Fun reference I tested on 32 GB ram laptop with no extra GPU: llama.cpp: “what is ls”, almost immediate starts answering at one ~word/sec. Ask opencode with same model (some gwen e4b or something) to check what’s in its working directory: 20 min to response.

Opencode system prompt contains a lot of stuff but even worse is oh-my-pi where their long prompt looks like random garbage hallucinated by a 2023 LLM:

https://m.youtube.com/watch?v=c_fQoDkULl0 (see around 8:00)

Hmm, by some bug my back arrow first slows me down from approaching c, but then will sudddenly spead me back up (while going backwards), so I get stuck in a dark world. (Firefox on Linux).

Edit, maybe it's because the speed shift is so coarse I cross past 0 km/h from back to forwards and reverse too fast...

Other than this, very very nice. Love this kind of thing for developing intuiting outside the normal range in which we experience the laws of Physics. Well done.

> my back arrow first slows me down

That's exactly what would happen ins space.

On some more testing, I don't seem to alter direction, just suddenly start to accelerate again. Ah well, it worked mostly.

The part you're excluding with "mostly" is still just the laws of physics. And it works according to those.

Looking forward to the first broken screens, and Apple's official reply:

"You're folding it wrong."

Hey that’s a Fermi Paradox solution.

Yes and it's also compatible with thermodynamics: it's much easier to destroy (increase entropy) than to create (locally decrease entropy)

A variant of the Great Filter :)

They are really creating a lot of buzz recently. I love it, I hope they succeed beyond just us geeks here. It is imho imperative for a fair and balanced technological society that it retains the means for truly private and secure communications.

My devices help me think, and my thoughts are my own. That is a fundamental condition of human-hood. Sure we can change that, if that's what we want, but do we? If we do, do we make it a conscious choice per person? Or do we let big tech/corporatocracy dictate what it is to be human?

For me there are novels with such big pay-offs that they are worth 100s of pages of setup to me. I usually only start them when I read online that they are worth it. I had this with Atlas Shrugged (500-600 pages of setup?).

Some are simply such a nice world (and indeed writing style) that they don’t even really need a Volta. I had that with the Commonwealth Saga by Hamilton. I just wanted it to go on and on (and it does for >2400 pages). (I was sadly disappointed by the follow-ups which are much less "sci"). Same with the Bobiverse books, eagerly awaiting part 6, although I still want to be in that world, it also feels like it should come to some beautiful end before dragging on too long...

Yeah same. I recently read the whole series to my son after he had begged me to stop after a couple of pages about 2 years ago. I told him: just keep your mind open for 100 pages.

The start is so cold, and weird. You can only feel the warmth and beauty that is coming on your second read, and part 1 takes some time to pay-off. What I like about Harry Potter is that many things are not what they seem when later looking back. It's clever like that.

On my own, 20 years ago, I couldn’t stop after starting. Sure the first chapters are a bit boring and different from the body of the books, but I have always given books about 100 pages to grab me, never consciously though about a Volta though, just almost always started to enjoy books after some time.

The follow-ups in a series can be enticing because of part one, but sometimes, as I had with 3 body problem, they disappoint so much or are so different that I just stop.

I've got the opposite problem with "Infinite Jest". I have had a hundred pages left for several years. I just need to sit down and soldier through those to say that I have read the bloody thing. It did not grip me.

I’ve had that as well haha. I have read many books for about 50%. I try not to feel guilty about it. Glass half full, right?

Yeah. Was hoping Strange New Worlds would be more TNG like but with ethical dilemmas of our time (ie black mirror but much more optimistic) but, meh. Not as meh as Discovery, but the Orville does it better.

I think everybody should read the two Economic Hitmen books by John Perkins.

It’s mostly the anti AI crowd that thinks that people that work LLMs think this way. I’ve never met anyone that actually does think this way.