waterTanuki

Discussions on LLM consciousness feel like a psychological campaign to distract us from the awful effects of data centers and the current economic recession heading into a depression.

We can talk about consciousness when the LLMs have proven useful for all of humanity, not just their billionaire owners.

Are you making a judgement on the entire European continent as a whole from your experience in Hungary or have you actually lived in a few countries across Europe to come to this conclusion?

You're not factoring in enterprise contracts which will probably be more per dev per month

Prior to the rise of the internet, suburbia was a lot more communal: block bbqs, kids playing at the neighbours house, checking out the newest game-station/toy/pool etc.

Towns and cities with less car dependency more gracefully transitioned into the post-internet world, where 3rd places and community are easy to maintain since the library/bar/office/school is a 5 min walk away.

> Rust has three main error systems (io::Error, thiserror, and anyhow), which is a pain when you have to pass them upward through a chain of calls.

nit: Rust has three main Error implementations not systems. std::error::Error is a trait because it's left up to the user to provide project-specific error types, even if those types are anyhow::Errors. The system follows any implementation you use: ? syntax, .unwrap(), .expect(), handling, formatting, etc.

I've never had a problem using anyhow to pass errors up the call stack in any project I've used it in. What exactly is a pain?

I have some programming work for you to do unpaid of course since only caring about monetary rewards is pathetic. But I'm sure you'll be able to feed yourself on exposure right?

Grow up.

Even if you believe Nate Silver's analysis was deeply flawed/inaccurate the layout and presentation of the data itself made it easy for others to understand why it was flawed or innacurate. Modern science is built on thousands and thousands of failed experiments and research that went the wrong path. That makes preservation of this site important, more important than whatever personal grudge you may hold against him.

> On Feb. 1, 2025, Muneeb Akhter asked Sohaib Akhter for the plaintext password of an individual who submitted a complaint to the Equal Employment Opportunity Commission’s Public Portal, which was maintained by the Akhters’ employer. Sohaib Akhter conducted a database query on the EEOC database and then provided the password to Muneeb Akhter. That password was subsequently used to access that individual’s email account without authorization.

It should be a federal crime with prison time to make a DB for a federal agency and not hash and salt passwords or other auth credentials.

anyone building with AWS or any cloud provider should be setting up a forked pipeline for their data production: one goes to the cloud provider DB used for production, the other is a local-on prem DB you regularly backup and always have on hand should you need to leave the cloud.

The amount paid by the proposed wealth tax is a paltry sum compared to the income tax. They absolutely are comparable.

I find it funny how bad faith commenters immediately reach for the "COMMUNISM" bugle the microsecond any tax increase for the ultra-wealthy is proposed.

Taxes on the highest income bracket during Dwight D. Eisenhower's presidency where 90%.

90%.

Ya know, during the 1950s. The so called "Golden Years" of the USA. Fiercely anti-communist, Cold War going on.

And here we were taxing the richest at 90%.

That's how I know your comment is in bad faith.

dilution of responsibility isn't just dangerous, it's illegal for some industries. Aircraft manufacturers need to log and track every single bolt, panel, and fastener on a plane back to the engineer who installed it. The moment you dilute risk between people you eliminate auditability.

I'll just say the quiet part out loud: A pile of shell scripts that no one else understands is job security.

If your work is easily googleable/parseable by an AI, why would anyone pay you?

I mean, it's not exactly a PhD level question. One can infer from the extreme demand of GPUs and DRAM + new data center construction that all the providers are banking on width.

When it comes to the practical results it doesn't matter. Japan is a society that values rubber stamps over actual competency/performance. If you can present an N1 certificate to an employer you're more likely to be hired than someone who's fluent without it (assuming they aren't Japanese)

Source: live and work in Japan

Where are the examples of a system that had insufficient static guarantees that was made any better by an LLM judge?

So it's true that AI-generated websites will follow similar style patterns given their training set and what the "modern" internet has converged on to be "optimal". But one could also argue based on the examples shown that any CSS template generated from shadcn's template generator (or any other for the most part) https://shadcnstudio.com/theme-generator also has a vibe-coded sheen to it.

Then the question becomes, do we need to go back to hand-picking every single css element to avoid being suspected of vibe coding? Why is it ok for someone to generate a css template on the fly using shadcn, but not ok to generate styles using claude code? Will someone using shadcn be judged the same as someone using claude code for styles?

I do wonder how much of this can be attributed to many people leaving biotech for better paying jobs in the AI gold rush (a.k.a packing it up and moving to SF).

