musicSpace: Using and Evaluating e-Science Design Methods and Technologies to Improve Access to Heterogeneous Music Resources for Musicology

Summary
Information resources for Arts and Humanities researchers are increasingly being made available via the Internet, while at the same time, researchers themselves are often seeking to break down traditional subject barriers and ask research questions that are interdisciplinary in nature. However, current on-line resources remain isolated leaving researchers to carry out multiple searches across multiple systems in order obtain materials corresponding to well-defined research questions. Within a single source the search options are often limited, giving insufficient control over criteria for the effective framing of research questions.

In this project, we take proven e-science-based technology and design approach, embodied in the mSpace software framework, and apply it to a range of significant music data collections, providing a single point of entry with exceptional ease of navigation. Working with Arts and Humanities Research Council resources Cecilia, EPSRC CHARM, MIMAS's COPAC and with our commercial partners, the Oxford University Press's New Grove Dictionary of Music and Musicians Online, Naxos and RILM, the system will support research in 3 diverse areas of musicology - 19th-century Italian opera, Monteverdi performance, and electroacoustic works of the 20th century. This range of sources will provide information on many different aspects of music, including fundamental data on composers and works, historical recordings, collections and library holdings.

Our approach integrates these heterogeneous sources, and provides a powerful, intuitive interface for complex exploration. Research questions are typically bound by a range of criteria along different dimensions (e.g. Italian comic Opera dating from 1830-1840 composed in Bologna) and mSpace provides an intuitive interface allowing the user to combine different slices through information, and to adjust and reorganise these slices easily. The boundaries of a particular search, engaging multiple data sources from databases to the Web, can be framed and reframed rapidly, such that queries that previously took days, weeks or longer to assemble can be answered within a few minutes. This fluidity encourages the asking of a variety of questions (given the low time cost involved) and further exploration of the information, so that previously un-discovered connections may be found.

A key goal of the project is to evaluate use of the system over a period of real use for real research outcomes. To this end we have three participant communities involved: we will use the system to carry out research specified in the three areas above that would not be tractable without this system; we have also partnered with the Integra project for them to use the system for their research, and we will be doing a community release. From this range of participants, we will investigate the ways in which the system supports and effects research practice in order to tune it to improve that support for innovation and discovery.

The project will therefore meet several related goals: to use eScience technologies and design methods (1) to automate the association of heterogeneous research resources (2) to provide interfaces that will reduce significantly the time needed to discover, access and coordinate such resources; (3) to enable new kinds of research questions to be addressed through new models of resource integration and interaction, and (4) to demonstrate an extensible, maintainable infrastructure that will support integration of new resources in other AH domains (5) to demonstrate the efficacy of these approaches for real innovative research.