brundolf

To my understanding the M1 and M2 have very functional GPU drivers. I don't know about the other specifics you asked

Asahi on previous M-series chips supports the GPU, they just don't support it on M3 yet (they have to reverse-engineer the GPU to write the driver)

More info: https://alyssarosenzweig.ca/blog/asahi-gpu-part-n.html

The entire administration has been operating on empty threats (see Brendan Carr's FCC speech policing). But most companies don't call them out on it, they just roll over

I didn't even see it was a brand blog. Sheesh

What a weird hill to die on

One thing people have pointed out is that well-specified (even if huge and tedious) projects are an ideal fit for AI, because the loop can be fully closed and it can test and verify the artifact by itself with certainty. Someone was saying they had it generate a rudimentary JS engine because the available test suite is so comprehensive

Not to invalidate this! But it's toward the "well-suited for AI" end of the spectrum

Very funny for this to come from the Warhammer studio, specifically

std::simd is so nice and easy to use, even as someone who's never done SIMD before. I wonder why it's stuck as nightly-only

I had something similar happen after sleeping a few times with my first-gen airpods pro. I think something must have gotten filled with earwax or something, because the noise cancelling got more and more broken and had more sudden sharp sounds until they were unusable

But since then I've had two pairs of airpods pro 2 and bought the 3s on the first day, and I've had no such issues with any of them, including sleeping, and flying twice with the 3's so far. YMMV

Super cool! I assume it plays nice with rust-analyzer?

I've heard that VSCode gets some special treatment and integrations with the typescript server that go deeper than normal LSP

I had the same experience and the same outcome. Zed was super fast for editing but slow for rich features, which on the net slowed me down compared with VSCode

Last week someone wrote a blog post saying "We dodged a bullet" because it was only a browser-based crypto wallet scrape

Guess we didn't dodge this one

I honestly don't think this administration wants economic allies

I feel like Dropbox has never figured out how to expand beyond folder sync. Which, it's still a fantastic folder-sync client and I'm a very happy customer of that product. But I mostly just have to swat away all the other things that pop up because I don't need them. And folder sync seems like a dying market, as fewer and fewer people work with files on disk

When Samsung came out with its ultra-thin phone earlier this year, reviewers said you can't really tell from pictures but it really does feel different in-hand, and is substantially lighter. This one is slightly thinner than Samsung's

Not enough for me to upgrade, but I would consider this one if I were buying this year

The rumors are also strong for a folding iPhone next year, in which case this may just be them using the same thinness work they already had to do for that. A foldable would prompt me to upgrade

I like Rust's approach, where common "errors" (eg unexpected responses from other systems) have to be handled, while true unrecoverable errors don't affect the signature

I got this confused with the Bear note-taking app for a minute (https://bear.app/), since it's in a closely adjacent domain and even has similar value statements. Unfortunate naming collision

It's still remarkable because so many people with that amount of money are still on the grind, still chasing the siren's call, still feeling miserable because it's not enough, still feeling tiny because they compare themselves to the billionaires

Very few people get to that point and then choose happiness instead of ambition

It makes Apple's charts look rigorous and transparent

Scaling aside, "without thinking" vs "with thinking" will never not be funny to me

Yeah. I think I've only ever found one situation where offloading work to a worker saved more time than was lost through serializing/deserializing. Doing heavy work often means working with a huge set of data- which means the cost of passing that data via messages scales with the benefits of parallelizing the work.

I find the "killer app" right now is anything where you need to integrate information you don't already have in your brain. A new language or framework, a third-party API, etc. Something straightforward but foreign, and well-documented. You'll save so much time because Claude has already read the docs

Same

I also just prefer CC's UX. I've tried to make myself use Copilot and Roo and I just couldn't. The extra mental overhead and UI context-switching took me out of the flow. And tab completion has never felt valuable to me.

But the chat UX is so simple it doesn't take up any extra brain-cycles. It's easier to alt-tab to and from; it feels like slacking a coworker. I can have one or more terminal windows open with agents I'm managing, and still monitor/intervene in my editor as they work. Fits much nicer with my brain, and accelerates my flow instead of disrupting it

There's something starkly different for me about not having to think about exactly what context to feed to the tool, which text to highlight or tabs to open, which predefined agent to select, which IDE button to press

Just formulate my concepts and intent and then express those in words. If I need to be more precise in my words then I will be, but I stay in a concepts + words headspace. That's very important for conserving my own mental context window

You're spot-on that excel managed to break through because it's one of the few kinds of rigidness that was both legible and familiar enough to the average person. But very little else is, because the average person doesn't think in precision.

> they lack the training to fully specify their ideas (it's tedious for a reason)

> almost no one wants to spend the day trying to get it out a probability machine

These are opposing statements. What AI does best is taking fuzziness or under-specification and making sense out of it. For the first time in history, computers don't need precise instruction to be useful.

Amazing that there's no mention of AI in this post. People have been trying and failing to blur this line since the beginning of computing, and the only real success story has been excel. And it's because rigid computing systems have to draw a line somewhere between user and developer, and if that line is in the wrong place, people will either get hampered or lost. And the correct threshold is different for every user and use-case

AI is going to finally be the realization of this dream. I don't think it could have happened any other way

Hahahahaha

I wonder if the software developer mindset plays into this. We're really good at over-reporting all possibly-relevant information for "debugging" purposes

It's not about being flummoxed, it's about being annoyed enough not to give it a chance

How much less adoption would Rust have gotten if it looked like OCaml instead of C++?

Its adopters are not stupid, they could have figured out alien syntax if they were already convinced of the benefit. But selling someone on an entirely new substrate for their professional work is a huge ask. You need to make it as immediately-palatable as possible, or they'll just keep on sailing without giving it a chance.