Re: Measure-free scores

>
> The appearance of measures as we know them is a somewhat modern and western
> circumstance. The music that I engrave for clients today uses them either
> rarely or in methods that are not rooted in 1600-1900 ideology. A major
> commission just turned in (from a composer I advise) uses timings and a
> continuous scroll of score where the only use of pages is for printing,
> cutting up, and taping together.
>
> I am sure these are considered by some to be minor or disposable
> approaches,
> and not to be taken to heart when devising a large-scale system.
>

I assure everyone that the chairs do not consider these approaches to be
either minor or disposable. I really hope that we can all put away the idea
that there is some sort of fundamental conflict here in terms of scope. The
work of this group does not begin and end with CWMN, and we all acknowledge
that. Talking as if this conflict existed, is not a good use of our time.

If there is a conflict, it is about priorities, not scope. Yes, CWMN is a
current focus -- in deference to the many members of the group, large *and*
small, who are working in this area and desire an improved semantic
approach to its representation. But it is only a current focus, not a
direction that irrevocably limits our future accomplishments.

I for one look forward to our developing a separate (and, philosophically,
equal) encoding that is general enough to accommodate a far wider range of
music -- not limited to CWMN, or Western musical culture, or left-to-right,
or indeed anything except some kind of two-dimensional canvas. I think we
will do that successfully in the end. BUT -- speaking as a software
architect -- I think this requires a lot more careful discussion of goals
and of the variety of music and accompanying assets that are to be treated.

To further this discussion, even though it's not top of mind for many in
the CG, I intend to prepare a straw-man proposal to present to the group.
It will emphatically not be based on MIDI, nor will it generalize using
CWMN as a basis. In particular I'll look at ideas that address a number of
fundamental questions that so far have gotten submerged in the noise of the
discussion. Some of these include:

- the nature of the applications which might generate such a format
- arbitrary mapping of graphics to time sequences (since there is not
necessarily a rigid notion of time)
- arbitrary mapping of time sequences to multiple performances (since no
single performance may suffice to capture or illustrate a score)
- highlighting ranges or points in time within an arbitrary graphical score
(since viewers will desire to know what part of the score a heard
performance corresponds to)
- arbitrary musical forms (any part of the score may be instantiated at
multiple points in time, not just one -- think of repeats)

Please: as professionals, let's respect the difficulties of solving these
problems and take our time solving them. Not everyone in the group will
want to, but I will -- even given the present emphasis on CWMN.

In the meantime, please let's honor the multiple interests of this group's
members, and consider that it makes sense to have multiple solutions in
scope, and that we must necessarily deliver some solutions before others.
At least one can be past-facing (CWMN), and leverage the most common
understandings that have emerged from Western music notation. Others can be
more general in terms of graphics and sound and -- in a tradeoff that few
so far seem willing to recognize -- more vague in terms of semantics,
making them unsuitable for some applications. Yet other dialects can be
semantically oriented, but address notational systems of other [past]
cultures that are sufficiently stable to be treated in this way.

Best,
...Joe