Page  00000001 USING CODES: COOPERATIVE MUSIC PROTOTYPING AND EDUCATIONAL PERSPECTIVES Evandro Manara Miletto Marcelo Soares Pimenta Rosa Maria Vicari Universidade Federal do Rio Grande do Sul (UFRGS) Instituto de Informatica PO.Box 15.064 - 91.501-970 - Porto Alegre - RS - Brazil. Phone: +55 51 3316-6168 {miletto,mpimenta,rosa} ABSTRACT This paper discusses how to use CODES - Cooperative Music Prototyping Design in music classes. CODES is a web-based environment for cooperative music prototyping that aims to provide users (musicians or non-specialists in music) with the possibility of creating musical examples (prototypes) that can be tested, modified and constantly played, both by their initial creator and by their partners, who will cooperate for the refining of the initial musical prototype. CODES characteristics are presented and the experiments in learning situations are briefly discussed. 1. INTRODUCTION Internet based networked music has gained wider acceptance and the existing applications have evolved towards more sophisticated projects and concepts including, for example, real-time performance systems and different systems for multiplayer interaction and collaboration. CODES is a Web-based environment for cooperative music prototyping whose aim is to allow users (either interested lay people or experienced musicians) to make music experiences and interact with each other in order to create simple musical pieces (here called music prototypes or simply prototypes). CODES comprises the concepts of Computer Music, Human-Computer Interaction (HCI), and Computer Supported Cooperative Work (CSCW) to allow lay people interested in music to experiment and reinterpret the sense of creating and developing their own musical culture and skills through the web. Like Weinberg [1], we are interested in providing any user (from lay people like novices and children to experienced musicians) with the possibility of accessing meaningful and engaging musical experiences. By using CODES, interested lay people can play the role - like experienced musicians - of actors of their own musical experiences, i.e., they can create musical examples (prototypes) that can be tested, modified, and constantly listened, both by the creator and partners that will be cooperate for the refinement of this musical prototype. This short paper is organized as it follows. The motivation for this study is presented in section 2. Section 3 presents some related tools or systems supporting cooperative musical composition. The CODES system, its architecture, and main characteristics are presented in section 4. Some characteristics of CODES interface are presented in section 5. Section 6 discusses the educational perspectives for cooperative music prototyping and finally the section 7 concludes the paper. 2. MOTIVATION The main motivation for this paper (apart from the technical challenges posed by the environment implementation) is the intention of overcoming three barriers that people who wish to engage themselves collectively in a process of musical prototyping usually have to face. The initial barrier is related to the possession of musical instruments. Currently, the availability of virtual musical instruments and remote tutoring and development over the Internet, allied to interactions among users, allow people to have the real possibility of creating musical material, instead of only acquiring something that comes to them ready. Another barrier is the physical distance between people in these groups, which are essentially small and spread. The creation of virtual communities on the Internet can overcome this barrier minimizing interaction problems among users. At last, another barrier to be broken is the necessity of knowing common music notation (CMN) [2]. This notation (score and symbols) is fundamental for a deep and complete learning of musical theory, despite its complexity. But it may pose some resistance for lay people curiosity not interested in learning a great number of symbols and forms in order to produce musical material [16]. By eliminating this obstacle, users can have all their creativity focused on sounds experiments and creation. 3. RELATED WORK This section summarizes the characteristics of some environments found in literature for collective musical composition. Clearly, they are environments for music composition, not music prototyping. In a survey about IMN - Interconnected Musical Networks [1] - the author proposes four different levels of interconnectivity among participants and the role of the computer in enhancing their interdependent social relations: "The Server", "The Bridge", "The Shaper" and "The Construction Kit". Most of Internet based

Page  00000002 systems for music composition described here are cited in that survey within the last level, "The Construction Kit". In this level, there is high interconnectivity among participants, which are usually experienced musicians. Participants are allowed to contribute with their own material and manipulate (listening, altering, refining, etc) others' contributions, usually in an asynchronous interaction and off-line material manipulation. The PIWeCS [3] system is a complex composition system based on a dialogue between human/nonhuman agency. The FMOL System [4] is related to real-time collaborative musical composition on the web. Using a plug-in, the system allows many distributed users to work together in one or more musical works. The EduMusical System [5] supports collaborative and interactive distance learning, aiming at teaching music to children and teenagers, oriented by music instructors. The TransJam System [6] aims at allowing musicians who are connected to the web site to make musical performances together, selecting the intervals (loops) of the instruments that will be played. Daisyphone [7] is an environment for remote group music improvisation presenting a novel design for more engaging, social, and serendipitous musical environments, focusing on the representation of looping music, providing support for remote collaboration, and ideas formulation. PitchWeb [8] is a multi-user musical instrument specifically designed for the Internet that uses plane geometric shapes (figures representing sound samples) that can be selected and combined collectively. There are other systems that deal with composition like CreatingMusic [9] and HyperScore [10] - that enable non-musicians to compose - collectively or not. Rolf Woehrmann and Guillaume Ballet's [11] article examines client/server architectures for computer music, drawing upon the example of IRCAM's Studio Online project. Some aspects - mainly regarding some technological concerns - are common to most of these systems, like a) adoption of a client-server architecture, b) usage of MIDI sound format, c) implementation in a platformindependent Java language, except FMOL, implemented in C++ and available only for Windows or Linux operating systems, and d) unrestricted access to ordinary users, i.e., any user may login or access the system without additional fee. Most of them may ideally be used by nonexperienced users, however, more skilled users in practice can obtain better results if they have specific knowledge about sound synthesis, for instance. 