“If man chooses oblivion, he can go right on leaving his fate to his political leaders. If he chooses Utopia, he must initiate an enormous education program - immediately, if not sooner.”
-R B Fuller
Guest Network in OpenWRT prevents inter-device communication on the LAN. So the thing can talk to its cloud service or whatever but can’t see my phone/laptop/PCs.
This means that untrusted devices which get pw0ned (or are surveillance-by-design) can’t then get into your more important machines. Most people use guest networks for IoT devices like Ring cams and smart washing machines but I don’t even own shit like that.
OpenWRT all the way.
Skynet firewall
PiHole
Stable IP addresses and a private interface for family phones/laptops/pcs
Guest network for nasty Roku TV/peleton
EDIT: my favorite thing about Skynet is the new-ish ability to filter by country
Not in Quebec. Straight to jail!
using Mate M8 desktop, obvs
all the talk about LLMs looking like they are “thinking” is based purely on vibes rather than reality
You seem to ignore the quantitative, evidence-based studies mentioned. Ex: https://aclanthology.org/2026.acl-long.155.pdf
They are massive autocomplete at scale
Rather a lot of the cognitive science community would argue that is what human brains are.
Anticipating the next sensory input is literally a basic driver for nervous systems in general and neurons very specifically.
We train in bursts because we need a ground truth/scoring function to guide the learning
These are completely orthogonal things.
We train in bursts (“epochs”) because empirically it leads to more rapid convergence on an acceptable error minimum.
The fact that a scoring function is needed is almost a tautology. You have to have some differentiable function to say what direction is lower error. Typically you segment your training corpus, and use the scoring function to get the error for one batch at a time to a certain threshold before proceeding to the next batch. Then go to the first batch and push the error bar even lower, and so on.
Kinda like tightening nuts on a car wheel. You don’t fully tighten the first one first; you have to wiggle the system into its final configuration.
Can you prove that humans don’t?
This, this this. This is something the prevailing armchair opinion is really missing. Their blatantly-wrong assumptions about human cognition.
It feels as proudly ignorant as my grandmother’s generation who proudly declared Evolution to be patently ridiculous (“my grerat-gramma was no monkey swingin’ from a tree!”)
I am not a researcher in the area, but I am the kind of nerd who watches university colloquia and sometimes does in fact read the paper.
“I don’t hallucinate all the time”
YES, YES YOU DO, do you not understand how the sensory system works? You interpolate between inputs. This is why optical illusions work! Your fovea is a tiny tiny spot and is the only area that’s actually in focus!
Learning by statistical sampling, reward/repress feedback loops, production system-based computation giving rise to “fast” (System I) thinking, “hallucinating” to interpolate between sensory inputs … the list goes on and on
Bookmarking this. I am low-key low-priority background-task setting up a Jellyfin with yt-dlp integration somehow and I love this simple brilliant idea, playlist with a cronjob. Same basic principle as Invidious.
Ditto. I use Codex and Claude Code on a daily basis. Copilot feel like it’s from the 80s, a few generations behind
Edge is just themed Chrome.
It’s edging, but it’s kinda like edging to the softcore titty movie channels on 1990’s cable when you couldn’t pay to descramble them
Unix principals include Dennis Ritchie, Ken Thompson, and Brian Kernighan
Unix principles include tools that do one thing and do it well, casting directories and sockets as file abstractions, and clean separation between kernel and userspace modes.
alias sudo fucking
fucking systemctl restart firewalld.service
fucking shutdown
deleted by creator