Early this year, a couple of new geographic vocabularies became freely available, to be accessed by machines and humans via web services.

The “Library of Congress Authorities and Vocabularies Service” (http://id.loc.gov) made available three new vocabularies, the MARC Code Lists for: a) Geographic Areas, b) Countries (with mappings to equivalent ISO 3166 codes) and c) Languages (mapped to ISO 639-1, 639-2 and 639-5). The same site offers Library of Congress Subject Headings (LCSH) since 2009.
The main goal is to provide machine access to LC data and selected links to other vocabularies, e.g. in the context of Linked Data. The data is encoded in SKOS/RDF.
Individual concepts are accessible via a web browsing and searching interface for human users or programmatically via content-negotiation.

OCLC offers a demonstration Web Service for FAST geographic headings. The complete FAST (Faceted Application of Subject Terminology) vocabulary contains over 1,6 million authority records, reworking LCSH for easier use and application (http://www.oclc.org/research/activities/fast/).
The Web Service (RESTful standard, http://www.oclc.org/developer/services/MapFAST) takes the chosen geographic coordinates and returns a ranked list of FAST headings near the specified location. Alternative name forms, the type of geographic feature, selected events at the location and other information from the authority records is displayed. Developers can use the Web Service to develop their own applications (mobile, geolocation services).

The MapFAST demonstrator (http://experimental.worldcat.org/mapfast), using the same Web Service, is a mashup prototype that uses a Google Maps interface to present FAST Geographic authority records and, via links, allows geographic subject searching in WorldCat.org or Google Books. The prototype demonstrates a strength of the subject faceting approach of FAST over coordinated subject headings.

This information is based on press releases and mails from LC and OCLC, respectively.

Autor: Kai Eckert, Universitätsbibliothek Mannheim

Das Semantic Web – bzw. Linked Data – hat das Potenzial, die Verfügbarkeit von Daten und Wissen, sowie den Zugriff darauf zu revolutionieren. Einen großen Beitrag dazu können Wissensorganisationssysteme wie Thesauri leisten, die die Daten inhaltlich erschließen und strukturieren. Leider sind immer noch viele dieser Systeme lediglich in Buchform oder in speziellen Anwendungen verfügbar. Wie also lassen sie sich für das Semantic Web nutzen? Das Simple Knowledge Organization System (SKOS) bietet eine Möglichkeit, die Wissensorganisationssysteme in eine Form zu “übersetzen”, die im Web zitiert und mit anderen Resourcen verknüpft werden kann. (more…)

Autoren: Stefan Gradmann (Humboldt-Universität zu Berlin) & Wolfram Horstmann (Universität Bielefeld)

Abstract: Die Spezifikation ORE der Open Archives Initiative (OAI) definiert Standards für die Beschreibung und den Austausch von komplexen Einheiten aggregierter WWW-Entitäten (’resources’). Einzelne resources können multiplen Medientypen angehören und in verteilten Datenspeichern abgelegt sein. Das Ziel von ORE ist es, den Inhalt und die Binnenstruktur von komplexen Aggregationen zu modellieren, um sie im WWW maschinell verarbeitbar und austauschbar zu machen. Der Artikel gibt einen Überblick über die ORE-Spezifikation, zeigt Anwendungsbeispiele und diskutiert abschließend das Potential von ORE. (more…)

In 2005, IFLA started “Functional Requirements for Subject Authority Records (FRSAR)” as a working group in the FRBR (Functional Requirements for Bibliographic Records) family. The group was supposed to focus on subject authority data (information about subjects from authority files) and its use in a wide range of applications, the semantics, structures and interoperability issues of such data, independent from any implementation or specific context.

In June this year, the FRSAR working group published a second draft of a “Conceptual Model” [1]. It focuses on general functional requirements and the potential of subject authority data for broad sharing and use.

This draft was open for comments and review until the end of July in order for discussions by the Working Group during the IFLA 2009 conference in mid-August in Milan. Further comments can be sent to mzeng@kent.edu.

The core of the model on the aboutness of works is the following:

work <<has as subject/is subject of>> thema <<has appellation/is appellation of>> nomen

The relationships between the three entities are many-to-many relationships and bi-directional. However, in a given controlled vocabulary and within a domain, a nomen should be an appellation of only one thema.

“Thema” is defined as “any entity that can be subject of a work”. Thema includes any of the entities which are originally defined by FRBR: work, expression, manifestation, item; person, corporate body; concept, object, event, place and all other subjects “work” might have.
The entity “Nomen” and the relationships ‘Thema has appellation Nomen/ Nomen is appellation of Thema’ are new proposals of the working group. “Nomen” is any sign or sequence of signs (alphanumeric characters, symbols etc.) by which a “Thema” is known, referred to or addressed as.

Two co-chairs of the working group, Marcia Zeng and Maja Zumer, compare the FRSAR model in a paper presented at IFLA 2009 [2] with related models (new thesauri standards BS8723 and ISO 25964-1; SKOS, OWL and the DCMI Abstract Model). They conclude that these models match rather well with the FRASAR conceptual model, and thus, that subject authority data that are modeled according to FRSAD and encoded in SKOS or OWL will have a high potential of interoperability and contribute to linked data and the semantic web.

[1] IFLA (2009). Functional Requirements for Subject Authority Data (FRSAD). A Conceptual Model. IFLA Working Group on FRSAR. 2nd draft 2009-06-10.http://nkos.slis.kent.edu/FRSAR/report090623.pdf

[2] Zeng, Marcia and Zumer, Maja (2009). Introducing FRSAD and mapping it with SKOS and other models. 75th IFLA General Conference, Papers, 23-27 August 2009, Milan, Italy. Available in 5 languages.
http://nkos.slis.kent.edu/FRSAR/index.html