Case-sensitivity: why -- or why not? (was Re: Damnation!)

[Neel Krishnaswami]

Martijn Faassen <m.faassen at vet.uu.nl> wrote:
> neelk at cswcasa.com wrote:
> [snip]
> > I hope (but fear that I'm wrong) that the same is not true of the
> > iteration protocol -- there's a critical need to be able to iterate 
> > over trees and linked-lists cleanly. This is IMO more significant
> > than GC or even type-class unification, and it's the major reason I'm
> > rooting for Stackless, since it permits Sather-style coroutine-based
> > iterators to be implemented in Python. 
> 
> I'd like to see something like that. For other inputs on this and
> other things, I think the algorithms part of C++'s STL may be worth
> investigating (keep the few neat bits, throw away the rest), and for
> a really out there thing I've recently been playing a bit with the
> language K (that is, I've read some of the manual), at www.kx.com
> K's a bit in the style of APL (though I don't know APL). All I've
> been doing is translating some of the operators to Python classes,
> but since I can't think in K yet I haven't accomplished a lot. :)

I've been meaning to learn J or K for a while. My interest was
intiated the language Charity. It was the claim of the Charity authors
that a large subset of interesting programs could be written as
compositions of a fixed set of combinator fns, and that programs
written in this way are much easier to optimize because you can have a
lot of information about the structure of the combinators. Charity is
really a toy language though (numbers are built from Peano's axioms!).

J and K are apparently languages that really stress this, *and* are
real languages, so I'm interested in learning one of them. However,
they look enough like line noise that I've put it off. I shall have to
rectify this RSN...


Neel