The Project of Software Is Complete(freddiedeboer.substack.com)
6 pointsby pbmonsteran hour ago2 comments
  • xyzsparetimexyz25 minutes ago
    Great article, though a bit long winded. I'd love a browser extension that read posts like this via llm and greyed out the less interesting half.

    But I feel like that's the only qualifier I'd add. The project of _traditional_ software is complete. AI and hybrid (heavily AI-integrated) software is still in progress.

  • pbmonsteran hour ago
    The article poses the question "What do you want your phone/computer to do that it can’t", and suggests that there really is nothing left - software is done, all remaining problems are hardware.

    I can't think of many things, but there are a few: I want an LLM teaching 1 on 1 language lessons. Good ones, with tailored vocab, tailored reading, tailored listening, grammar lessons, and (most importantly): conversation training including it correcting my pronunciation and grammar.

    Current LLMs can do most of that when prompted correctly (and supplied with textbooks and podcasts), just the last part is missing.

    But what else is there?

    • xyzsparetimexyz22 minutes ago
      That falls into the qualifier I added here: https://news.ycombinator.com/item?id=49099233. Any kind of software frontier is AI-specific.
    • PaulHoulean hour ago
      I liked it because he shows the logistic curve, something that's forgotten today when people are enamored with the exponential and the "intelligence explosion" model of

        dx    2
  -- = x
  dt
      even though you never see people actually write that one down.
      • xyzsparetimexyz28 minutes ago
        Idk, people frequently say 'every exponential is actually a sigmoid'.