JCDL'05 presentation on The NEUMES Project

Our paper was placed in the Conference session on Creating Information Representations for the Humanities. There were several concurrent sessions: this one was not the largest, but it was well attended. Notable among those in the audience were representatives from the U.S. Library of Congress and OCLC (Online Computer Library Center).

The framework of my talk was a PowerPoint presentation with heavy emphasis on graphics. I chose not to attempt to cover the entire paper, since it is dense with material, and presenters were allowed only a half hour. Instead, I focused on a few of the most important points, as follows.

From the viewpoint of data-encoding, Gregorian chant is best classified with other types of chant that were written in neume notations -- not with classical or 'art' music. Early forms of chant notations are fundamentally incompatible with modern musical notation, and they pose different data problems. Furthermore, Gregorian chant is properly viewed as the culmination of medieval chant literature, not as the beginning of modern music.
Western (Latin) and Eastern (Byzantine) neume notation systems ought to be covered by a unified data representation, because they are related by common, liturgical functions and by some common texts.
A distributed or decentralized digital library is a better architectural model than building a centralized database at one site. This model allows greater sharing of the workload of transcription and greater freedom of access to the data, and it is more likely to be self-sustaining economically.

During the presentation and the question/answer time, I tried to engage the audience's interest in the topic so they might be motivated to read the full paper. I thought the presentation was well received, and I was impressed by how insightful some of the audience questions were.

A conference like this is, of course, a 'two-way street' for information exchange. My own perspective on digital libraries was broadened by some of the presentations that I attended. My thinking was especially stimulated by a keynote address, "Digital Libraries Initiatives: What I Learned (and Didn't) in 10 Years," by Hector Garcia-Molina, who is a professor of computer science at Stanford University and whose former students founded Google.com and other important digital-library initiatives [Dr. Garcia-Molina's PowerPoint slides can be downloaded here].

- L.B.

The Fifth ACM/IEEE Joint Conference on Digital Libraries (JCDL'05) took place in Denver, Colorado, 7-11 June 2005. The theme of this year's conference was "Cyberinfrastructure for Research and Education." The conference was well attended by representatives of industry, universities, and non-profit organizations. The attendees generally came from two main areas of interest: those concerned with digital library content; and those concerned with software infrastructure and standards. Louis Barton, John Caldwell, and Peter Jeavons submitted a paper for the conference [see, "Project Reports" section of this website], which was accepted by peer review and was published in the Conference Proceedings. Louis Barton attended the conference in order to present this paper. What follows, below, is his report on the presentation.
Louis Barton delivering a PowerPoint presentation about the NEUMES data representation at JCDL'05.	There was active discussion during the Conference session on Creating Information Representations for the Humanities.
Our paper was placed in the Conference session on Creating Information Representations for the Humanities. There were several concurrent sessions: this one was not the largest, but it was well attended. Notable among those in the audience were representatives from the U.S. Library of Congress and OCLC (Online Computer Library Center). The framework of my talk was a PowerPoint presentation with heavy emphasis on graphics. I chose not to attempt to cover the entire paper, since it is dense with material, and presenters were allowed only a half hour. Instead, I focused on a few of the most important points, as follows. From the viewpoint of data-encoding, Gregorian chant is best classified with other types of chant that were written in neume notations -- not with classical or 'art' music. Early forms of chant notations are fundamentally incompatible with modern musical notation, and they pose different data problems. Furthermore, Gregorian chant is properly viewed as the culmination of medieval chant literature, not as the beginning of modern music. Western (Latin) and Eastern (Byzantine) neume notation systems ought to be covered by a unified data representation, because they are related by common, liturgical functions and by some common texts. A distributed or decentralized digital library is a better architectural model than building a centralized database at one site. This model allows greater sharing of the workload of transcription and greater freedom of access to the data, and it is more likely to be self-sustaining economically. During the presentation and the question/answer time, I tried to engage the audience's interest in the topic so they might be motivated to read the full paper. I thought the presentation was well received, and I was impressed by how insightful some of the audience questions were. A conference like this is, of course, a 'two-way street' for information exchange. My own perspective on digital libraries was broadened by some of the presentations that I attended. My thinking was especially stimulated by a keynote address, "Digital Libraries Initiatives: What I Learned (and Didn't) in 10 Years," by Hector Garcia-Molina, who is a professor of computer science at Stanford University and whose former students founded Google.com and other important digital-library initiatives [Dr. Garcia-Molina's PowerPoint slides can be downloaded here]. - L.B.