4. SYSTEM ARCHITECTURE CODES implementation is based on the classical client server architecture (figure 1), composed basically of the Applet Sonic Manipulation Manager, on the client side; and of the Cooperation Manager, the Sonic Pattern Manager, the User Manager, the Database Manager, and a file database, on the server side. The functions are to control the sound manipulation, to control the actions of inviting, adding comment to musical pieces and perception events, to control the organization of sonic patterns in the server, to control all activities related to the user, and to establish the connection with the database, respectively. CODES implementation follows a free software philosophy aiming at providing easy access to software developing tools. The authoring software used was Java Sound API, the Apache web server, and the MySQL database. The file format chosen was Standard MIDI Files because of the easy manipulation and loading it provides. The sounds of synthesized MIDI files, played on most PCs, still have a low quality; however, future possibilities will be made easier by this choice (conversion FROM MIDI to CMN - music sheet). In addition to characteristics discussed in section 3, CODES provides three other important aspects to consider in a collaborative environment for music composition/prototyping: a) Group Memory; b) Support to long prototyping sessions; and c) Sound Format for Export/import. 5. CODES USER INTERFACE The user interface of CODES was designed aiming at covering aspects related to interaction flexibility, robustness and easiness, as well as carefully presenting an adequate support when delivering complex musical information, in order to provide an effective interaction of the users with the environment itself. The environment was designed to reach a balance between those GUIs that are so "easy" for the user that they end up depleting his or her expressiveness and others that are so hard that they discourage beginners [12]. Usually, these users do not know about CMN, which symbolizes mainly the pitch and duration of notes, and the fact that it was not designed to represent a general view of a high level musical structure [13]. CODES considers that the musical prototype is formed by Lines (tracks) of instruments, arrangements,

Page  00000003 effects, etc, such as bass, notes, and rhythm lines, etc. Each line belongs to a user who has the privilege of editing (selecting other sonic patterns). However, the nuCr i0 RllnwiT to rrente mnre thin nne line (timiure 7\ Figure 2. A screen shot of CODES editing window. Sonic patterns are high-level musical structures (small parts of musical files in MIDI format) that make it easier the process of sound choice and prototyping. The patterns, which form a sequence in a cell, have, besides the difference in name, small sonic difference of the same style and same duration, which facilitates the adaptation of the user with his or her piece. The alternate representation of the music structures offered by CODES includes, therefore, concepts such as rhythm, tempo, melody, harmony and timbre, elaborated in advance, making it possible for the user to experiment, choose, and combine sonic patterns in an easy and interactive way, providing a more immediate result in this process, and reducing the difficulty that CMN could present at an initial stage [14]. 6. EDUCATIONAL PERSPECTIVES FOR COOPERATIVE MUSIC PROTOTYPING Cooperative music prototyping is here defined as an activity that involves people working together in a musical prototype. Cooperation in CODES is asynchronous, and thus users can access the prototype, doing their experiments and writing comments at different times. CODES considers that a musical prototype is initiated by someone, the prototype owner. The prototype owner uses CODES to elaborate an initial musical prototype and to ask the collaboration of other "partners" by sending explicit invitations (typically using email facilities). Partners who accept the invitation can participate of collaborative musical refinement and manipulation. The coordination of all activities in such musical context can happen naturally when the group recognizes one member as someone having more musical abilities or being more experienced. Usually, in music classes the more experienced user is the music teacher and his/her opinions and actions as an explicit coordinator role may stimulate the reactions (questions, doubts, etc) of students. CODES usage can provide interesting alternatives for beginners in music. By means of interactions with, and advices from, more experienced users, CODES provides support for learning and positive interdependency, encouraging collaborative actions, argumentation and discussion during the multisession long-term development of a cooperative musical prototype. In learning environments that lack resources (musical instruments, for instance), the use of CODES through a computer connected to the Internet can provide interesting alternatives for beginners in music. The group can carry out sonic experiments creating a musical prototyping where each student plays a defined role and an activity to be developed in this prototype. The group, through interactions and advices of the teacher, decides which musical gender will be studied, as well as the number and the kind of instruments and music structures that will be put together in the prototype. Then, it is possible to work in music creation collectively, using the metaphor of a