Comments on killerstorm: Lisp syntax is great!

I'm a mathematician just beginning to try to l...

2010-02-09T17:55:48.957+02:00

I'm a mathematician just beginning to try to learn Lisp; I just recently (in the last few months) discovered Haskell, and in the process, discovered just why Lisp is so special. Both are highly mathematical languages!

I'm attracted to Haskell because it has such a mathematical notation--proof notation, and not just formula notation--but ultimately, I'm attracted to Lisp because of its macro system, and because of its dynamic nature (I've used a lot of Python, and C++/Java before that; I've always appreciated the dynamic nature of Python :-).

I love Lisp syntax, but primarily for different re...

2008-09-29T23:43:00.000+03:00

I love Lisp syntax, but primarily for different reasons than stated here. Not having to worry about precedence rules is a nice side-benefit, but the real beauty of Lisp syntax is the blurring of the line between what is data and what is code, allowing you to do all sorts of powerful things (like macros). I also like that special syntax is not used for things that are actually the same. For example, the mathematical operators are simply functions that are applied in the same way as any function is.

Overall, I think that the precedence rules argument is not a strong enough reason by itself, as I think okke's sarcasm clearly shows.

sysprv: even if you insert parentheses in your own...

2008-09-26T03:06:00.000+03:00

sysprv: even if you insert parentheses in your own code, you might still need a precedence table when you'll read some other's code

shevegen: indeed syntax is not that importa...

2008-09-26T03:03:00.000+03:00

shevegen: indeed syntax is not that important, but still people have their preferences.

i don't get Forth syntax (or lack of thereof) either, but at least it is simple and has no quirks.

on the other hand, as far as i know Ruby has quite arcane syntax, and while it's quite powerful, it's not a language i'd like to program in.

As for me, I don't bother learning operator preced...

2008-09-26T01:59:00.000+03:00

As for me, I don't bother learning operator precedence of infix languages. I just use parentheses to explicitly say what I mean ;))

Well, syntax issue is overrated. I think how capab...

2008-09-26T01:29:00.000+03:00

Well, syntax issue is overrated. I think how capable a language is should not be restricted ENTIRELY down to syntax.

I hate Forth Syntax but I saw really awesome stuff performed in it.

Anyway, what happens if you compare lisp syntax to ruby syntax? Ruby syntax beats javascript syntax hands down with ease - wearing a blindfold.

if you ever tried teaching people math you probabl...

2008-09-25T23:11:00.000+03:00

if you ever tried teaching people math you probably know that considerable effor (perhaps most of effort) goes into teaching how to parse damn notation.

i had to draw parse trees for my sister to understand solving equations (with logarithms, iirc), i'm no kidding! not seeing parentheses she could not understand why you can move one isolated part to the right hand side while you can't another.

Wow, you are totally right. Mathematicians are cra...

2008-09-25T22:32:00.000+03:00

Wow, you are totally right. Mathematicians are crazy as well. Why don't they just use parenthesis instead of remembering that strange order?

Its much easier to read (((2*3)+(2*3))+(5*5)) than (2*3+2*3)+5*5. I mean, I might end up with 65 instead of 37 if I didn't know the precedence rules!

No, I'm happy to learn a couple of precedence rules instead of being verbose all the time. I don't think your argument holds.

eadmund: no, but i've reported bug to a mailing li...

2008-09-25T22:02:00.000+03:00

eadmund: no, but i've reported bug to a mailing list. i guess guys there can fix this faster than i can, if they consider this a bug, or course..

noumeno: yep, IMHO JavaScript has relatively nice ...

2008-09-25T21:58:00.000+03:00

noumeno: yep, IMHO JavaScript has relatively nice syntax if we compare it to Ocaml. maybe it's just me, but parsing all those pattern mathings and let .. ins is PITA.

So, did you submit a patch for Parenscript to avoi...

2008-09-25T20:58:00.000+03:00

So, did you submit a patch for Parenscript to avoid the bug in the future?

Basically, just slap parens around the operands and life is good.

Hi. I kind of wrote a post about memoization in ja...

2008-09-25T20:42:00.000+03:00

Hi. I kind of wrote a post about memoization in javascript, and the syntax didn't turn out to be "ugly" (from my subjective point of view):
function memo(f) {
f.memo = f.memo || {};
return function (x) {
return (x in f.memo)? f.memo[x] : f.memo[x] = f(x);
};
}

I compared this solution to an OCaml one, and although my comparison wasn't that "objective", I believe that javascript's syntax can be quite beautiful to. You can read more about that in my blog post

fwiw, i've never run into operator precedence issu...

2008-09-25T20:13:00.000+03:00

fwiw, i've never run into operator precedence issues in my code because I use lots of parens to make things clear (as i would when writing a mathematical formula). since the code will be read by many other programmers over its life i tend to try and make it as readable as possible.

saving a few keystrokes at code creation just isn't on the top of my list.

i'm not saying that the lisp approach successfully address this issue, mind you, but i don't think it's never been a significant issue in my code or code I've maintained (which is lots).

And another great thing about Lisp syntax is emacs...

2008-09-25T19:48:00.000+03:00

And another great thing about Lisp syntax is emacs+paredit, which allow editing code in somewhat structured way, rather than having lines as primary structure feature.