Metadata Interoperability Among LAMs
The following post was submitted by students enrolled in LIS2407 – Metadata at the University of Pittsburgh School of Information Sciences. For more information on the series, see the introductory post.
By Samantha Cabo and Sara Purifoy
As data that describes digital or non-digital resources, metadata is integral to the management of items within institutional media repositories such as libraries, archives, and museums (LAMs). These institutions have long used the MARC standard as it includes an effective metadata schema in addition to encoding, storage, and exchange systems. However, the recent growth of digitized and born-digital resources complicates the metadata creation process across information communities. This is particularly prevalent in terms of metadata interoperability or “the ability of two or more information systems to exchange metadata with minimal loss of information,” which is beneficial because it can be used by interconnected computer systems to communicate and transfer data (Neiswender, 2009). Web-scale interoperability of metadata eliminates the isolation factor inhibiting information systems from maintaining and integrating their local database systems; descriptive information can now be stored in the form of a Hypertext Transfer Protocol (HTTP) Uniform Resource Identifier (URI), along with institutional holdings and access information (Seeman & Goddard, 2015).
The move toward implementing linked data within LAM databases is gaining momentum, mostly because of the varying degrees of inconsistency among databases and catalogs. However, with the constant uncertainty of what standards the future may hold regarding metadata interoperability, this transition has largely been theoretical rather than practical. Luckily, there are LAMs who have taken the plunge and emerged triumphant, or well on their way to success. Although this analysis references only a few of these examples, it aims to discern what best practices may look like for LAMs interested in bridging the gap.
One of the biggest challenges LAMs face when it comes to interoperability is the issue of systems integration. LAMs are generally used to operating as silos, each with a localized catalog or database, and little time or manpower to dedicate to importing and reformatting records from other institutions. To facilitate maintenance and integration of local systems, and support the need for cross-domain integrated searches, time should be invested in the development of excellent metadata (to be adopted across all LAMs) in lieu of updating local records that only serve their parent institution and users (Seeman and Goddard, 2015). As this metadata is developed, domains will become more linked across the web. Other institutions, projects, and LAMs will begin to reuse these links, or URIs, thereby improving the interoperability of data within the domain. On a high semantic level, this means that both humans and computers must be able to read and interpret the metadata correctly across platforms (Haslhofer and Klas, 2010). A call to arms aimed more particularly at LAMs, however, is to advantageously use this process as an opportunity to expose their unique resources and special collections not only to one another, but also to a wider group of users (Seeman and Goddard, 2015).
In order to gain wider exposure for LAM resources, structural and semantic heterogeneities, or the differences between metadata models across institutions, need to be eliminated. However unlikely, this could be achieved if a powerful organization such as the World Wide Web Consortium (W3C) were to recommend a standardized metadata model to be used across LAMs. Haslhofer and Klas suggest that metadata mapping is the best technique for achieving metadata interoperability (2010). Metadata mapping is the process by which metadata from one repository is successfully copied, translated, and then used within a second sister repository. However, metadata interoperability requires a mapping schema that can account for all technical levels of metadata: M2 schema definition languages, M1 metadata schemes, and M0 metadata instances (Haslhofer and Klas, 2010). If a metadata mapping schema can incorporate techniques that recognize and address these instance levels, it is the best (albeit most technical, time consuming, and costly) bet for a consistent and comprehensive solution.
Although the transition to linked data may seem daunting and clouded in risk, it is up to LAMs to invest in their expertise in metadata generation and management, and to look to one another for help in this effort. There has been great technological change and innovation in a brief period of time. LAMs must consider where they want to stand in the new information society, and how they are going to get there. For LAMs to survive, their practices, services, and resources need to flow to their users on the web. “[T]hey must be of the web not just pass through the web but live and interact with web resources” (Zengenene, 2013). The mission of LAMs has always been the unbiased provision of information and resources; Zengenene (2013) quotes Sir Tim Berners Lee (creator of the World Wide Web) as saying, “data isn’t worth much until it’s free…freed from the silos in which it is locked up, and used in a mash-up that creates valuable new resources for you and others.”
Haslehofer, Bernhard and Wolfgang Klas. 2010. “A Survey of Techniques for Achieving Metadata Interoperability.” ACM Computing Surveys, 42(2): 1-37. DOI:10.1145/1667062.1667064.
Neiswender, C and E. Montgomery. 2009. “Metadata Interoperability–What Is it, and Why Is It Important?” In the MMI Guides: Navigating the World of Marine Metadata. http://marinemetadata.org/guides/mdataintro/mdatainteroperability. Accessed June 15, 2015.
Seeman, Dean and Lisa Goddard. 2015. “Preparing the Way: Creating Future Compatible Cataloging Data in a Transitional Environment.” Cataloging & Classification Quarterly, 53(3-4): 331-340. DOI: 10.1080/01639374.2014.946573.
Zengenene, Dydimus. 2013. “Global interoperability and linked data in libraries.” New Library World, 114(1-2):84-87. http://dx.doi.org/10.1108/03074801311291992.