re: Solution
@Absinthe
Cool. A cons pair implemented as a closure. cons(a, b) returns a function closing over a and b, which takes a function as argument, and calls it on a and b.
Cool. A cons pair implemented as a closure. cons(a, b) returns a function closing over a and b, which takes a function as argument, and calls it on a and b.
Follow
re: Solution
@billstclair yep, that is what it seems. But what is the use/benefit of this technology?
re: Solution
@billstclair I am not sure I am seeing what power they are trying to illustrate...
re: Solution
@Absinthe
I've seen object systems done as closures. In most languages, a closure can modify its state, so it's an interesting way to close over that state, in a way that is invisible introspection by any other means but the published interface (or a debugger that understands the closure representation).
I've seen object systems done as closures. In most languages, a closure can modify its state, so it's an interesting way to close over that state, in a way that is invisible introspection by any other means but the published interface (or a debugger that understands the closure representation).
re: Solution
@billstclair @Absinthe Except they do. Partial computation and first-class functions are the bedrock of most asynchronous programming (i.e., most Web content.)
re: Solution
No use except to illustrate the power of closures. Nobody would build a real system that way.