If I scavenge any machine today, how likely would I be to find a 6502 vs something more modern? I’d argue that some people might have a NES at home and one could get a 2A03 from it, but in a hypothetical scenario where I need to scavenge some computational power, I’d find an Android phone
You're much more likely to stumble one something more modern, but that modern something is also much less repairable. It's great if it works and if it can run Linux or Dusk OS, but when it can't, you're out of luck.
With a 6502 or other such CPU, the machines you scavenge them from are much more repairable and adaptable. You can use those components like lego blocks. It breaks? either repair it or strip the working parts to use in another frankenstein computer.
I get the idea of making a frankenstein computer, I just disagree that 6502 is THE platform to do it on. Practically, there's no way for me to find it. Other comment mentions ARM, which is a much more interesting proposal to me
ARM is an interesting proposal if you want to order a SBC online and run software on it. Soldering an ARM CPU with low tech tools? That's something else.
I have ported zForth to an even weaker chip, the famous 10c risc-v micro ch32v003 (16k flash, 2k ram) so no issue running on this: https://github.com/BogdanTheGeek/zForth
Allow me to brag about romforth (https://github.com/romforth/romforth) which I ported to the "3c" Padauk and can run on really small rom/ram microcontrollers.
Caveats:
- tested only on an emulator SDCC/ucsim_pdk, not on real hardware
- given how small the ram is, there is no user dictionary but new words can be defined and tested using what the Forth folks refer to as "umbilical hosting".
Even for a Forth, 3KB of RAM is rather tight. Dusk OS intentionally de-prioritize compactness and it couldn't run on that amount of RAM. It can get a C compiler loaded in about 100KB of RAM, but 3? not enough to boot.
OK, so we'd play with zForth then, as BogdanTheGeek notes here. That reminds me, I have a Scamp board sitting here on my desk that I really should play with more.
https://udamonic.com/what-is-a-scamp.html
"we'd", you mean in a collapse scenario? Forths are, by nature, "collapse-friendly", but one particularity with Collapse OS and Dusk OS is that they are fully self-hosted. This includes the tools necessary to improve upon themselves.
From a quick glance, it looks like BogdanTheGeek's Forth is written in C, which means that it's not self-hosted. If all you have is that disposable vape with this Forth in it, you lack the tools to deploy it on another machine or to improve it in place.
One could also port Collapse OS to ARM. I guess it wouldn't be a very big effort.
Good points! Really, I should start with learning Forth on the devices I have first, before getting to concerned about others. ARM does seem like a useful target though, given that they're basically everywhere these days.
Try this prompt: Create a detailed step by step plan to implement a boilerplate Zephyr project skeleton for Pi Pico with configured st7789 SPI display drivers
Ask Opus or Gemini 2.5 Pro to write a plan. Then ask the other to critique it and fix mistakes. Then ask Sonnet to implement
I tried this myself and IMO, this might be basic and day-to-day for you, with unambiguous correct paths to follow, but this is pretty niche nevertheless. LLMs thrive when there's a wealth of examples and I struggle to Google what you asked myself, meaning that LLM will perform even worse than my try.
I found that second line works well for image prompts too. Tell one AI to help you with a prompt, and then take it back to the others to generate images.
Why do you care to connect with another human? Try to feel his emotions, what he tried to express? If you see no value in that, there's no discussion to have, honestly. For most people I know there's value in connecting with others and emphasizing with their emotions
Art is a difficult, subjective matter sometimes. I don't think we can expect everyone to "get" every piece of art. If the poster upthread wanted to, they could read more about the painting, in detail, where perhaps someone writes about various specific features of it and what people believe those features mean. Maybe that would provide more understanding, and they could feel his emotions that way.
I'm not saying they have to or should do that; maybe they just don't care enough. And that's fine. But the option is there.
If someone prompts an AI, "generate an image in the style of Picasso's Guernica", then the result of that, by definition, has no deeper meaning. No emotion went into creating it. The person who prompted the AI could make something up, but it's hard to say what's "real" there. Even if they were to guide the image generation by describing their own emotions, the result wouldn't really be their own expression of their emotions. It would be the AI's probabilistic guess as to what those emotions look like on paper, when rendered using Guernica's style, based on a mish-mash of thousands of different artists and art history research. Ultimately it just doesn't mean anything.
I accept the idea that a talented artist could guide the AI with much deeper specifics about what to "draw", how to draw it, etc. And maybe -- maybe -- that's something that would convey the human's emotions faithfully. But I don't think that's what we're talking about here.
> But I don't think that's what we're talking about here.
Actually that is exactly what I'm talking about. I'm not talking about AI beginners putting in some words into a text box, I'm talking about creatives who use workflow managers like ComfyUI to create exactly the output they envision in their minds. In this way, the AI generation is merely a tool to get out whatever is in their head via synthesized means rather than manual (literally, hand) means. For example, this is a list of node work flows, it's similar to game programming in that you have inputs and you want to transform them to certain outputs, and that transformation work is thoughtful by the human and is what I imbue the creative aspect to.
Many things require one to reject self-imposed boundaries. For example[1]:
> There's a story that, IIRC, was told by Brian Enos, where he was practicing timed drills with the goal of practicing until he could complete a specific task at or under his usual time. He was having a hard time hitting his normal time and was annoyed at himself because he was slower than usual and kept at it until he hit his target, at which point he realized he misremembered the target and was accidentally targeting a new personal best time that was better than he thought was possible. While it's too simple to say that we can achieve anything if we put our minds to it, almost none of us are operating at anywhere near our capacity and what we think we can achieve is often a major limiting factor.
---
Art is nothing like shooting. My first instinct looking at Guernica is that I also feel nothing, but one can limit oneself and say: if I feel nothing initially, I will feel nothing at all. If you prime yourself to be open to an experience of putting yourself into the shoes of the author, you might start feeling something.
As a place for meaningful discussion HN fell to Reddit level. Very little substance and it became acceptable to circlejerk about how bad your enemies are.
Okay, propaganda, sure. Propaganda is a different argument - it compromises the quality of search - and I will happily agree with you. I want to emphasize, that the claim was: using Yandex through Kagi is supporting war criminals. Which war criminals? Which war crimes is Yandex complicit in? I plainly do not accept that propaganda === war crimes.
>I plainly do not accept that propaganda === war crimes
I'd say propaganda is often necessary, and usually successful, at allowing governments to execute war crimes. I think most modern war crimes are lied out of existence by propaganda outlets like Yandex (or any number of US news outlets at other points in recent history).
They are accused of sending Ukrainian users data to the Russian state.
Also, ‘In August 2024, the company was accused of collaborating with the Federal Security Service (FSB) to direct users to fake websites for the Freedom of Russia Legion and the Russian Volunteer Corps, groups composed of Russian supporters of Ukraine. The websites collected data from Russians who may be interested in joining the organisations and supporting Ukraine.’
Following that reasoning, is Gazprom invading a country? No? Then I guess we should buy their gas. Why not send the latest and greatest ASML lithography machine to Mikron? It's not like they're actually launching the missile using the processor they manufacture.
It would be interesting to know people’s views on sanctions against Russia, as these are a key tool against Russia and the logic is much as you have described here.
My guess is that HN users don’t support sanctions.
Lying used to be something that people are ashamed of