dzhar11

From FAB's AGENT.MD: > - Avoid magic numbers and strings by extracting recurring or meaningful values into descriptive constants (const) or enums. ---

I've been seeing the same thing with models like GPT5.6 and Opus4.8 in GH Cop CLI. They still introduce magic numbers, and in Scala they often put an entire 10-line Spark expression inside an if condition instead of extracting it into a meaningfully named value to keep "if" readable. I wonder when common sense instructions will be baked into the models.

This article somewhat reflects my experience with autonomous agentic coding. I've run several experiments with similar results: the agent burns through all my tokens while making very little progress, or produces something unacceptable.

So I'd rather micromanage the process step by step. It takes more of my time, but the result is much, much closer to what I actually wanted.

The issue is that the Claude's C Compiler repository does not explicitly identify the project as a proof of concept. It was largely produced by one person directing Claude, and Bun's Rust rewrite also seems to have been driven by roughly one person using Claude Code. That similarity is what worries me. Bun's rewrite could also turn out to be proof of concept.

Jarred, thank you for working on Bun. Many "vibe coded" :D projects start strong and are later abandoned (like potentially Anthropic C), so I understand why people worry about Bun's future. I hope Bun lasts for many years, like GCC. Bun is fast and great to use.

People! What are you talking about? Who said Clippy was hostile? How? Clippy was always there when I had to write an annoying essay, a long letter, or something else for school. It reassured me that everything was saving correctly on my 1.44" diskette. The real nightmare back then was bad blocks on your HDD or diskette -- not Clippy. Clippy is my friend.

Sentiment toward Clippy-as-software (approximate, opinionated comments only)

    Clippy hostile       |######################......|  ~70 %
    Clippy not hostile   |########....................|  ~30 %

Scale: 30 columns; '#' ≈ 2–3 % of clearly opinionated comments

I appreciate the author's notes on managing ADHD. I was glad to find I didn't learn much new, because I'm already applying many of these practices.

I've tried several to-do apps, but centralized systems didn't work for me. I now use multiple to-do lists across different media: some on paper, some on my phone, others in markdown files within project folders... sending email to myself. It may seem messy, but it works for me. One system doesn't fit everyone. And any customization and tweaks are encouraged.

Books that helped me:

Atomic Habits by James Clear emphasizes small, consistent changes. Over time, they build into significant improvement. It’s better to improve your system gradually than attempt a major overhaul.

The Now Habit by Neil Fiore offers tools to overcome procrastination and start tasks. It helped me understand my resistance and find ways to move forward.

Getting Things Done by David Allen focuses on reviews and planning. I struggle to make time for them regularly, but even occasional reviews help.

Marcus Aurelius' Meditations, especially annotated versions, provide a Stoic perspective. His reflections on virtue and responsibility inspire me. He seemed to dislike being emperor, preferring philosophy, but accepted his role out of duty. That example helps me do necessary tasks, even when I don't feel like it.

Please, rewrite it in Rust. :D

For my personal single-html projects, I prefer a straightforward coding style that avoids external dependencies. I implement what I need in a simple, quick, and efficient manner using only native browser APIs, supplemented with a few helpers like those in 'Hn.js' to make the rest of the code shorter.

I am trying to learn and use Emacs, but I am not an expert. Surprisingly, Emacs 29 runs very good on Windows. I am happy with eshell, dired, trying to use emacs more for some simple tasks. I am experimenting with sharing emacs init.el configuration via local git + symlinks between 3 machines: 2 Windows and 1 Linux. Package configuration is challenging; sometimes, it works on Windows but breaks on Linux. Only nyan-cat never caused any issues ;)

I am in my late 30s ;) I understand you. We are... singularity. This good feeling when ChatGPT does what I want. I am working on my pet project and struggling to convince my friends and coworkers how cool my project is, but ChatGPT understands what I am doing and is ready to help. It reminds me of my addiction to online video games: World of Tanks, Team Fortress 2, PUBG, Dead by Daylight... It is a shot of dopamine when I win and anger when I lose. Burnout after some time... can't play anymore, but I have to... Oh, I have wasted (or invested) a few full years (like when you add all hours on Steam and divide by hours in a year) in video games (instead of sleep). At least now I can feel when burnout is coming. I have learned when to stop. No magic. Just time.

Wow, this is a fantastic project; thank you for sharing! For a few years, I've been thinking about automating something with remote NumPad. I've also got "Velocifire" numpad and used it as MouseKeys. This project is a huge inspiration ;)