Workshop 3

STERNA & Rijksmuseum

Title: Semantic Web

This workshop will be about the options for cultural heritage institutions to semantically enrich their digital collections. The moderators of the workshop have extended experience with both creating semantic information systems and building end user services. During the workshop, the entire work process of creating, handling and sharing semantically rich information will be discussed and exemplified.

Content enrichtment is an important aspect of making the most of your valuable digital resources. Metadata are a key asset of heritage institutions. Through the metadata, they combine their own content with those of others and connect to central reference structures such as taxonomies, classification schemes and thesauri. The result may be heritage content that is multi modeled, multilingual and stored on servers across various countries.

How do you handle such heterogeneous cultural and scientific heritage content? That is the central question in STERNA, a best practice network with 12 European organisations with the aim to build a digital library on birds. A digital infrastructure will be built that fits the project partners, and that can be used by smaller institutions as well. This semantic infrastructure with its tools enables smaller organizations to join large scale digital library initiatives such as Europeana, even if they lack the technical expertise and/or financial means to do so. Through the STERNA tools, the metadata become available in a context that is much richer than their original local environment.

However, in many cultural instituations the basic subject metadata are poor and far from complete. Enrichment is needed to strengthen its semantic qualities prior to sharing it through tools and services. During the workshop, the Rijksmuseum Amsterdam will provide input about their experiences with enriching rather straigthforward object metadata through semantic technologies. With the use of several internally and externally developed thesauri, a prototype annotation tool has been developed to support subject matter annotation in a rich and standardised way. With this tool, cataloguers can quickly find an appropriate term from multiple heterogeneously structured thesauri. The Rijksmuseum Amsterdam will report on the planning and organisation of such a tool and on the importance of collaboration between art historians and IT researchers.

Once the metadata have been given proper semantic qualities, they can be utilised in digital infrastructures that can handle different data modelling and multilingualism, such as developed in STERNA. Key element in the STERNA approach is an innovative and highly flexible information architecture based on open source software and semantic web technologies. The end-result of creating, enriching and sharing metadata is an intricate findability layer that serves as an intermediate layer between the local content and web interfaces that make use of state-of-the-art search methods.

At the workshop, there will be demonstrations of the various technologies used from creation to presentation of heritage information in a semantic web environment, with a special eye on the viability of these technologies for smaller organizations. The participants at the workshop are encouraged to discuss how the implementation of semantic technologies would affect their organisations and their workflow, what it takes in their own organisations to enrich and share their own metadata and what the role is of the thesauri they might manage. They are also invited to discuss options to set up and maintain strategic partnerships between heritage institutions, ICT researchers and companies as an important condition for creating semantic services of value.

Program

1. Introduction to the use cases- 15 minutes

2. What is the semantic web? - 10 minutes

3. Remodelling data for findability - 15 minutes

4. Reusing web resources for subject annotation - 15 minutes

Short break - 5 minutes

5. An interface to integrated collections- 15 minutes

6. Organizational aspects - 15 minutes

7. Lessons learned and discussion - 30 minutes

------------------------------------------------------------------------------

Speakers

Michiel Hildebrand. Michiel is a researcher in the Web&Media group at the VU University Amsterdam. He is finishing his PhD at the centre Mathematics & Computer Science (CWI) in Amsterdam on interactive access to heterogeneous linked data. As a member of the MultimediaN E-Culture project he has investigated faceted browsing, annotation and text- based search on integrated data from the Cultural Heritage domain. He received his master's degree in Cognitive Artificial Intelligence (cum laude) from Utrecht University.

Geertje Jacobs. Geertje Studied Art History in Utrecht and in Florence. Since 2007 she works as project manager of the registration and digitization project "Print Room Online" in the Rijksmuseum.

Maarten Heerlien studied history at the University of Groningen and has worked for several years as an information specialist in the public domain. He currently works at the Dutch national museum of natural history Naturalis where he is project manager of the European digital library project STERNA.

Hans Nederbragt is CEO of Trezorix, an IT company that is active in the field of information architecture and that is specialized in connecting knowledge resources and in making the information in these resources findable and retrievable in a precise and integrated way.