On Thu, 2006-09-07 at 23:02 +0200, Thomas wrote:
> I'm not a computer scientist, but isn't uniqueness essential for an
> interpreter/compiler to work?
Uniqueness of what?
> Also, what does this have to do with business logic?
I was speculating that a language designed for representing such things
as business logic might not be ideal for representing music, and that
syntax errors may not be such an issue in a language designed for
livecoding.
> Isn't a fault tolerant programming/computing system the thing
> everybody would pay lots of money for?
Sure but when you're making music, tolerating faults means something
quite different from when you're modelling flight navigation systems or
proving mathematical models.
> Integrating interpretors into editors with some kind of automatic
> syntax correction is surely useful to avoid errors, but far away from
> allowing errors (like the guitar).
I think we're getting confused between syntax errors (instructions that
do not strictly conform to grammatical rules) and semantic errors
(playing a bum note). It is all too easy to write a syntactically valid
program that plays bum notes - ie not the notes we were aiming it to
make.
Avoiding syntax errors therefore allows unrestricted semantic error
making.
> Your last point leads away from the actual discussion, as these errors
> are not directly related to the act of playing.
I was drawing an analogy between the constraints of languages and the
constraints of drumkits, trying to work out what the syntax of drum
playing would be. I might well be talking nonsense there though.
alex
Received on Thu Sep 07 2006 - 21:41:34 BST