Notices by Zach Weinersmith (zachweinersmith@mastodon.social)

So, it seems like at a policy level the US basically accepted somewhat higher inflation than other developed countries in exchange for a thriving economy. Successful according to most economists, but it may have cost the administration a second term. Do you agree, and how do you think it'll affect future decisions?

Anyone read Bostrom's new book? I thought "Superintelligence" was a bit out there, but it was a pretty straightforward book about philosophy of AI. "Deep Utopia" feels like a philosopher having a stroke. I don't know that I've ever read a more confusing and repetitive book.

One of the weirdest responses for A City on Mars is the people who think we're *up to something*, usually meaning some kind of sneaky scheme against technology or freedom or something. Mostly it's weird because we're super open about our position? Like, we're not with-holding secret communist leanings. In fact, we spend a while talking about how newspace tech is really cool and Elon, personal faults notwithstanding, made a massive difference.

I guess basically... like if people think "you suck," well, by all means say so. But the "I see what you're up to, you sneaky sneaky Weinersmiths" is bizarre for a 400+ page book where we lay out our policy position.

Working on some of my famous realistic dialog

Has anyone read the new Ta-Nahesi Coates book? I've loved his past work, but his interview with Ezra Klein was borderline bizarre. It felt like a professor trying to politely steer a smart grad student who clearly hadn't used his whole brain on a draft dissertation. And partway through, Coates literally says something like "It's a fair point that I didn't talk to anyone who might've disagreed with me. I should grapple with that more."

Also not concerned about movement because the little God runs around the world 24 hours a day. He had a six pack at age five. He can do situps from a hanging position.

I don't think he generally understands or cares about the gaming aspects of things, so we're more interested in open worlds, especially aquatic ones, that he could get immersed in.

So, our autistic 7 year old (mostly non-verbal) rarely plays with any particular thing for a long time, but the teachers at school put him in a VR set and he watched water for 45+ minutes.

We're thinking of getting one for his birthday. He hardly wants anything, so it'd be nice to do something special. Any recommendations? Thinking a meta quest 2?

Something I find really counterintuitive, and didn't believe till I tried it: learning more than one language at once seems to make acquisition easier for the main language I'm learning? Like, I don't have a spreadsheet or anything, but my main language learning is French right now, and it feels like if I start the day with a little basic German I score higher on the French. All I can figure is there's some brain part saying "okay fine, we're doing languages now." Like a warm-up?

Anyone wanna link me to a thoughtful podcast or debate or whatever on Israel/Palestine.

(Snarky rejoinders less appreciated than sharing the thing requested)

Full comic here: http://smbc-comics.com/comic/brown

Bonus panel here: http://smbc-comics.com/comic/blink

@sophieschmieg Very small. Stock market's still up 10%ish on the year. I think everyone's just waiting for the AI bubble to pop valuations, but part of the problem is the relevant companies are raking in money hand over fist for non-LLM stuff.

Today the stock market has crashed... down to the level it was at 2 months ago.

At some point someone's going to coin "Kamalanomics." Can we pre-empt this with Economic Kamalicy? It's more expansive because you can divide it into both fiscal and monetary Kamalicy, not to mention trade and foreign Kamalicy.

Anyone have a go-to pizza recipe? I've tried a bunch but nothing blew me away. Part of the problem is it has to work in my regular-ass 500 degree oven.

And the model I use could be fairly stupid. Just a sort of underground box. No need for deep physics or even an understanding of what the point of a basement is.

Like, GPT-3 failed questions of the form "I'm in the basement and look at the sky. What do I see?" GPT-4 fixed this by having humans correct its mistakes. I imagine if I were a kid getting this question for the first time, especially in a place where there aren't typically basements, what I'd do is probably imagine being in a basement.

Are there any LLMs yet that are able to kick questions over to a physics model? Like, it seems that at least for some questions, the way we get an answer isn't by thinking about what we've seen or learned or said before, but literally imagining the world. For kids, this seems to include things like finger counting for addition.