XML scores and music representation

John T. Payne:

>I realized that XML is very well suited to serve as the
>foundation layer for a generalized musical notation capable of
>representing anything and everything about music susceptible to
>codification. In conventional terminology, this would be musical
>notation as an XML application....

>Using XML it would be possible to create a language in which everything >currently expressible in any musical notation could be written...

I like the idea. In fact, I already figured out how to
do that. My system is phrased in terms of a large Java
package (object-oriented class library), but I had always
thought of it as a sort of standard and wanted to look into
an XML version someday. My representation for scores can
incorporate all sorts of objects not normally considered as
"surface level". I think your notion of XML-izing "scores"
(i.e., performance instructions) is a bit too limiting---
it would be more useful to represent musical ideas in general.

If you're interested, I've a page on it:
http://members.fortunecity.com/odradek5/pp/rustyprogress.html

There is a brief overview in the paper and a long exposition
in the thesis. The part dealing with the representation of
scores is on pp. 143--168, but is part of a much larger set
of objects, any of which may appear in a "score" or form.
Tuning, of course, is covered as well.

-m

_________________________________________________________________
Get your FREE download of MSN Explorer at http://explorer.msn.com

M,

--- In tuning@y..., "Michael Saunders" <michaelsaunders7@h...> wrote:
> If you're interested, I've a page on it:
> http://members.fortunecity.com/odradek5/pp/rustyprogress.html

I get "page not found" -- is this URL correct?

Jon

> I like the idea. In fact, I already figured out how to do that. My
> system is phrased in terms of a large Java package (object-oriented
> class library)

Well, don't throw any of that away, because that would be a good start
on building processors to work with the application language.

> My representation for scores can incorporate all sorts of objects not
> normally considered as "surface level". I think your notion of
> XML-izing "scores" (i.e., performance instructions) is a bit too
> limiting---it would be more useful to represent musical ideas in
> general.

Good! Someone else to help pull in the direction of generality!
You've definitely got a big headstart!

I hope that it's my use of "scores" you find limiting, and not the
suggestion of using XML. It should be possible to translate directly
from the object classes you've already defined to XML structures that
would hold the same information, and which could be used by processors
written in any language. (Perl, Python, ...)

> http://members.fortunecity.com/odradek5/pp/rustyprogress.html
> OI:R
>
> Rusty Progress Report
> ----------------------------------------------------------------------
>
> Rusty is a computer language for the representation of musical ideas and
> the writing of musical programs. An introductory paper (from the
> Proceedings of the First Annual Symposium for Systems Research and
> Cybernetics in the Arts, Baden-Baden, 1999) is available here in MS Word
> and plain text formats. The entire dissertation is available for
> download here in Postscript (2.9Mb) and PDF (1.7 Mb) (viewable with
> Adobe Acrobat Reader) formats, as well as two browsable versions (with
> less than optimal handling of figures, equations, footnotes, etc.) here
> and here. The old site at Cardiff is available here, including the "What
> is Rusty" page and pictures of the old demo.
>
> The above materials describe the plan for the language in great
> detail. It now remains to code it but, as resources have been extremely
> meager, little progress has been made. This page will serve as the
> progress report, updated as progress occurs.
>
> August 28, 2000---The first classes which must be coded are those which
> represent mathematical objects on which most musical objects depend.
> Principally these are one-dimensional functions. The class structure of
> this section has been altered somewhat from that described in the
> thesis, but it's still based on analytical combinations of complex
> functions represented as piecewise near-minimax polynomials on a
> partitioned domain. The interfaces DomainValued and Evaluable are
> complete as are the classes Domain and Subdomain. The class NearMinimax,
> which represents a near minimax polynomials and calculates the expanded
> Chebychev points, is very nearly complete. A survey of possible Rusty
> generators (a "Map of Parnassus") is underway, and may soon have its own
> webpage.
>
> May 4, 2001---I have provided online versions of the thesis and the
> Baden-Baden paper in various formats. Work on the mathematical
> foundations continues slowly; I'm currently reviewing methods for
> regression (least-squares using optimal or given basis polynomials or
> trigonametric polynomials), various types of piecewise interpolation
> (splines, etc., as a method of function definition), etc.

I don't understand a lot of that, but that's as it should be... ;-)

[] John T. Payne - PO Box 4534 - Boulder, CO 80306
[] http://www.well.com/user/satyr/ - satyr@well.com