musical orchestra: each student has a defined role in the final result. In addition, the teacher can enable many patterns related to the same instrument for different students and all students can compare the different contributions, choosing or mixing alternatives. The teacher can also apply concepts of musical dynamics and expressiveness, indicating different sonic structures in different moments of the prototyped musical discourse. Thus, the combination of both music prototyping and music education is promising and merits additional research. Particularly, our intention is to provide computer-supported features for music education following the TEC(L)A model [15]. This perspective is very interesting because it creates a new orality to the teaching of music, using collectively Internet technology, not by the adoption of a theoretical or written model, but by the achieved agreement reached inside the group. Likewise, the perspective of musical understanding by Swanwick [15] called "intercultural", says that "music is free to travel and, just like language, is continually being refashioned, adapted, reinterpreted-to create 'new human values', to 'organize thought', to 'transcend' the limits of local culture and personal self." The musical accomplishment in groups is natural, for instance, for teenagers, who usually form musical bands because of particular individual needs to complement each one musically with the help of other friends who play different musical instruments, in a process of mutual help where people learn with each another. In the other hand, an aspect remaining to be explored is the implementation of real-time interaction in CODES architecture. Interaction among users in a synchronous mode may bring new features and properties not considered until now. 7. FINAL CONSIDERATIONS This paper presented the most important characteristics of CODES environment project, focusing on HCI,

Page  00000004 CSCW aspects, and mainly on the educational perspectives of its use. Educational possibilities of the use of CODES make possible that the teacher use it as a learning and teaching tool, once it is inserted in a guided tutored? educational context. Thus, the environment can be the catalyst of the musical culture, motivating and making users experiment with deep learning and by means of interactions among CODES users. Our approach for cooperation among users in order to build collective music prototypes is an example of a very promising educational tool for musicians and lay people because it enables the knowledge sharing by means of rich interaction and argumentation mechanisms associated to each prototype modification. Consequently, each participant may understand the principles and the rules involved in the complex process of music creation and experimentation. CODES has been used in a restrict context to validate and consolidate its functionalities, identify and correct failures, and establish new requirements - functional and non-functional (performance, for example). We hope to amplify its audience soon by publishing the system and its results. Our cooperative approach for music prototyping has been applied in actual case study and the results obtained from evaluation methods until now express the relative success of our work but we intend to continue our research towards better improvements. It is worth acknowledging that we can involve not only more musicians and music educators but also DJs in the improvement process itself. In a further work, we intend to develop more (qualitative and quantitative) assessment experiments in order to evaluate the scalability and richness of our approach. 8. REFERENCES [1] Weinberg, G. "The aesthetics, history, and future challenges of interconnected music networks." Proceedings of the International Computer Music Association Conference, G6teborg, Sweden, 2002. [2] Roads, C. The Computer Music Tutorial. Cambridge: MIT Press, c1996. [3] Whalley, I. PIWeCS: Enhancing human/machine agency in an interactive composition system. Organised Sound, 2004. 9(2):167-174. [4] Jorddi, S. Faust Music On Line: An approach to real-time collective composition on the Internet. Leonardo Music Journal, 1999.Vol 9, 5-12. [5] Ficheman, I. K. Collaborative Distance Learning Supported by Interactive Electronic Media: A case study. Master Thesis. Escola Polit6cnica da Universidade de Sho Paulo. Sho Paulo, 2002. (In Portuguese). [6] Burk, P. "Jammin'on the Web - a new Client/Server Architecture for Multi-User Musical Performance.Visual". In Proceedings of the International Computer Music Conference 2000, Berlin, 117-120. Association [7] Bryan-Kinns, N. "Daisyphone: The Design and Impact of a Novel Environment for Remote Group Music Improvisation" in Proceedings of DIS 2004 - ACM Symposium on Designing Interactive Systems. 2004,135 - 144. [8] Duckworth, W. "Making Music on the Web". Leonardo Music Journal, Vol. 9, 2003. 13 - 18, MIT Press. [9] Subotnick, "M. Creating Music". Available in the web at, accessed in June/2004. [10] Farbood, M.; Pasztor, E.; Jennings, K. "Hyperscore: A Graphical Sketchpad for Novice Composers". IEEE Computer Graphics and Applications, Volume: 24, Issue: 1, Year: Jan.-Feb. 2004. [1 1] W6hrmann, R. Ballet, G. "Design and Architecture of Distributed Sound Processing Systems for WebBased Computer Music Applications". Computer Music Journal 23, 73-84. 2002. [12] D'arcangelo, G. "Creating a Context for Musical Innovation: A NIME Curriculum." In Proceedings of the Conference on New Instrument for Musical Expression - NIME2002, Ireland. 2002. [13] Suthers, D., Toth, E.E., Weiner, A. "An Integrated Approach to Implementing Collaborative Inquiry in the Classroom". in Proceedings of Computer Supported for Collaborative Learning. Toronto, Canada. 1997. [14] Miletto, E. M.; Pimenta, M. S. Towards a Webbased Environment for Cooperative Musical Composition. In. Proceedings of the IX International Symposium on Computer Music, Campinas, Brazil. 2003 (In Portuguese). [15] Swanwick, Keith. "Music, mind, and education". Music Education in a Pluralist Society 1988, International Journal of Music Education, No. 12. 1988. [16] Miletto, E.M.; Pimenta, M.S.; Costalong, L. Using the Web-Based Cooperative Music Prototyping Environment CODES in Learning Situations. Proceedings of 7th Conference on Intelligent Tutoring Systems (ITS 2004), Springer-Verlag, 835-837.