In the early 1990s I was teaching electroacoustic music composition at EMS in Stockholm. In that class I tried to implement the usage of scores in my own lectures and also encouraged the students to draw their own scores for their assignments. These scores were at first handwritten and not built on any formal system. As inspiration I presented some of the few graphic scores that exited for electroacoustic music made by e.g. Stockhausen, Messiaen, and Ligeti. When presenting electroacoustic works lacking a score, I drew my own to aid in following the music and analyzing form.
In 1993 Lasse Thoresen and Peter Tornquist from Norway held a lecture at EMS in Stockholm presenting Aural Sonology. Among the things they presented was a set of graphical symbols filling the categories of Schaeffer’s typology (TARTYP). Peter Tornquist made these symbols in a graphics program (Claris Works for Macintosh). After this event I began producing scores using Tornquist’s symbols, copying and pasting them in to new analyses. But getting symbols and lines to match up on the screen in a graphics program and look the same when printing them wasn’t easy. Further, there were many programs competing and equally used for different purposes (Microsoft Word, Quark Express, Photoshop etc.), and no easy way to share graphic symbols between them.
The idea to use a font began to grow as a solution, that way the system could be used in all programs handling text, including notation programs like Finale.
Around 1998 on a grant from the Royal Music high school in Stockholm I began developing a font called “Addenda”. At the same time, I began experimenting with making “electronic scores” (then presented as QuickTime movies) with the use of this font.
The Addenda font was still based directly on the categories in Schaeffer’s TARTYP which had apparent shortcomings. The TARTYP had less than 30 categories for describing sound-objects; far too few to describe all possible sounds accurately. Worse still: many real world sounds didn’t seem to fit well in any of the categories.
When drawing scores by hand, however, the basic arsenal of circles, squares and lines could be combined freely to draw the actual components of sound-as-heard without being restricted by a limited set of categories. This experience gave clues to how Schaeffer’s system could be revised. It took some time to sort out the details and still further time to really try out the revised system with analysis of actual music. Lasse Thoresen was drawing scores by hand and I was trying to come up with ways to render the symbols using a font. Minimizing the variation in these scores to fit in a limited amount of signs that could be produced by a font was a large part of the task.
In the end, I feel, the limitations of a font actually helped to make the notation more standardized.
Eventually this led to a new font where the symbols could be put together piece by piece. For a limited increase in signs in the font, billions of combinations could now be made.
At that time a standard font could hold 128 signs defined by the ASCII standard, and an additional 128 non-standardized signs. The latter had symbols whose placement differed both for different languages and for different operating systems (e.g. Windows and Macintosh). As the list of symbols in Aural Sonology grew, I ended up creating four different fonts, Sonova (spectromorphology), Agros (fields), Forma (formal elements), and Strata (layers). But Sonova was really running out of space to put in new symbols. Also, compatibility problems began to be difficult to control when transferring documents between Windows and Macintosh computers.
Worst of all, symbols where directly tied to a letter on the keyboard; a change in the keyboard layout meant that all earlier documents would have their symbols switched! I wrote scripts swapping letters in Word to translate all previously made documents each time a major change of the keyboard layout had to be made. A horrible situation!
Then, one day, it looked like the transition to Unicode was really going forward. Operating systems started to support it, even Microsoft Word could handle it. So, I started to make a Unicode version, merging the four fonts into the “Sonova UC” font. For many years, though, the old fonts had to be used in parallel awaiting the implementation of Unicode in essential programs (‘Finale’ didn’t get it until 2012).
In Unicode each symbol is standardized by a consortiumIt is not likely that Sonova ever will be included in the Unicode standard, even standard musical symbols are only given a very limited support. But Unicode has a special ‘Private Use area’ that’s ideal for adding custom symbols. To access the symbols from this area special Keyboard Layouts had to be constructed. The four fonts from before now corresponds to four Keyboard Layouts.
Today we have one font, working on all operating systems in all countries. There is one logical layout of the symbols in the font itself, separate from the keyboard layouts that can be independently changed at will, any time, without breaking existing documents. And plenty of space for adding new symbols without breaking backwards compatibility.
So, when using the Sonova font you have to change the font to Sonova AND switch the keyboard layout. A bit annoying, but now you know the history behind it.