University of Virginia
University of Illinois, Urbana-Champaign
Introduction
The MEI (Music Encoding Initiative) is an encoding toolset
meant for modeling music information (Roland, 2006).
This does not refer to audio, but to the symbolic representation
of Common Music notation (CMN), the “standard” form of
music notation used between 1700 and 1935 (see Fig. 1). MEI
is an XML application expressed in DTD form (for the present),
with a TEI-like header (Fig. 2), a <music> element instead
of <body>, and <front> and <back> to accommodate text.
The goals for MEI are to encode CMN “out of the box”, to limit
verbosity without compromising the self-documenting aspect
of XML, to support repertoires other than CMN, and to support
creation of multi-lingual interfaces by allowing generic
identifier names to be changed, all to better enable creation of
scholarly editions. In this poster presentation we highlight the
recent developments made to improve and extend MEI with a
special emphasis on the impacts these developments can have
on digital musicology and scholarship.
Figure 1. Sample CMN rendition of "Quem queritis" <?xml version="1.0" encoding="UTF-8"
standalone="no">
< ! D O C T Y P E m e i S Y S T E M
"http://www.lib.virginia.edu/digital/resndev/mei/mei17b/
mei17b.dtd">
<mei version="1.7b">
<meihead>
<meiid/>
<filedesc>
<titlestmt> <title>"Quem queritis"</title>
</titlestmt>
<notesstmt>
<bibnote type="encoding-date">2003-03-15</
bibnote>
</notesstmt>
</filedesc>
<profiledesc>
<langusage>
<language id="la"/>
</langusage>
</profiledesc>
<revisiondesc>
<change>
<changedesc>
<p>Transcoded from MusicXML version 1.0</p>
</changedesc>
<date>2006-09-26-04:00</date>
</change>
</revisiondesc>
</meihead>
Figure 2.Sample of the TEI-like MEI header for "Quem
queritis".
Recent Developments
MEI is extending its notation encoding capacity beyond
CMN. While most representations of notation have
been limited to CMN, MEI has a TEI P4-like
extension/restriction mechanism which can be used to expand
the universe of sources to which it can be applied. Work is
progressing in two particular areas: “White mensural” notation
and “Medieval neumatic” notation. Both of these developments
also better situate MEI as an encoding mechanism for serious
digital musicology (see Section 3).
Mensural notation is a musical notation system used from the
later part of the 13th century until about 1600. The name
“mensural” refers to the capacity of this system to notate
complex rhythms with great exactness and flexibility. Mensural
notation was the first system used for European music that
systematically used individual note shapes to denote temporal
durations. Measure-less music requires structural reorganization,
or score-by-staff encoding rather than score-by-measure-by-staff
encoding: the staff must directly allow features previously
available only within <measure>, such as for timed events.
MEI is introducing new concepts to enable such encoding:
events -- ligature, mensuration and proportion signs;
orthographic vs. semantic accidentals; barlines, and new
attributes and attribute values.
Neumes are the basic elements of Western and Eastern systems
of musical notation prior to the invention of five-line staff
notation. Neumatic notation eventually evolved into modern
musical notation, but remains standard in modern editions of
plainchant. MEI is introducing new concepts to enable encoding
of this system: single and compound neumes, ligatures,
mensuration and proportion indicators, orthographic accidentals,
custos, interpretative marks (episema, liquescent neumes,
quilisma) that can be treated as timed events, mora (an
augmentation dot) that may be handled the same as a modern
augmentation dot, and new attributes and attribute values.
MEI is now supporting data collection via MusicXML (Good,
2002), a translator and interchange format for common Western
musical notation from the 17th century onwards. A new 2mei
XSL style sheet facilitates transformation of MusicXML to
MEI, creating an input path from any software that supports
export of MusicXML. On average the resulting MEI files are
only 51% as large as the original MusicXML files.
Enhancing Support for Digital
Musicology and Scholarly Editions
MEI has been gaining favorable reception among those
engaged in digital music scholarship. For example, the
MeTAMuSe project has said, "In contrast to MusicXML, which
is the de facto industry standard, but which is rather limited in
the representation of musicological concepts such as multiple
divergent sources, MEI has definite advantages in the
musicological context" (Byrd et al., 2006). MEI has also been
described as one of "two really serious contenders" in this
problem space (Kay, 2004).
Digital critical editions of music start with the encoding of the
musical sources, and then add layers for presentation and
meaning. MEI's improved capacity to encode multiple types of
notation, provide support for non-transcriptional text
commentary/annotation, as well as improved methods of data
capture into the encoding format and data export will hopefully
foster new efforts to create scholarly critical editions using
XML. Such content-based encoding modeled on text encoding
formats is not only best-suited to the development of these
digital editions, but can potentially best document the
intellectual process of the development of the corpus, making
the critical work better suited for verification and scholarly
argument.
