1. 4

  2. 1

    I found this article to be very interesting. I’ve never spent time in languages which are by default lazily evaluated, so I’ve never run into this problem—I’ll occasionally turn it on, but usually I don’t, so this is an interesting concern to me. For people who have spent time in Haskell/Clojure/something lazily evaluated by default, do you find it harder to reason about thunk leaks than regular, eager-evaluated memory leaks?

    1. 1

      Yes, but I suspect that’s largely a (barely existent) tools problem.