Software Development

869 Articles

Local LLMs Can Work Better Than Claude, At Least For Some

September 9, 2026 by 52 Comments

This is going to be a very personal question, because when you’re talking cloud vs local anything, it comes down to this: how much are you willing to pay for independence? The local option might save you long term, or it might never pay off the capital investment. It will almost certainly cost you your time to set up and maintain your own system — but what you get back is independence. With LLMs, traditionally you lose quite a bit of performance, but as [Anurag Singh] points out on XDA Developers, a lesser model might actually let you get more done, depending on your workflow.

[Anurag] had been on the 20$/month plan with Anthropic when he decided that the scratch just wasn’t worth the sniff– he was hitting usage limits he couldn’t stand at that level, but couldn’t justify a higher tier of access. So he decided to try a local LLM, even though all he had was a 16 GB MacBook Air M5, not a beefy workstation. Since his workflow isn’t so much ‘vibe code the whole thing for me’ as ‘help me find where I went wrong here, electronic rubber duck’, Qwen2.5 Coder 14B proved more than adequate for his use case.

It can’t understand all the moving parts of a large project as well as Claude can — not surprising given how old it is and how much memory it has to work with — but that’s [Anurag]’s job. He’s the programmer, it’s just the assistant. For his use case, he can make use of his existing hardware and having the the LLM right in VS Code is allows for a speedy workflow.

Your millage may vary, but if you want to get into locally running LLMs, we can point you at the easy ways to get started. Depending on your hardware, you might want to grab another GPU.

A 1024 Byte Python Interpreter

September 8, 2026 by 9 Comments

Like many of us, [Austin] finds writing code satisfying — especially if it is challenging. His latest challenge: shoehorn something that looks like Python into 512 bytes. Ok, that didn’t work out, but would you believe 1024 bytes of source code?

The goal was to properly execute a fizzbuzz program using a decidedly Python-like syntax. Since he is only interpreting Python, some things were simplified. In addition, he tried hard to minimize things like whitespace and variable names. Still, there were other things to do, and he borrowed from tips for code golfing.

The resulting code is essentially illegible, but it handles integer variables and literals, assignment, arithmetic, many control structures, functions, and even print. Not bad for 1 K.

Despite being tiny, the parser does a lot, but it does cut corners. No bytecode, no syntax tree, and no error handling. Variables have to be a single lowercase letter. So, ok, it isn’t really Python. But it is something.

Naturally, this reminded us of the obfuscated C code contest. If you really want to go small, we appreciate small Forth.

From A Ten-Line Script To A Real Utility With Codex

September 3, 2026 by 67 Comments

I’m an experienced programmer, and I’ve worked in many different languages. Sometimes being a programmer is a two-edged sword. You want to accomplish something, and you can do it easily — but it can be a lot of work to do it right. Maybe more work than you want to do.

Normally, I’ll kick out a few lines of script for something I want and be done, accepting that it isn’t production-hardened. This time, however, I decided to try an AI tool to see whether they could do the work I was too lazy to do myself. While I’ve played with chatbots, I wanted to try one of the dedicated coding agents, in this case, Codex. Outside of asking ChatGPT to write a simple function or find the cause of an error message, I haven’t done much coding with AI assistance, so I was interested to see what these agents brought to the table.

A Radio Problem

The problem was simple: I wanted an easy way to put buttons on my Linux desktop that launched Internet radio stations. Sure, I could open a player and paste in a long URL, but I’m far too lazy to remember all those URLs.

I searched for a way to make Shortwave — an Internet radio player — open a URL from the command line. Apparently, you can’t. Google Gemini suggested writing a script that launches cvlc, the command-line VLC player, with the URL as an argument.

That’s easy, so I did it. Of course, then I had to find the stream URLs for all my favorite stations. It turns out that Radio Browser maintains an extensive database of stations. I considered scraping the site or using its API, but honestly, the little script was becoming too much of a project.

Besides, I was already struggling to manage the media player’s lifetime. I didn’t want a new station playing on top of one that was already running, and I wanted a command to stop playback, so the script had already grown larger than I first imagined.

My first version used a temporary file containing the player’s process ID so a future script execution could kill the old player. That usually works, but it isn’t very robust, and I knew it. But how much work did I really want to do here? I decided I had done enough and turned the rest over to Codex, OpenAI’s coding assistant.

Continue reading “From A Ten-Line Script To A Real Utility With Codex” →

ShieldFont: Bludgeoning AI Scrapers That Disrespect Robots.txt

August 14, 2026 by 57 Comments

In the more innocent days of the World Wide Web you could simply put a robots.txt file in the root of your website that search engine indexing bots and similar would consult for the indexing wishes of the site owner. In this brave new world of LLM training data indexing such pleasantries are however rarely respected, leaving site owners to resort to increasingly more involved ways to bludgeon so-called AI scrapers, with ShieldFont being one of the most recent methods.

Its basic functioning is detailed in the white paper, explaining their use of ligatures. These are normally used to join multiple graphemes or letters into a single glyph which are rendered in the final text. By substituting about a quarter of the words in a text with such ligature-based versions in an intelligent, dictionary-based manner, the HTML version – as typically parsed by a scraper – will read as grammatically valid but nonsensical text, while the rendered font version will look normal.

