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

RE: The Accessibility of Type Theory Research

James McCartney wrote:
> On Nov 20, 2003, at 9:22 PM, Anton van Straaten wrote:
> > Hey, you should check out Haskore, a computer music system embedded in
> > Haskell:
> > http://haskell.cs.yale.edu/haskore/
> >
> > It might provide an incentive to learn something about advanced
> > statically typed systems... :)
> I am aware of it. Seems like the last release was in 2000. But I've not
> heard of many folks using it, and I don't find much interesting about
> it. The pattern classes in SuperCollider are more powerful and more
> concise. In fact I find that Haskore tries too hard to make a type out
> of everything and so overly constrains the musical concepts in a
> preconceived way.

I'm glad to finally see a meaningful comment like this backed up by some
actual experience, rather than the earlier speculation on what ST systems
might not be able to do.

I think you're right, embedding DSLs in Haskell can suffer from the kind of
problem you describe, since types are the major vehicle for extending the
language.  It's one reason I like Scheme and its macros, which provide more
flexibility in extending the language.

> My language has some popularity, is taught in many computer music
> programs in universities, and has had quite a few pieces written and
> performed with it. Musicians don't really care about type systems..

Nevertheless, if you're working with language design, you're working with a
type system whether you realize it or not.  As a mere consumer of music, I
could claim that I don't care about music being out of key, but I'd probably
be wrong about that.

If I began posting on a composer's mailing list, saying I was composing
something, and claiming I didn't need to know anything about music
terminology or notation, what kind of reaction would I get?  "Programmers
don't really care about musical notation", I might say.

> > You do know that magic carpets aren't real, right?
> Literature must be lost on you.

No, rather I distinguish quite clearly between fantasy and computing
science.  It's hard to give much real-world credence to an example involving
magic carpets, unless your point is that the unwashed masses you mentioned
sometimes delude themselves into thinking they're flying on magic carpets,
when in fact they're wallowing in mud.  (Mud features quite prominently in
Marquez stories.)

> > Why "in spite of", though?  More information is available to us today,
> > more easily, than ever before.  The concepts finding their way into
> > languages like Io today were pioneered by academics, decades ago.
> > I'm not an academic, but I've benefitted enormously from the academic
> > information that's available, including many of the references that
> > have been given on this list.  If you "don't even bother" with any
> > of that, that's your choice.
> >
> > The only "spite" here is in the sense of "cutting off your nose to
> > spite your face".  It's hardly surprising that you can't find many
> > academics willing to spend time giving free tutoring.
> I think you presume that I am taking a role here that I am not.

Perhaps in my last sentence about tutoring, I was really thinking of some of
the things Steve Dekorte has said.  The rest was a direct response to what
you wrote in the message I responded to.  You might think you're doing
something "in spite of" the academics, but it seems to me you're only able
to do it because of the academics.