Comments URL: https://news.ycombinator.com/item?id=47226757

Points: 3

# Comments: 2

So not only do you have a closed loop system that has objective/automatic pass-fail criteria you also don't even have to supply the instructions about what the function is supposed to do or the test cases!

Obviously this isn't going to be 100% reliable (especially for edge cases) but you should be able to get an enormous speed up. And in many cases you should be able to supply the edge case tests and have the LLM fix it.

(Codex is still free for the next few days if you want to try their "High"/"Extra high" thinking models)

Anthropic said that mass surveillance was per se prohibited even if the government self-certified that it was lawful.

Anthropic said that mass surveillance was per se prohibited even if the government self-certified that it was lawful.

    ps aux | grep nginx | grep root | grep -v grep

    ps -u root -C nginx

    psc 'process.name == "nginx" && process.user == "root"'

Comments URL: https://news.ycombinator.com/item?id=46264595

Points: 3

# Comments: 0

These are what are claimed to be the onerous regulations slowing down innovation.

Copying and pasting tends to be a very tedious operation in the shell, which usually requires moving your hands away from the keyboard to the mouse (there are terminals which allow you to quick-select and insert lines but they are still more tedious than simply pressing enter to have the command on the line editor). Maybe try using llm-cmd-comp for a while.

> I do not see how editing the command is a good tradeoff here in terms of complexity+UI.

I don't find it a tradeoff, I think it's strictly superior in every way including complexity. llm-cmd-comp is probably the way I most often interface with llms (maybe second to basic search-engine-replacement) and I almost always either 1. don't have the file glob or the file names themselves ready (they may not exist yet!) at the time when I want to start writing the command or they are easier to enter using a fuzzy selector like fzf 2. don't want the llm to do weird things with globs when I pass them directly and having the shell expand them is usually difficult because the prompt is not a command (so the completion system won't do the right thing).

But even in your own demo it is faster to use llm-cmd-comp and you also get the benefit that the command goes into the history and you can optionally edit it if you want or further revise the prompt! It does require pressing enter twice instead of "y" but I don't find that a huge inconvenience especially since I almost always edit the command anyway.

Again, try installing llm-cmd-comp and try out your demo case.

    - it puts the command in the shell editor line so you can edit it (for example to specify filenames using the line editor after the fact and make use of the shell tools like glob expansion etc.) 
    - it goes into the history. 
    - It can use a binding so you can start writing something without remembering to prefix it with a command and invoke the cmd completion at any place in the line editor. 
    - It also allows you to refine the command interactively.

https://github.com/CGamesPlay/llm-cmd-comp

    - corfu+marginalia+vertico+embark+orderless is the standard completion stack now
    - magit (maybe also see also the "casual" the transient front end for other modes)
    - avy
    - (fset 'yes-or-no-p 'y-or-n-p)           ; 'Y' or 'N' instead of 'Yes' or 'No'
    - (setq confirm-kill-emacs 'y-or-n-p)
    - evil (optionally if you like vim keybinds, you still need to know basic emacs))
    - if using evil: evil-collection, evil-args, evil-goggles, evil-traces, evil-escape, evil-nerd-commenter, evil-lion, evil-surround, etc are not "standard" but still useful 
    - configure melpa source
    - which-key
    - helpful
    - undo-fu
    - gptel
    - projectile
    - eglot
    - saveplace
    - desktop
    - uniquify
    - dired/wdired
    - flycheck/flymake
    - treesitter
      

    - general (for making your own keybinds)
    - some kind of multicursor mode (I use iedit but it's simple)
    - yasnippet
    - org (usefulness depends on the person)

Also, very important, learn the help system (C-h ), especially C-h f, C-h k, C-h w, C-h c. And the info system

1. quick-select the output in the terminal (select files/paths, urls, etc and either copy or paste them at the cursor). This is very useful as you often don't have the foresight to pipe the command to pipe the output of the command to some selection mechanism and even placing text on the line-editor is not that easy by default.

2. view images (sixel or kitty protocol). This is pretty useful visual analog to cat that doesn't require opening another program and works over ssh. Also for video.

There are some other nice utilities for doing things like downloading files directly in the terminal (it2dl for iterm and kitten transfer for kitty).

kitty doesn't work out of the box on macos if I remember; you have to set configuration for option/command etc.