I generally agree with you that the projects could be more accessible. I also believe that these projects exist to do one thing, which is create a solution to a problem and share it with others.

As for your learning and explaining what the various bash command chains do: Explaining bash commands is actually one of the few things I would say AI (especially local/self-hosted LLMs) are REALLY good at.

There are lots of obscure options for every bash command, and having used a lot of LLMs for work, most LLM models can use and explain these bash commands, piping, and exit conditions quite well.

Not everyone wants to use AI, which I can understand. But, if you’re going to use AI, use it to break down more complex things and explain them to you.

  • source
  • parent
  • context
  • [–] 4 points 6 days ago (3 children)

    Without expertise in computer equipment and long term storage with humidity specifically:

    Generally targeting less than 60% humidity is sufficient for long term storage of many items.

    As for dust, depending on the form factor you are targeting, some cases come with built in dust filters, while you may also be able to buy some dust-specific air intake material. If you have any intake fans, then you can cover those with that material and clean it from time to time.

    Dust is not likely to be your biggest concern.

  • source
  • [–] 24 points 1 month ago (1 child)

    This sounds like a fork of OpenClaw with an extra embedding memory system built in. A bit like hindsight or one of the other vector memory stores with the ability to overwrite a memory.

    I’m betting there’s a lot of formerly open source projects here that got vendored to make this.

  • source
  • [–] 4 points 1 month ago (2 children)

    American here: my high school used to sell lunch milk in clear hand-sized square plastic bags. We were expected to successfully pierce only one side of the bag with a pointed straw.

    The technique was… learned. Some people took a while to master it.

  • source
  • parent
  • context
  • [–] 2 points 2 months ago

    After testing via api for a few days, this model is genuinely impressive for its size/price.

    Feels a little like running opus 4.6 on medium think. Quick enough, clean enough. Still makes some minor errors once in a while, but it’s fast enough to be useful and cheap enough that you can use it for everything and still pay very little.

  • source
  • view more: next ›