I refuse to try these proprietary SaaSS GPT services so I have no idea what I'm missing out on, but LLaMa at least can be run locally, and while the results aren't probably much better than a markov chain, it's fun to play around with. Unfortunately, being easily able to train models yourself is probably still quite far away.

From playing around with Llama for programming tasks, I'm guessing that sometimes typing the question and reading the GPT generated response might end up being faster than searching and then checking individual pages and trying to decipher what you want. Probably works best when you have a very specific question that relates to "folk" knowledge that people don't write about.

I'm probably shopping at the wrong places I guess --- after spending the afternoon searching through vivocity, $16 dollars was the lowest price that I could find.

I went shopping at MUJI the other day for the first time, and wow things were super cheap --- what's the catch? Slippers for $16!!!

To this day, I refuse to actually learn what the flags actually stand for, and simply remember it as xzf = "eXtract Ze Files"

Ferris got nothing on this

This really hits home today...

Been playing around with Rhombus programming language for a bit to build a small game --- got a little demo running (assets from LPC)

huh, never had that problem with Minetest