News for hackers. Stuff that matters.
Re:Leaving Haskell Behind by AnimalMuppet on Thursday August 24, 2023 @08:57AM (#37248221)
Some programs live for decades. Some programs get worked on after the original authors are gone. Code is written once, but (if the program is worth keeping) read many times. Optimizing for "easy to write" is optimizing the 10% and ignoring the 90%. Optimizing for "easy to read and understand by someone who is not the original author" is critical for important, long-lived programs.

It's not "laziness". It's understanding that maintaining software long-term is a really significant problem.

Re:Leaving Haskell Behind by HelloNurse on Thursday August 24, 2023 @12:27PM (#37250861)
Even when the maintenance is short term, there is also optimization for making the expected type of changes quick and/or easy to perform, something that appears particularly hit or miss with language specific problems: having to fit monad square pegs into typeclass round holes when two libraries need to be used together, unexpectedly needing a significant increase of sophistication and/or boilerplate to support a generalization that should be small, finding a compromise between conflicting preludes or incompatible extensions.
Re:Leaving Haskell Behind by erik_seaberg on Thursday August 24, 2023 @12:46PM (#37251199)
The thing to optimize for is letting experts communicate with each other clearly and concisely. I’m only going to be a newcomer to a system for a short time, and any work while I'm ramping up is not going to be very valuable.