News for hackers. Stuff that matters.
Re:Alan Kay on Lisp by spc476 on Friday October 27, 2017 @08:56PM (#15573423)
Godel's Incompleteness Theory states that within an axiomatic system, there are statements that are true that you can't prove true. You can only prove them true if you go up a level.

In programming (assuming a "computer language" as an analog to "axiomatic system"), there are abstractions that cannot be implemented. You can only implement them by going up a level.

In LISP, you have this "going up a level" ability with macros. The only reason this works for LISP is that the language is written in terms of data LISP can manipulate, meaning you can manipulate code at compile time. Forth has an ability to "go up a level" as well, with a defined way to extend the compiler.

Re:Alan Kay on Lisp by Retra on Friday October 27, 2017 @10:26PM (#15573659)
LISP macros are formalized in LISP. You don't "go up a level" by using them. You go up a level by going into the human brain and thinking about your code in a way the computer doesn't.
Re:Alan Kay on Lisp by hxegon on Friday October 27, 2017 @09:22PM (#15573488)
Wow. That's a really interesting way of explaining macros.
Re:Alan Kay on Lisp by andai on Friday October 27, 2017 @09:28PM (#15573502)
I wonder if LISP and LSD encourage similar ways of thinking.
Re:Alan Kay on Lisp by tempodox on Saturday October 28, 2017 @08:39AM (#15574955)
Based on my own experiences with both, I'd say: Yes. Although I'm sure you couldn't prove it mathematically (yet).
Re:Alan Kay on Lisp by physicsyogi on Friday October 27, 2017 @09:53PM (#15573578)
I’ve been using Clojure off and on for a few years but skipped learning about macros because I haven’t (yet) needed to write my own. Your comment makes me regret that decision.