[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Cognitive "lightness" of a language

On Thu, 6 Dec 2001, Simon Cozens wrote:
> On Thu, Dec 06, 2001 at 01:27:59PM -0800, Niall Dalton wrote:
> > I'd like to propose an alternative view on what a lightweight language is.
> Why bother? Once you've decided what "lightweight" means, where does
> that get you?

For a start, this view of LLs does not mean that they are suited to just a
certain set of tasks - an LL should be suited to the cognitive process
that a programmer or designer goes through when using the language. 

I think that this has implications for the design of LLs. If some existing
language meets these goals then its probably accidental. I haven't seen
any design justified on cognitive grounds, nor off-hand can I think of a
language to perfectly meets all the goals.

Such a language may be quite different from existing ones, or it may
be a variation on well-known themes. It may, for instance, have
implications on the implementation that are worth exploring. 

So, don't you think its worth considering?

On the subject of implementations, some options exist even for existing
languages, lightweight or not. For example, why go with variations on
bytecode? If it will be further compiled for faster execution then there
are better formats for distribution.