Re: Suggestion : Future directions, zoom out etc

[David Jeske <jeske@...>]

On Tue, Feb 05, 2002 at 05:21:56PM -0800, Steve Dekorte wrote:
> On Tuesday, February 5, 2002, at 03:25  PM, paulmatthews wrote:
> > It seems there is an amply sufficiency of little languages that lack
> > rigour. ...There is an existing heavyweight language that has
> > most of these features, i.e. Java.
> 
> It's interesting that are no languages that are both lightweight and 
> strongly typed. (At least none that I know of) Could this be more than 
> coincidence?

(hey steve :)

I would guess the reason for this is that coming up with a strong type
system which is staticly checked and is at the same time both usefully
expressive and unobtrusive is a "hard problem", so most little
languages (in fact most languages) punt and ignore that problem
entirely.

To paraphrase an ML presentation to a bunch of Perl programmers:

  "Given that static typing, as implemented by C and Pascal, 
   is a failure, what can we do about it?

   One strategy is to simply give up...and Perl gives up more
   than other languages."

   http://perl.plover.com/yak/typing/typing.html

It's pretty to easy to see why having great static type-checking is
going to make the language implementation bigger, because writing
software smart enough to save you from yourself and others -- without
getting in your way too much of the time -- dosn't come for free.

It seems to me that most staticly typechecked languages have opted to
make the problem easier by getting in your way more often.

Those ML and Haskell folks keep chanting on about how they've solved
this problem, but I can never understand how to get enough of that
functional programming and pattern matching stuff out of my way to get
some simple UI widgets on the screen. Ohh well.

-- 
David Jeske (N9LCA) + http://www.chat.net/~jeske/ + jeske@chat.net