Álvaro Mendes Barbosa
PhD Thesis, April 2006

Dissertation submitted to the Department of Technology of the Pompeu Fabra University for
the Program in Computer Science and Digital Communication, in partial fulfillment of the
requirements of the degree:

Doctor per la Universitat Pompeu Fabra with Mention of European Doctor

Dissertation directed by:
Dr. Xavier Serra and co-directed by Dr. Sergi Jordà

Doctoral panel members:
President: Dr. Josep Blat Gimeno (Pompeu Fabra University, Spain)
First board member: Dr. Christopher David Chafe (Stanford University, USA)
Second board member: Dr. Paulo Ferreira Lopes (Portuguese Catholic University, Portugal)
Third board member: Dr. Daniela Coimbra (Royal College of Music, UK)
Secretary: Dr. Rafael Ramírez Meléndez (Pompeu Fabra University, Spain)
First substitute: Dr. Fabien Gouyon (Research Center for Artificial Intelligence, Austria)
Second substitute: Dr. Jason Freeman (Georgia Institute of Technology, USA)

Universitat Pompeu Fabra
Departamento de Tecnologia
Estació de França
Passeig de Circumvallació, 8
08003 Barcelona, España

Research leading to this Dissertation was conducted by the author at:


This Doctorate Research Work was supported through the award of a Doctorate Scholarship by:


(SFRH/BD/5192/2001)

The Author is affiliated with:

To Sofia

Abstract
This dissertation derives from research on musical practices mediated by computer
networks conducted from 2001 to 2005 in the Music Technology Group of the Pompeu
Fabra University in Barcelona, Spain. It departs from work carried out over the last
decades in the field of Computer-Supported Cooperative Work (CSCW), which
provides us with collaborative communication mechanisms that can be regarded from a
music perspective in diverse scenarios: Composition, Performance, Improvisation or
Education.
The first contribution originated from this research work is an extensive survey and
systematic classification of Computer-Supported Cooperative Work for Music
Applications. This survey led to the identification of innovative approaches, models and
applications, with special emphasis on the shared nature of geographically displaced
communication over the Internet. The notion of Shared Sonic Environments was
introduced and implemented in a proof-of-concept application entitled Public Sound
Objects (PSOs).
A second major contribution of this dissertation concerns methods that reduce the
disrupting effect of network latency in musical communication over long distance
networks. From laboratorial experimentation and evaluation, the techniques of Network
Latency Adaptive Tempo and Individual Delayed Feed-Back were proposed and
implemented in the PSOs prototype.
Over the course of the PSOs development other relevant and inspirational issues were
addressed, such as: behavioral-driven interface design applied to interface decoupled
applications; overcoming network technology security features; system scalability for
various applications in audio web services.
Throughout this dissertation conceptual perspectives, of related issues to computer-
mediated musical practices, were widely discussed conveying different standpoints
ranging from a Psycho-Social study of collaborative music processes to the Computer
Science and Music Technology point of view.
i

Resum
Aquesta tesi recull la recerca al voltant de les pràctiques musicals mitjançant xarxes
d’ordinadors realitzada al Grup de Tecnologia Musical de la Universitat Pompeu Fabra
a Barcelona entre l’any 2001 i el 2005. Parteix del treball dut a terme durant la última
dècada dins del camp del Treball Cooperatiu amb Ordinadors (Computer-Supported
Cooperative Work, CSCW) el qual aporta els mecanismes de col·laboració els quals, des
de un punt de vista musical, poden ser estudiats en diversos escenaris: composició,
interpretació, improvisació i educació.
La primera contribució d’aquest treball és un anàlisi exhaustiu i una classificació
sistemàtica del Treball Cooperatiu amb Ordinadors per Aplicacions Musicals. Aquest
anàlisi es va centrar en la identificació de propostes innovadores, models i aplicacions,
amb un especial èmfasi en la natura compartida de la comunicació mitjançant internet.
El concepte d'Entorns Sonors Compartits va ser presentat i implementat en una
aplicació prototip anomenada Public Sound Objects (PSOs).
La segona gran contribució d’aquesta tesi consisteix en l’estudi del possibles mètodes
per reduir les interrupcions degudes als retards inherents en la comunicació musical
entre xarxes molt allunyades. A partir de l’experimentació i avaluació al laboratori les
tècniques Network Latency Adaptive Tempo i Individual Delayed Feed-Back van ser
definides i implementades dins del prototip PSOs.
Al llarg del desenvolupament del PSOs es van haver de resoldre altres problemes, com
per exemple, el disseny d’interfícies en funció del comportament per a aplicacions amb
interfícies desacoblades, la superació dels diversos sistemes de seguretat de les xarxes
informàtiques i les possibilitats d'escalabilitat de diverses aplicacions d’àudio per a web.
Durant l’elaboració d’aquesta tesi es van discutir diferents perspectives per resoldre
problemes relacionats amb la pràctica musical mitjançant ordinadors, aplicant diferents
punts de vista provinents de l’estudi psicosocial dels processos de col·laboració musical
al món de la informàtica i de la tecnologia musical.
ii

Acknowledgments
I would first like express my gratitude to Professor Francisco Carvalho Guerra for his
trust and encouragement over the last five years, as well as the opportunities he has
provided for me and all my colleagues at the School of Arts at the Portuguese Catholic
University in Porto, Portugal.
I am deeply grateful to my doctorate advisors Xavier Serra and Sergi Jordà for their
guidance, unconditional support and most of all for providing a role-model of character
and knowledge, which will always be a reference to me.
For their support, fruitful discussions about my ideas and collaboration in the work
which gave birth to this dissertation, I am also particularly grateful to:
Josep Blat, Alexandro Ramirez, Martin Kaltenbrunner, Günter Geiger,
Alexander Carôt, Fabien Gouyon, Pedro Cano, Diego Dall’Osto,
Perfecto Herrera, Ross Bencina, Marcos Alonso, Rafael Ramirez
(Pompeu Fabra University, Barcelona);
Jorge Cardoso, João Seabra, Paulo Ferreira-Lopes, Guilhermina Castro,
Joana Cunha e Costa, Daniela Coimbra, Luís Gustavo Martins, Carlos
Barreiros, Mafalda Barbosa, Mariana Madaíl and Kurt Stewart
(Portuguese Catholic University, Porto).
For extremely helpful discussions, advices and inspiring ideas related with the main
topics of my dissertation, I am tremendously indebted to my friend Carlos Baquero
Moreno (Minho University, Portugal).
For their acknowledgement and interest in my work I would like to express my
gratitude to Curtis Roads (University of Santa Barbara), Antonio Camurri (University
of Genoa), Chris Chafe (Stanford University), Atau Tanaka (Sony CSL Paris), Dante
Tanzi (University of Milan), Kiyoshi Furukawa (Tokyo National University), Jason
Freeman (Georgia Institute of Technology), Scot Gresham-Lancaster (Cogswell College
Sunnyvale, CA), Suguru Goto (Paris VIII University), Nicolas Collins (Editor of
Leonardo Music Journal) and Leigh Landy (Editor of Organised Sound).
iii

For their help, proof-reading and bearing with me, while writing my thesis, I would like
to thank my friends Carla Almeida, Cristina Sá, Sahra Kunz, Helena Figueiredo,
Cinthia Ruiz and Joana Martins.
I also owe a great debt of thanks to my colleagues at the Portuguese Catholic
University, with whom I’ve shared a rich work experience over the last years: José
Paulo Antunes, Teresa Macedo, Gonçalo Vasconcelos, Maria Lopes Cardoso, Amilcar
Sousa, José Miguel Cadilhe, Miguel Lobo, Carlos Caires, Luis Teixeira, Paulo da
Rosária, Hélder Dias, Adriano Nazareth, Baltazar Torres, Mónica Monteiro, Miguel
Rodrigues and Jaime Neves.
I would like to mention my appreciation to the people that previously trusted and
encouraged me in my professional career: Cónego Ferreira dos Santos, Luis Proença,
Tiago Azevedo Fernandes and Armando Batista.
I would also like to acknowledge that my Doctorate research was funded by the
Portuguese institution Fundação para Ciência e Tecnologia, through the award of a
Doctorate Scholarship (SFRH/BD/5192/2001).


