Getaneh Alemu

Linked Data for libraries Benefits of a conceptual shift from library-specific record structures to RDF-based data models

Contemporary metadata principles and standards tended to result in document-centric rather than data-centric; human-readable rather than machine-processable metadata. In order for libraries to create and harness shareable, mashable and re-usable metadata, a conceptual shift can be achieved by adjusting current library models such as Resource Description and Access (RDA) and Functional Requirements for Bibliographic Records (FRBR) to models based on Linked Data principles. In relation to technical formats, libraries can leapfrog to Linked Data technical formats such as the Resource Description Framework (RDF), without disrupting current library metadata operations. This paper provides six key recommendations for libraries and standards agencies. These include rising to the challenges and embracing the opportunities presented by current technological trends, adopting minimal requirements of Linked Data principles, developing ontologies, deciding on what needs to be retained from current library models, becoming part of the Linked Data cloud, and developing mixed-metadata (standards-based and socially-constructed) approaches. Finally, the paper concludes by identifying and discussing five major benefits of such metadata re-conceptualisation. The benefits include metadata openness and sharing, serendipitous discovery of information resources, identification of zeitgeist and emergent metadata, facet-based navigation and metadata enriched with links.

Towards a conceptual framework for user‐driven semantic metadata interoperability in digital libraries: A social constructivist approach

Purpose – With the aim of developing a conceptual framework which aims to facilitate semantic metadata interoperability, this paper explores overarching conceptual issues on how traditional library information organisation schemes such as online public access catalogues (OPACs), taxonomies, thesauri, and ontologies on the one hand versus Web 2.0 technologies such as social tagging (folksonomies) can be harnessed to provide users with satisfying experiences.Design/methodology/approach – This paper reviews works in relation to current metadata creation, utilisation and interoperability approaches, focusing on how a social constructivist philosophical perspective can be employed to underpin metadata decisions in digital libraries. Articles are retrieved from databases such as EBSCO host and Emerald and online magazines such as D‐Lib and Ariadne. Books, news articles and blog posts that are deemed relevant are also used to support the arguments put forward in this paper.Findings – Current metadata approaches ...

The Social Space of Metadata: Perspectives of LIS Academics and Postgraduates on Standards-Based and Socially Constructed Metadata Approaches

This article aims to contribute to the discussion on user-generated metadata. It presents the perspectives of eleven purposefully selected library and information science (LIS) academics and postgraduates on optimality of standards-based versus socially constructed metadata approaches and develops conceptual themes that emerged, which include the obsolescence of the Online Public Access Catalogue (OPAC) in relation to web search engines, the preference of users for searching versus browsing, and the proliferation of extant metadata standards resulting in interoperability challenges. This study also reveals the role ascribed to Web 2.0 and to semantic web technologies in libraries and the importance of enriching information objects with metadata that better conveys the various perspectives of users. Finally, the importance of underpinning metadata decisions with sound theoretical foundations is identified.

Toward an Emerging Principle of Linking Socially-Constructed Metadata

Current discussions on the use of Linked Data for digital libraries favor standards-based (expert-created) metadata approaches. The inclusion of socially-constructed (user-created) metadata such as tagging, ratings, reviews, and recommendations is scant or absent. While efforts by the worlds major national and academic libraries to release their bibliographic data (metadata) are an important step toward the adoption of Linked Data principles, as this article argues, without the metadata richness that would be obtained by including socially-constructed metadata (Web 2.0), the current efforts may result in sub-optimal utilization of the potential of Linked Data. With the aim of exploring and investigating the potential roles of metadata linking, this article presents the results of 21 in-depth interviews conducted with Library and Information Science (LIS) professionals. A constructivist grounded theory analysis of the interview data resulted in the emergence of four metadata principles—namely, the principles of metadata enriching, metadata linking, metadata openness, and metadata filtering. Considering the evidence from data analysis, it was found to be essential that a mixed-metadata approach should be considered, so that socially-constructed metadata approaches augment and enhance standards-based ones through proactive user engagement and a platform of collaboration afforded through the adoption of Web 2.0 technologies. After highlighting each of these principles, this article focuses on discussions of the principle of metadata linking.

8 – The principle of metadata filtering

Metadata filtering represents the importance of providing contextualised metadata views to users based on individual requirements. This is important in a mixed metadata paradigm as metadata contains a multitude of perspectives and interpretations from various groups of potential users. Thus, in order for users to effectively utilise metadata, and hence enhance the findability and discoverability of information objects, efficient filtering is required. This should be achieved via a flexible, contextual, personalised and reconfigurable interface.

The theory of metadata enriching and filtering

The Theory of Metadata Enriching and Filtering is underpinned by four overarching metadata principles, namely metadata enriching, linking, openness and filtering. Thus, by providing a holistic theory and demonstrating the interdependence between these principles, this book develops and presents an emergent metadata theory, which is considered to have wider implications in the way metadata is created, utilised and managed in libraries. This results from the suggested benefits of a mixed metadata approach, where the case was made that a priori (before publication) metadata and post-hoc (after publication) metadata are considered complementary to each other.

The principle of metadata enriching

The emergence of a mixed metadata approach raises the issue of addressing the key principles of simplicity versus diversity. For this new paradigm, a new Principle of Metadata Enriching is suggested. This refers to the process of enriching existing simple, a priori, expert structured, metadata with diverse, post-hoc, socially-constructed metadata. Metadata enriching indicates a continuous process of user metadata co-creation, and thus metadata is never complete. It is an ongoing process of enriching.

The Web 2.0 paradigm and the emergence of socially-constructed metadata approaches

The development of the World Wide Web, and in particular the Web 2.0 paradigm, presents libraries with opportunities to transform their services. This is characterised as a move from one-way delivery to a two-way collaboration, from closed systems to open access, from a priori (pre-defined) metadata to post-hoc (after publication) metadata. The Web 2.0 paradigm supports the re-conceptualisation of library users as metadata co-creators rather than passive consumers. This socially-constructed metadata provides libraries with a new approach to describe information objects that supports diversity and meets the challenges of growing library collections.

The emergence of socially-constructed metadata in a mixed metadata approach

Socially-constructed metadata provides a platform for two-way collaboration using current technologies such as tagging, user-reviews, recommendations and metadata crowdsourcing. This results in post-hoc metadata with each user contributing, via a network effect, to metadata diversity. However, standards-based, a priori, metadata restricts users’ involvement and ensures consistent structures and quality. These approaches though, are not necessarily in opposition. Combining both, into a mixed metadata system, unlocks the potential for metadata diversity and scalability whilst maintaining hierarchical and semantic structure. This changes the underlying principles for metadata systems, to consider variable user participation, aggregation and provenance, along with a change in the definition of quality so that it focuses on findability or usefulness.

The principle of metadata linking

Metadata enriching is the underlying principle that supports the overall aim of adopting a mixed metadata approach, which combines the strengths of both standards-based and socially-constructed metadata approaches. However, a single institution, of whatever size and resources, cannot continually enrich metadata on its own. The concept of linking is a powerful means for creating seamless connections between disparate sources of data, which supports enriching by making library metadata reusable. The principle of metadata enriching can thus only be optimally utilised if the metadata is linked in-house (internal linking – within a particular library metadata set) as well as to external metadata sources (external linking). Linking metadata within and outside the library domain is crucial for efficient metadata enriching.