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

Re: [zzdev] A plan: Geek Clang

Antti-Juhani Kaijanaho schrieb:
> Hi,
> I was assigned the task of developing Clang.  My plan is as follows:
> I will build several orthogonal prototypes of Clang named after Greek
> philosophers.  I except that none of them will be the ultimately chosen
> Clang design, but I hope that these ventures into the design space
> will give us experience and ideas that help in the ultimate design.
> These versions are parallel to each other, and separate from the current
> prototype Clang.  I will develop them as GZZ modules.

Nice idea to COUNTER namelessness. ;)

> The Greek Clang will have minimal implementations mainly allowing testing
> of the design ideas.
> My first design will be called Thales Clang, after Thales of Miletus
> (624-547) who was arguably one of the founders of modern mathematics.
> The basic idea of the design is namelessness for everything,
> weak typing, and a procedural/functional framework for abstraction.
> After implementation, Thales Clang should be powerful enough to express
> anything that is computable.  (I will of course implement primitives for
> manipulating ZZ space.)
> Comments?

I think it's a good idea to try out some alternative designs first. One
thing, though: You may want to talk to Ted about HIS idea for Clang, as
expressed in the Floating World design and the original ZZ design (the
two seem to be a bit different to me, but who am I to say); not to cut
out other possibilities, but because there's some chance he has ideas
we're not even *thinking* of...

Are you going to have Ted's triggers / result cells in Thales?

Is there anything in particular I could do so that my applitude /
scripting system will be compatible to the Greek Clangs?

I suppose in Thales you just don't NEED names, but can assign them if
you wish -- or not?

Looking forward to try your stuff out,
- Benja