The Musici Database is a research platform that focuses on European musicians who migrated to Italy in the 17th and 18th century. It supports the work of the international research group Musici,1 whose main interest is to compare the cultural impacts and the social roles of foreign musicians in the cities of Rome, Naples and Venice. As the sources on that historical period are dispersed over a number of archives, the researchers decided to share their data on a common platform. For the duration of the project, the Musici Database serves as a research infrastructure, which is intended to be published at the end of the project. The database will contain information on approximately 1,000 persons.
The database uses an existing repository architecture that provides a broad support of typical research workflows, which can be configured individually. This architecture is known as the Person Data Repository (PDR).2 The aim is to create a musicological database that is fed with data by a number of researchers, and is also capable of interoperating with other data sources (see Roeder 2009). As the quantity of accessible historical data can be critical for the results of the project, it is a main interest to find collaboration partners whose interest is to share research data, helping to enrich the information available to the researchers. The work is facilitated by the client software Archive Editor, which is a substantial element of the PDR.
The configuration of the Musici Database revealed some data constellations that are typical for the field of digital musicology. Further specialties were accounted for by the focus on migrant biographies. An approach to this was presented by Roeder and Plutte (2010a).
Notation and Performance
The focus on music requires some special information types to be taken into consideration. Usually, music needs to be represented on two different levels: the musical performance, which can be described as the event when one or more musicians produce music; and the composition, usually written in musical notation, which makes it easier to remember and reproduce musical ideas of higher complexity.
Both these forms of musical representation are of relevance for the study of musical biographies. While the written representation is well-preserved in the archives, the performative representation did not exist before the 1860s and can only be reconstructed by coeval descriptions and historical performance practice. As the analysis of musical performance can serve as a key to the social role of music, while the analysis of written music offers insights into creation processes, musical ideas and inspirations, the Musici Database records descriptions of musical events and instruments as well as the various creation levels of compositions from draft to print.
Since the project takes biographies of musicians and composers into consideration and compares personal motivations and ethnic backgrounds to social roles they had abroad, a basic question is how musicians integrated their personal background and their expertise in a foreign structure. Minority cultures existed in most Italian cities, thus giving musicians an opportunity to adapt to a familiar local structure and to the individual musical scene of the city. Many musicians did not only work in the field of music, and in some cases, the migrant musicians’ language skills supported second careers in other areas.
The Musici database was prepared to record personal background data and personal relationships as well as itineraries, residences and various occupations; the vocabulary was configured with a broad range of musical professions.
Quantity of Traces
A major challenge of the historical period of 1650-1750 and of the research subject of the Musici project is the relatively little quantity of sources, compared to the 19th and 20th century. Most musicians did not leave many traces on their itineraries, and the available information is more fragmentary. When a person has been identified as musician, sometimes it is not possible to find out his or her name. The Musici Database supports the combination of research data, which helps to identify identical persons by logical comparison of the available information.
Another essential problem is the lack of officially accepted proper names. The German musician Heinrich Schütz was known as Enrico Sagittario in Italy; the settlement of Aachen was known as Aquisgrana; the region of Sachsen was known as Sassonia. Much depended on the language and the cultural background of the writer. The Musici Database collocates various noun forms of the same entity and maintains the original form.
This also meets the requirements of the multilingual research group, where Italian, German and French are the principal languages, and of the final publication of the database, which is also intended to be internationally accessible.
The PDR architecture, which is used for the Musici Database, provides an environment that allows researchers to store, edit and share information on historical persons. It serves as a platform for both research and publication, and it functions as an infrastructure that is able to interoperate with other data sources.
A Data Model for Historical Persons
The PDR data model is centered on information snippets, which can be connected to corresponding persons and sources. An information snippet contains information on one or more persons, usually one descriptive phrase or a quotation from a source (described in MODS). All information can be marked-up with proper names, keywords and dates.3 It is possible to configure the mark-up vocabulary in different languages, while original text and a language-independent standard form can be maintained.
Adapting the Data Model for Musicians
The PDR data model has proven its capability to meet most of the requirements for the musicological research, as it is well suited to describe both events, seen as information that is connected to all involved persons, and compositions, seen as source objects (see Roeder & Plutte 2010b). However, for the future it might be of interest to extend the data model by other types than persons, in order to describe also places, events and relics in more detail.
