Managing Metadata in Web-scale Discovery Systems

This book shows you how to harness the power of linked data and web-scale discovery systems to manage and link widely varied content across your library collection.

Libraries are increasingly using web-scale discovery systems to help clients find a wide assortment of library materials, including books, journal articles, special collections, archival collections, videos, music and open access collections. Depending on the library material catalogued, the discovery system might need to negotiate different metadata standards, such as AACR, RDA, RAD, FOAF, VRA Core, METS, MODS, RDF and more.

In Managing Metadata in Web-scale Discovery Systems, editor Louise Spiteri and a range of international experts show you how to:

• maximize the effectiveness of web-scale discovery systems
• provide a smooth and seamless discovery experience to your users
• help users conduct searches that yield relevant results
• manage the sheer volume of items to which you can provide access, so your users can actually find what they need
• maintain shared records that reflect the needs, languages, and identities of culturally and ethnically varied communities
• manage metadata both within, across, and outside, library discovery tools by converting your library metadata to linked open data that all systems can access
• manage user generated metadata from external services such as Goodreads and LibraryThing 
• mine user generated metadata to better serve your users in areas such as collection development or readers’ advisory.

Readership: The book will be essential reading for cataloguers, technical services and systems librarians and library and information science students studying modules on metadata, cataloguing, systems design, data management, and digital libraries. The book will also be of interest to those managing metadata in archives, museums and other cultural heritage institutions.

1. Introduction: the landscape of web-scale discovery - Louise Spiteri
2. Sharing metadata across discovery systems - Marshall Breeding, Angela Kroeger and Heather Moulaison Sandy
3. Managing linked open data across discovery systems - Ali Shiri and Danoosh Davoodi
4. Redefining library resources in discovery systems - Christine DeZelar-Tiedman
5. Managing volume in discovery systems - Aaron Tay
6. Managing outsourced metadata in discovery systems - Laurel Tarulli
7. Managing user-generated metadata in discovery systems - Louise Spiteri

"It was interesting to learn about the leading projects, primarily from national libraries, and see some real examples of linked open data sets. Big data, the Semantic Web and BIBFRAME are also covered but nothing gets too complicated, making it a great introduction to this area."
- Catalogue and Index 
 

Dr Louise Spiteri is Director of the School of Information Management, Dalhousie University, Halifax, Nova Scotia, Canada. Dr. Spiteri has published extensively in the areas of social tagging, folksonomies, social discovery systems, and library cataloguing. Dr. Spiteri teaches in the areas of information management, metadata, records management, cataloguing, classification, and taxonomies.

Contributors:

Marshall Breeding, independent consultant, speaker and author

Danoosh Davoodi, information science and technology professional

Christine DeZelar-Tiedman, Manager of the Archives and Special Collections Metadata Unit, University of Minnesota Libraries, USA

Angela Kroeger, Archives and Special Collections Associate, University of Nebraska at Omaha, USA

Heather Moulaison Sandy, Assistant Professor, iSchool, University of Missouri, USA

Ali Shiri, Professor, School of Library and Information Studies, University of Alberta, USA

Laurel Tarulli, Library and Information Services Manager, Sacred Heart School of Halifax, Canada

Aaron Tay, Library Analytics Manager, Singapore Management University

1. Introduction: the landscape of web-scale discovery – Louise Spiteri

2. Sharing metadata across discovery systems – Marshall Breeding, Angela Kroeger and Heather Moulaison Sandy

As we increasingly use web-scale discovery systems to help clients find a wide assortment of library materials, how do we manage the different metadata schemes used to describe these different materials? This involves the concept of mapping within and across collections, depending on the scope of the collection. How do we manage these mappings to provide seamless discovery? How do we create, maintain and share records that reflect the needs, languages and identities of culturally and ethnically varied communities?

3. Managing linked open data across discovery systems – Ali Shiri and Danoosh Davoodi

By converting our library catalogue records to linked open data, people can discover library resources across different discovery tools, within and external to the library, which take advantage of full text search and Semantic Web technology and standards (e.g., RDF). This involves going beyond integrated discovery systems, which bring together varied resources within libraries, to what can be called ‘pan discovery systems’, which allow us to link our metadata records across different discovery systems. In a library context, practical linked open data applications are as of yet few in number, of limited maturity, and relatively untested. There are different semantic options available for encoding catalogue records in RDF, and to date there is little consistency in metadata schemas selected (beyond the use of RDF itself). This semantic heterogeneity complicates interoperability.

4. Redefining library resources in discovery systems – Christine DeZelar-Tiedman

Given the increased scope of web-scale discovery systems, it is important to reconsider and redefine what we mean by library resources. Should the web-scale discovery systems link also to human resources in the library, such as expertise of staff, research projects conducted under the purview of the library, and related social media links (e.g., tags, hashtags and so on, related to relevant content)? Should we continue to rely on the library website to provide access to services, as distinct from the catalogue, which provides access to collections? If we want truly integrated discovery systems, should we consider making these connections seamless – services and collections accessed via the catalogue? What would this mean to managing metadata? How would we incorporate metadata about library services in the catalogue?

5. Managing volume in discovery systems – Aaron Tay

The well established measures of recall and precision are becoming increasingly relevant in web-scale discovery systems. Given the way that most people search, which is the simple keyword box that searches all text anywhere in the record, web-scale discovery systems will lead to increasingly large recall as we provide access to more linked items. Do we need to be careful about the sheer volume of items to which we can provide access via web-scale discovery systems? Do we want these systems to become another Google, where precision of results is not always as accurate as we would like? Are we too obsessed with the notion of providing access to everything at the expense of the quality of the results?

6. Managing outsourced metadata in discovery systems – Laurel Tarulli

The increasing reliance of vendors to provide metadata and indexing is something we need to examine. Vendors who provide both metadata and discovery layers might not share their data with another discovery vendor; so, for example, if EBSCO produces metadata for its records but your institution does not use the EBSCO web-scale discovery system, you may not receive any metadata records. How do we negotiate metadata outsourcing for web-scale discovery systems, and how do we ensure the provision of accurate and comprehensive metadata?

7. Managing user-generated metadata in discovery systems – Louise Spiteri

Web-scale discovery systems allow for the blending of metadata generated by actual library clients, and reviews or tags imported from outside sources such as Goodreads and LibraryThing. How are we mining this information? Are we simply uploading these data into our discovery systems and leaving them there purely for viewing purposes, or are we actually mining the data they contain to help us connect with our users, help build more user-centric vocabularies, create community-based reading suggestions and so forth? Our focus tends to be on managing library metadata that we create, but how are we managing user-contributed metadata, including any we import from services such as Goodreads and LibraryThing? Are we looking at trends revealed by this social data? Are we using these metadata to improve library-contributed metadata, connect with our clients, and create shared communities?