On April 29, 2014, I gave a presentation at the second international Maria Szymanowska et son temps Symposium in Paris, France. The event was sponsored by the Polish Academy of Sciences and organized by Elżbieta Zapolska-Chapelle (Board President) of the Société Maria Szymanowska. My presentation was entitled: “Szymanowska Scholarship: Ideas for Access and Discovery through Collaborative Efforts,” and was meant to present examples of how libraries, archives, and other cultural heritage institutions have been working to make their collections accessible, discoverable, and open to their users. I briefly explored ways in which the Szymanowska scholarly community could think about future research, which would make use of open data, linked open data, and tools associated with digital scholarship. My slides can be viewed within this post or directly via Google Drive.
Here is the main body of my presentation: “Szymanowska Scholarship: Ideas for Access and Discovery through Collaborative Efforts”
My interest in Maria Szymanowska’s music and life began during my undergraduate studies, when I discovered that she was largely an unknown figure to scholars in the United States. As a pianist, I had performed several of her compositions, including Le Murmure and Vingt Exercices et Préludes, and by the time I was in graduate school, I had decided to conduct my research on her life and career as a pianist and composer. I knew that there was a need for a resource, which could act as a guide to primary and secondary source materials, many of which were not located in the United States. My main interest was in those materials held by archives, libraries, and museums, which can be elusive and difficult to track down, even by experienced scholars. Why was it and largely still is, difficult to locate and access these primary sources? The reality is that there have always been and continues to be a percentage of primary source materials, which are never fully indexed or documented (contextually) in finding aids or library catalogs, for various reasons, but often due to lack of funding or staffing. One of the ways in which librarians and music scholars have addressed this issue is through the creation of resources, such as bibliographies, indexes, thematic catalogs, or other research guides, which focus on primary and secondary sources made up of early and modern musical editions, discographies, literature, and other relevant materials, which are then identified and annotated for use by other scholars, students, and the public.
While completing my research for my thesis, I realized that there was a need for such a resource on Szymanowska and this led me to write a bio-bibliography, which was published in 2010.
As a librarian and scholar, I think that bio-bibliographies are still important for scholars and students doing research on subjects or in areas that are not well represented in traditional scholarship, however, since I first started my research on Szymanowska in the early 2000s, I have realized that there needs to be a greater emphasis placed on enabling access and discovery of materials, especially primary sources, through modes beyond print. Book-length annotated bibliographies are still being published and are largely done in print format with e-book equivalents. Although these are useful in identifying unique or unknown primary sources, as well as secondary literature, they continue to present this information in a structure that is static. [Slide 2: A few definitions]
While the author of a print book can provide narrative and contextualize the material presented to the reader, the information is static and confined to those pages. It is unable to demonstrate relationships between the identified sources and subjects, or create networks beyond the page. Due to these limitations, it is important that librarians and scholars look beyond the print mode and create, either in addition to, or in lieu of print, digital resources that are built on linked open data, which many international libraries, archives, and museums have begun to implement. Here is a brief video created by Europeana (which I’ll discuss shortly) about linked open data and its importance.
My talk today will focus on two things: first, I will discuss how libraries, archives, and museums (LAMs) are beginning to make information or data, as well as the content itself, available to academics and the greater public, through the use of linked open data, which enriches the data and makes it visible. Second, I will explore and propose several ways in which scholars can use this data or digital content to become curators and collaborators with LAMs. I will also present several examples of projects, which are creating digital resources or building applications on this data.
Paper-based materials = paper-based networks, so how can we make it easier to access the data, which is described in finding aids and kept in archival boxes? And how can we identify or visualize relationships within these networks that may then lead us to additional information? [Slide 4: Paper based materials = Paper based networks]
A traditional finding aid, whether in print or digital form, can only provide us with a static layer of information related to our subject or person. It is then up to the researcher to continue digging through archival records, in order to find connections that may exist within or across collections. Rather than solely relying on these paper trails, what if we were able to access this hidden archival data using digital tools built on linked open data? Libraries, archives, and museums employ standards in order to make it easier to access and discover materials critical to our research. Using information already present in our catalog or discovery systems, such as authority headings (i.e. Szymanowska, Maria), which allow you to identify persons or works, you can then begin to connect records together and form a network within your collection or across multiple content providers.
For example, the New York Public Library Archives & Manuscripts department is beta testing a digital tool, Term Explorer, which allows a researcher to visualize relationships or networks between a person, subject, topic, or work.