iv

Table of Contents
Abstract……………………………………..…………………………………………... i
Abstract (Catalan) — Resum…………………………………………………………. ii
Acknowledgements…………………………………………………………………….. iii
Table of Contents……………………………………………………………………… v
List of Figures………………………………………………………………………….. x
List of Graphics and Tables……………………………………………………….. xiv

Chapter 1 - Introduction.................................................................. 1
1.1 Motivation................................................................................ 2
1.2 Objective of this Dissertation ................................................... 3
1.3 Structure of this Dissertation ................................................... 4

Chapter 2 - Survey of Computer-Supported Cooperative Work for
Music Applications ............................................................... 5
2.1 Computer-Supported Cooperative Work .................................. 5
2.1.1 Operation Modes in CSCW......................................................................... 7
2.1.2 Synchronous and Asynchronous Modes in CSCW....................................... 8
2.1.3 The CSCW Classification Space................................................................ 10
2.1.4 Shared Virtual Environments .................................................................... 11
v

Figure 20. Diagram for a remote collaborative musical performance for a pianist (using
Yamaha Disklavier Pianos) and live electronics performed on a Laptop. ...... 60
Figure 21. Diagrams from the TransMIDI System showing possible group topologies . 63
Figure 22. Musicians at McGill University (Dan Levitin — sax and Ives Levesque —
trombone) Jamming with remote Musicians at Stanford University projected
on Screen (Alexander Carôt — Bass and Estabin Wilson — sax). ................... 64
Figure 23. Screen-Shot from the Peersynth network synthesizer. ................................ 66
Figure 24. Screen-Shot from the Qintet.net client interface. ........................................ 67
Figure 25. NINJAM client interface ............................................................................ 68
Figure 26. Shared Sonic Environments in the Networked Music Classification Space.. 69
Figure 27. Screenshot from the demonstration video Documentary on the Auracle ..... 74
Figure 28. Graphical Representation of different computational instances in the Co-
Audicle ......................................................................................................... 75
Figure 29. Transmission Control Protocol (TCP) and the User Datagram Protocol
(UDP) .......................................................................................................... 80
Figure 30. Centralized Network Model........................................................................ 83
Figure 31. Centralized Network Model with Multiple Communicating Servers............ 83
Figure 32. Distributed Network Models ...................................................................... 84
Figure 33. Group Abstraction for Broadcast Transmission.......................................... 85
Figure 34. Centralized Shared Sonic Environment Model with Local Feed-Back......... 89
Figure 35. Centralized Sonic Shared Environment Model with Local Feed-Back......... 90
Figure 36. Traditional Music Instrument Interaction Model........................................ 94
Figure 37. Violin Interaction Model ............................................................................ 94
Figure 38. Virtual Music Instrument Interaction Model .............................................. 95
Figure 39. Nomadic Virtual Music Instrument Model ................................................. 97
Figure 40. Oskar Fischinger’s device for producing light effects. ................................. 99
Figure 41. Multi-Touch Sensing through Frustrated Total Internal Reflection...........100
Figure 42. A Compaq IPAQ Running a PD Patch, Gunter Geiger performing with a
PDA ............................................................................................................101
Figure 43. The Public Sound Objects Client Interface running on a PDA..................102
Figure 44. The ReacTable* architecture (illustration by Ross Bencina) .....................103
Figure 45. “TeleSon” Performance September 04, 2005: Chris Brown and Gunter Geiger
at ICMC 2005 in Barcelona, Spain (on stage ate SGAE auditorium); Martin
Kaltenbrunner and Marcos Alonso at Ars Electronica Festival in Linz, Austria
(on screen)...................................................................................................104
xi

Figure 73. Locations of the PSO performance on March the 31st of 2005 ..................158
Figure 74. João Seabra, Jorge Cardoso and Álvaro Barbosa performing simultaneously
with PSOs respectively in Toronto, Porto and Barcelona ............................158
Figure 75. PSOs Installation at NIME 2005 — New Interfaces for Musical Expression
Conference, 26-28 of May Vancouver, Canada. ............................................159
Figure 76. PSOs Installation at ICMC 2005 — International Computer Music
Conference, 5-9 of September Barcelona, Spain. .........................................159
Figure 77. PSOs Trial Installation at Porto School of Arts, 7-14 October 2004..........160

xiii

Chapter 1. Introduction

1.3 Structure of this Dissertation
This doctorate work followed a methodology that departed from a contextualization
and survey of the field, followed by a definition of concepts and ideas which represent
advances in the field and concluded by a test-proof implementation of a software
prototype and respective evaluation.
Three distinct parts can be defined within the main contents of this dissertation:
Part I (Chapter 2)
Systematic study and classification of state of the art systems for computer-supported
cooperative work, with particular emphasis on geographically displaced musical
practices.
Part II (Chapters 3 and 4)
Detailed analysis and discussion of experimental proposals, concepts and methods,
based upon the contextualization studies.
Part III (Chapter 5)
Discussion about the implementation and evaluation of a proof-of-concept software
prototype entitled Public Sound Objects.



4

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

conferences were both technically and financially healthy, and they would only require
the oversight of ACM and other SIGs.
The term CSCW was introduced by the computer scientists Irene Greif of
Massachusetts Institute of Technology (MIT) and Paul Cashman of Digital in the early
eighties.
According to the article written in 1992 by Liam J. Bannon’s about CSCW (Bannon, J.,
1992), Cashman and Greif came up with the term Computer-Supported Cooperative
Work to describe the object of interest for a small workshop they organized in
Massachusetts (August 1984), concerning the development of computer systems that
would support people in their work activities. The workshop brought together people
from different areas, such as office information systems, hypertext and computer-
mediated communication.
Since the first Computer-Supported Collaborative Work conference organized in
December 1986 in Austin, Texas, the enthusiasm for the topic continued to grow until
today, with an increasing activity in the research and development of systems and
applications as well as the publication of research work.
Besides the major CSCW conferences, in recent years there have been a number of
CSCW-related conferences and workshops on collaboration technology, group decision
support systems and multi-user systems both in Europe and North America, and in
addition, several journals include CSCW in their list of topics.
Finding a commonly accepted definition of CSCW and its scope has been a difficult
task mostly due to the multidisciplinary nature of the field which brings together
people across a range of different backgrounds like computer science, psychology,
sociology, organizational theory, and anthropology, just to mention a few.
A very general notion can be defined in the sense that “it is focused on the design of
computer-based technologies with explicit concern for the socially organized practices of
their intended users” (Suchman, L., 1989) (Bannon, J., 1994), but when designing
specific applications to suit the specificities of certain work contexts, major differences
between the resulting solutions might come along.
6

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

It is therefore not surprising that different tendencies emerged in the field. On one
hand, there are groups focused on modeling and designing office communication
systems. On the other hand there are those interested in developing a richer
understanding of cooperative work practices.
This last approach is also followed in even more specific areas of interest, oriented
towards general artistic creation and in particular to musical and sonic expression.
2.1.1 Operation Modes in CSCW
In general a group of users operating in joint projects will follow a specific approach to
their contribution according to the project requirements or technological constrains of
the system being used.
In a 1991 survey of CSCW systems (Rodden, T., 1991), Tom Rodden presents a
systematic approach defining different operation modes in CSCW starting by
classifying existing systems/applications and identifying their functional roles:
Message systems — In these systems the users operate in Transference mode
interchanging information documents, but the work development is done individually,
without a common sense of the global information structure (e.g. e-mail based systems).
Computer Conferencing — Information regarding a certain topic is broadcast towards an
interested community. All the users participate and cooperate at the same level in the
joint event and information is normally held with conference messages within one
central database rather than the individual mailbox approach used in messaging
systems. The development of reliable high speed communications has led to the
emergence of new real-time conferencing systems, allowing conference members to
communicate in real-time, and enhanced the scope and power of this class of
applications (e.g. video/audio conference; news groups).
Coordination Systems — Addresses the problem of integrating and adjusting in a
harmonious fashion the synergies of a group of people working together in the same
physical space, by introducing the support of computer systems (e.g. white boards;
7

