> I think Framework's in a hard place with the Framework 12. Because it's an odd dimension, and because they wanted a full 360° hinge for tablet mode, they had to compromise on the display.

Seems a bit weird that framework went with a 360 hinge w/ sub-par display & sub-par stylus. I wonder if there's any demand for that and what's the use case?

I bought one because as far as I could find it is basically one of two 2-in-1's on the market that officially support Linux. (It replaced an old Android tablet and one of the original FW13's, both of which were getting old.)

not exactly a dupe but definitely closely related.

>That’s not an accident; it was the point.

I wonder if interest in programming books is plummeting at this point.

If anything I find books even more valuable now. They are the source for your LLMs after all. I like to read the author's own voice rather than have it regurgitated in generic form by an LLM.

Most people didn't read before LLMs and they'll still not read now, of course.

I agree, but basically my main use case for LLMs is just asking for clarification about programming questions/concepts since books don't always explain things in the most immediately understandable manner for all readers and you often have follow-up questions that the book doesn't proactively anticipate or answer.

An Isac Asimov book about a spaceship stranded in space without anyone on the team having the knowledge to repair it comes to mind.

I would imagine so, for most developers.

Ironically it's because of AI-coding that I'm spending more time practicing Haskell. And since it's still helpful to read references, I bought this bundle.

think about how much faster it would've been with a small local model!

what if other people want to use that table? is it okay to move their laptop/bags off?

I actually find it annoying if people put stuff on tables and then walk away for an hour or more, hogging a spot that could be used by others.

nope, u don't touch it. everyone does it and accepts it and it's not that annoying i guess.

indicator of being detail oriented

I liked this article's definition of Full Self Driving (Level 4 autonomy), it is very clear - when Tesla directly takes on the legal liability for unsupervised driving.

Interesting, this tech is already in some chinese EV car models since 3 years ago, so it's not super new.

Seems like HW3 has been pretty good since FSD v12.3+ came out.

If we describe HW4 as 99% of the way to true FSD, then HW3 is probably 95% of the way.

Though approaching 100% (maybe 2x the human standard) is going to be exponentially harder to get to.

I live in a suburban area in a cold climate. Based on what I've seen of "FSD," it's essentially unusable on most of the roads near me, and doubly so in winter. This is even true on larger highways/freeways, as when snow falls the camera systems can't see the lane markings. Not to mention the fact that some of those roads are so badly maintained that the lane markings are faint to nonexistent.

I don't think Tesla can honestly claim 99%, or 95%, or even 50% of the way to FSD until they solve these issues. Until they do, it's just a fun toy. After all, years ago they were claiming that you'd be able to "summon" a Tesla sans driver from LA to NYC. What happens if there's a winter storm on the way?

it is not even a fun toy because fun toys are fun for kids and no sane person would use FSD wirh a kid in a car

> If we describe HW4 as 99% of the way to true FSD

What part of that final one per cent includes "will not blow through railroad crossing gates when a train is approaching"?

Or "will work in a Pittsburgh winter's night on a snow covered, poorly line marked road"?

> HW4 as 99% of the way to true FSD

If we describe current LLMs 99% of the way to AGI and full sentience then https://en.wikipedia.org/wiki/ELIZA was probably 90% of the way.

I'm reminded of the adage about getting 80% of the way through shipping a product, and then doing the other 80%...

Yeah that's what I was referring to, the last 10% takes 90% of the time. Or the difficulty of getting from five 9s to nine 9s.

In this case the final 1% is 200x harder than getting to the initial 95%.