This first slide (slide 5) positions content related to 19th-century musician Teresa Carreño to the Carl V. Lachmund collection at the New York Public Library and the next slide (slide 6) represents the connection between Carreño and Eugen d’Albert (her third husband) in relation to the Lachmund collection.
Adding linked open data into the mix allows cultural heritage portals, such as Europeana and the Digital Public Library of America (DPLA) to serve up content that is being created by individual galleries, libraries, archives, and museums (GLAMs) on a national and international level. [Slide 7: Europeana; Slide 8: DPLA]
In addition to acting as a portal, Europeana and the DPLA, each provide new ways for researchers to engage with the content that has been digitised or described.
For example, the DPLA uses temporal (dates) information to allow researchers to display content relevant to their search in a timeline view, which promotes further discovery of related content. For example, my search for content by or about Adam Mickiewicz uses the publication dates for works printed since 1829 and visualizes them on a timeline with bars of varying length, which represent the number of works published or written in a given year. You can then further explore the content by selecting the bar for a specific year, such as 1890, and gain access to the record with digital content (when available). Both Europeana and DPLA use linked open data and have an open API (application programming interface), which allows researchers and the public to re-use the content created by GLAMs in innovative ways.
Beyond the library (archive, gallery, or museum) – What does linked open data mean for scholars?
The increase in digitization of materials does not necessarily make it easier to do research, discover or access these sources. In their article, “Using Linked Open Data to Enhance Subject Access in Online Primary Sources,” Thea Lindquist, Michael Dulock, et al, point out the various barriers, which include:
- decontextualization of sources (inadequate metadata for searching according to subject, time, period, and geographical area);
- impenetrability of institutional databases; and
- sheer magnitude of results.
One of the ways around these barriers, then, is to use data being created by libraries and contributed to services, such as VIAF (Virtual International Authority File) in order to enhance records and link them using linked open data. [Slide 11 – 13: VIAF]
VIAF is a service that combines all known name authority files into a single record by matching and linking them across the web. For example, an authority record for Maria Szymanowska, will show variant spellings of her name, each linked to authority records at a specific international institution (identified by a flag).
The same idea is applied to works, hence, her compositions are identified and linked to authority records. The majority of VIAF authority headings for Maria Szymanowska are linked to modern musical editions, and secondary literature or works, therefore there is still a need for primary source records held across international institutions to be created or contributed along with linked open data to VIAF. These primary sources, such as her Divertissement pour le pianoforte avec accompagnement de violon, held by Wellesley College, could then be accessed and discovered more easily. Once records are enhanced with linked open data, they can be connected across the Web and more easily discovered. [Slide 14: Divertissement]
Within the last seven years, American and European libraries, in particular the Library of Congress, Bibliothèque national de France, Biblioteka Narodowa (Warsaw, Poland), and Deutsche National Bibliothek, have worked towards making their bibliographic data accessible through linked open data. There is still much work to be done before all of this data is linked and discoverable, however the effort from these and other institutions is becoming more visible. For example, the BnF has made their data available as an open data project on data.bnf.fr, giving their users access to data about an author, work, or topic through one portal. [Slide 15: Data.bnf.fr]
For example, when I search for the author: Maria Szymanowska, I am brought to a page which provides biographical data, as well as information about documents by or about her, which are held at the BnF. You’ll also see that external references or sources are listed on this page.
The user can then access the item record through the BnF catalog from this data portal. One of the things that would make these records even more useful to scholars and users interested in Maria Szymanowska would be to add digitized content that can be linked. Access to digital images of early editions, correspondence, or manuscripts would allow people to view these sources, share them, and even possibly re-use them for scholarly purposes. [Slide 16: Data.bnf.fr; Slide 17: Ballade par Mme. de. St. Onge]
In this part of my talk, I will explore several ways in which scholars can become curators and engage in collaborative research. In order for there to be greater access to and discovery of Szymanowska sources, scholars should collaborate with libraries, archives, and museums on several levels, which can include identifying collections or materials in need of attention, providing additional context to materials through curation in online exhibits, as well as proposing collaborative research projects, which could lead to grant funding. This type of work can ensure that sources, such as correspondence or diaries are digitized, described, indexed and made available through portals, such as Europeana, or re-used in digital projects.
For example, in Nineteenth Century Disability: Cultures & Contexts (a project led by Karen Bourrier from Boston University), primary source texts and images on physical and cognitive disability from different library and archival collections are curated and presented together. [Slide 18 – 19: Nineteenth Century Disability]
In Documenting Teresa Carreño, a project I am leading, primary source texts and images related to her musical career are curated, transcribed, linked to original records, and visualized spatially on a map. [Slide 20 – 22: Documenting Teresa Carreño]
The BnF open data project, mentioned earlier, offers data for download, which can be used to create visualizations specific to one person, topic, or work, or combined with other data in order to represent relationships or connections across collections. For example, using data from the BnF, it is possible to explore a composer’s oeuvre on a timeline, such as this one for Chopin.
