E : josh.saunders@which.net Qualifications 2:1 BA(Hons) Music and Visual Performance Dartington College of Arts, Totnes, Devon 1994-1997 MA with Distinction Hypermedia Studies Hypermedia Research Centre, The University of Westminster, Harrow Campus 1997-1998 Provisional Title TOWARDS THE DEVELOPMENT OF SOFTWARE INTERFACES AND SYSTEMS WHICH ALLOW THE USER/COMPOSER TO 'AUTOMATE' THE PROCESS OF MUSICAL COMPOSITION. THIS MUSICAL MATERIAL WOULD BE THEREFORE BE MANIPULATED THROUGH MIDI, DIGITAL-AUDIO, TAPE, AND OTHER ELECTRO-ACOUSTIC RESOURCES. Stage I Year 1 (approximately six months) Intensive reading research into precedents, and current work, in the field of algorithmic, generative, 'intermorphic', stochastic, electro-acoustic, concrete (et al) composition. This reading would include a study of a 'trajectory' of European and American composers in electronic music history, such as Varese, Stockhausen, Reich, Nancarrow, Xenakis, Venao, Wishart, Aphex Twin, et al. I would examine seemingly oxymoronical design and engineering interactions, in "open/exclusive" and "closed/inclusive" software systems meant for the creation of music for the layman and specialist alike. This can be seen as twofold, in the form of : (a) "open" systems are often more specialist, and as a result more exclusive - for example, object-orient design in IRCAM/Opcode's 'visual' algorithmic music tool, MAX, and ; (b) "drag-and-drop" systems of music creation, for example, "Muzik" for the Sony Playstation, are more inclusive, and more akin to games - it is however what I would posit to be a 'closed' system, in that it presents pre-set 'palette' of sounds and rhythms which cannot be altered. From these examples, we can ask : what are the pros and cons of both the programmatical and abstract, and the appropriative and literal, in interactive music systems design? How have designers and engineers worked together, successfully and unsuccessfuly, to find solutions to putting a console onto real-time audio data manipulation? Questions at this point : why has Steinberg's Cubase time-axis system become so popular amongst the 'casual' or non-specialist/non-institutionalised composer? Stage II Year 1/3 The collation and analysis of data, output from experiments into number generative systems as applied to systems of music software for Win32/DOS, Open Source (Linux), and Macintosh architectures. These preliminary exercises could perhaps be undertaken using (where appropriate) : Macromedia Director XTRAs, and the Lingo scripting language, for "front-end" systems design, and manipulation of MIDI. This could also incorporate experiments with ActionScript, the Lingo variant in use with Macromedia Flash 4/5. The use of programs such as CSOUND or CECILIA. This initial practical research will work towards examining and experimentating with 'sculpting' digital audio, manipulated from PC soundcards, and samplers (i.e, E-MU, AKAI) To make computational life easier when analysing the musical data that gets thrown out by initial experiments into generative systems, number "filtration" and calculation systems will be developed (see point 3 above). These will measure and log occurences of, for example, pitches, within given durations of the system ouputting data in runtime. Through these experiments, it is hoped that valuable relationships between different numbers (in, for argument's sake, music created by changing parameters in sine or cosine equations, which then play MIDI), will be found in generative compositions. These will throw up questions I would wish to address about the nature of 'randomness' and probability in, for example, an equation which triggers MIDI notes, being passed a random, or interactively driven integer, at regular intervals during run-time. Once these relationships have been established, and different equations, or interative (eg, mouse-movement driven) systems of musical generation have been set into place, the logical next step would be investigate the incorporation of heuristics, in order to establish systems or 'rules of thumb' for different styles of generative musical composition. Style would therefore be reduced to a system or repository of defining parameters, which can then be re-deployed by the application, and manipulated 'on-the-fly' by a succesful GUI or command-line interface. Questions asked at this juncture will include whether the system can be made to 'learn' sequences or musical "motifs" as number relationships (for example, the IV V I musical cadence). The key question is therefore : can the output of musical algorithms mimic styles or genres, and if so, how can this be subverted by the user? Eventually, I would hope to fine-tune and inevitably 'enculture' these number outputting systems of generative musical 'engines', which our ears could then recognise as, or associate with, being in certain idioms, genres or styles. The final and ultimate quest of this stage would be to discover and log patterns of number useage in the parameters of musical material (eg, pitch, meter, volume/velocity) and make this work towards the construction of something essentially musical. As opposed to a chaotic automata, the interface would eventually be designed around the parameters of the 'engine' so that the user could have a degree of access to the program as a vehicle of expression. This would avoid the program become victim to its number crunching force and not become just a number/equation value generator in soundwaves. I would therefore, in stage three, be examining whether complex ideas involving relationships between the sciences and the humanities, could be semiotically and symbolically compressed by the interface designer into a console or device based GUI. Stage III Year 3/5 As well as the main focus of interface design, the culmination of research in stage three will examine parameter changing in the musical 'engine(s)', the analysis of data output by it at runtime, and possibilities of storage libraries which could enable the software to 'learn' or repeat patterns it has generated. Stage three will therefore concentrated in