Since the MusicXML project started, we knew we would face some resistance because the format was owned, developed, and maintained by a single, small software company. Recordare joined both OASIS and the MIDI Manufacturers Association with the idea of networking with these industry standards communities, and possibly submitting MusicXML to a standards process in the future.
This activity took a big step forward in November 2003 when Recordare, Indiana University, and Matthew Dovey started an OASIS discussion list on music notation. As part of this discussion process, we met with as many industry members as we could at the 2004 NAMM show to gauge their interest and willingness to participate in the standards process.
We received very sage advice from our industry supporters, who strongly urged us to avoid the path of a standards organization and maintain the PDF model of an open standard supported and controlled by a single company. They counseled that those not yet using MusicXML were not doing so because of its provenance, but for their own business reasons. Moving MusicXML to a standards organization would not have any positive effect on its adoption. If anything, it would make things worse because MusicXML would not be able to adapt to changing circumstances as quickly as before. Based on this advice from our customers, we decided not to move MusicXML into a standards organization like OASIS.
The wisdom of this advice was amply demonstrated in 2005. Major business opportunities led to the rapid development of the MusicXML 1.1, with over 70 new features to better support music formatting and digital sheet music preparation. These features needed to be designed, implemented, and tested very quickly, as MusicXML 1.1 had to ship sufficiently in advance of immovable product deadlines from three separate companies. This would never have been possible in the higher-overhead environment of a standards organization, even one as industry-friendly as OASIS. The MIDI Manufacturers Association provides a more lightweight process than OASIS, but typically at the cost of active participation by contributors from outside the music products industry.
Standards organizations work well for mature, well understood technologies. They provide a valuable organizational framework for complex negotiations between diverse stakeholders in a technology with a large potential market. Formal standardization in large markets can save money overall, and in life-critical applications it can save lives. But in small vertical markets like symbolic music software, these organizations are too costly in both time and money for even the largest industry members. (A large industry member in symbolic music software is one that has a developer staff in the double digits.) It is particularly costly given that, by itself, the backing of a standards organization is no guarantee that a format will become more widely used.
The lack of relevance of standards organizations for music notation software was demonstrated by SMDL, which was never implemented by any commercial product despite the ISO 10743 seal of approval. History appears to be repeating itself with MPEG’s recent project on Symbolic Music Representation, which is busy integrating WEDELMUSIC-based technology into MPEG despite a near-unanimous show of industry non-support.
Getting an XML format adopted as an interchange standard is a social and technical process. For the largest, most well-understood areas, standards organizations can help in both processes. But for smaller industry areas like music notation, the overhead of standards organizations can hinder these processes, rather than help them.
Many factors aid the adoption of an interchange standard. It is very important for the user community to have an active voice in the development and maintenance of the standard – a voice that is truly powerful, not simply for show. But there are many means to that goal, and individual companies can provide a more effective mechanism than standards organizations in some circumstances.