One of the ways we can already visualize networks linking sources related to Szymanowska is through a visualization tool, such as LODLive. [Slide 24: LodLive] LODLive is a project, which demonstrates the use of linked open data standards and proves that resources published according to semantic web standards are easy to access and understand. Entering the VIAF URI for Szymanowska into http://en.lodlive.it/ will create a central node, which holds the information from the VIAF record. From this node, other URIs and linked open data connect to form a network.
The user will see that the different nodes represent data linked to records from providers, such as Wikipedia or the Bibliothèque national de France. This type of visualization is not comprehensive and will only show connections that exist which meet the linked open data standards. However, it gives us a starting point for thinking about how not only LAMs, but also scholars, can participate in improving existent data and context about persons (such as Szymanowska, Adam Mickiewicz, or Johann Wolfgang von Goethe), works, and topics.
Another example, based on this same concept, is Linked Jazz, a project, which is using linked open data to enhance the discovery and visibility of digital cultural heritage materials. Data from digital archives of jazz history has enabled this project team to visualize relationships between jazz musicians. [Slide 27 – 28: Linked Jazz]
This symposium has inspired me to imagine that perhaps this kind of visualization of Szymanowska’s social network would allow us to demonstrate the important, as well as peripheral, relationships with persons, such as Bertel Thorvaldsen or Mickiewicz. A project could be undertaken, such as Mapping the Republic of Letters (Stanford University), which would allow us to build a collaborative database of sources about Szymanowska’s social network, and also visualize it. [Slide 29: Republic of Letters]
Content from LAMs who are not implementing URIs or linked open data will not be represented in these types of networks. Therefore, it is important to raise awareness about the capabilities of linked open data and to collaborate with librarians and technology staff to create projects, which are centered around standards that encourage open data. Such collaborative efforts can ensure that Szymanowska sources at various institutions are connected and made not only accessible, but discoverable, by academics and the public. As a lesser known figure in musical history, Szymanowska is not as well-represented in VIAF or other open data projects, as other musicians may be (i.e. Chopin or Liszt).
Imagine being able to access data about Szymanowska’s musical output, correspondence, literature, as well as digital content, through one portal, such as Europeana. This data could then allow users to re-use content and represent the data in different ways, including timelines, maps, or networks, which could display connections or relationships between Szymanowska and other artists or prominent figures of her time, such as Goethe or Thorvaldsen.
As more institutions begin to make their raw bibliographic data public and implement linked open data, the networks and relationships between persons and collections will become more enhanced and richer, allowing scholars and the public to more easily access content, which had been previously hidden within institutional database frameworks, as well as discover additional content through the new connections created across institutional collections. It is thus important, that librarians, archivists, scholars, and IT professionals work together on recommendations, as well as actual projects, in order to ensure that research and scholarly needs are being met.
It is especially important that the legacies of musical figures, such as Szymanowska, who may not hold a central place in the western music canon are preserved across archival collections, libraries, and museums. Moreover, it is essential that descriptive data about sources related to Szymanowska’s life and profession is made re-useable to encourage future research and scholarship, but more importantly, accessible, in order to preserve the cultural heritage and memory of a notable Polish composer and pianist.
[I would like to thank the University of Connecticut Libraries, Polish Academy of Sciences, and Société Maria Szymanowska for the opportunity to attend and present at this important event. This presentation will be revised and printed in the proceedings of the 2nd International Symposium.]
Correspondence between Bertelt Thorvaldsen and Maria Szymanowska (written between 1824-1831). Thorvaldsens Museum, http://arkivet.thorvaldsensmuseum.dk/dokumenter/korrespondance/szymanowska-maria.
Digital Public Library of America (DPLA), http://dp.la/info/developers/.
Europeana Labs, http://labs.europeana.eu/.
Lindquist, Thea, Michael Dulock, Juha Törnroos, Eero Hyvönen, Eetu Mäkelä. “Using Linked Open Data to Enhance Subject Access in Online Primary Sources,” Cataloging & Classification Quarterly Vol. 51, Iss. 8, 2013. http://www.tandfonline.com/doi/abs/10.1080/01639374.2013.823583.
Linked Data, http://www.w3.org/standards/semanticweb/data.
Linked Data Tools, http://www.linkeddatatools.com/.