mhitza

Ha! That's pure nonsense propaganda. Show me where it is 10x easier to take disability than establish a company. Then show me how EU destroyed Germany's nuclear energy, while doing well in other countries.

If you can't you are no better than the russian and right-wing americans that push anti-EU propaganda.

For me htmx is a utility library to spruce up a couple of interfaces here and there that benefit from a bit of reactivity. Anything more complex I really don't feel that it's the right tool. Or that there is any great tool to build SPAs.

There is sometimes this trends on projects of using anything like a universal hammer. And then they SPA their project.

If it's technically.implemented uniformely without thought of user access patterns, then yes it's a bad idea causing friction and frustration.

If I drag&drop a file in an app or it's icon, or by right click. The permission would be implicit.

For configuration files they could be part of allow rules, since they are so common if they follow the XDG specification.

For saving files, flatpak apps could be allowed to save anywhere in the users home sub-directories (not home top-level) as long as it doesn't overwrite any existing file. There are downloads, public, documents pictures, videos, that are barely used by default.

There is plenty of room for better desktop based integration, while of course lacking funding and care for users (thinking of particular devs in this case).

Currently it's a mishmash of separately using flatseal (and equivalents) and having your fingers crossed if you're likely to exfiltrate or not your data.

Premature optimization argument fits right in. Now that memory is up to 10x more expensive it is worth considering optimizing programs with large memory footprint.

Use of the term of direct democracy feels dilution in that case. It is a significant step in that direction, though for me (and wikipedia) direct democracy is without the middleman. Switzerland still has parties, elected government and so forth, no?

If someone else hasn't given a thought yet to Liquid Democracy, I propose they give it a couple of minutes of imaginary thought https://en.wikipedia.org/wiki/Liquid_democracy

For some real world data there are some studies about (I think) German branch of the Pirate Party using it in internal decision making and there is interesting insight for those curious.

It be nice if we all where allowed to vote on particular laws if interested, and defer certain topics to our delegated experts. (Chat Control for example, would be a non issue then.)

The technology for that approach has been mature enough since the rise of social media and private crypto currencies (as a means of electronic private vote, particularly around usage of smart contracts for delegation). What about the interest and spreading of these ideas? Pretty limited, to non-existent. Some people like to follow leaders, though naturally wouldn't expect it to be the silent majority.

I think this regulation that comes into effect February of next year concerns all kinds of batteries, not only those of smartphones https://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CEL... . I think, because I didn't doublecheck right now further than reading the first matching "appliances" paragraph (ctrl-f).

What I know off the top of my head, from reading the legislation a while ago. Is that technical reasons, such as waterproof devices, will still be able to retain non user replaceable batteries if it is needed to keep the device waterproof.

If you are refering in terms of the code of practice part of the EU AI Act

> I've read that watermarking should in theory be impossible to detect except by the entity that watermarked it

This is a carveout exception, for watermarking. In the spirit of those terms it should be machine identifiable.

In my opinion they should have thought better about this, paricularly for text, because in its current forms it is easy to lead next to a new "tamper-proof" requirement, which in practice is DRM. And we do not need more DRM.

For images, music there is metadata already where such information can be stored. And if end users are found using unlabeled AI their accounts could be ban from these platforms. Not something the social platforms might want, but it's a saner approach than trying to reinvent the secret printer dots on all generated media.

They had large increase in volume, and trying to move to Azure at the same time. I don't envy them for either work they need to do. But also don't feel pithy because it's Microslop, at the end of the day.

What do you use instead of llama.cpp? With vllm for example most models don't seem to be supported out of the box.

Yes, it is very impressive what they can do by recycling copyrighted material. They are more impressive for me when they are not used in agentic contexts. Though that doesn't sell hype anymore to inflate valuation.

For more than a year now I was renting a limited GPU server for ~300$/month to learn, experiment, research and build internal tooling around open weight models. Thinking they are tools with potential and buying the exaggerated marketing are different things.

My history of comments on HN lands often on both providing what I believe to be my insights working with LLMs and calling out exaggerations, stupid terms of service, and the other mishaps in the field. You are free to browse them if you'd like to see my broader opinion.

I'm a pleb developer, don't have the smarts, the prestige or the salary of those working at BigTech.

When AI firms ate more than half of global VC private investment in 2025 https://www.oecd.org/en/about/news/announcements/2026/02/ai-... I would expect better results than what we have today.

The most well paid people in the industry brought us here. And "here" is very much as fuzzy as last year with better harnessing towards the local optima. And I say local optima because even the perceived capabilities have slowed down, nevermind the benchmark numbers which are in aggrement.

The best paid engineers in the world, with almost no practical budget limit, still deliver shoddy quality software with AI. Is that not fact? And if it is what does that say for the rest of us.

You are allowed to believe. I'm still waiting for the beneficial results, not only those that benefit griefters, hackers and scammers. AI has been a huge boon there.

Reality will materialize and markets will redress hopefully once they go public. Which they very much seem to be hesitant to do right now.

You can generally understand your AWS cloud usage, and waste can be self evident with their existing tools. Not at all with llms.

