I used tmsu (“tag my shit up”) for a while, but it required too much discipline and then I dropped it.

In addition, tools like fzf for fuzzy file-search (comes with shell integration to e.g. replace the default history search in bash) and ripgrep-all made this kind of organization unnecessary for me. It now suffices to have a vague idea where a thing is located and I can do a brute-force search in a few seconds.

The next-level filesystem argument is brought forward every few months, but I’m not buying it.

and then suddenly there’s the perfect use case

Yeah but like WHAT?

It was 2006 or 2007 when I sent a girl from my class a funny pdf.exe on ICQ, which simply restarted her computer when she started it. I didn’t know that she would lose a whole day’s work that way, but eh what are you gonna do if programs don’t have autosafe.

Also, anyone remember “dialers”? Fun times.

I think you generally can’t know if someone shared their code with the intention that others may use it, but it’s a reasonable assumption.

I find this unnecessarily derisive. There are good reasons for a UI or README not being user-friendly, the top-most one being (imo) that it is really, really hard to get right, takes a lot of time and doesn’t primarily solve the problem the project was started for.