• @refalo
    link
    4
    edit-2
    1 hour ago

    While it’s great to have the ability to write proofs, I feel like the article makes it sound like it’s some magic bullet that solves all the problems you have with mistakes and bugs in other languages. But really, there’s nothing forcing you to even write said proofs, or define them properly, or turn spark mode on… so you can still make very large mistakes IMO that won’t be caught by the compiler.

    • @[email protected]
      link
      fedilink
      18 hours ago

      agreed, and you have to get the proofs perfect, in whatever language in which they are expressed, to be useful. like you have to get the code correct twice to make sure it works once.

      i think having safe patterns/apis where the compiler can automatically spot classes of potential error is a better approach