Re:Why Reliability Demands Functional Programming
by jancsika on Sunday December 28, 2025 @01:35PM
(#46413304)
> The arguments against pure functions appear to be somewhat contingent: current adoption levels, practice, convenience, and taste.
At least with many GUIs, the time it would take to know, document and/or test all combinations of UI behaviors quickly exceeds the time available to humans. It's at least tractable to start with vastly more states than could ever be measured, and then gate or special-case certain ones to try to make things minimally safe/sane/reliable for the user.
Re:Why Reliability Demands Functional Programming
by xpe on Monday December 29, 2025 @08:42PM
(#46428508)
I can't tell if you are agreeing (by elaborating) or disagreeing (by giving a counter-example) or saying something else. Mind hitting me over the head with it?
At least with many GUIs, the time it would take to know, document and/or test all combinations of UI behaviors quickly exceeds the time available to humans. It's at least tractable to start with vastly more states than could ever be measured, and then gate or special-case certain ones to try to make things minimally safe/sane/reliable for the user.