3 ms·
I wonder why until now nobody wrote any comment about the content of this paper, a very simple implementation of lambda calculus showing as simply as possible h
by martyalain 6y ago
I wonder why until now nobody wrote any comment about the content of this paper, a very simple implementation of lambda calculus showing as simply as possible how on two operations, abstraction and application, can be built the data and control structures at the bottom of a programming language: booleans, pairs, lists, recursion, numbers... And so, no answer to my question.
- martyalain 6y agoAbout this paper Simon Peyton Jones (Haskell, ...) mailed me: "It makes sense, yes, but it’s not clear what is new. This is just the untyped lambda calculus and church numerals, isn’t it? All good, but not new. Of course, old is good – it has stood the test of time. Lambda calculus is fantastically small yet it is Turing complete. Sometimes the implementation makes a difference." And my answer was: " Yes "Sometimes the implementation makes a difference" and here we are, the implementation of metatalk is absolutely different. As it can be seen at the end of this page http://lambdaway.free.fr/lambdawalks/?view=meta3 http://lambdaway.free.fr/lambdawalks/?view=meta3 in the "Annexe" section. Nothing but a hundred of vanilla javascript lines, which are the kernel of lambdatalk. Your opinion is welcome.