Arc Forumnew | comments | leaders | submitlogin
2 points by evanrmurphy 2484 days ago | link | parent

> I'm sure you're right; we just need to figure it out for ourselves.

Maybe. :) These ideas aren't totally unheard of or untested. For example, PicoLisp has auto-quote for the number case, and it's been around since the 80s! XD


Instead of '(1 2 3) you can just write (1 2 3). This is possible because a number never makes sense as a function name, and has to be checked at runtime anyway.

PicoLisp also takes advantage of the initial bindings idea from my #3 to some extent. In PicoLisp, transient symbols are initially bound to themselves, and ordinary symbols are initially bound to nil.

2 points by akkartik 2484 days ago | link

I think I'd be more likely to feel positively toward completely unheard-of ideas[1]. For example, I mistrust dynamic scoping more because some variants of lisp have had it. I felt positively toward ssyntax and open defs from day 1 even though they were unheard of.

But all this is just gut instinct. I'll try anything once :)

[1] Though it was good to find out about the parentage of autoquoting, thanks. I wasn't aware of that.


4 points by Pauan 2481 days ago | link

I dislike default dynamic scoping because it seems like a bad idea to me. The fact that other Lisps already have it seems pretty irrelevant, except insofar that it gives us a chance to try out dynamic scope and see what it's like.

In other words: I think old ideas are fine, and new ideas are fine. Ideas should stand on their own merits. The only difference between an old idea and a new idea, is that the old idea has been tested somewhat, so we can have a better idea of whether it's a good idea or not.