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

automated meeting rooms with a network structure to support voting systems; multi-
user software based on analytical decision techniques).
Co-Authoring Systems — General class of systems that supports the co-authoring of a
product, designed to address the specificities and requirements of the product following
a structured development of the content (e.g. Systems for joint development of software,
with characteristics like maintaining up-to-date versions of the code produced by each
project member and integration mechanisms for their partial contributions).
Even though this classification is quite accurate in the context of office communication
systems, when considering an application out of this scope, like a scenario where a
computer network is used for a collective artistic creation, it is not totally applicable3.
On the other hand, Rodden also presents in his survey more general characteristics
common to CSCW systems, which could be considered as the environmental facets of
cooperative work. The geographical nature of the user is considered as the Space
Dimension environmental facets and can be “Remote” or “Co-located”.
The form of interaction provides the Time Dimension environmental facets, and it can
be “Synchronous” or “Asynchronous”.
The aspects of synchronicity are extremely relevant when characterizing the operation
mode of a joint system and it is a topic that requires special attention.
2.1.2 Synchronous and Asynchronous Modes in CSCW
The users operation in a joint project can be performed in a synchronous or
asynchronous mode.
In the synchronous mode, all the participants are active simultaneously on the common
document.

3 The classification of different systems for collective artistic creation in the context of music
and sonic arts is discussed in this chapter in section 2.4.
8

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

A more extensive and detailed classification of web based collaborative systems was
published in 2001 by Georgia Bafoutsou and Gregory Mentzas from the National
Technical University of Athens, Greece at the 12th International Conference on
Database and Expert Systems Applications (Bafoutsou, G. and Mentzas, G., 2001).
2.1.4 Shared Virtual Environments
Shared Virtual Environments (SVEs) go beyond the typical CSCW applications
mentioned so far. These systems are not conceived with the main purpose of
maximizing the synergies of a group in order to achieve better and faster results in the
common tasks performed by the users.
In a SVE it is intended to create a shared space in a computer network, inside which
users can achieve a certain degree of immersion and flexibility in their behavior. This
virtual space provides a context in which the interaction outcome is somehow unique
and highly influenced by the characteristics of this media. Even though SVEs are more
about simulations and experimentation than improving the efficiency of specific tasks,
the flexibility and scalability of these systems often have proven to be effective when
applied to contexts like education support or conferencing.
The pioneer internet systems that convey the essence of a virtual community space
started of with the original Multiple-User Domain/Dungeon (MUD) software developed
in the early 80’s by Richard Bartle and Roy Trubshaw at the University of Essex in
England. A MUD is a real-time structured textual chat forum. It has multiple
“locations” like an adventure game, and may include combat, traps, puzzles, magic, a
simple economic system, and the capability for characters to build more structure onto
the database that represents the existing world.
In the late 80’s students on the European academic networks quickly improved the
MUD concept, creating several new MUDs (VAXMUD, AberMUD, LPMUD). In many
of these systems research was done to include bulletin-boards and social interaction
mechanisms which added academic value to the projects. This, along with the fact that
Usenet feeds were often unreliable and difficult to get in the U.K., made the MUDs a
major form of social interaction in the online community at this time.
11

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

computer graphics, like the Habbo Hotel 8 launched in January 2001, tend to have
excellent acceptance amongst the online-community.
2.2 Computer Mediated Communication and
Networked Music
“What I want to say about Networked Music in general is that All
Music Is Networked. You can think about an Orchestra as client-
server network, where a conductor is “serving” visual information to
the “client” musicians, or a peer-to-peer networking model in an
improvising Jazz Combo, where there is no one directing, and the
musicians are all interacting, so, any performance context we can
think of in some way there is a network connecting the performers
(…). Networked Music with capital N and capital M (the kind we
are talking about) is about performance situations where traditional
aural and visual connections between participants are augmented,
mediated or replaced by electronically-controlled connections.”
(From Jason Freeman’s lecture opening at the 1st Networked Music
Workshop during the International Computer Music Conference
2005)
From his definition of Networked Music Jason Freeman clearly illustrates why
collaborative music paradigms can easily be approached in the context of computer
mediated communication.

8 The Habbo Hotel (http://www.habbohotel.com/) Habbo Hotel is a graphical chat environment
for the Internet, holding a community of nearly two million members. Built in Macromedia
Shockwave, the website takes the form of a virtual 3D hotel displayed in 2D Graphics.
14

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

It is basically supported by communication processes of continuity. These spaces are
currently created by means of a diverse set of technological possibilities previously
discussed in section 2.1 of this chapter.
Virtual Communities are normally formed on the basis of a specific interest shared by
its members, which is the reason for their affiliation. Each person thus gains a sense of
affinity with the rest of the community members as well as a sense of collective identity
that distinguishes him from non-members. So, more than the majority of communities
established through face-to-face interaction, virtual communities result from a conscious
choice of each member to participate. A final inherent characteristic of virtual
communities is the fact that individuals are creators and not consumers. Virtual
Communities’ members often are highly specialized in the specific subject conveyed
even in the most common community activities (Bogazzi, R. P. and Dholakia, U. M.,
2002).
Nevertheless, Virtual Communities also present differentiated characteristics amongst
them:
ƒ Asynchronous communication (as in mailing lists), synchronous (as in chat-
rooms) or both;
ƒ Verbal language (still the majority of the cases), visual language (static or
moving) or acoustic language;
ƒ Open participation or participation subject to pre-defined conditions;
ƒ Functional (useful to the lives of its participants) or hedonistic goals (based on
the pleasure derived from communication and content creation).
One can also consider the level of the community’s internal structure.
ƒ A high structural level creates in the communities a strong interdependence
among its members, originating phenomena typical of the processes of group
formation: the development of sanction norms and its mechanisms, affective
bonds or group identity.
16

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

In 2003 during the International Computer Music Conference, held at the National
University of Singapore, some awareness was raised about this type of research
burgeoning to become one of the acknowledged topics of Music Technology.
In his Keynote Speech from ICMC 2003 Roger Dannenberg mentioned “Networked
Music” as one of the promising research topics and at least four papers (Barbosa, A.,
Kaltenbrunner, M. and Geiger, G., 2003), (Stelkens, J., 2003), (Hajdu, G., 2003) and
(Obu, Y., Kato, T. and Yonekura, T., 2003) were centered on this topic, even though
they were scattered over different panels, instead of one distinct session.
Since then the term Networked Music has become increasingly consensual in defining
the area, and according to Jason Freeman’s definition: it is about music practice
situations where traditional aural and visual connections between participants are
augmented, mediated or replaced by electronically-controlled connections.
2.2.3.1 Landmarks in Networked Music Research
In order to have a broad view over the scientific dissemination of Networked Music
research I present the following Landmarks in the field over the last six years (1999-
2005):
(1) The ANET Summit; (2) The Networked Music Workshop at ICMC; (3) Four
published Doctorate Dissertations; (4) Six surveys and partial overviews published in
journals about Networked Music; (5) A dedicated issue to Networked Music from
Cambridge Press’ Organised Sound Journal.
(1) The ANET Summit (August 20-24, 2004)
The summit was organized by Stanford University’s Center for Computer Research in
Music and Acoustics (CCRMA) and held at the Banff Center in Canada, was the first
Workshop event addressing the topic of High quality Audio over Computer Networks.
The guest lecturers were Chris Chafe, Jeremy Cooperstock, Theresa Leonard, Bob
Moses and Wieslaw Woszczyk.
The Workshop Syllabus stated:
25

