RE: https://labyrinth.social/@nash/116178591588359360
you ever write code so inefficient you have to secure 80% of the world’s DRAM production
| pronouns | he/him |
| mood | :kirby_sleep: intensifies |
| webbed site | https://themoep.at |
| gif is pronounced | /χif/ (voiceless uvular fricative) |
RE: https://labyrinth.social/@nash/116178591588359360
you ever write code so inefficient you have to secure 80% of the world’s DRAM production
going through my notes and found this gem from my dad (who's not that much into computers much) that better captures threat modeling than any classes I took as an undergrad:
"what data do you have on your computer and who might want to have that? what time and resources would they be willing to invest in order to get it?"
good news: I fixed an issue with my synth setup that made the right channel not work properly
bad news: I spent an hour taking a module apart only to realize the right channel cable was not plugged in

@gunstick @jollysea I second this view. My usual progression is: bash one-liner (mostly pipes and loops, a little bit of subshell), then a collection of these oneliners in an executable file (usually Makefiles for my own weird reasons. I'd recommend justfiles nowadays even though I don't use them yet) and then python scripts and finally python projects.
benefit of bash is: quickly run something in a lot of places, calling random simple things like bash -c "for f in $(ls .config/); do echo $f | ts; done" is powerful.
benefit of python: way more helpful to debug, way less painful mistakes, and way better handling of complex data
I learned a lot of nice tricks through the wizard zines and comics for bash: https://wizardzines.com/comics/