Meeting of the TEI Music SIG: November 8, 2008, London
The first meeting of the Special Interest Group in Music was held on the 8th of November 2008 within the TEI Members Meeting at King's College London.
The SIG will concentrate on MEI first, as it is designed for critical music editions. This decision does not exclude the other formats from further evaluations.
The SIG has to provide recommendations and examples for the integration of music encoding within TEI. This will start with music examples in Common Western Notation, but shall be extended to other notations as much as possible.
The question of integration v. collaboration of MEI and TEI was left open until further evaluations; it is not yet certain if the (digital-)musicological community (which is not as large as the TEI community) will agree to definitively incorporate MEI into TEI. This political question will be addressed later.
In the long term the SIG will work on determining the actual similarity of the header parts and the textCrit / meiCrit modules. It would be desirable to allow TEI encodings in a MEI document (at appropriate locations) and vice versa. For example, a musical document with text considered worthy to be encoded in TEI could be represented as an MEI document with TEI inclusions, or it could be represented within TEI, where the <teiHeader> is mandatory and the MEI element <mei:work> could substitute or go alongside <text> and/or <facsimile> . On the other hand, a mainly textual document with music embedded could make use of the latter example. In order to reach this interoperability, the technical improvement of MEI is a first objective to be reached.
For all of these tasks, Johannes Kepper suggested that the SIG might cooperate with the research project called Digital Music Notation Data Model and Prototype Delivery System (conducted by Roland and Kepper). This project will translate MEI from a one-man-show to a community-based project, and will discuss the political issues concerned with the collaboration/integration with TEI.
MEI will have to consider referring in its documentation to the authority lists and controlled values defined by the MPEG4.23 standard wherever appropriate. This has no consequences for the concepts and model of MEI, but helps to standardize the interpretation of certain basic symbols and elements. For the MPEG4.23 standard documentation please refer to: http://www.interactivemusicnetwork.org/mpeg-ahg/w8632-MPEG-SMR-part-23-rev-public.pdf . Furthermore, the development of tools for WYSIWYG generation and dynamic rendering of MEI should be supported by the SIG as much as possible.
Short-term agenda - first steps
For the further discussion about MEI's independence it is important to compare the headers. A first question to be addressed is: "Is it possible for MEI to use an extended TEI header?"
A first attempt to answer that question might be to "oddify" MEI, as this allows an easier comparison of individual parts. At the same time, it allows the generation of different schemas as well as DTDs.
At the same time, the primitive objects and values of MEI shall be compared to their equivalents (if existing) in the MPEG standard.