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

“This three-day summit is an exploration of the state-of-the-art in
ethernet-based professional audio networks. Developers, engineers,
musicians and others interested in the growing practice of high-
resolution audio over ethernets will gather to focus on the new
technology. The scope includes IP-based systems and systems with
dedicated protocols.
A 1998 AES whitepaper on "Networking Audio and Music Using
Internet2 and Next-Generation Internet Capabilities" expressed a
vision of the future and challenges that lay ahead. Six years later,
with technical developments continuing, musical collaborations of
various kinds have been tested and the Internet has evolved.
Predicted application areas which are now taking off include audio
production, music education, broadening musical participation, and
scientific and engineering data representation (sonification). The
summit offers an opportunity to compare today's reality with what
was foreseen and to look ahead to what's next.
The summit is a "neck-ties removed" working group that brings
together academic and commercial interests, developers and users,
audio specialists and network engineers. The program includes
hands-on demonstrations in The Banff Centre's concert and
recording facilities, a "how-to" covering representative open-source
software-based systems, demos of products, presentations, a tutorial,
and a panel discussion.
Continued topics from the 1998 vision of audio over next-generation
networks include current and future quality of service, implications
of end-to-end design, cost and complexity of bridge devices, formats
and adherence to audio industry standards, and scalability
requirements. New topics will include but are not limited to Internet
signal processing, User studies, and new artistic forms.” (Syllabus
from the ANET Summit)
26

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

expression, as well as systems for practicing musicians to extend the
boundaries of performance.
After a broad overview of historical projects, like from the work of
the League of Automatic Music Composer and the HUB from the
1970's and 1980's and other important approaches to networked
music, the workshop will focus on case studies of particular projects
and the tools they use. Ross Bencina's network transport
infrastructure, OSCgroups, which uses a centralized name lookup
server and peer-to-peer data interchange via Open Sound Control,
will be discussed in connection with recent work by Scot Gresham-
Lancaster and "The Hub." Phil Burk's TransJam, a Java-based
server for real-time collaboration over the Internet, will be explored
in connection with the Auracle networked sound instrument.
Specific issues to be addressed include the logistics of event
coordination, the mediation between transparency and complexity in
the system, the handling of timing and latency issues, human
interface design, and the maintenance and monitoring of client
reliability.” (Syllabus from the Networked Music Workshop at
ICMC 2005)
(3) Published Doctorate Dissertations
ƒ 2002 Golo Föllmer “Musikmachen im Netz Elektronische, ästhetische und soziale
Strukturen einer partizipativen Musik (Making Music on the Net, social and
aesthetic structures in participative music)” — Martin Luther Universität Halle-
Wittenberg — Germany.
ƒ 2002 Nathan Schuett “The Effects of Latency on Ensemble Performance” —
Stanford University, California — USA.
ƒ 2003 Jörg Stelkens “Netzwerk-Synthesizer (Network Synthesizer)” — Ludwig
Maximilians Universität, München — Germany.
28

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

ƒ 2003 Gil Weinberg “Interconnected Musical Networks — Bringing Expression
and Thoughtfulness to Collaborative Music Making” - Massachusetts Institute of
Technology, Massachusetts — USA.
(4) Surveys and partial overviews published in journals about Networked Music
ƒ 1999 Sergi Jordà, “Faust Music On Line (FMOL): An approach to Real-time
Collective Composition on the Internet”, Leonardo Music Journal, Volume 9,
pp.5-12.
ƒ 2001 Dante Tanzi, “Observations about Music and Decentralized Environments”,
Leonardo Music Journal, Volume 34, Issue 5, pp.431-436.
ƒ 2002 Gil Weinberg, “The Aesthetics, History, and Future Challenges of
Interconnected Music Networks”, Proceedings of the International Computer
Music Conference, pp.349-356.
ƒ 2003, Álvaro Barbosa, “Displaced Soundscapes: A Survey of Network Systems
for Music and Sonic Art Creation”, Leonardo Music Journal, Volume 13, Issue
1, pp.53-59.
ƒ 2005, Gil Weinberg, “Interconnected Musical Networks: Toward a Theoretical
Framework”, Computer Music Journal, Vol. 29, Issue 2, pp.23-29.
ƒ 2005, Peter Traub, “Sounding the Net: Recent Sonic Works for the Internet and
Computer Networks”, Contemporary Music Review, Vol. 24, No. 6, December
2005, pp. 459 — 481.
(5) Organised Sound Journal Volume 10, Number 3 (December 2005) — Dedicated Issue
to topic of Networked Music
This Issue of Organised Sound, edited by Leigh Landy, is entirely dedicated to
Networked Music.
The Call for contributions stated:
“Interconnection has always been a fundamental principle of music,
prompting experimental artists to explore the implications of linking
29

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

2.3 Systematic Study of Networked Music
Systems
Over the last decades artists have used cutting edge computer technology to maximize
the aesthetics and conceptual value of their work, not only by enhancing the way they
traditionally create, but also by using technology as a media in itself to express
meaningful artistic work.
The idea of using computer networks as an element in collective artistic creation and
performance (or when both come together in improvisation) was no exception, since it
provides a particularly engaging opportunity to achieve stylistic and conceptual
originality.
On the other hand, identifying the influence of aesthetic and conceptual values in
specific techniques related with the media or technology with which an art piece is
created is often a hard and unclear task.
The Berlin based writer Florian Cramer, who has published in the area of code poetry,
comparative studies in literature and art, refers to this precise task in the context of
Combinatory Poetry and Literature on the Internet.
“(…)Although it is difficult to distinguish a combinatory literature
from other forms of literature ever since linguistics defined language
as a combinatory system itself, combinatory poetry nevertheless
could be formally defined as a literature that openly exposes and
addresses its combinatorics by changing and permuting its text
according to fixed rules (…)” (Cramer, F., 2000)
One could equally argue that it is also hard to differentiate the artistic influence of a
computer-supported collaborative system in the music creation process, but since
collaboration in itself is part of the traditional music language, this new form of
communication will influence the paradigm according to new rules.
These new rules will not only be based on the new possibilities of geographical
displacement and asynchronous collaboration, but they will also be strongly dependent
31

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

on technological constrains, thus defining a different stylistic and conceptual way to
create music.
In this section we will discuss many of the systems and ideas that emerged in the
context of Music and Sonic Arts collaboration over computer networks.
These systems somehow adapted to the rules imposed by the new media even if
developers were not always totally aware of this fact, since most of the times they use
methods of development based on a project oriented perspective, and not towards
systematic research methods.
As we will see in practical terms theses rules are dependent on the type of system and
to what extent it uses the possibilities presented by large or small scale computer
networks.
2.3.1 Early Experiments with Musical Networks
The idea of the communication media influencing musical practice is by no means new
nor bounded to technology.
One of the most remarkable examples in western music of the media’s influence in
music performance leading to stylistic novelty is the Venetian polychoral music style.
This is a type of music of the late Renaissance and early Baroque eras which involved
spatially separate choirs singing in alternation. It represented a major stylistic shift
from the prevailing polyphonic writing of the middle Renaissance, and was one of the
major stylistic developments which led directly to the formation of what we now know
as the Baroque style.
The style arose from the architectural peculiarities of the imposing Basilica San Marco
di Venezia in Italy. Aware of the sound delay caused by the distance between opposing
choir lofts, composers began to take advantage of that as a useful special effect. Since it
was difficult to get widely separated choirs to sing the same music simultaneously
(especially before modern techniques of conducting were developed), composers such as
Adrian Willaert, the maestro di cappella of St. Mark's in the 1540s, solved the problem
by writing antiphonal music where opposing choirs would sing successive, often
32

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

texture with delicate rhythmic patterns, resulting in an evolving rhythmic experience
dependent on the way each musician relates to the orchestra (Perlman, M., 2004).
Even though this notable examples show us that some of the present ideas in Music are
inherited from a rich past culture, it was with the advent of electronics and computer
technology that the concept of an Interconnected Music Network (IMNs) was taken
further, allowing multiple ways of crossed control between performers and instruments.
An early example of group communication using electronic technology in the field of
performing arts is John Cage’s 1951 famous “Imaginary Landscapes No.4” for twelve
radios played by 24 performers (Cage, J., 1961).
In this piece Cage unleashed the expressive potential of technology to enhance acoustic
group interdependency by using the then recently invented commercial transistor radio
as a musical instrument providing a sonic medium for collaboration, procedures and
rules in his piece. The composition score indicated the exact tuning and volume settings
for each performer but with no foreknowledge of what might be broadcast at any
specific time, or whether a station even existed at any given dial setting.
The explorations of the transistor radio as an infrastructure for collaboration opened
the door for other explorations with the electronic media, which were not necessarily
based on external sound production.
In the late 1970’s the commercialization of personal computers in the United States,
allowing fine tune network topologies, enabled the first groups of experimental
musicians to create musical computer networks at a local area scale.
In the mid-1970s, from the San Francisco Bay Area, emerged the first ensemble to
investigate the unique potentials of computer networks as a medium for musical
composition and performance entitled “The League of Automatic Music Composers”
(Brown, C. and Bischoff, J., 2005) (Bischoff, J., Gold, R. and Horton, J., 1978)
(Chadabe, J., 1997).
Originaly the “League” came together through the mutual interest of Jim Horton, John
Bishoff and Rich Gold, naming their new genre of musical performance “Network
Computer Music”.
34

