein's work.

I really hate how society continues to argue about whether math is invented or discovered because society doesn't have a consistent definition of math. So you can hear someone saying something that makes sense in one frame and is absolutely ridiculous in another frame and they don't even realize that they cross the boundary from sensibility to ridiculousness.

@mattly Yeah I am privileged to still get to talk to Phil and it's his project. He's a phenomenal Lisp dev and Fennel shows that.

There's almost no exciting programming languages with good communities these days and it's sad.

We have lots of formerly good options where the community has gone really sour (e.g., Clojure and Scala). We have options where the community seems fine but the language is a total mess (E.g., Kotlin, Swift). We've got a bunch of legacy contenders that can't really modernize (e.g., C++). We have languages that both have difficult communities and the language isn't very good (Go).

I shouldn't sleep on Lua, Lua is not an awful language and it's actually got a really vibrant community.

So we have like Rust, maybe Gleam and maybe Elixir, Julia, maybe Racket? Zig possibly maybe but it feels hard to make a call there yet on quality... I can't really say many of the ML family of languages went anywhere (as much as I love and wish we'd see more dependent and linearly typed MLs).

@mattly @danilo I dunno about the former but I'm just arguing the TE folks have *range* :)

@mattly @danilo Wes Anderson? For something like this? https://teenage.engineering/store/po-14/

@pmonks @mattly @janl The reason the cryptocurrency grifters had such an easy time transferring from one sinking ship to another is that they were already existing in a world of pure FOMO pandering. They're like the Bane of FOMO, they grew up with it rather than simply adopting it.

@pmonks @mattly @janl May I submit a slightly different take on the "AI" bit?

It's that the investment class read one surprising paper and suddenly realized they've been sleeping on a field that is *already mined for explosive growth opportunities.* Chatbots are not new and not substantially improved by LLMs. Diffusion models are just a different take on already existent technologies (although they can be pushed incrementally further). They *missed the boat* and are having a fomo fit.

I think in an effort to combat elitism and credentialism the world of software engineering may have leaned too far into the "education doesn't matter" department. We see this in a lot of ways.

For one, the sociopathic interview hazing ritual. It is irredeemably bad and even good people who know it's bad feel obligated to do it. For two, the externalization of all data science from product and in house engineering has forced the privacy crisis, because folks actively disdain basic math for devs.

Blender is the emacs of visual data except that unlike emacs it hasn't even peaked yet.

Nix is absolutely a supply chain attack waiting to happen and it'll be incredibly obnoxious to unwind if it's done halfway competently, for example.

You might argue it never will be, but I think a post xz-attempt world shows that's kinda copium.

I think it's sorta interesting how Nix is completely and irrevocably at odds with a lot of modern software best practices and business decisions, and yet the folks who normally would point that out seem to like Nix all the more.