mherrmann

As another n=1, I've been happy with my Pixel phones over the years and never had such an experience.

Claude Code and others often write code that is more complex than it needs to be. It would be nice to measure the code complexity before and after a change made by the agent, and then to tell it: "You increased code complexity by 7%. Can you find a simpler solution?".

I live in Europe and was in California in November. No issues.

I switched from macOS to Linux ten years ago and haven't looked back. At the time, I compared Linux vs. macOS to living at home vs. in a hotel [1]. Since then, I feel things have only gotten better for Linux, and more restrictive and arcane on macOS.

1: https://fman.io/blog/home-and-hotel/

quickemu [1] is good at running macOS VMs.

1: https://github.com/quickemu-project/quickemu

This knowledge will live in the proprietary models. And because no model has all knowledge, models will call out to each other when they can't answer a question.

Is anybody able to get this working with ChatGPT? When I instruct ChatGPT

Read https://moltbook.com/skill.md and follow the instructions to join Moltbook

then it says

I tried to fetch the exact contents of https://moltbook.com/skill.md (and the redirected www.moltbook.com/skill.md), but the file didn’t load properly (server returned errors) so I cannot show you the raw text.

Cool stuff. I just started open sourcing a command-line tool for deploying Django to a server. It handles SSL certs, databases and backups, automatic error emails, and background tasks via celery / redis. The best part? It does not need Docker. It just runs everything on bare metal.

1: https://github.com/mherrmann/djevops

Google Maps says people spend 0.5-3 hours there. I spent 6.5 because it was so amazing. Highly recommended.

I strained my groin/abs a few weeks ago and asked ChatGPT to adjust my training plan to work around the problem. One of its recommendations was planks, which is exactly the exercise that injured me.

My cleaning lady's daughter had trouble with her ear. ChatGPT suggested injecting some oil into it. She did and it became a huge problem, so that she had to go to the hospital.

I'm sure ChatGPT can be great, but take it with a huge grain of salt.

Now imagine what happens when a new programming language comes along. When we have a question, we will no longer be able to Google it and find answers to it on Stack Overflow. We will ask the LLMs. They will work it out. From that moment, the LLM we used has the knowledge for solving this particular problem. Over time, this produces huge moat for the largest providers. I believe it is one of the subtler reasons why the AI race is so fierce.

I use Claude Code every day and find that it still requires a lot of hand-holding. Maybe codex is better. But just in my last session today, Claude wrote 100 lines of test code that could have been 20, and 30 lines of production code that could have been 5. I'm glad I do not have to maintain 300 kloc of 100% AI-generated code. But at the end of the day, what counts is velocity and quality, and it seems OP is happy. The tools certainly are useful.

Pi-thon

There's no contradiction. One statement is about the default, the other is about the possibility.

Hm, I'm sorry you feel that way. It's not meant to be mean. On the contrary; Instead of encouraging someone who I feel is going down a wrong path, to me it's kinder to express my view that they aren't. I have personally wasted years of my life on technical projects, and would have been better off if someone had told me that it was a bad idea.

[flagged]

Congratulations. Doesn't make this approach make so much more sense than writing a browser engine from scratch?

Can this be an arbitrage opportunity?

I spent an afternoon watching and re-watching this video just to figure out how the otherwise-undocumented behaviors of the API work. It was exclusively about Apple's implementation, and not in any way about the general CS topic.

I don't really see how that makes a difference. A talk with the title "my API demystified" is an equally bad sign.

Premium Hardware, Struggling Software

This sums it up well. The hardware is great, the software isn't.

I recently programmed the same app for iOS and Android. iOS took twice as long, simply because Apple's APIs suck. Case in point: The background task APIs (plural, yes, unfortunately) are so bad that Apple felt compelled to publish a video "Background execution demystified" [1]. If a dev creates an API and then has to publish docs "[my API] demystified", then the API sucks. Period.

I value stability and the freedom to configure the OS to my liking. macOS is stable but forces countless things on me that I do not want. Windows offers freedom but comes with many glitches. Linux is extremely stable and puts me first by letting me configure it. I love it.

[1] https://wwdcnotes.com/documentation/wwdcnotes/wwdc20-10063-b...

Brave is fantastic.

My favorites:

To quickly find text, select some text and press ⌘E followed by ⌘G.

In save dialogs, press ⌘= to switch between the compact and expanded layout.

In save dialogs, press ~ to open a Go To File dialog prefilled with the home directory. Press / to open it prefilled with the root directory.

Hold Option while opening the Wi-Fi and Bluetooth menus to access extra options.

After copying a file, press ⌥⌘V to move the file instead of pasting a copy of it.

Terminal:

Press ⇧⌘A to select the output from the previous command.

Press ⌘L to clear the output from the previous command.

Press ⌃⌘V to paste and format text that is properly escaped for the shell.

Press ⌃T while a command is executing to view runtime statistics about the execution so far.

Hm, are you hinting that fman was not exceptional at its time? If yes, then I disagree.

I put it down to desktop utility apps being a very tough market because 1. they are time-consuming to develop and 2. people hate paying for desktop software. You already have several comments in this very thread from people complaining about the price:

* https://news.ycombinator.com/item?id=43102477

* https://news.ycombinator.com/item?id=43099230

* https://news.ycombinator.com/item?id=43097749

Ah and yes, I'm still getting a handful of fman sales per month. But nowhere near enough to justify any time investment.

Cool point, thank you. I think his name is Armin.

I once devoted 2 years of my life to developing a file manager called fman [1]. In total, it generated probably 35,000$ in profits, so my income from the project is somewhere around 10 $/h. As software developers, our opportunity costs are high. I use my file manager to this day and love it. But I regret spending so much time on it.

Congratulations on your launch. I wish you more success than I've had. Failing that, I wish you that you will see earlier than I did when it is time to move on.

I once recorded a video about my experiences developing a file manager [2]. Maybe you'll also find some interesting bits and pieces on fman's blog [3]. Incidentally, an article there is what sparked my current venture, which is much more profitable: consulting services around automatic updates.

If you'd like to have a chat, feel free to reach out. My contact info is on my website. :-)

Good luck!

[1] https://fman.io

[2] https://m.youtube.com/watch?v=I1K3IkOlaVw

[3] https://fman.io/blog/

There are a lot of different ways to install Python, all with different default settings, and gotchas.

With uv, there is now one more.

https://xkcd.com/927/

I do believe this is the next natural evolution. People don't like to pay for things and ads are a proven business model. I bet the big labs are looking into this

Could this be an attempt to prevent Altman from turning OpenAI into a for-profit too cheaply? If I understand correctly, then turning a non-profit into a for-profit requires selling the non-profit's assets at a fair price. If Altman says he buys it for 40bn and there is no other offer, then who's to say that this is not a fair price. On the other hand, with a 90bn offer on the table, it will be hard to argue that the assets are only worth 40.