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

This Field Map of Networked Music Systems is centered on criteria based on how
computer mediation facilitates Musical Interconnection. Therefore, some of the Systems
included in the same category might have different approaches in regard to its
architecture, functionality or other musical aspects, such as, adequacy for performance
or composition practices.
For example the FMOL System (Jordà, S. and Aguilar, T., 1998) and the Webdrum
System (Burk, P., 2000b) are both included in the “Construction Kit Approach”
(Weinberg, G., 2005), even though they are totally different systems in terms of the
music creation process.
FMOL is an asynchronous software oriented towards iterative off-line composition (in
the iterative database repository usage of this system real-time processing only occurs
in the off-line client side), while Webdrum is a Synchronous system oriented towards
online jamming and performance (real-time collaboration on the web).
In fact, there are many possible approaches when proposing a systematic field
classification and its result will always depend on what information is intended for the
reader to extract from this categorization.
One interesting proposal for a high level categorization of Networked Music Systems
was suggested by Scot Gresham-Lancaster during an extensive exchange of e-mails
between the lecturers of the 1st Networked Music Workshop held during ICMC 2005 in
Barcelona (Ross Bencina, Jason Freeman, Álvaro Barbosa and Scot Gresham-
Lancaster).
“(…) there are two distinct streams here. One that is regarding
network music that is intended for a general and possibly non-
musician user, and then techniques for practicing musicians to
extend the boundaries of performance. (…) they need to be
addressed within their separate contexts.” (From an e-mail by Scot
G-L June, 17th of 2005)
This proposal contemplates only two categories of systems: (1) Systems and techniques
for practicing musicians; (2) Systems for a general and possibly non-musician users.
Even though each category covers a very broad spectrum of different features in terms
39

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

equivalent to a system able to create several instances of the same instrument allowing
different users to play together.
It is clear though, that live performance system, which allow players to influence, share,
and shape each other’s music in real-time or synchronously, are based on a Multi-user
Musical Instrument.
In this case, a high degree of interdependence between performances is expected in
order to achieve virtuous results, and therefore real-time communication requirements
are a critical point. This is the main reason why this approach has been constrained to
Local Area Networks, where communication latency allows real-time immediate
connections.
2.4.2 Music Composition Support System
LOCATION
L o c a l Remote
Music Composition
Support Systems

Figure 8. Music Composition Support Systems in Networked Music Classification Space
The primary function that emerged from the use of internet technology in the musical
context was to provide mechanisms that assist the composition of music pieces by
means of network communication.
Composing music by two or more authors is a process that traditionally can be
accomplished in different ways.
Conventionally, to compose a musical piece, a composer individually conceives the
music and registers his ideas with symbolic musical notation — a score — (usually in
44

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

D., 2005) and these two forms of composing music, in fact, correspond to High Level
Models proposed by these authors.
The Endogenous creative trajectory correspond to the traditional composition approach
mentioned first at section 2.4.2 and the Exogenous creative trajectory corresponds to
compositional approach often practiced in recording studio environments.

Figure 9. Draft Model of Endogenous and Exogenous Creative Trajectory
For the majority of internet users interested in creating music this process is inevitably
more engaging and from this observation a new class of internet applications emerged
for music creation aiming to materialize the idea of an on-line Recording Studio.
To take advantage of internet’s global communication possibilities in this context, new
systems came up based on the idea of collaboration between geographical displaced
users in one common project developed in a virtual studio environment.
As a concept these are distributed systems, however there also exists a centralized
server that is part of the system and that manages the organization of users into
multiple session groups.
Typically the interface layer resembles typical multi-track software, like Digidesign’s
Protools, Steinberg’s Cubase or Nuendo and allows the users to lay down tracks of
MIDI and digital audio either in a synchronous or asynchronous mode collaborating
with other users that have access to the session.
A pioneer system which followed this approach was the ResRocket Surfer. It was a
successfully freely distributed application, released in 1994 (Moller, M. and others,
1994), with a reasonable amount of users forming a community of musicians that
actually created music cooperatively over the Internet. The system allowed performers
46

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

long distance collaboration on digital non editing recording software packages, being
currently supported by: Digidesign’s Protools, Emagic’s Logic Audio and Steinberg’s
Cubase VST. Despite the relative success in the early days of Internet global usage, the
Rocket Networks Company unexpectedly ceased activities in 2000.
In the following year a similar system entitled TONOS-TC8 was introduced as a
commercial Application. Software (TONOS Company, 2001).

Figure 12. TONOS-TC8 Interface
TONOS TC-8 followed the same principles of the basic ResRocket application, but it
was restricted to 8 tracks of audio and MIDI recording, allowing users to access a
centralized account with 40Mbytes of Hard-Disk Space. Similarly to Rocket Networks
this company suddenly ceased activity in 2004.
Even though these systems did not make it as viable commercial products, the basic
principles behind their implementation are still very attractive to Popular Music
Recording Professionals. Therefore, it is not surprising that in 2005 two new systems
based on this paradigm were released.
In April 2005, during the Frankfurt Musikmesse (International Trade Fair for Musical
Hardware and Software) the German based company DigitalMusician.Net (DMN)17 ,
presented a software prototype that offers similar features to those of the previously

17 DMN is available from http://www.digitalmusician.net/
48

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

How do we do it? eJamming™ algorithms delay the sounding of your
instrument until you receive data from your fellow eJammers. So
from the time you hit your keyboard, strum a guitar string or strike
a drum skin, the time it takes to hear that note and those of the
other players on your stage varies from 15mS (milliseconds) within
a city, 25-40mS within a 1500 mile jam and 40-50mS cross country.
eJamming™ has even connected musicians from NYC to London at
49mS (…)”

Figure 14. eJaming Stage interface
All these systems are suited for a Music Studio Composition Process, since they have a
common option of Multi-Track Recording incorporated with long distance audio
communication.
In this experimental composition process composers are often required to develop some
sort of written notation to automate the edition and sequences (effects, samples, loops,
etc). This practice has the disadvantage of being less universal, since either it is a
personal technique developed by experience in studio technology, or a proprietary form
of notation imposed by hardware and software manufacturers.
50

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

Attending to this requirement in September 2005 another commercial product was
released. The VSTunnel 19 Software is a VST Plugging (compatible with any audio
production software that supports VSTs) used like an insert effect in a sequencer's
master out channel, which allows an audio connection over the Internet to other
VSTunnel enabled clients.

Figure 15. The VST-Tunnel Plugging Interface
2.4.2.2 Experimental Collective Composition Systems
All previous systems mentioned in the category of Music Composition Support Systems
were designed to address a very specific target audience, practicing musicians and
music producers which have preceding knowledge in working with Digital Non-linear
Multi-Track recoding software. However, with the Internet also came a sense of
democratization in information access, which empowered both regular and expert users
to create and participate in experimental music composition practices through systems
specifically designed for this purpose.
The earliest examples of On-Line Composition Environment, designed for collective
music creation by non-practicing musicians came up in the late 90s.

