If your antenna is strong enough, you can pick up a lot of lower power devices from a long ways off.

This is problem with platform lock in. Be it discord, X, FB groups, Reddit. Communities form and then fragment as the members who realize there are better open options leave.

The users of the web have enough desktop and laptop cpu cycles to spare for passing a few decentralized messages around without killing battery life or performance. We don’t need these experience degrading/dehumanizing ad platforms to underpin our internet. That’s just a narrative peddled by companies who aim to make obscene profit on top of their already obscene profits.

They’re basically the same as regular, but the wattage rating (usually called tdp) is lower.

Correct. Micro-ATX is the smaller version of the larger ATX and still larger EATX (extended atx). Your old case probably fits micro atx if it’s not OEM. You can populate it with a mb, cpu, ram, ssd, and power supply (don’t need more than 500w for your use case) and eventually move to a nicer case like that Node if/when you fall in love with the hobby. My Rpis are collecting dust since switching to a low power server.

It’s a whole different experience when general advice applies to your hardware vs the Rpi ecosystem. Many more options. In 2024, ATX offers no real benefit over the smaller form factor beyond better heat management for high power builds with spaced out components.

And a correction: node 304 supports 6HD, the 804 supports 8

I would recommend getting a “forever” case like the Node 304. You won’t regret the purchase and you can use for any future upgrades. It stores 8 (correction, 6) 3.5 drives, so you can add on as you grow.

Find a used a motherboard like this: https://www.ebay.com/itm/235546915389?mkcid=16&mkevt=1&mkrid=711-127632-2357-0&ssspo=GP45S9r5R6-&sssrc=4429486&ssuid=uaLd2h3oTQO&var=&widget_ver=artemis&media=COPY

With a cheap GE (low power version) AMD processor and 16/32gb ram and whatever nvme ssd you can scrounge.

It will cost you maybe $100 over some alternatives, but you can use it for years and keep upgrading as you go.

Most Dell and OEM parts won’t work on standard cases, FYI.

My uncle bought a used car built in communist east Germany. He always emphasized how it was built like a tank to last. Capitalism is great and all, but it promotes waste. Companies have an incentive to make products that fail and need to be repurchased. Planned obsolescence is fine if it was only about people craving something better. As it stands, it’s more of a forced switch with breakable parts.

deleted by creator

deleted by creator

This is why we need more websites to adopt secure client side scripting.

JavaScript may or may not be it, but the web needs to be reachable/archivable. It should also have attribution, but that’s a tangent.

If that were the case, the scientific method would not exist.

You should read a clockwork orange.

Most people here don’t understand what this is saying.

We’ve had “pure” human generated data, verifiably so since LLMs and ImageGen didn’t exist. Any bot generated data was easily filterable due to lack of sophistication.

ChatGPT and SD3 enter the chat, generate nearly indistinguishable data from humans, but with a few errors here and there. These errors while few, are spectacular and make no sense to the training data.

2 years later, the internet is saturated with generated content. The old datasets are like gold now, since non of the new data is verifiably human.

This matters when you’ve played with local machine learning and understand how these machines “think”. If you feed an AI generated set to an AI as training data, it learns the mistakes as well as the data. Every generation it’s like mutations form until eventually it just produces garbage.

Training models on generated sets slowly by surely fail without a human touch. Scale this concept to the net fractionally. When 50% of your dataset is machine generated, 50% of your new model trained on it will begin to deteriorate. Do this long enough and that 50% becomes 60 to 70 and beyond.

Human creativity and thought have yet to be replicated. These models have no human ability to be discerning or sleep to recover errors. They simple learn imperfectly and generate new less perfect data in a digestible form.