For debugging sake, put the full path to the binary, eg /usr/bin/gotop (check with which gotop)

Which I can totally understand, but I would like to spend more time with my family and less time writing code.

This also allows me to iterate faster and identify useful ideas That justify deeper effort.

Yeah, that’s always a risk, but as you said, humans make mistakes too. And if you change your approach to software development by writing more tests and using strict interfaces or type annotations, etc., it is pretty reliable and definitely saves time.

They can also be really good for quickly writing code if you line up a whole bunch of tests and line up all the types and then copy and paste that a few times, maybe with a macro in Vim.

The LLM will fill in the middle correctly, like 90% of the time. Compare it in git, make sure the tests pass, and then that’s an extra 20 minutes I get to spend with my wife and kids.

It would be on the order of aN intensive video game, maybe. Depends on the size of the model, etc.

Training is definitely expensive but you are right in that it’s a one-time cost.

Overall, the challenge is that it’s very inefficient. To use a machine learning algorithm to do something that could be implemented deductively is not ideal (On the other hand, if it saves human effort…)

To a degree, trained models can also be retrained on newer data (eg freezing layers, LoRa, GaLore, Hypernetworks etc). Also newer data can be injected into a prompt to make sure that the responses are aligned with newer versions of software, for example.

The electricity consumption is a concern, but it’s probably not going to be the end of the world.

Yeah but wine helps lol

Or a makefile / justfile would be good too.

I put those on each directory and do just run to pick up the thing I was working on quickly.

I’ve always had an easier time jumping into an oop code base, then eg a lisp one.

I hear people when they say they don’t want their data mixed in with their logic but The pressure to structure code Is very nice.

I found reading through the rust book was a nice walkthrough of problems one can hit and how that language elected to solve them.

In terms of practice:

• Write a vim config
  • Shell out to python if you’d like
• Learn a bit of elisp and org-mode
• Rewrite all your shell scripts into a python CLI
• Write a pyqt6 GUI for tasks and notes on the exact way you’ve always wanted it to work
• Write an AI tool to auto-format links etc with phi3
  • Very exciting how much these smaller models can do!

I agree, it’s not often considered a systems programming language and it may not be the perfect tool here.

However, it is worth mentioning that cgo Can serve as a escape hatch depending on the use case.

Ah my bad, didn’t read.

Odin is a nice choice then, beef is another small bespoke language.

There is also zig, Go and Rust.

No language is perfect, but those languages have some features that are nice.

Windows may be easier for games, they’re exclusively written for Microsoft so that’s to be expected ( although Valve has done a lot here).

Generally speaking, modern distributions like Fedora will be no more difficult than Windows or Mac. The important distinction is that it will be different.

Microsoft has spent a lot of effort putting their operating system into every single school and business on the face of the Earth and as a result many have decades of training with that OS. That doesn’t mean their operating system is better or easier. It just means it’s familiar. If you used Android for two decades and then picked up an iPhone, I’m sure that would be just as difficult.

In the scientific space, we’ve been using *nix systems since well before Microsoft was even around so our tooling doesn’t typically support Microsoft. For us Microsoft is more difficult because that’s the training that we have.

So, it’s not that Linux has a worse user experience per se, rather it provides a different user experience. Some may consider shell scripts worse than control panel, but that’s a preference. One isn’t worse than the other. They are just different.

In my opinion:

• Web browser
  • exactly the same
    • slight edge to in terms of privacy and security (there’s a reason Tails isn’t written in Windows)
• Media – Movies
  • exactly the same

The difference is in work, If your workflow is heavily Microsoft focused, Is a truly awful experience and you’ll feel like a second-class citizen. But if you’re working on technical things, the inverse is true, eg

For document production:

• TeX
  • Linux is much easier
• HTML / Markdown / pandoc
  • Linux is much Easier
• Microsoft Office
  • Windows is much better here
    • I don’t use Microsoft Office though so YMMV

Finally, it’s not really fair to lump all the next distributions into the same bucket, Is over 1,000 distributions and they are all quite different, Only common element is the kernel.

Gentoo is very technical but it’s also very interesting, Arch is similar. Fedora OTOH we’ll usually walk out of the box And you have your choice of desktop environment with Good support for alternative window managers like sway/Hyprland etc.

Perhaps you’re simply more familiar with Microsoft / Apple, maybe it’s not more difficult?

I too use Linux for work, but I have limited experience on Microsoft systems and have been on Linux based systems for over a decade. For me windows is a chore.

In my opinion, it’s a matter of perspective and experience. Yours is aligned with something different, that’s all.

Couldn’t agree more.

Important benefits include:

• types (eg dates)
• easy import into duckdb

Can be viewed with visidata

Perhaps another perspective is where to draw the line in terms of expected expertise.

The book is really good and offers a snapshot of those inner workings of the language.

For using it, LLMs (open weight and otherwise) perform very well and may fill the gap.

Yeah it’s great! Check it out.

Use Quartz and Obsidian because it’s easy. If not mcdocs.

I think I’ll keep doing it. It’s worked fine for the past decade 🤷