19 VSTunnel is available from http://www.vstunnel.com/
51

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

Bamboo was designed bearing in mind that its control can be fully mastered, visually
resembling a rectangular web, where the horizontal lines correspond to the sound
generators, and the vertical lines to the processors. It behaves like a guitar or a harp,
as its strings can be plucked or fretted with the mouse, and it also behaves like a multi-
channel oscilloscope, since every vertical string continuously draws the sound it is
generating.
In Golo Föllmer’s on-line essay “Soft Music” (Föllmer, G., 2001), Sergi Jordà refers to
the musical ideas that led him through the process of programming FMOL:
“(…) I keep changing them while I write them and they always
surprise me, and the more they surprise me the more I like them.
That’s why I like FMOL. It created a musical style that I didn’t
know before I started the program (…)”. (Sergi Jordá from a Video
Interview in Crossfade, December 2000)
As a live performance instrument FMOL has been used in events by the FMOL Trio in
a setup based on two FMOLs, a saxophone player and in many cases other invited
musicians. The FMOL experience as a performance and improvisation instrument both
in interdependent communication with other traditional instruments or with other
FMOL instruments has resulted in refined and exciting electronic music pieces (Feller,
R., 2002).
Another requirement of La Fura dels Baus’ commission to Sergi Jordà was that the
system would allow collective participation of internet users in composition of pieces
with the FMOL software which would later be included in La Fura’s play F@ust 3.0
and in fragments of the multimedia opera Don Quijote en Barcelona, which premiered
at the Gran Teatre del Liceu of Barcelona in October 2000. The original system was
built following a client server model, allowing composers using the FMOL client
software to log into a central web based server, in order to download pieces stored in a
song tree-structure database.
For the collective interaction model a "vertical" approach was preferred instead of a
more typical sequential approach which would consist of pasting small fragments one
after the other.
53

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

composers, performers and audience, and inviting everyone visiting the site to be a
creative participant.
The components of this interactive paradigm were the Web site, PitchWeb, and an
Internet Band. The Web site features a variety of interactive musical, artistic, and
Text-Based experiences; the PitchWeb allows listeners to participate actively and
creatively; and the Cathedral Band, which gives periodic live performances and offers
listeners focused moments in which to come together and play music in community on-
line (Duckworth, W., 2005). The system has been maintained und technologically
updated until this day 20.
One other possible scenario in the context of collective composition of music by
communities of users is the case that existing communities with focused interests in
Audio and Music end up stimulating mechanisms of interaction to enhance their
communication paradigm leading to compositional environments.
This is the case of the Free-Sound Project21, created in 2004 by Bram Jhong at the
Music Technology Group of the Pompeu Fabra University in Barcelona, with the
purpose of serving as groundwork for the International Computer Music Conference
2005, which was dedicated to the Free Sound Theme. The project is a collaborative
database of Creative Common22 licensed sounds, with several mechanisms of sound
Surfing, Downloading and Uploading. It became rapidly successful, leading to further
experimentation in terms of the available features on the web site, which resulted in a
particular collaborative mechanism, which became a collective sonic composition tool,
the Remix! Tree.

20 The present version of the Cathedral in 2005 is heavily based on Macromedia Flash
technology. It is available from http://www.monroestreet.com/Cathedral/
21 The Fee-Sound Project is available from http://freesound.iua.upf.edu/
22 Creative Commons is a nonprofit organization offering flexible copyright for creative work.
(http://creativecommons.org/)
55

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

The basic idea is very similar to the previously mentioned iterative database model
from FMOL. When the users add a sample which is a remix of another sample, it will
appear in a tree structure. Remixed samples appear as branches in the tree.

Figure 18. The Free Sound Project Remix! Tree Interface
These systems provide effective enhancements in the process of music production. Yet,
they are mostly oriented towards composition perspective, leaving little space for more
experimental forms of performative Arts, and thus constraining the potential of what
the Internet can offer as a medium for artistic expression in itself.
56

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

2.4.3 Remote Music Performance Systems
LOCATION
L o c a l Remote
Remote Music
Performance System

Figure 19. Remote Music Performance Systems in the Networked Music Classification Space
Many examples can be found in which the Internet’s potential is explored in order to
provide a connection between physical spaces geographically apart.
Nonetheless, there are crucial differences in terms of the complexity of a working
system that serves a Broadcast or Unicast scenario in which it is only intended to
provide one-way Tele-Presence, or a Multicast23 communication setup that links two,
or more, Collaborative Performative Spaces.
2.4.3.1 Tele-Presence Systems
The idea of having the presence of one or more remote performers from anywhere in
the world in events taking place in physical spaces, facing live audiences, during fixed
periods of time is an exciting one.
Of course considerations must be made in terms of when to present public events that
occur simultaneously in different places at a global level. If for example a concert would
be presented publicly at the East Coast of the United States during the afternoon, it is

23 The concepts of Unicast, Broadcast and Multicast are discussed in chapter 3
57

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

unlikely that it would be possible to associate this event with another public
presentation taking place in Europe, since it would occur in the middle of the night.
Such an event is also very complex in terms of bringing together all the required
logistics, since it deals with sensitive, and many times experimental, technology that
must comply and work together in distinct sites, separated by long distances. One
other concern is network bandwidth, which might be an impediment for a continuous
data deliver at the client side.
Different approaches were taken to address this issue in remote performances projects
realized over the last few years.
One approach has been to use cutting edge communication technology, like high speed
and broadband networks combined with streaming technology.
Over the last few years, leading research in this field was conducted by Jeremy R.
Cooperstock at the Centre for Interdisciplinary Research in Music Media and
Technology (CIRMMT)24 from McGill University in Montreal, Canada and by Chris
Chafe at the Center for Computer Research in Music and Acoustics (CCRMA)
SoundWIRE25 Group from Stanford University in San Francisco, USA.
Following the release of 1998 Audio Engineering Society (AES) whitepaper on
"Networking Audio and Music Using Internet2 and Next-Generation Internet
Capabilities" (Bargar, R. and others, 1998) 26, in collaboration with McGill University
the first landmark in Broadband Internet Audio Streaming was realized in September
26, 1999. A musical performance at McGill University in Montreal, was transmitted

24 The CIRMMT Group: http://www.music.mcgill.ca/cirmmt/
25 The CCRMA SoundWIRE Group: http://ccrma.stanford.edu/groups/soundwire/
26 Internet 2 (http://www.internet2.org) is a consortium being led by 200 universities working in
partnership with industry and government to develop and deploy advanced network applications
and technologies, accelerating the creation of tomorrow's Internet. Internet2 is recreating the
partnership among academia, industry and government that fostered today´s Internet in its
infancy.
58

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

This project implied a collaborative performance between Fundamenta Nuova Theatre
in Venice, Italy and Auditório Ilídio Pinho at the Portuguese Catholic University in
Porto, Portugal.
The Performance required high-end Musical Instruments (Yamaha Disklavier, MIDI
controllable, Piano), video synthesis and a Virtual Music Instrument Setup. It shows
that in this class of Remote Music performance Systems the logistics, hardware
requirements and the overall complexity of getting everything to work properly is not
of a simplistic nature.
The 1987 “Clockworktower Concert” by The Hub28 is unquestionably one of the earlier,
if not the earliest, examples of a geographically displaced collaborative music
performance with computer technology (Brown, C. and Bischoff, J., 2005), and
throughout the decade similar experiments were carried out, but unfortunately few
were documented or disseminated in the Music Technology community.
Some known examples of the Collaborative Performance approach are29:
Eve Schooler: Distributed Music: A Foray into Net Performance (Sept. 1993):
Synchronized three real-time streams from different hosts; delays in the order of 200 ms
difficult for performers to be listeners.
Paul Hoffert: CyberSoiree (Feb. 1996): ATM-based technology for audio and video
streaming of a four-way jazz performance. Delay>0.5s delay but musicians learned to
compensate through extensive practice.
Dimitri Konstantas: Distributed Musical Rehearsal Studio (May 1996): ATM based
distributed rehearsal with conductor at different location from musicians. 80 ms one-
way delay for audio-video synch; echo resulted in "extreme confusion" (Konstantas, D.
and others, 1997) (Konstantas, D., 1998).

