Parallel Polis was a strategy how to survive and thrive in a somewhat dystopic society (also known as communism). Since it's starting to be pressing, it might be a good idea to revisit how to create a parallel society in case the main one gets broken. When it gets broken. https://fleebr.community/blogs/%CE%B6/parallel-polis

Best opening sentence of a book (or any other reading). Go! I'll start: "The sky above the port was the color of television tuned to a dead channel" (from W. Gibson: Neuromancer, still one of the best fiction books about AI, I highly recommend especially the ending, because that is still the most likely scenario in my mind).

Testing a new vibe coding tool - Crush. Claude Opus 4 with thinking - those credits add up! But it's credits that expire in a few days anyway, but it's good that I can see that. Very nice tool.

GitHub

GitHub - charmbracelet/crush: The glamourous AI coding agent for your favourite terminal 💘

The glamourous AI coding agent for your favourite terminal 💘 - charmbracelet/crush

I have started more than 15 companies in my lifetime. A few of them successful. I would do an AMA, but I'll answer three questions that you didn't dare to ask instead. Q: What is the best advice you have for young entrepreneurs? A: Stay away from EU. Don't incorporate there. Don't target EU customers. Q: What is the best target market that no one sees? A: The netizens that own cryptocurrency and can use them for payments (no ETF owning grandmas). Q: Where do I find investors or money? A: Don't. Bootstrap.

After listening to DHH at Lex Fridman's podcast I'm trying lazyvim instead of my pimped up vim config. It's pretty cool. I've setup avante with Venice since I have quite a lot of API credits and now all my machines are a bit AI enabled directly in my vi. $ cat .config/nvim/lua/plugins/ai.lua return { "yetone/avante.nvim", event = "VeryLazy", opts = { provider = "openai", system_prompt = [[ You are a great programmer/coder. Do not write excessive comments, do not comment on what you changed/inserted. Write clean, understandable code. If you are using Python, prefer poetry for dependency management. You are interacting with an expert in Linux, Python, networking, do not overexplain simple concepts. ]], providers = { openai = { endpoint = "https://api.venice.ai/api/v1", model = vim.env.VENICE_CODING_MODEL or "qwen3-235b:strip_thinking_response=true", api_key_name = "VENICE_API_KEY", }, }, }, }

S kým sa vidíme v Košiciach? :) View quoted note →

I had (thankfully indirect) interaction with a shameless scammer. We know his full name, he has 70k followers and just steals money (fiat, crypto) under various pretenses (OTC trades, he even tried to sell citizenships). He has a full on act, it's quite interesting. When people realize he's a scammer, he just laughs and tells them they scammed themselves because they are stupid. These people exist, it's super weird. I believe it's some kind of psychopathic disorder. Even his family knows, his brother helps, parents know and pretend they don't. Hundreds of thousands of euros. I don't think they have a long lifespan, so it's not an evolutionary adaptation. But we live in the same world as they do. For me, the main point is assuming other people's inner world is similar to mine (also called empathy) is super wrong. They're hacking this assumption. A friend of mine has a good heuristic for this. When someone is talking to you, and you get obligated to listen, ask "are they trying to hack my brain through empathy?". Imagine a beggar. You might or might not want to help them, but you should be able to identify if and when it's a hacking attempt.

It is now (somewhat) possible to run fully local (no cloud, no spyware) Alexa/assistant alternative with Home Assistant, Ollama (I run it on Apple Silicon) and as an endpoint Home Assistant Voice Preview Edition. I've written about my setup (and challenges), but TLDR: You can just say "Ok Nabu, turn on my bedroom AC" and it works. I've dreamt about this in Cypherpunk Visions 2023-2025 (a short book). Now I am going to write update for 2026-2029 (3 years is a good timeframe) and this will be "normal" for those who want it now, it's not just a dream.

Home Assistant Community

Blueprint on AI using ollama on Apple Silicon

Hey everyone! 👋 I wanted to share my setup for a completely local voice assistant using Home Assistant with a Mac Mini as the AI inference serv...

enhanceur /ˌɒnhɒnˈsɜː(r)/ noun Etymology: Early 21st century: from French, blend of enhance + -eur (agent suffix), modeled after saboteur. First documented usage in online forums discussing "reverse vandalism" and "guerrilla kindness." Definition: A person who secretly improves public spaces, systems, or situations without seeking recognition, often leaving others pleasantly confused about why things work better than expected. (informal) One who commits acts of clandestine betterment; a benevolent phantom who strikes when nobody is watching. Example sentences: "The office enhanceur had struck again—someone had mysteriously fixed the perpetually jammed printer, replaced the dried-up pens with working ones, and nobody had seen a thing." "After years of finding public toilets inexplicably cleaner than expected, Marcus began to suspect the existence of a secret society of bathroom enhanceurs." "The park's enhanceur worked only at night, planting flowers in bare patches and repainting faded benches, leaving morning joggers to wonder if they'd imagined the previous decay." Usage note: Unlike their destructive counterparts (saboteurs), enhanceurs suffer from a peculiar occupational hazard: nobody believes they exist. This has led to the common phrase "enhanceur's paradox"—the better you are at your job, the less evidence there is that you were ever needed. See also: ninja janitor, guerrilla gardener, phantom fixer, anonymous benefactor