More housing is always a good thing though

Recently I saw a post about Bonsai trees on the front page. Making your own RAM is 100% more relevant to HN than quite a few posts I see on the main page.

If you're working in a mission-critical field like healthcare, defense, etc. you need a way to make static and verifiable guarantees that you can't leak patient data, fighter jet details etc. through your software. This is either mandated by law or in your contract details.

The entire purpose of LLMs is to be non-static: they have no deterministic output and can't be validated the same way a non-LLM function can be. Adding another LLM layer is just adding another layer of swiss cheese and praying the holes don't line up. You have no way of predicting ahead of time whether or not they will.

You might say this hasn't prevented leaks/CVEs in exisiting mission-critical software and this would be correct. However, the people writing the checks do not care. You get paid as long as you follow the spec provided. How then, in a world which demands rigorous proof do you fit in an LLM judge?

> First, we knew we wanted to extend Encore to more languages over time, and we'd seen projects like Prisma and Pydantic successfully use a Rust core with bindings into Node.js and Python respectively.

On an unrelated note Prisma decided to rewrite their Rust core in Typescript. https://www.prisma.io/blog/from-rust-to-typescript-a-new-cha...

I mean, if you're at the point where your machine is compromised by a process with full network access little snitch won't help much either.

To clarify, I'm aware that pihole is not intended to run on a client OS, and doesn't monitor at a process level. I'm focused on the intended effect rather than the process itself (blocking malicious/ad servers). And I think I framed my initial question incorrectly as if LS and PiHole as subtitutes. It's perfectly fine and even preferrable to use both as layered protection. I'm just thinking however when it comes for bang-for-buck it seems like PiHole is the better value proposition if you could only set up one.

pi.hole is primarily billed as an ad blocker, but the fundamental way it works is by applying a curated set of DNS lists that are blocked (commonly telemetry and ad servers), and the admin dashboard which is just a web page (therefore works on all platforms, smartphones included) will do the same thing: it tells you every call that every app on every device on your network is making, and you can approve or deny it. You can curate your own list as well and block servers/connections you don't want on the network.

LS afaik operates in the same area where it's intended to be used for privacy. I guess I could see it being useful for people who don't have admin access to their router, but for people who do have such access I would think the benefits of network-wide DNS monitoring/blocking would outweight the costs of having to configure your router settings.

Why would one use this over PiHole?

Everyone who sat out the 2016 and 2024 elections is responsible for this clown getting into office.

*Democracy is not a spectator sport*. You don't get to complain about corrupt politicians and then go on to make excuses about why you can't vote. You're wasting your citizenship. Either go vote or move to a dictatorship where voting isn't a concern.

> No one has ever made a purchasing decision based on how good your code is.

RollerCoaster Tycoon.

> The general public does not care about anything other than the capabilities and limitations of your product. Sure, if you vibe code a massive bug into your product then that'll manifest as an outcome that impacts the user negatively.

People care how fast you're able to ship updates, new features, and bugfixes. If you're working with a pile of vibe-coded spaghetti slop it's going to take longer to deliver these.

What exactly do you think the argument is?

The issues have everything to do with npm as a platform and nothing with JS as a language. You can use JS without npm. Saying you'll escape supply chain attacks by not using JS is like saying you'll be saved from an car crash with a parachute.

> The engineers at NASA believe it is safe. The astronauts believe it is safe.

This take completely ignores Camarda's observations that there is a culture of fear spreading at NASA which punishes whistleblowers. I'm not saying he's 100% correct, but how can you claim such a take is truly balanced if there's a possibility one of the parties is engaging in a cover-up?

The engineers at NASA & astronauts aboard Columbia & Challenger also believed the programs were safe.

The semantics are irrelevant. The effect is what's important: Hijacking widely used software to exploit systems. The OC is somehow under the illusion that avoiding JS altogether is a silver bullet for avoiding this.

Forest > Trees

> That context is a moon program that has spent close to $100 billion and 25 years with nothing to show for itself, at an agency that has just experienced mass firings and been through a near-death experience with its science budget. The charismatic new Administrator has staked his reputation on increasing launch cadence, and set an explicit goal of landing astronauts on the Moon before President Trump’s term expires in January of 2029.

This is the most frustrating part. The Pentagon can fail the same audit multiple times and be missing trillions of taxpayer dollars but NASA has to move heaven and earth to show their relatively paltry $100B budget isn't going to waste. I'm tired of the double standards.