28 This concert is briefly referenced in section 2.3.1 of this chapter
29 From Jeremy Cooperstock’s notes on Internet Audio Landmarks — ANET Summit
Presentation, 2004.
61

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

Seiji Ozawa: Opening Ceremony Nagano Winter Olympics (1998): Conduct choruses on
5 continents: singers in Sydney, New York, Beijing, Berlin, False Bay. Time lag
adjustor used to eliminate satellite delay.
Still in 1998, during an interview for the Computer Music Journal with the Sensorband
ensemble , Zbigniew Karkowski refers to the group’s extensive experience in
collaborative performances over the internet (using ISDN connections) which they
called Network Concerts conveying the fundamental artistic issues raised by this set-up:
“Another artistic aspect of ISDN concerts is the idea of control.
Very often, composers use computers to achieve greater control. We
have found, after playing several concerts like this, that we could
never control the output 100 percent. Aspects like the delay become
unknown variables, which is interesting (…)” (Bongers, B., 1998)
In the same interview Atau Tanaka adds that:
“As artists, our first instinct is not to make technical improvements
to the system, but rather, to manipulate the technology in a creative
manner. The technical limitations become characteristics of the
composition. Doing this allows us not to be so worried about
transmission delay, rather, to be concerned about the general notion
of distance (…)” (Bongers, B., 1998)
The SensorBand concerts were based on synchronous collaboration in a peer to peer
model between two performers. Yet, other experimental systems focused on the idea of
having several synchronous performances as close as possible to a real-time situation.
This is the case of the 1998 TransMIDI system (Gang, D. and others, 1997),
implemented using the Transis multicast group communication layer for CSCW
applications (Amir, Y. and others, 1992). This system allows musical performers (and
listeners) who wish to play together to organize into multiple session groups.
62

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications


Figure 21. Diagrams from the TransMIDI System showing possible group topologies
Users interact in a synchronous mode (close to real-time) over the network, and may
dynamically join or leave a session group. Players contribute to the session by playing
on their MIDI controllers, using General MIDI protocol, and it is possible to have
different topologies including the formation of hybrid groups of participants and cores
with one or more leaders, also permitting access to listener groups.
Some of the most recent Collaborative Music Performance experiments, many of which
are in the context of music education, took advantage of broadband technology and
videoconference technology30:
Internet2 Initiative: World's First Remote Barbershop Quartet (Nov. 2000): Multi-
location quartet; each of the 4 singers in different cities, conductor in 5th. Network
delay variances prevented singers from hearing each other or the conductor.
Internet2 Initiative: Music Video Recording via Internet2 (Nov. 2000): Multi-location
music video recording session using real-time streaming video. Musicians
simultaneously connected via timing tracks to a mixing board.
Chris Chafe: QoS Enabled Audio Teleportation (Nov. 2000): CD quality sound (750
kbps) of 2 separated musicians in Dallas streamed to Stanford. Musicians played
"together" in same space (Stanford) but delay was severe.

30 Partly from (Woszczyk, W. and others, 2005) and from Jeremy Cooperstock’s notes on
Internet Audio Landmarks — ANET Summit Presentation, 2004.
63

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

Zukerman “Playing Together” Sessions (Dec. 2000): From New York-Ottawa, Pinchas
Zukerman teaches violin classes to McGill University students, in Montreal, using
broadband connectivity from the National Research Council in Ottawa to the McGill
University in Montreal.
John Wawrzynek: Network Musical Performance (May 2001): Gestural coding (e.g.
MIDI) used to manage data for distributed musical performance. Musicians at Berkeley
and CalTech, playing on MIDI keyboards; local feedback only.
Zukerman Music Master Classes (Feb. 2002): Again from New York-Ottawa, Pinchas
Zukerman teaches violin classes to McGill University students, in Montreal from the
National Research Council in Ottawa to the McGill University in Montreal, but this
time using broadband CA*net3 (Canadian fiber-optic network), capable of transmission
rates of up to 40-gigabits per second. This allowed the use of SDI video (High
resolution Digital Video) and multi-channel 96 kHz/24 bit audio and display on 50"
plasma screen (near life-size). Improvements in immersive perception were remarkable.
McGill-Stanford Jam Sessions (Jun. 2002): The UltraVideoconferencing technology
developed at McGill University was used in a cross-continental jam session between
Musicians at McGill University and Stanford University. The event featured full-screen
bidirectional video and multi-channel audio in what was the first demonstration of its
kind over IP networks.

Figure 22. Musicians at McGill University (Dan Levitin — sax and Ives Levesque — trombone)
Jamming with remote Musicians at Stanford University projected on Screen (Alexander Carôt
— Bass and Estabin Wilson — sax).
64

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

Furthermore, if we consider all the experimental Musical Practice that has been carried
out in the last decade using video and audio conference technology it is most certain
that this is the approach to Networked Music that has the largest repertoire of music
performed.
Innumerous references to this approach can be found in Sot Gresham-Lancaster article
“Video Conferencing Software as a Performance Medium” published in 2005 at the
Networked Performance Blog from the Turbulence.Org Web Site31.
A similar scenario can be found in communities of experimental Artists and Engineers
which develop custom made experimental music software, frequently adapted for
internet Joint Performance. A representative example is Sergi Jordà’s FMOL Peer-to-
Peer multi-user development based on the original software mentioned in the previous
section of this chapter.
This enhancement implemented in 2001 consists in providing MIDI connectivity over
the Internet between two Instances of the software running in remote machines. All the
sound computation is done by each peer.
The first FMOL Networked concert took place in October 2001, during the
Networkshop festival in Dresden, Germany, between the location of the festival and the
city of Barcelona, Spain. Attained delays were in the range of 100 ms using a
conventional 56 kb modem connection, and according to Sergi Jordà a very good feeling
of playability was achieved with this setup.
This condition of immunity to network delay in FMOL music is related to the nature
of its free and improvisatory Musical structure. The sound sequencing technique used
in this system, based on low frequency oscillators (LFOs) excitation of sound
generators, creates rhythmical and melodic progressions which, to some extent, support
flexible reaction times and short lacks of synchronicity from the performing partners.

31 The article can be found at http://www.turbulence.org/blog/archives/2005_04.html (accessed
2005/10/08)
65

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

tempo measures, allowing users to play together synchronously even though they will
not be playing in the same tempo interval as other player.

Figure 25. NINJAM client interface
NINJAM uses compressed audio which allows it to work with any instrument or
combination of instruments. It streams compressed audio to a NINJAM server, which
can then stream it to other performer in a jam session.
The Remote Music Performance approach to the development of Networked Music
Systems manly uses the Internet as a communication media that provides links
between performative spaces in an event driven perspective and performed by a well
specified group of users.
A different way to move towards Networked Music over Internet is by exploring its
shared nature by the means of distributed On-line shared spaces suitable for collective
sonic creation by anonymous, possibly non-musicians, on-line users. This approach
leads more towards improvisation paradigms as discussed in the following section.
68

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

2.4.4 Shared Sonic Environments
LOCATION
L o c a l Remote
Shared Sonic Environments

Figure 26. Shared Sonic Environments in the Networked Music Classification Space
Shared Sonic Environments based on the concept of Shared Virtual Environments
(SVEs) discussed in section 2.1.4 of this chapter and embraces several distinctive
features:
ƒ The focus is on synchronous collaboration between on-line users and usually
more than one user is connected at any given moment (locally or remotely).
ƒ It is based on a public shared space that is openly available to the online
community and therefore it must use the most disseminated and open
technology on the Internet.
ƒ People can be found on-line improvising in collective music pieces, given that
everyone should be able to choose either to participate as a performer or simply
as a member of the internet based audience.
ƒ No requirements are demanded from a regular user in terms of previous
knowledge of musical practice.
ƒ Each user is normally able to express himself by somehow manipulating or
transforming a sound or a musical structure.
ƒ It is suitable for a spontaneous improvisation approach to sonic creation.
69

