linked data

Linked Data for Libraries (LD4L) is a collaboration project of the Cornell University Library, the Harvard Library Innovation Lab, and the Stanford University Libraries, that began at 2014 with a two-year Andrew W. Mellon Foundation’s grant of $1 million dollar and this year has concluded its first research phase.

The news is that the Andrew W. Mellon Foundation has renewed this grant with $1.5 million dollar (2016-2018) and has extended it to two other projects: the Linked Data for Libraries Labs (LD4L Labs), headed by Cornell, and the Linked Data for Production (LD4P), headed by Stanford. In this way, there is a LD4L Gateway that includes the three projects.

The goal of the former LD4L project was to create triple stores called Scholarly Resource Semantic Information Store (SRSIS) model based on BIBFRAME, to link bibliographic data (MARC21 transformation), person data, and usage data, and to connect library resources with institutional and other data on the web.

On its behalf, the LD4L Labs project is focused on developing and support “tools for linked data creation and editing, the bulk conversion of existing metadata to linked data, and a common system to support initial work in entity resolution and reconciliation”.

Finally, the goal of the LD4P is to begin the transition of technical services production workflows to ones based in Linked Open Data (LOD). This first phase of the transition will focus on the development of the ability to produce metadata as LOD communally, the extension of the BIBFRAME ontology to encompass the many resource formats that libraries must process, and the engagement of the broader library community to ensure a sustainable and extensible environment.

As you see, three leading projects on the library Linked Data arena!

Enjoy it!

Andreu Sulé

University of Barcelona

In June 2014, at the American Library Association Conference in Las Vegas, Zepheira announced the Libhub Initiative, a project that aims to raise the visibility of libraries on the Web by “actively exploring” BIBFRAME and Linked Data. Its strategy is based on automatically exportation and conversion of library MARC21 records into BIBFRAME, and its transformation into Linked Data. Finally, publishing on the Web the transformed and connected content.

Nowadays, the Libhub Initiative is in an experimental phase: hearing from the broader community, gauging interest and willingness to get involved, and recruiting Active Supporter, Interested Participating Libraries, Partners, and Sponsors. During this experimental phase there are no costs to libraries because the main objective is to create a very big and collaborative database (cloud service) that allows library data to be discoverable and presented at or near the top of Web search engines page results.

It’s interesting to know that Eric Miller, Zepheira’s President, prior to founding Zepheira, led the Semantic Web Initiative for the World Wide Web Consortium (W3C) at MIT, and that in 2011 the Library of Congress contracted with Zepheira to define the way forward for moving library data into the Web. Actually, for the past 2 years, Zepheira has been defining BIBFRAME.

Another interesting information to evaluate the importance of this project is to know that amongst its sponsors and partners we can find EBSCO, Innovative, SirsiDynix, Deutsche Nationalbibliothek and Denver Public Library.

Without a doubt, Libhub Initiative is a project that deserve our special attention.

Enjoy it!

Andreu Sulé

University of Barcelona

We are very proud to announce the publication in the Revista española de Documentación Científica [Spanish Journal of Scientific Documentation] of our last paper Aplicación del modelo de datos RDF en las colecciones digitales de bibliotecas, archivos y museos de España [Implementation of the RDF data model in digital collections of Spanish libraries, archives and museums].

The article discusses how and to what extent the RDF data model is applied in major Spanish digital collections of heritage materials. With this objective, we analysed fifty-one digital repositories to determine whether they expressed their records in RDF, offered SPARQL query points searchable by external agents, and used references as property values.

Our main conclusion is that the use of RDF is unequal and excessively conditioned by the use of applications that automatically convert records into RDF triples. Actually, few of the collections analysed give SPARQL points for external queries.

Another finding is the very scarce use of linked data that connect data of the Spanish repositories with other datasets. In this sense, our recommendation is that these collections should enrich their data and define aggregation levels for generated RDF data in order to be disseminated, made accessible, and adapted to the semantic web.

Actually, nowadays we are researching ways to semi-automatically enrich data of DSpace repositories with external links to other datasets (VIAF, DBpedia, GeoNames, etc.). We’ll keep you updated.

Meanwhile… Enjoy it!

Andreu Sulé

University of Barcelona

This week we have known the National Library of the Netherlandshas adopted LodViewfor publishing its data as Linked Data.

LodView is an open source Java application that, in conjunction with a SPARQL endpoint, allows you to publish RDFdata according to all defined standards for Linked Open Data. LodView offers different data representations, one of which is a very interesting graphical view managed with the data viewer LodLive.

You can see several samples at http://lodview.it/. For example, if you choose dbpedia.org  as a “Esempi”, you can see the representation of “London” resource provided by LodView. If you click on the “view resource on lodlive” (right hand on the top page) you see the same information in a graphical way.

In fact, the LodView project was elected as one of the five finalists of the 2015 LODLAM Challenge, held in Sydney, Australia, 29-30 June.

Enjoy it!

Andreu Sulé

University of Barcelona

Last week I participated in the free webminar Implementing Linked Data in Low-Resource Conditions organized by the Dublin Core Metadata Initiative (DCMI). Presented by Johannes Keizer and Caterina Caracciolo (both membership of FAO), the webminar explained how small and medium-size institutions can publish their data on the Web. Following the example of Agri SA, a web-based resource linking agricultural datasets resulting from the collaboration of small institutions, Johannes and Caterina provided recommendations on how to proceed for making data publicly available. Special mention to their reference to Drupal as a tool to publish data as linked data. Very interesting and useful!

Enjoy it!

Andreu Sulé

Universitat de Barcelona

Few days ago, OCLC and Library of Congress have made to know a very interesting executive summary of a white paper, Common Ground: Exploring Compatibilities Between the Linked Data Models of the Library of Congress and OCLC, about the compatibility of linked data initiatives at both institutions. More specifically, between the Bibliographic Framework Initiative (BIBFRAME) at the Library of Congress and the Schema.org and BiblioGraph.net application profiles at OCLC.

Beyond the advance and future research lines of confluence between OCLC and the Library of Congress initiatives, this executive summary shows to what extent the linked data are in the forefront of the most important library organizations of the world.  How libraries are conscious that the future of information retrieval, discovery and access depends on the Web in a collaborative sense. That means they have to get out of its comfortable zone and to promote the interoperability of its data with agents of other communities. Linked data is here to stay!

Enjoy it!

Andreu Sulé

University of Barcelona