> For the same reason, plugging in an external keyboard is also a no-go since freshly updated iPhones are placed in what's known as a Before First Unlock state, which prevents wired accessories from working until the passcode is entered.

The user can't even enter their passcode, how do you expect them to perform code execution?

Having to write all the specs and tests just right so you can regenerate the code until you get the desired output just sounds like an expensive version of the infinite monkey theorem, but with LLMs instead of monkeys.

> Oh, and mozilla's off-line translate for private translation of web pages... that's another neat AI thing they added that I've found super helpful.

Yes, it's awesome! And one of my favorite additions to Firefox in many years, it's stuff like that they should focus on if they want AI, imo.

[1]: https://wiki.archlinux.org/title/Speech_dispatcher

I can't seem to find anything that mentions a Firefox integration though?

All of the small LLM models break down as soon as you try to do something that isn't written in English, because - surprise - they're just too small.

There would need to be a hardware breakthrough, or they would have to somehow solve the heavy cost of switching the models between pages.

Instead of useful AI stuff that is a clear improvement to accessibility, they're insistent on ham-fisting LLM solutions that no one have even asked for.

Off the top of my head, they could instead:

1. Integrate something like whisper to add automatic captions to videos or transcribe audio

2. Integrate one of the many really great text to speech models to read articles or text out loud

PostNord Denmark has been operating with massive losses for a while now, in part because they were required by law to be able to deliver everywhere in Denmark, when there were very little demand for it. The money just isn't there, which is why the law has been changed.

The cost of sending a letter was also just going up and up. In 2025, it cost $4.55 _per letter_.

I heard good things about Nomad (albeit from before Hashicorp changed their licenses): https://developer.hashicorp.com/nomad

I got the impression it was like a smaller, more opinionated k8s. Like a mix between Docker Swarm and k8s.

It's rare that I see it mentioned though, so I'm not sure how big the community is.

Ah, that makes sense, I guess. I also get the feeling that the language itself is still under very active development, so until 1.0 is released I don't think it matters too much what it's implemented in.

> Also, so much of the k8s ecosystem is in Go that it was a natural choice.

That might turn out to be a costly decision, imho. I wanted to use CUE to manage a repository of schema definitions, and from these I wanted to generate other formats, such as JSON schemas, with constraints hopefully taken from the high-level CUE.

I figured I'd try and hack something together, but it was a complete non-starter since I don't work within the Go ecosystem.

Projects like the cue language live and breathe from an active community with related tooling, so the decision still really boggles my mind.

I'll stay optimistic and hope that once it reaches 1.0, someone will write an implementation that is easily embedded for my use-cases. I won't hold my breath though, since the scope is getting quite big.

I've been trying to apply CUE to my work, but the tooling just isn't there for much of what I need yet. It also seems really short-sighted that it is implemented in Go which is notoriously bad for embedding.

There's also rustic, which supposedly is optimized for memory: https://rustic.cli.rs/docs/

There's just not enough capacity to build memory fast enough right now. Everyone needs the biggest and fastest modules they can get, since it directly impacts the performance of the models.

There's still a lot of happening to improve memory, like the latest Titans paper: https://arxiv.org/abs/2501.00663

So I think until a breakthrough happens or the fabs catch up, it'll be this painful race to build more datacenters.

How does this work? Does it rely on GPT to extract the data or does it actually generate a bunch of selectors? If it's the former, then the results aren't reliable since it can just hallucinate whole results or even just parts.

Comments URL: https://news.ycombinator.com/item?id=41385569

Points: 2

# Comments: 0

Took me a while before I actually realized this limitation as well. I started noticing it when I was searching for things I didn't know exactly how to find.

Kagi is great for when you know exactly what to find and have a general idea of where, but when you actually need those n-page results to help you refine your search, and you can't, it's a showstopper for me.

I'm not even sure why it's a limitation. It might not be intentional?

Yeah, that's been an increasing problem.. Google has been especially bad with this for a while now.

When I tried to search for something where I wasn't sure exactly what I was looking for, I could only change my query so many times until it stopped being what I wanted.