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

ƒ Due to the permanent availability of these systems it supports events which are
unlimited in time.
Before presenting some of the work done in this area we find it useful to discuss two of
the inherited concepts from outlining Shared Sonic Spaces: On-line Improvisation and
the Time scales of a permanent event.
2.4.4.1 On-Line Improvisation
Given that, regardless of the latency, Internet communication media is tolerably
suitable for synchronous collaboration, thus Remote Music Performance Systems,
discussed in previous section, fulfill the requirements for an event oriented towards
remote performance. Furthermore, the physical setup requirements for a remote
performance should be the same for musical interpretation or improvisation.
Yet, even though the setup requirements are physically the same in both cases, there
are major conceptual differences.
In the context presented in this dissertation, musical interpretation refers to the process
of playing a predetermined sequence of events in a musical instrument providing some
sort of synchronism with other musicians or audiovisual events. In a musical
interpretation there is a great deal of space left for individual expression and even for
an improvisation experience, however the events performed by the musicians are driven
by a prearranged sonic choreography to a very large extent.
In musical improvisation musicians are not coupled in such a systematic approach, and
much more space is available for spontaneity, free expression and continuous
development of elaborate interactive relationships between the participants. This
process is many times referred to as a Jam session. One can also think of improvisation
as the process that results from composition coming together with interpretation, in the
sense that when improvising the musician is creating musical structures with a sense of
awareness similar to the composition process, even though he is doing it in real-time,
reacting to an outside stimulus, like when interpreting music.
70

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

timescale, the Micro timescale, the Sample timescale, the Subsample timescale and the
Infinitesimal timescale.
Most musical creations driven by the concept of an event are situated in the Macro
timescale defined by Roads as “The time scale of overall music architecture or form,
measured in minutes or hours, or in extreme cases, days”. However, one could question
where an ongoing musical piece, permanently available for hybrid communities of
creators and listeners belongs.
Realistically, this scenario should fit in the Supra timescale, which Roads defined as “A
timescale beyond that of an individual composition and extending into months, years,
decades, and centuries”, since the Infinite timescale is in reality a mathematical
abstraction and it is beyond the time life of the present cultural and technological state
of development.
Some recent Artistic proposals approached the concept of a musical event that is
unlimited in time.
In 1999 Antoine Schmitt 34 created the Infinite CD for Unlimited Music, the first
infinite CD to be published and distributed. This CD produced by Epplay, Schmitt and
Icono, once inserted in a computer generates music infinitely, always different and
always similar, without any images or any form of interaction.
Another essential reference is on-line piece entitled “Eternal Network Music” by Chris
Brown and John Bischoff35, where several flexible music pieces permanently go on since
February 28, 2003.
In the same way that the Internet’s essence is to provide permanent connectivity, a
Shared Sonic Space event is also permanent and public, since it is continuously

34 More information about Antoine Schmitt’s infinite CD can be found at the web site
http://www.infiniteCD.org/.
35 These pieces are available from the Web Site: http://crossfade.walkerart.org/
72

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

available to the public both via the Internet providing a permanent choice for the users
to be in either the performer’s or the spectator’s role.
2.4.4.3 System Implementations
A very Early example of a Shared Sonic Environment System implementation, which
was significantly inspiring for the early research work leading to this dissertation, is
Atau Tanaka’s MP3Q piece on the web (Tanaka, A., 2000), classified by the author as
a shared on-line sound space. The web application streams multiple channels of mp3
audio from different servers and users can concurrently manipulate these mp3 sources
by actuating over graphical representation of the systems current behavior via a sort of
3D Cube.
An extremely significant couple of developments that led to a number of applications
centered on the idea of a shared Sonic Environment were two very specific technologies
envisioned by Phil Burk: An audio Software Synthesis Application Programming
Interface (API) for Java entitled JSyn (Burk, P., 1998), allowing multi-platform client
sound synthesis in web-browsers, and the TransJam System (Burk, P., 2000a), a Java
based server that can be incorporated into applications that allow people to interact
synchronously over the Internet.
Based on these Technologies three Shared Sonic Environment Systems were
implemented36:
ƒ The WebDrum, developed by Phil Burk, is a drum box that can be shared by
several people over the Internet. Users are not required to have any musical
experience and while sharing this software they can chat with each other at the
same time as editing drum patterns and listening to the music they create
together.
ƒ The Eternal Network Music, developed by Chris Brown and John Bischoff, is
based on two interactive music pieces, which are part of a historical

36 Links to this systems are available from the TransJam website: http://www.transjam.com/
73

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

retrospective of the Hub. In fact, the Hub’s work inspired the development of
the TransJam server.
ƒ The Auracle, developed by Max Neuhaus, Phil Burk, Jason Freeman, C.
Ramakrishnan and Kristjan Varnik, is a voice driven, interactive, collaborative
instrument. Working from Stuttgart Germany, they have used JSyn and
TransJam technology along with Linear Predictive Coding (LPC) speech
analysis, neural nets, evolutionary strategies and other techniques to create an
engaging sonic environment (Ramakrishnan, C., Freeman, J. and Varnik, K.,
2004).

Figure 27. Screenshot from the demonstration video Documentary on the Auracle
An even more extreme example of custom made music software, oriented towards a
very specific community of musicians, are live coding environments (programming
languages oriented to live music generation), such as Supercollider, or ChucK (Wang,
G. and Cook, P., 2003).
74

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

The most recent development in the ChucK framework aims to make this collaborative
system oriented towards geographical displacement following the approach of a Shared
Sonic Environment. The Co-Audicle (Wang, G. and others, 2005) is defined by the
authors as a collaborative audio programming space, for collaborative, multi-user
interaction based around the ChucK language.
It operates either in client/server mode or as part of a peer-to-peer network. The
different instances conveyed in a Co-Audicle session are represented graphically at the
client’s interface through engaging metaphors.

Figure 28. Graphical Representation of different computational instances in the Co-Audicle
Shared Sonic Environments is a central concept in this dissertation. In this context
most of the experimental research developed between 2001 and 2005 was focused on the
development of a Proof-of-Concept system prototype entitled Public Sound Objects
(PSOs). The project is a web-based Shared Sonic Environment, which has been an
experimental framework to implement and test different approaches and concepts for
on-line music communication, discussed in Chapter 3 of this dissertation, such as the
notion of a network-music instrument incorporating latency as a software function, by
dynamically adapting its tempo to the communication delay measured in real-time.
PSOs first released was in 2002 by Álvaro Barbosa and Martin Kaltenbrunner at the
Music Technology Group from the Pompeu Fabra University in Barcelona — Spain
(Barbosa, A. and Kaltenbrunner, M., 2002). The System is extensively discussed in
Chapter 4 of this dissertation.

75

Chapter 2. Survey of Computer-Supported Cooperative Work for Music Applications

Like similar paradigms oriented towards visual or textual communication (MUDs,
MOOs, IRC, Active Worlds, etc) tend to lead to new mechanisms of interaction not
usually seen in “real life” (Curtis, P., 1992), a similar result can be expected in
paradigms oriented to music or sonic arts, suggesting that the sonic outcome of such
systems could express interesting new artistic results.
It is clear that this area of sonic creation is quite promising, not only by the fact that it
allows the enhancement of known paradigms to make music, but also since it provides
a context for stylistic novelty.
The results from this survey led directly to the research and developments in the
Shared Sonic Environment project Public Sound Objects introduced in Chapter 5 of
this dissertation.
77

Chapter 3
Networked Music Practice Topologies
In his book “Networking the World”, 1974-2000, Armand Mattelart characterizes
communication networks as:
“(…) an eternal promise symbolizing a world that is better because
it is united. From road to rail to information highways, the belief
has been revived with each Technological generation (…)”
For Mattelart Networks are systems that facilitate the movement of persons, material
and symbolic goods, which can have diverse structures (linear, radial, centripetal,
rhizomatic) but do not require a bidirectional stream of movement in each channel
(Chandler, A. and Neumark, N., 2005).
Similarly to cooperative work mediated by computer technology, Music Practice
Networking should be regarded as a paradigm which requires this bidirectional flow of
information. In this chapter, general ideas and proposals for generic topologies are
presented in the perspective of providing orientation and reference concepts to project
development of Computer Mediated Networked Music Practices.
3.1 Networked Models for Collaborative Music
Practice
The ubiquitous nature of communication in computer networks, firmly manifested in
the Internet era, has greatly contributed to a favorable environment in which joint
editing systems accomplished exponential acceptance by the on-line community.
78