Eleanor Selfridge-Field and Edmund Correia, Jr.
© 1994 Center for Computer Assisted Research in the Humanities
Reproduced by the IEEE Technical Committee on Computer Generated Music by permission

Musical Information in Music Desktop Publishing

When technologists first contemplate the problems of coordination that the printing of musical scores poses, they tend to focus on what is regular and predictable: beat counts that add up to the sum suggested by the time signature, pitches that remain within the tonality specified by the key signature, texts that are set one syllable to a note, articulations that are specific to one note, and so forth.

The more the representation of musical scores is studied, however, the more it is appreciated that they contain a great many logical compromises and depend in part on oral tradition for uniform interpretation. Pedagogy is essential to the preservation of the tradition that scores represent.

Because of the inconsistencies and idiosyncrasies that occur from piece to piece, aberrations of musical logic seem to be ubiquitous in the classical repertory. When we base definitions of what is "common" or "regular" on Western music of the eighteenth and nineteenth centuries, we inherently create a bias against the perceived "abnormality" of other repertories be they drawn from earlier or later art music in the European tradition or folk music from Western or non- Western traditions.

To explore the competency of music notation programs, Computing in Musicology, a publication of the Center for Computer Assisted Research in the Humanities, has distributed each year since 1986 a number of set pieces to illustrate a wide range of potential problems encountered in the repertories from the Middle Ages to the later twentieth century. While this list is by no means exhaustive, it is representative of the kinds of problems that are often overlooked by those attempting to write, as well as use, programs for desktop publishing.

Each software contributor is also welcome to submit one free-choice example, and this option has often been used by those wishing to display a really unusual capability. Our set pieces have explored European music from the fifteenth century to the present, although for copyright reasons, we omit recent music from this presentation.

When we began this survey in 1986, platform options were very different from those of today, and it would serve no useful purpose to recapitulate their history. Most programs in use today are designed to work with graphical user interfaces, such as those provided by the Macintosh and PC Windows. However, applications in musical analysis are for the most part more easily run with information that is presented in the ASCII format, rather than with graphic information devoid of intellectual meaning. This puts methodologies for scholarly applications other than printing somewhat at odds with those that support desktop publishing.

Another complicating issue is that of sound files, sound protocols such as MIDI, and languages such as Csound for representing sound. The same problems of coordination of informational domains exist.

In the ensuing examples, problems are organized by type. To assist those lacking extensive performing experience, we indicate the expectations that would be made of a generalized representation.

Settings from many different programs were available for each example of common music notation. We attempted to balance the coverage of diverse systems in our choices. Users should note that most of these examples were presented at much higher resolutions in print than is possible in a CD-ROM reproduction. Bar numbers are local to the example.

[ Index | Previous Paragraph | Next Paragraph ]