HN user

3form

647 karma
Posts1
Comments239
View on HN

The problem being that I don't think there's a definite proof that any of human thinking is more than a sum of high-granularity partial solutions that can be put together.

It could be that with enough tokens, big enough context window, and ability to dig out the relevant partials, many such thought processes could be simulated.

One example I can think of: privacy preserving ML models. Company wants to keep their model safe and also perhaps not to force you to run some classification on your phone for battery savings. You want your inference features to be private, because they e.g. contain your location data. You send them encrypted, you get encrypted results back and decrypt them.

It could have been obfuscated, but assuming we have HmE, obfuscation is more tricky to be done right.

Do we change the definition of encryption meaningfully in this process, though? If so, I don't see how really. It's just that out of set of potential encryption algorithms, for this purpose we would pick ones that are fit for it. It wouldn't be AES I guess... But it would still be encryption. Maybe a weaker one, but that weakness could be coincidental. I'm not familiar with any theorems in that space.

EDIT: I see a note that we lose ability to verify integrity of the data with HmE, but I guess it's still not really changing the definitions - just most of encryption used today also provides integrity guarantees as a kinda nice and desirable side effect, but it doesn't change that the encryption that doesn't still is encryption.

The point is to not have subkeys or any other way to access any part of the plaintex - as the goal is generally to offset the computation and the costs of it to another party. And yes, most likely it will require a very particular format of the data, and in all likelihood will allow only for a limited set of operations.

The idea of homomorphic encryption is to do things without the knowledge, and not gaining the knowledge. If ciphertext contains a number, and you don't need to know what number it does to always be able to multiply it by 2, you succeeded - as a simple example.

I don't know what's worse: that they did that, or that the operating system allowed them to do that. On both macOS and Windows according to my understanding that should require admin rights, normally, not to mention the degree to which Apple made macOS immutable (I'm not familiar with the details, to be honest).

This really does - I don't mean it in a disparaging way - read like a mid-life crisis writing. Having to come to terms with the fact that the "old world" is not the current world anymore, which is too bad, because you like the old one, and you don't quite fit in the new one either.

This feeling has existed across generations and most of us have to go through it eventually.

The world, however, is not any less real than it has always been and is not collapsing.

Maybe it's trivial what I'm saying, but hiring the most skilled available person does not solve the skill disparity.

I saw this with screenshots already for like 10-15 years now? It's some overly conservative policy to comply with fair use. I recall a page outlining that you should basically pick the minimum possible resolution that still allows you to distinguish the features of interest. I get why they do this, but it's really horrible from archiving and accessibility perspectives... And I would like to treat Wikipedia as one of the main archives of the world's collective knowledge.

All this goes to my "world has gone insane over IP law" bucket. Similar to people disallowing their games being streamed or even shared in screenshots.

Definitely AI generated. But the project is interesting, because that space felt a bit dry to me. netboot.xyz and iventoy are cool, but for most basic use cases I always felt these things could be yet further simplified. So I guess I'll go and review the code when I find some time.

EDIT: Found the disclosure in the repo: >I've used Claude CLI to help with some parts of this project - mostly making the web UI pretty, as I'm NOT a frontend developer. I also used it to generate the docs, but I review them manually - no automatically-generated AI code goes into the project without review from myself.

I guess that's fair.

Something that I'm missing from both the article and the comments - I would remap the sleep button to F4 at OS level. Repeat analogous steps for other keys.

Granted, you'll lose these functions, and likely switching to another keyboard will drive you mad, but I guess this is a good stopgap software based solution.

I am not well versed in Android or iPhone software development, but yes, I don't believe that making a non-bloated mobile app is pushing the frontier of software engineering.

There could be some arguments made somewhere as to where R&D money could go, perhaps somewhere in the backbone that billions could use, but the UI is not it.

All that said, I don't know how it furthers your initial argument exactly, as the DMA "beneficiaries" benefit from this lightweightness in zero percent. If anything, it's a negative, because one could assume they have to do better than that with what they're offering.

R&D spend? In messaging product?

Sorry, but these companies spend much more effort on making sure their product is walled off and incompatible with everything than giving it any actual quality.

The company is very small, and they're doing a lot with what they have. Steam alone is full of arcane features that I keep discovering. There's a lot of backend stuff. They're making games and hardware.

Perhaps some of this is contracted, similar to the Linux compat and drivers, but it's still impressive to me, compared to the orgs like Spotify, order of magnitude larger with barely any features at all. (I understand there's legal, huge backend, and I didn't see many bugs over time, but still)

I guess from my perspective there are even more dire problems in the US that I'm surprised people accept. But it seems they don't know, or care, or know that they should care.

Perhaps it's the lack of proper authoritarian regime in the US' past that drives this. I believe the temporal proximity of such makes people aware of, and angry against, the many traps that such systems leave in their "law", so you can be imprisoned anytime for anything. EU has a bunch of countries with varying degree of such past.

Feature A exists for years. New feature B enters. If you want to opt out of feature B, you now need to opt out of feature A as well, which you probably liked much more.

I don't know a better term than blackmail to describe this. Thesaurus seems to imply blackmail is for money, but so is extortion, and it doesn't give me good suggestions.

EDIT: Coercion is more correct, but it's way too mild in my perception.