The callous part of me thinks that was the intention from the start: wash out the capable engineers who were competently producing and hire ethnic cohorts who played the game.

It's Amazon. Anything that gets them the metric that they can put in their promo doc is a good thing.

6-12 months later, when they've been promoted, they can nix it and add it to their next promodoc as company savings in efficiency.

Are you honestly summarizing the current state of advancement in LLM capabilities as "somewhat better search"?

It's apparent from that statement that you've not even been paying a slight bit of attention to state of the art. It isn't 2023; we're a far sight further than you conceptualize, and you'd be well suited to start familiarizing.

No, that's when you want a RAID system with multiple disks.

This is only for small corruption, like a failing drive might cause, or due to cosmic rays/bitflips.

This is just EC for a non-redundant drive.

That's such a dumb, smoothbrained description I doubted the capabilities of the person who wrote it - at first.

Then I realized he was right. There isn't a filesystem that does RS or, really, any sort of single-device EC. This is a genuinely good idea, at least for spinning disks, where sector corruption is more likely.

This may be personal anecdote only, but I've only seen 1 SSD get any sort of data corruption as a failure mode, and it was a very early iteration - maybe a 8 or 16GB one. SSDs seem to simply fail outright - the FRU or something else failing for one reason or another. The storage is fine.

I read that as "land asteroids in the next 2 years" and I was initially much more excited than the actual headline suggested I should be.

Landing asteroids to mine would be much cooler than just a mere moon base.

Let's keep in mind who Elizabeth Barron is: she's the prior Community Manager at GitHub who was best known for DEI initiatives. Before that, she was a Community Manager at Pivotal. Before that, she was the PHP Community Organizer at Engine Yard. Geeknet (of SourceForge and prior Slashdot relation)? You'll never guess! (OK, yes, it's, Community Organizer, er, no, "Development Manager".)

Her only personal github anything is an "about me" blog. She has no repos of "scripts I find useful" or "little cool thing I did". She has no forks for things she's contributed to or features she's working on. Judging by her commits, comments, etc. she's still basically just doing that: it's just her, the marketeer. She doesn't have any code contributions anywhere.

She also appears to be a current, touring professional photographer (though, not frequent) and to have done web development before "the Internet".

She is wearing PHP like a digital marketing meatsuit, and while the project starves for technical direction and purpose, she's manually doing PRs about *checks notes* bunny.net. Seriously.

This is a language which should have died years and years ago but has been kept publicly alive by a bunch of low-tech-skill web developers who likely survive almost solely on inertia and having a corner people with both "Grandma, we told you to stop clicking on the links in your email" energy and entirely too much money.

I'm still using a 40" 1080p monitor I bought for $50 in 2011. It's the only TV in the house and we just use it for things mostly made before 2005.

The new stuff isn't worth the time.

You mean software features - which now exist?

Explain how you'd fake a 3000lb+ car made of non-magnetic metals being able to suspend itself upside down, do an in-place flip, a barrel roll, and an inverted barrel roll, and have a 0.8s quarter mile while being co-witnessed by no fewer than 4 cameras (including multiple fly-over drones) at a live event.

Let's hope so, at least with OSes.

Realistically speaking, there's really no reason to NOT improve the performance of your software, now - especially when selling your software depends on the cost of the hardware being cheap, as has always been the case for Microsoft. LLMs have made this kind of bughunt/performance optimization obscenely cheap.

There is no good reason for the OS itself to use 4+GB of memory and interfere with application operation. It's just laziness on the part of the developers.

More broadly, I think the preponderance of shitstack slop written in python and typescript are more likely to be an impediment to long term optimization. And as long as it's the default choice by LLMs (as it seems to be), we'll be getting a lot more of it.

Let me know when they can get a useable install down to 4GB, officially. The last time I used Windows 11, it was atrocious even on 32GB.

You should be skipping unit tests, actually, and just plastering the project with e2e tests. Unit tests don't really help, unless you've had them written adversarially by a different model/agent based on the original spec. But if you're also able to be productive, you probably already know this.

That was the old, original business plan. You may not have noticed, but there's not been much talk of an IPO in the past month. The administration knocked down their attempts at regulatory capture (as did much of the industry). They missed the boat.

Not in decades, no. lol

Gemma 4 12B IT is a great model, but I'm increasingly disinterested in both Google's models and cloud models in particular.

I can run minicpm-v 4.7 at 300 tok/s on my local equipment (versus under 150 tok/s on qwen3.8 based models). LFM2.5 just released a 24B model that surpasses Qwen 3.6 in most benchmarks and does almost 180 tok/s on my M2 Ultra.

Local models are coming faster and deeper and it's almost impossible to even keep up with benchmark evaluations.