The thousands of chant manuscripts extant from medieval Europe provide a rich legacy of the liturgical traditions in monasteries, cathedrals and parish churches. The melodies to which the sacred texts of the Mass and the Offices were intoned are represented by a great variety of neumatic symbols in dozens of distinct notational types. The complexities involved in analyzing the surviving manuscript sources are remarkable.
Digital Transcription is a translation of the semantic content of an artefact to a digital data stream. It is not a replication or facsimile of the artefact, but rather, a representation of the artefact's symbols in a form that can be processed by computers.
Attempts at the transferrence of chant notation to a digital format has, for several decades, assumed many forms. Some chant scholars have used modern notation programmes (like Finale or Sibelius); others have devised fonts which display square neume shapes or represent melodic pitches. There are obvious benefits to many of these methods. Chant musicologists generally have adopted those which have provided the fewest obstacles in their research or those which have been available to them (i.e., a font within an existing word-processing programme). However, few of these formats effectively cross computer platforms, can be used in alternate applications (i.e., word-processor to database to website, to printed publication), or contain complete information about each chant (e.g., pitches, melodic motion, ligation, manuscript information, etc.).
The way in which the content of a transcription is represented as data (i.e., the data representation) ultimately limits the kinds of processes that can be effectively performed on the content. As is often the case in computer applications, the availability of new, automated tools that have radically different capabilities, as compared to manual procedures, may occasion a reconsideration of techniques that were worked out for hand-processing.
Digital transcription is the crucial point of contact between humanities scholars and a computer software system. Digital photographs of manuscript leaves (of which many are online) serve one's need for consulting the original notation. Digital transcriptions complement images of chants, and those which are linked to online source images, such as NeumesXML transcriptions, provide the researcher with engaging tools for study.
The "NEUMES project" (Neumed and Ecphonetic Universal Manuscript Encoding Standard) proposes another option for chant scholars as a system for the transcription of medieval notations. This project is directed towards creating a software infrastructure for digital transcription of medieval western and Byzantine chant manuscripts. Transcriptions of individual chants in the NEUMES data language1 form the base data for the Project, and XML metadata describe the chants and their sources.
The NEUMES Project can be found on the Internet at http://purl.oclc.org/SCRIBE/NEUMES/. It is funded at The University of Oxford by the Eduserv Foundation.
 In this document, NEUMES (in italics) refers to the data representation created by the "NEUMES Project", that is, the data language which consists of characters (e.g. "&punctum;") for transcribing a chant into a digital format. NeumesXML, on the other hand, is a 'wrapper' for the NEUMES character data which provides metadata (in the form of tags) that include description and editorial commentary about the content (i.e., about the chant).
The NEUMES data representation allows chant researchers to create, share, and analyze their transcriptions in a permanent, interoperable data format. A uniform means by which the content of medieval manuscripts can be viewed, searched and analyzed by computer offers enormous benefits to chant scholars and researchers in related fields.