• 0 Posts
  • 28 Comments
Joined 2 months ago
cake
Cake day: January 15th, 2026

help-circle


  • There’s some history behind it. First the author didn’t want to have it open source, iirc due to some bad experiences with having their source code open? Don’t remember exactly. The repo you linked existed for bug reporting basically. Then later on this changed and the source code got released. I still don’t know if the author is interested in getting public contributions to the codebase.

    Regardless, the author makes some very pleasant software, and is super responsive at fixing issues/implementing feature requests. Hats entirely off to them.




  • I’m not going to watch a YouTube video that could be a few paragraphs of textual explanation, because it’ll no doubt be eight times longer than it needs to be for the benefit of more ad money or promotion in the almighty algorithm.

    The linked one is a short video with a duration of 02:37. There’s no padding in this one. Naturally, you can’t actually get all of the nuances of the full-duration video, which also can’t cover the full nuances of the study itself that it’s based on (https://iopscience.iop.org/article/10.1088/2634-4505/ad310c).

    Pop science videos making studies accessible to the general public are good, actually. I recommend that you stop being dismissive of them. Had you actually put in the time, you wouldn’t have posted things that are in direct contradiction with the latest science on the subject, spreading misinformation in the process.













  • So far, there is serious cognitive step needed that LLM just can’t do to get productive. They can output code but they don’t understand what’s going on. They don’t grasp architecture. Large projects don’t fit on their token window.

    There’s a remarkably effective solution for this, that helps both humans and models alike - write documentation.

    It’s actually kind of funny how the LLM wave has sparked a renaissance of high-quality documentation. Who would have thought?