MEI is well-suited for the creation of scholarly editions that
document the creation and revision history of a single musical
composition. In MEI, a single file supports encoding of the data
common to all sources only once, rather than requiring
redundant markup and encoding in multiple files. The
<source> element holds bibliographic and physical description of a single source document and can be linked to
specific data via its data attribute, while data can be linked to
the source via the source attribute, a mechanism not unlike the
“declarable/ declaring” attributes in TEI. When the meiCrit
parameter entity is enabled, parallel alternative encodings are
possible at the score, measure, and staff levels. This feature
would be particularly useful for the construction of such
scholarly editions as the “Online Chopin Variorum Edition”
(<http://www.ocve.org.uk/>). In the case of
manuscript music, the <handlist> element in the header
and the “hand” attribute (available on most music content
elements) allow one to track the scribes, copyists, etc. who
notated the music.
MEI now supports non-transcriptional text
commentary/annotation. The <annot> element provides a
way to group participating events, the notes that form a
descending bass line, for example, and provide a label for the
group. An editorial or analytical observation, encoded
elsewhere, may be pointed to using the linking attributes.
Alternatively, the observation may be included directly within
the <annot> element.
Ongoing and Future Work
Work is progressing to support two modes of visualization
and export. The first is a conversion from MEI to
MusicXML which will allow any software that reads
MusicXML to display/manipulate the data. This is analogous
to the method, i.e., conversion to HTML which is used to
display SGML/XML. This MEI-to-MusicXML conversion, is,
however, lossy. The second mode, direct conversions to other
internal representations, requires writing a filter for each
existing data format. This is a time-consuming task, but will
greatly reduce data loss.
Also, additional elements necessary for manuscript encoding
are planned: <add> (for something added by another person
or at a later date), <del> (for something marked out),
<unclear> (for illegible passages), <damage>(damage to
the carrier), <supplied> (for data supplied by the editor),
and <handshift> to indicate a change in scribal hands.
Appendix
<work>
<music>
<mdiv>
<score>
<scoredef>
<staffgrp>
<staffdef n="1" id="P1" label.full="Voice"
clef.line="2" clef.shape="G" midi.div="2"/>
</staffgrp>
</scoredef>
<section>
<measure n="1" id="d1e18" right="dbl">
<staff def="1">
<layer def="1">
<note id="d1e32" tstamp="0" pname="g"
oct="4" dur="4" dur.ges="2" stem.dir="up">
<verse n="1">
<syl>Quem</syl>
</verse>
</note>
<note id="d1e53" tstamp="2" pname="f"
oct="4" dur="4" dur.ges="2" stem.dir="up"/>
<note id="d1e69" tstamp="4" pname="d"
oct="4" dur="4" dur.ges="2" stem.dir="up">
<verse n="1">
<syl wordpos="i" con="d">que</syl>
</verse>
</note>
<note id="d1e90" tstamp="6" pname="f"
oct="4" dur="4" dur.ges="2" stem.dir="up"/>
<note id="d1e104" tstamp="8" pname="e"
oct="4" dur="4" dur.ges="2" stem.dir="up"/>
<note id="d1e120" tstamp="10" pname="f"
oct="4" dur="4" dur.ges="2" stem.dir="up">
<verse n="1">
<syl wordpos="m" con="d">ri</syl>
</verse>
</note>
{CODE DELETED FOR SPACE}
</measure>
</section>
</score>
</mdiv>
</music>
</work>
</mei>
Bibliography
Byrd, Donald, Time Crawford, and Geraint Wiggins.
MeTAMuSE: Methodologies and Technologies for Advances
Musical Score Encoding. Project proposal submitted to The
Andrew W. Mellon Foundation. Final Version, 19 May 2006,
Typescript. 2006.
Good, Michael. "MusicXML in Practice: Issues in Translation
and Analysis." Proceedings of First International Conference MAX 2002: Musical Application Using XML, Milan, September
19-20, 2002. 2002. 47-54.
Kay, Michael. XSLT 2.0 Programmer's Reference. 3rd edition.
Indianapolis, IN: Wiley, 2004.
Roland, Perry. The Music Encoding Initiative (MEI) . 2006.
<http://www.lib.virginia.edu/digital/resn
dev/mei/>
If this content appears in violation of your intellectual property rights, or you see errors or omissions, please reach out to Scott B. Weingart to discuss removing or amending the materials.
Complete
Hosted at University of Illinois, Urbana-Champaign
Urbana-Champaign, Illinois, United States
June 2, 2007 - June 8, 2007
106 works by 213 authors indexed
Conference website: http://www.digitalhumanities.org/dh2007/
References: http://web.archive.org/web/20070810143343/http://digitalhumanities.org/dh2007/DH2007.detail.html http://web.archive.org/web/20080703194728/http://www.digitalhumanities.org/dh2007/abstracts/titles.xq
Series: ADHO (2)
Organizers: ADHO