TehPers

Poor lonelycpp. They became a target for major nations worldwide over such a dumb reason.

If I were them, I’d do anything I can to break the app immediately and force them to use anything else. I wouldn’t want to be in the supply chain for an app like this where being compromised is immediately a RCE incident for a government app.

What I usually push for is that every CI task either sets up the environment or executes that one command™ for that task. For example, that command can be uv run ruff check or cargo fmt --all -- --check or whatever.

Where the CI-runs-one-script-only (or no-CI) approach falls apart for me is when you want to have a deployment pipeline. It’s usually best not to have deployment secrets stored in any dev machine, so a good place to keep them is in your CI configs (and all major platforms support secrets stored with an environment, variable groups, etc). Of course, I’m referring here to work on a larger team, where permission to deploy needs to be transferrable, but you don’t really want to be rotating deployment secrets all the time either. This means you’re running code in the pipeline that you can’t run locally in order to deploy it.

It also doesn’t work well when you build for multiple platforms. For example, I have Rust projects that build and test on Windows, MacOS, and Linux which is only possible by running those on multiple runners (each on a different OS and, in MacOS’s case, CPU architecture).

The compromise of one-script-per-task can usually work even in these situations, from my experience. You still get to use things like GitHub’s matrix, for example, to run multiple runners in parallel. It just means you have different commands for different things now.

This is exactly what I was thinking. They aren’t programmed to follow the user’s instructions to begin with. Why is it a surprise when they deviate from them?

It’s a fundamental misunderstanding of the ML that goes into these LLMs. They are prediction machines. They might have “specialist” submodels or whatever that are better at predicting specific areas, but that’s about it.

Don’t worry. I’ll sanitize his statement:

That god damn required age bullshit’s going to fuck up Arch and every other distro worth a damn.

Hope I was helpful!

But what caused this girl to sue? The article didn’t mention a big event afaik.

Based on the article, there are supposedly hundreds of these lawsuits. While I couldn’t say what caused her to sue, the outcome opens the floodgates. If anything, it’s accountability for some of the harms these platforms have profited on if it means they can now be sued for millions per person for it.

So is this enough for you then?

If entire distros default to 1/1/1970 then it might not be super helpful.

Might be interesting to see if some anti-tracking distros outside of these jurisdictions are interested in doing something like that.

What’s wrong with the title? It’s the article title.

Also, Beehaw has no downvotes, so you can downvote all you want on your instance if it makes you feel better, but it won’t federate the downvote.

As a first language, JS is too much. They would need to learn three languages to make websites (JS, CSS, and HTML).

I’d start with Python. It’s easy to learn, and modern Python gives you the tools to write code that’s easy to read and follow without being too verbose.

uv should make things very easy to setup too. Install uv, then give them a starter repo with the Python version set. uv run should just work after that, no manual venv/conda/etc nonsense involved.

My school didn’t have a course, but the test was entirely in Java.

C would have been a lot more useful.

LTS for anything around LLM lasts about 3 months these days, so not really surprising.

Honestly, it’s shocking Sora was around for how long it was. Remember Dall-e before it got consumed by ChatGPT?

Don’t worry. This will all get reverted real quick once it makes its way into a user prompt for headless installations. Imagine needing to pass a DOB in through stdin somehow every time you docker run ubuntu lol.

Like ISO 8601. /s

Sounds like Google decided to take dead internet theory into their own hands.

I recently built a voice-to-text agent in Rust

Agent…? Uh, okay, let’s just use that word for everything now.

I did not have the Rust toolchain installed on my system. I simply told the coding agent that I use Nix, and it figured out how to pull in the entire Rust toolchain through Nix, compile the project inside an isolated shell and produce a working binary.

Sorry, where is the part where you built something?

Anyway, NixOS gets a lot of praise. Maybe it’s something I should try if Manjaro doesn’t survive its current drama (though it seems like they have a path forward now).

What’s with all these articles assuming some fictional reality that coding is replaced by LLMs now? Find me a tool that can build any software anyone wants to build, then ask yourself if you trust that tool to do your taxes for you (if you’re in the US anyway), book an entire 3 week $10k vacation for you, manage your finances and set budgets for you, and so on.

What software does the author think people will build with these tools that’s “fast fashion” style slop? Even assuming a non-coder wants some random one-off software for something, in what world would they ask a LLM to write it for them rather than look for an existing program that does what they want?

In the dev world, LLMs have proven that they cannot build production-quality software by themselves. They can build pretty demos maybe, but everything from Amazon’s tool recreating prod to whatever the fuck the Windows devs are doing shows that trusting an LLM to work on real software will eventually backfire.

Working with ML libraries is a lot different than traditional software dev since those tend to be a lot more loosely typed and grounded in mathematics (linear algebra, calculus, etc).

To answer all of your questions at once: just keep writing code. Work on a lot of projects! Try reading the source code for other projects, like the libraries and tools you use.

Try to use LLMs as little as possible when writing code as well. It won’t help you practice.

This is neat! It might be useful to turn this into more of a CLI utility by having a mode for accepting some Gregorian date via stdin or CLI option, and outputting a parseable format for the Symmetry454 date via stdout. I’d recommend looking at the argparse module if you’re interested, though there are some third party packages that can make it even easier to parse program arguments.

They take something like “please sort your imports. We agreed to use isort last week” as a personal attack.

I would take this personally as well, to be honest. Using isort over Ruff? Blasphemy.

Is that little Bobby Tables? I guess he’d be grown up now, but he’ll always be little to me.