Antonio Camurri, Giovanni De Poli, and Davide Rocchesso

[The following is meant to accompany the taxonomy presented in the
Editor's Notes of Computer Music Journal 19(2). It is based on several
months of discussions involving a broad selection of practitioners in
the field.--STP]

Two interesting debates on the relationship between computer music,
computer science, and multimedia art and technology were held at the
1994 International Computer Music Conference (ICMC) in September 1994 in
Århus, Denmark and at the IEEE International Conference on Multimedia
Computing and Systems in May 1994 in Boston, Massachusetts, USA. This
discussion focuses on the topic of "Didactics of Computer Music to
Computer Engineering Students and Professionals," from the panel
discussion at ICMC 1994.


Introduction

The aim of this panel was two-fold. On the one hand, it focused on the
didactics of computer music (CM) in the scientific community, with
particular regard to computer science and engineering (CS&E); on the
other hand, it discussed the relevance of CM in CS&E.

Before presenting the different viewpoints from the panelists, a
possible misunderstanding must be clarified. We believe the term
"computer music" is traditionally interpreted primarily from a musical
perspective rather than a scientific one. For us, however, the goal is
not to discuss how and why to teach music (CM) to engineers, but to
bring to light relevant scientific work from the last 30 years that
gives rise to an effective, new, scientific CS&E discipline. For this
reason, in this note we refer not to CM, but instead to "Sound and Music
Computing," or SMC. Indeed, this important difference emerged during the
panel discussion--in particular from an observation by Carla Scaletti--and
SMC has been proposed as the new name of the engineering discipline
related to CM.


The Didactics of SMC

The panelists presented an overview of their experiences from three
continents and different viewpoints. In the USA, didactics of SMC is
carried on in CS&E and music departments. Historically, the discipline
has been cultivated in two different areas: either CS&E, artificial
intelligence, multimedia, etc.; or electrical engineering, with
particular regard to signal processing. This is analogous to the origin
of the CS discipline of computer graphics.

Roger Dannenberg's experience is an excellent example of the first
approach, as it is being conducted at Carnegie Mellon University, where
they are trying to identify electronic media concerns in the areas of
psychology, art, and CS. Such a large program implies a diverse student
body, and severe problems arise. One issue is the prerequisites that
such a program should expect. While it should be open to students with
different backgrounds, a two-year-long master's degree program is
probably too short to ensure that all students share a comprehensive set
of core skills.

Another issue is the industry's expectations. Academic research must
deal with topics of interest to the industrial market, especially in the
case that companies pay tuitions in the form of scholarships or grants.

This is sometimes a problem in CS&E, although companies are showing a
growing interest in multimedia, virtual environments, and related topics
in which SMC is of primary importance. The market for video games and
multimedia entertainment and "edutainment" is growing fast, and SMC
should be one of the leading disciplines for these applications.

The other side of didactics in the USA is well represented by Julius
Smith of the Center for Computer Research in Music and Acoustics at
Stanford University. His activities are focused on signal processing,
traditionally a component of electrical engineering programs. He teaches
three courses in signal processing for music and musical acoustics. The
entry-level course requires only a basic knowledge of calculus, so the
three-course series is almost self-contained. Julius Smith noticed there
is support from the market for their research in musical signal
processing, which leads him to forecast a good future for the field of
SMC.

The state of the art of didactics in Europe varies strongly, depending
on the policy of each country, as stated by Giovanni De Poli of the
University of Padua in Italy. The teaching for music practitioners is
provided mainly in conservatories. The SMC research is carried on in the
electrical engineering, CS, and physics departments. In each country,
different traditions determine which is the leading discipline.
Seminars, symposiums, short courses, and thesis projects can be found in
many European universities, and regular courses and master's programs
are being set up in several significant sites (e.g., Padua, Milan,
Paris, and Barcelona). Giovanni De Poli suggested that in order to get
more support for these activities from the world-wide technical
community, specific programs for providing SMC education to graduate
students (e.g., the French master's degree program "acoustics, signal
processing, and informatics applied to music" [DEA ATIAM|) should be
fostered by the International Computer Music Association (ICMA).
Moreover, the ICMA is a significant example of cooperation among
different institutions.

Shuji Hashimoto from Waseda University in Tokyo, Japan is one of the
leading exponents of Japanese SMC research and didactics. He explained
that in Japan, SMC is associated with the kansei concept (pleasure,
emotion, sensuous intelligence), thus leading to a wide variety of
applications and interactions with other disciplines. The Ministry of
Education supervises "kansei information processing," while the Ministry
of International Trade and Industry is responsible for the "kansei
industry." Several commercial markets find their place in the kansei
framework, including computer graphics, computer music, computer games,
and karaoke systems. There is a noted lack of professional education on
the engineering side, with the unique exception of the Kyushu Institute
of Art and Technology. The School of Science and Engineering at Waseda
University started an interdisciplinary course in 1995 with two majors:
science and technology policy; and expression technology. For the latter
major, a multimedia studio is planned, and classes in art and
technology, computer art, artistic symbolics, musicology, and
composition are being considered.