A postgres index post is unlikely to reach front page. It's already part if the docs, and should include more context to be read worthy.

They are not equal comparison.

This before the fact that there is no guarantee that a model follows your agent instructions (plenty of easy to reach for research on it), and you also get suggestions by devs at these companies to wipe parts of your model's instructions because the model is better now tm.

If cloud providers change their billing quasi monthly, and if you'd need to fiddle with your indexes every couple of days. I'm not sure we'd be using them as much.

There is interesting information about the inference pipeline, but almost too late to the party (by at least a year), and for which audience? Techies understand in broad strokes the tech if they are interested, normies will definitely not read it.

All that to say, that yes, it's worth having a laugh. If for nothing else, as a release valve for all the problems they create in the real non-VC world.

Anthropic is IPOing in October according to news, you might be interested in investing.

Yes, llama.cpp supports presets (an ini format). Though it's not always clear which llama-server flags it does not support.

Thanks for the laugh with the netflix logo. I feel that is very apt when all their recent shows I've glanced at seem to have their scripts brought from Temu or from a genai prompt.

There are indeed potentials for annonymous service providers, or ad-hoc services outside governmental control.

However for anything consumer protection related, they must not be only transparent, but I'd argue, clearly obvious who is behind the service.

In the context of the previous comment there was an implied equivalence between individual and corporation rights to privacy.

Large corporations already get lenient passes for breaking the laws and regulations. Let us not propagate ideas that champion for even more things that benefit corporations.

For individuals, yes. For service providers, no.

> AI to write code with recent verdict over the IP rights of AI generated code.

Which recent verdict?

> It would be interesting to see how this effects the copyleft Licences with contributors are using AI for the PRs

I know Linux and GCC have been diligent about tagging and tracking LLM based contributions. In the worst case they can chuck it all out and handwrite it back.

General outage was uncommon, but unicorns and octocat popping up for particular pages/repos was common enough for me in the past.

Relinking the recent study which argues that many instructions in the context are not followed through https://news.ycombinator.com/item?id=49096969

Many of these skills, rules, "playbooks", and such are kitchensink attempts at steering the model. It augments the model to frame it's reasoning according to project rules and flows, but cannot be really trusted to adhere to it. More of a vibe guideline.

But could also be a good software tester if given any kind of remote connection support feature.

One of my early experiments last year with open source models and context size was with GPT-OSS 20B (the mxfp4, the "smart" 4-bit quantization). Even though it boasted a 128k context size it was bad at recall around 32k characters (didn't bother to implement the tokenizer for counting).

The recall text was a simple hash generated, filler text from a dictionary file and a request at the end to return only the hash from the beginning of the prompt. Past 32k characters the response contained hallucinations of characters or full hashes.

Just having large context size doesn't paint a full picture of capabilities, prompt adherence and other quality metrics.

Alternative source https://finance.yahoo.com/technology/ai/articles/nvidia-plan...

> The concern is familiar: NVIDIA money funds customers who then buy NVIDIA chips.

That's an uncharitable interpretation of the initial comment.

It can also be read as health-sensible reminder given the climate warnings. AC units demand is already up 26% year over year in my country (according to our biggest online retailer, a month ago). Will be a highly profitable year for manufacturers.

Are you young and raised with the cult of personality, like so many gen Z and alpha?

Because I find your words surprising since critique is what I was raised with in the late 90s early 2000s. Offline and in the media.

At the same time your definition of criticizing is very lax, when I'm stating banal opinions and a broad observation.

Love the typo.

I think Musk's luck is running out, he's been too political this past year and he seems to also have run out of geek "street cred". Remember youtubers visiting (whatever the Tesla trade show was called) were more perplexed than impressed with what has been shown. Similar sentiment seen across trade shows (like CES) where everything was AI and no one seemed to ask themselves "why are we putting AI in this?". The answer, of course, for startups and stock listed companies, to make themselves more valuable on paper.

Investors would probably be aware of that if they run any kind of sentiment analysis on online social media content. And at the same time there's some pull-out from tech stock. So maybe some are starting to realise that things are too shaky for their risk profile?

I'm in agreement with your claim, though if you recall some of these micro SaaSes went hyperfocused on niches. The value proposition is then smaller, with larger data liability than with some behemoth's of SaaS.

Though for the later ones, one might also ask themselves if they are worth their cost when using 20-30% of provided functionality.

H) is my favorite. Nicely detailed birds, less crowded designs.

Anything with a face, I'm tired of that trope of banknote design.

Why would anyone use your micro SaaS when they can build their own?

Have you seen a different perspective? Cause in my informational bubble SaaS ain't doing too well since vibecoding became more common.

Another practical application would have been if they would have charged for CI - by the minutes - for projects with no user or contributor community.

"Forward deployed engineers" starts to seem like a trend with some of these AI companies.

Selling high-rate outsourcing services with surely guaranteed lock-in mechanics, under the umbrella of accelerating AI adoption within the industry.

While before you at least owned the code and the know-how, now you can outsource that too while you're getting in return systems that share data with third parties and might also be generally vibecoded.

What a sweet deal! /s