Archive Editor, the client software, is available for different platforms and facilitates the process of data ingestion.4 It is based on Eclipse RCP, which allows extensibility and customization through plug-ins.5 For the multilingual Musici project, localization packages were developed for German, English, French and Italian (presented by Roeder & Plutte 2010b).
Adapted Vocabulary for Musicological Research
To meet the requirements of musicological research, a vast vocabulary of some hundred semantic tags from musical and social history was developed before the researchers started the data ingestion. This project vocabulary was developed by the researchers with Archive Editor. In addition, a number of special source types were defined by the researchers, facilitating the source recording of archive documents like diaries, letters, contracts, bills, drafts, compositions, and etcetera.
Linking to other Resources
Archive Editor allows connecting and uniting data from different projects. The PDR data model utilizes standard identifiers defined by national libraries (PND, ICCU, BNF, LCCN). Data sets bearing the same identifiers can be united by choice. This improves the interrelation of directly and semantically connected personal data inside the Musici Database as well as among other person data repositories or other data sources using the same identifiers and standards.
During the first phase of the project, a small quantity of data (about 150 musicians) that had already been collected had to be converted to the PDR data format. The project vocabulary was used to enrich the semantic layer of the data. Additionally, some services provided by PDR, like date and place identification tools, were applied during the automatic conversion.6
Standard visualization methods for the repository data are available through the PDR basic infrastructure, although other visualizations need to be developed to meet the interests of the project. These are chronological and geographical views of itineraries and settlements of musicians, as well as social network views of musicians and other social groups, such as clerics or merchants. This could also be achieved collaboratively with other musicological projects.
It is desired to connect the Musici Data Repository to other repositories and databases to exchange data. Contacts have been maintained with French and German musical database projects. The repository can exchange data through RDF and is prepared to be integrated in the Linked Open Data network.
The results of the Musici Database are twofold. Firstly, an infrastructure for musicological research was created by configuring the PDR system. It became clear that international projects require some special services which probably would not have appeared in a monolingual project. Secondly, a semantically rich database was created, which by its flexible vocabulary allows for structured searches and various forms of visualizations, configurable to correspond directly to specific research questions of the project. It is planned to publish all data together with the research results at the end of the project in December 2012.
Figure 1: The Musici Database
Körner, F. (2010). Datenarchäologie und Datenaufbereitung. digiversity, September 30, 2010, http://digiversity.net/2010/datenarchaeologie/.
Plutte, Ch. (2011). Archiv-Editor – Software for Personal Data, TPDL Conference, September 26-28, 2011.
Roeder, T. (2009). Kooperationsmöglichkeiten mit dem Personendaten-Repositorium der BBAW. Berlin: Berlin-Brandenburgische Akademie der Wissenschaften, urn:nbn:de:kobv:b4-opus-9231
Roeder, T., and Ch. Plutte (2010a). Un repositorio per i musici stranieri nell‘Italia dal 1650 al 1750. Concezione del database del progetto ANR/DFG. Deutsches Historisches Institut in Rom, January 27, 2010.
Roeder, T., and Ch. Plutte (2010b). Die MUSICI-Datenbank. Personendaten-Repositorium und Archiv-Editor. Jahrestagung der Gesellschaft für Musikforschung, Deutsches Historisches Institut in Rom und Ècole Française de Rome, November 04–05, 2010, http://www.gfm-dhi-rom2010.de/programm/hauptsymposien/hauptsymposium-iii/#AbstractPlutte
Walkowski, N.-O. (2009). Zur Problematik der Strukturierung und Abbildung von Personendaten in digitalen Systemen. Berlin: Berlin-Brandenburgische Akademie der Wissenschaften, 2009, urn:nbn:de:kobv:b4-opus-9221
zur Nieden, G. (2010). Musici. Poster-Präsentation auf dem Workshop ‘Personen – Daten – Repositorien’. Berlin, September 27-29, 2009, http://pdr.bbaw.de/workshop/poster/musici
3.The approach of this data model is precisely described in Walkowski (2009)
4.Archive Editor is available for Windows, Mac OS and Linux. http://pdr.bbaw.de/software
5.The software includes a complex help system and a software update mechanism (Plutte 2011). It was extended by a search interface that connects directly to various online research services. Scenarios to integrate the Archive Editor into other research environments through plug-ins are currently being examined.
6.The methodolocigal approach for data conversion was described by Körner (2010).