fine-tuning the 'engine' through creating the interface. Firstly, how could a 'hands-on' parameter-altering interface be designed that would be at first intuitive to the user? How could 'rules-of-thumb' be expressed algorithmically when expressing style in composition? What systems of signing and/or graphic design would be most suitable for the software? Would the software be more suitable in a high-art, avant-gardist context ; would the system be eventually geared towards the more functional uses in music therapy for people with disabilities? Ultimately, I would hope to make the decision as to where to contextually place and fine-tune the applications as a result of putting these theories into practice and analysing the results achieved in Stages 1 and 2. If a learning system of reference library files is put into practice, logging information at run-time specific to user input, could the user teach the 'engine' to learn, and create something of a musical automata? Composition Folio [ NB - other compositions not yet uploaded! ] Three Guitar Studies Played by George Theophanadis, Dartington Hall, 1997 Andante Ritmico | Page 1 Of Score Andante Ritmico | Page 2 Of Score Andante Ritmico | Page 3 Of Score Click To Hear [ MP3 ] Largo | View Score Click To Hear [ MP3 ] Lento | Page 1 Of Score Lento | Page 2 Of Score Click To Hear [ MP3 ] Personal Experience/Previous Research Relevance 1993-4 Undertook a thirty week nightschool course in Electro-Acoustics at the Welsh College of Music and Drama. 1994 Passed Grade V Royal Associated Board Music Theory Examination Passed Grade VIII Royal Associated Board Classical Guitar with Merit 1994-1997 During my degree, I specialised in the use of MIDI sequencing (Cubase/StudioVision) and DSP/Digital Audio software (DigiDesign Pro Tools/SoundDesigner), often combining these arranged software compositions with live musician interaction. This I either multitracked from a booth, or mixed live through a PA. In Winter 1996, as a course requirement to spend three months "offsite", I opted to work on an Internship for a Multimedia production house in downtown New York City, called Harvestworks (www.harvestworks.org). Harvestworks, receiving grants from the US NationalEndowment For The Arts, provided budget training courses in audio and multimedia software. I assisted archiving many of theartist's in residence on DAT using Retrospect on a Mac OS platform, and was allowed to attend Multimedia software courses free of charge. My portfolio of work at the end of my BA(Hons) comprised of work which drew on BigEye, a piece of Quicktime-to-MIDI software developed by Tom Demeyer at STEIM studios in Amsterdam. I created a piece using a projected video piece, which acted as a "score" triggering "hotspots" (MIDI notes) allocated on a camera lens which videod the projection. The recording video camera, routed through BigEye, therefore acted as the means of playing MIDI notes - in this case, an AKAI S3000 sampler. 1997-1998 On my time at the Hypermedia Research Centre (www.hrc.wmin.ac.uk), my practical tutoring was taken on by Andy Allenson, Andy Cameron, and Tomas Roope, formerly ANTIROM(now TOMATO MULTIMEDIA). Their "sound toys", programmed in the Lingo scripting language on Macromedia Director, and sold to their clients such as Levis, Guinness and the Health Education Authority have influenced many other Multimedia houses. Andy Allenson has also since gone on to create a piece of "phase shifting" audio software, called "Phase", made using the YAMAHA MIDI Xtra within Macromedia Director, which will be a point of reference for some inquiries in my proposed PHD. This work has also run in parallel (although separate) to the work of AudioRom, who exclusively specialise in multimedia "sound toys", and who were also affiliated to the University of Westminster through their "Contemporary Media Practice" BA (Hons). Related Theorists and Practitioners Genetically modelled algorithms Bruce L Jacob, University of Michigan www.ee.umd.edu/~blj/algorithmic_composition/ The "GenJam" software application for Jazz improvisation John A.Biles, Rochester Institute of Technology, New York www.it.rit.edu/~jab/ Computer 'Artists' John Whitney www.siggraph.org/artdesign/profile/whitney/nowhitney.html http://panushka.absolutvodka.com/panushka/history/profiles/whitney_john/indexdd.html The KOAN engine (KOAN pro/KOAN X/KOAN web-browser plug-in/KOAN templates) www.sseyo.com "Phase-shifting" rhythmic and melodic patterns using MIDI Andy Allenson, was ANTIROM, now ROMANDSON www.antirom.com Graphic driven sequencers and "sound toys" AUDIO-ROM www.audiorom.com Abstract "sound toys" James Tindall www.thesquarerootof-1.com Jamming with MIDI over modems through TCP/IP (using MIDI as a network and application layer) Resrocket www.resrocket.com Composer's Desktop Project Trevor Wishart www.bath.ac.uk/~masjpf/CDP Bibliography (Abridged cross-sectional) Attali, Jacques, "Noise : The Political Economy of Music", Manchester University Press, Manchester, 1977 Carrol, John (Ed.), "Designing Interaction : Psychology And The Human-Computer Interface", Cambridge University Press, 1991 Ching, Francis D.K, "Architectural Graphics (Third Edition)", John Wiley and Sons, New York, 1996 Emberton, David J. and Hamlin, Scott J., "(Macromedia) Flash 4 Magic", New Riders, Indiana, 1999 Hollis, Richard, "Graphic Design : A Concise History", Thames And Hudson, London, 1994 Manning, Peter, "Electronic And Computer Music", Clarendon Press, Oxford, 1992 Read, Gardner, "Music Notation : A Manual of Modern Practice", Taplinger, New York, 1979 Rowe, Robert, "Interactive Music Systems", Massachussets Institute of Technology Press, Cambridge, Massachussets, 1993 Selfridge-Field, Eleanor (Ed.), "Beyond MIDI : The Handbook of Musical Codes", Massachussets Institute of Technology Press, Cambridge, Massachussets, 1997 Wildbur, Peter and Burke, Michael, "Information Graphics : Innovative Solutions In Contemporary Design", Thames and Hudson, 1998 Wishart, Trevor, "On Sonic Art", Emmerson, Simon(Ed.), Harwood Academic Publishers, Amsterdam, 1996 ŠJosh Saunders 2001