[Top] [All Lists]

Re: [Haskell-cafe] Language simplicity

Subject: Re: [Haskell-cafe] Language simplicity
From: Martin Coxall
Date: Wed, 13 Jan 2010 08:49:40 +0000
On 12 Jan 2010, at 21:25, Andrew Coppin wrote:

> OK people, it's random statistics time!
> Haskell '98 apparently features 25 reserved words. (Not counting "forall" and 
> "mdo" and so on, which AFAIK are not in Haskell '98.) So how does that 
> compare to other languages?
> C: 32
> C++: 62
> Borland Turbo Pascal: ~50 [without the OOP extensions added later]
> Eiffel: 59
> VB: The source I checked listed in excess of 120 reserved words, but I'm 
> dubious as to how "reserved" they really are. (Is CInt really reserved? I 
> doubt it!) It also depends wildly on which of the bazillion VB dialects you 
> mean.
> Java: 50
> JavaScript: 36
> Smalltalk: 0

There are six singleton pseudo-variables that act as reserved words: 
true,false, nil, self, super and thisContext.

> Lisp: AFAIK, there are no truly reserved words in Lisp, only predefined 
> functions. (??)

All Lisps have "special forms" which are evaluated uniquely and differently 
from function application and are therefore reserved words by another name. For 
example, Clojure has def, if, do, let, var, quote, fn, loop, recur, throw, try, 
monitor-enter, monitor-exit, dot, new and set!.

> Python: 31
> Ruby: 38
> Tcl: Same analysis as for Lisp I believe.

COBOL: Over 400 (!)

> As you can see, this conclusively proves... something.

Generally speaking, the most widely used languages seem to be near the upper 
end of the range.

I don't think it really tells you that much. Possibly that a little superficial 
complexity through syntactic sugar can make your language a lot more 
human-friendly, but that it's possible to go too far and end up like C++.

Haskell-Cafe mailing list

<Prev in Thread] Current Thread [Next in Thread>