The Relevance of SMC

The main questions we are trying to address are: (1) is SMC eligible for
the rank of a true discipline in CS&E (like, for example, computer
graphics)? (2) what is the current situation for SMC in CS&E
departments? and (3) what is the role of professional organizations, in
bringing SMC to a higher consideration in the scientific community? As
we noted above, the discipline's first problem is its name. Since the
root of CM is music, the scientific community doesn't have a clear
picture of the real aims of the field. Several proposals for a better
definition have been advanced: musical informatics, sound and music
signals, sound and music computing (the term used in this note),
computing methodologies for sound, and musical signal processing. Others
will probably be generated when the content and the classification of
the discipline are clearer. For the definition of the area spanned by
SMC, two starting points are those outlined by Antonio Camurri
(University of Genoa): the taxonomy of Computer Graphics given in the
January 1994 issue of ACM Computing Review and the CM taxonomy printed
in Computer Music Journal 18(1), pages 5-8. On this basis, a new
classification for SMC must be developed. [See the Editor's Note in the
previous issue of Computer Music Journal--STP]

An active role in promoting SMC is taken on by organizations such as
ICMA, more recently the IEEE Computer Society Technical Committee on
Computer Generated Music (IEEE CS TC on CGM), and ACM SIGSound. These
three associations have distinct areas of interest, although they
partially overlap. Table 1 helps to define the different ranges of action.




Table 1

 

  
Organization
  
Topical Focus
 

 
ICMA
 
Independent organization with an interdisciplinary CM/SMC focus

 
ACM SIGSound
 
Sound Processing within a CS organization


 
IEEE CS TC on CGM
 
SMC within an electrical engineering organization





Some initiatives must be taken to better coordinate the activities of
these three groups. Antonio Camurri proposed to establish a working
group (e.g., the panelists and other interested people)--with the support
of ICMA, IEEE CS TC on CGM, and ACM SIGSound--with the goal of refining
an SMC curriculum for CS&E university courses, and a classification of
the discipline. The result should be a publication in a leading
(probably ACM or IEEE) CS&E journal, for a wider impact in the CS&E
community. Any effort to better establish the discipline should be taken
into account and eventually pursued. Efforts to increase the relevance
of CM in the scientific community will be enhanced by our individual
experiences within our own countries.

Shuji Hashimoto depicted the web of societies related with the kansei
concept in Japan, including a Special Interest Group on Computer Music
(former JMACS) in the Japanese Information Processing Society, a Human
Communication Group in the Institute of Electronics and Informatics, a
Communication Engineering Human Interface Technical Committee in the
Society of Instrumentation and Control Engineering, and a Virtual
Reality Technical Committee in the Society of Mechanical Engineering.
Toward a global view, Mr. Hashimoto proposed a Joint (ICMA/ACM/IEEE)
International Conference on CM and SMC, to be held every three or four
years, which would be based on the ICMC. Such an initiative might represent
growth for the whole community. Mr. De Poli presented the Italian
approach as an example to be considered: a CM/SMC association (AIMI),
promotes specific topies and interdisciplinary eooperation. Every two
years, AIMI organizes a Colloquium on Musieal Informatics (CIM),
intended to be an interdisciplinary event. In international workshops,
smaller groups of researehers discuss hot topics, and special-interest
groups (e.g., on music analysis) function as permanent fora for
exchanging information and experiences. The result is one community that
presents different faeets to different diseiplines.

Extending the diseourse to a wider domain, Giovanni De Poli proposed
trying to improve the quality of SMC seientifie researeh, to eonvey the
most signifieant eonferenee papers to teehnieal journals. He also
observed that it would be useful to eoneentrate efforts and improve
eooperation among different organizations, thus achieving a "critical
mass" for a more effeetive presenee within the seientifie eommunity.
Johan Sundberg of the Royal Institute of Teehnology (KTH) of Stoekholm,
Sweden, after his experience in ICMC paper-selection panel, noticed that
many good papers had to be rejected for lack of space. On a larger
scale, it is our impression that the quantity and quality of computer
music research are well beyond the space available in existing journals.

An active debate on the topics raised here can only further our common
goals, and we would like to invite Computer Music lournal readers to
participate, in the pages of the Journal as well as in the media of the
ICMA, the IEEE CS TC on CGM, and the ACM SIGSound.