1. 3
  1. 1

    This is great, and along the same lines I’ve been thinking for the past few months. I started writing an AST editor (https://github.com/christianbundy/tri) but it’s honestly not a quick and easy side project.

    I’m looking forward to seeing more developments in this space, I’m convinced that text editors <<< code editors.

    1. 0

      Video isn’t loading well for me, and I can’t find a transcript, so consider this a blind take based on the title:

      If you can convince a computer to compute, and the “shape” of that computation is under your influence and not set in stone, then whatever you used to do the convincing is code, whether it’s C, Lua, an Excel sheet, an arrangement of Minecraft blocks, a drag-and-drop flowgraph, or wires in a plugboard.

      There have been many, many attempts at “telling the computer what to do without writing code” over the years, and all of them end up in one of two places:

      1. They stay too simple to do much of anything useful, or
      2. They gain enough complexity to become programming languages in their own right, albeit weird ones. The people who work with them become programmers, and face their own equivalents of every challenge that programmers deal with on a daily basis, plus the challenge of figuring out how to apply good programming principles to a system that has chosen a lot of non-mainstream metaphors.