Naturally, there are some disadvantages to this, such as screen readers for the visually impaired needing to also use the rendered font version, and it’s just as effective on legitimate search engine indexing bots. That said, if you apply this to static, archived content, or content marked as ‘do not follow’ in said robots.txt, then it might just be one way to make ChatGPT and friends spit out really funny output in the future now that the novelty of wood glue on pizza and eating rocks has somewhat worn off.

While LLM scrapers can adapt to this by also parsing the rendered text, this makes the scraping effort significantly more expensive. Together with maze traps like Nepenthes and Cloudflare’s offerings that seek to keep these scrapers busy scraping dynamically generated content through infinite linked pages, the tools available to combat the menace of these scrapers keep developing.

DOOM Using Regular Expressions

July 28, 2026 by 16 Comments

Regular expressions (regexes) are an amazingly powerful way to perform operations on collections of e.g. text. Regexss can also be considered to be a programming language, even a Turing complete one. Ergo it’s perfectly acceptable to thus design a way to use regexes to run and play a game of DOOM, as [Artem Lytkin] recently did.

The GitHub project page can be found here, containing the Python-based code that allows the demonstration to run, as well as the other components, including the C runtime and the 96.6 MB text string that defines a CPU’s registers, RAM, a framebuffer, the DOOM engine compiled to this custom CPU’s instruction set and the WAD file for the game itself.

The C-based driver applies the fixed, ordered list of find-and-replace rules to this string, which after more than ten-thousand of such substitutions later results in a single frame of the game. At about 80,000 substitutions per second on the given test system, that gets you to a sort-of playable framerate, even.

Naturally, the practical value of playing DOOM like this is pretty low, but as a demonstration of why regexes are awesome it’s hard to beat.

FreeBSD Just Removed The Last Of Its GPL-Licensed Code

July 28, 2026 by 46 Comments

With a recent change that removed dialog in favor of bsddialog, FreeBSD has now retired the last piece of GPL-licensed code in its base system. Although the impact of this change will be minor for users, it does highlight once again the Open Source divide that has split developers since the 1990s, when [Linus Torvalds] attached a scribbled set of notes to the v0.01 Linux kernel release that would later be replaced with the very similar GPL license and its derivatives.

This history and its impacts are also the subject of a recent video by [Brodie Robertson]. For the FreeBSD project the biggest change here is probably that the entire GNU subtree in the codebase is now gone, As detailed in the video, this is the end of a very long-running project, whereby FreeBSD in its early days incorporated GNU code for userland tools. These components and its replacements are detailed in the FreeBSD wiki.

Naturally, this change has upset some people for reasons best known to themselves, but from a project management perspective this makes a lot of sense. Not having BSD-incompatible licenses like GPL in your source code massively simplifies matters. The main difference being that the GPL is reciprocal, requiring that derivative works also be released under the GPL — a feature that is often at odds with commercial projects. Meanwhile, the BSD licenses merely require the use of BSD-licensed code to be declared.

The upshot of the BSD-license in the case of FreeBSD is that it has found it and its components used in many commercial products, including MacOS/OS X, the PlayStation’s Orbis OS, and of course the BSD networking stack is happily used in Windows, macOS and just about anywhere else. Meanwhile the GPL forced Linksys to open the firmware to its WRT54G series of routers, ultimately leading to the development of OpenWrt and similar projects.

Although the OSS licensing flamewars will likely never end, it’s hard to disagree that FreeBSD is pretty healthy at over thirty years old, even if using it as a desktop OS comes with a few asterisks.

Continue reading “FreeBSD Just Removed The Last Of Its GPL-Licensed Code” →

Codeberg Bans Cryptocurrency And LLM-Generated Code Projects

July 24, 2026 by 88 Comments

Community-led open source project hosting site Codeberg has formally announced that projects whose code is largely or fully machine-generated through LLMs and other ‘AI’ tools will no longer be welcome. This follows on the heels of a similar ban on cryptocurrency-related projects.

The community vote was on two issues, the first being the notion that scraping of project code for the use in LLMs should be forbidden, which was a motion that easily passed. The second motion was on disallowing projects whose code was substantially generated by LLMs like Claude, OpenAI Codex, and similar. This motion passed with 358 in favor versus 144 against.

In the earlier linked blog post the reasoning behind especially this second issue is expanded upon, covering not only ‘license whitewashing’, but also the direct and indirect hardware costs, with the expanding ‘AI’ datacenter hyperscaling having massively increased hardware costs for Codeberg over the past years, as the costs have been largely externalized.

Also covered is the aspect of these LLM-based tools destroying the OSS community, which is something that is backed up by recent studies. Even if we ignore that such LLM-tools are destroying the cognitive abilities of its users, there’s an argument to be made that if LLM-scraping is disallowed, then it’s consistent to also not allow LLM-generated code.

In the Terms of Use you can see these changes, both for LLMs and for cryptocurrency projects.

Thanks to [mk-fg] for the tip.