News for hackers. Stuff that matters.
Re:The unreasonable effectiveness of declarative programming by reuben364 on Tuesday May 19, 2020 @10:57AM (#23235550)
One thing I realized that seq'ing too many things might be bad for performance. As I understand it,

   foldr1 (\x y -> ((x `par` delay 1) `seq` y)) $ map pure [1..]
Would require more and more comparisons over time. Is there some sort of codensity like trick that you can do? Or is this something the user of the library would worry about if it came to that. Or just have a redundant API.

There are no comments on this comment yet.