News for hackers. Stuff that matters.
Re:Propositions as Types [pdf] by edef on Tuesday May 06, 2025 @12:03PM (#43906667)
> I get to take for granted that a boolean is true or false (and not some brilliant third value!)

  [True, False, undefined] :: [Bool]
Re:Propositions as Types [pdf] by immibis on Tuesday May 06, 2025 @01:35PM (#43907654)
The traditional third value is actually FileNotFound: https://thedailywtf.com/articles/what_is_truth_0x3f_

but in Haskell, yes, it's undefined. Which isn't a real value! For example, infinite loops are undefined. Theorists like to call it a value of every type, but in practical terms, it's more like a computation that never produces a value. The builtin "undefined" can be written as an infinite loop ("undefined = undefined") and many other pure infinite loops can also act as undefined values. The runtime is able to catch some and crash instead of hanging, but not others.

Re:Propositions as Types [pdf] by ddellacosta on Tuesday May 06, 2025 @05:33PM (#43909877)
See, you can even do quantum computing very naturally in Haskell.
Re:Propositions as Types [pdf] by mrkeen on Wednesday May 07, 2025 @03:27AM (#43913068)
Nah.

  check :: Bool -> a
  check  True = undefined
  check False = undefined
  check     _ = undefined


  7:1: warning: [GHC-53633] [-Woverlapping-patterns]
      Pattern match is redundant
      In an equation for ‘check’: check _ = ...
    |
  7 | check     _ = undefined
    | ^^^^^^^^^^^^^^^^^^^^^^^