Tuesday, September 4, 2012

Linked Open Data

Already during our first seminar about Semantic Web and Linked Open Data organised by Digisam, it stood clear that there is a lot about these concepts that needs to be clarified. What is really needed for making digital cultural heritage information open and a part of semantic web? Today there are few practical examples of linked open data but instead a lot of questions, discussions and visions.

Digitisation has, until now, been focused on reaching a critical mass of information on the Internet and building content-rich portals and thematic applications that attract users' interest. Today, when the digital cultural heritage is increasing in volume, cultural heritage information must be presented in a way that optimises the potential of the information. This means for example that the online search for information needs to move from users getting only accidental results to providing them with structured results that are based on individual search needs and search patterns. It is important that the the increasing amount of cultural information does not overwhelm the users so that they stop looking if what they want is not among the top search results. We therefore need more effective mechanisms for linking information, building applications and performing searches in the cultural heritage information as relevant and useful as possible, and that is exactly what linked open data is about.

There seems, however, to be some disagreement about the concept of linked open data and what the term Linked Open Data really means. Using the Internet to distribute data does not automatically mean that data becomes available to all web users (for example e-mail communication is distribution of data over the internet). Similarly, the data that is made ​​available on the Internet is not "open" by default just because it's there. Publishing information in the form of open data means that data is licensed for free use and it is easier to make structured links to other information.To be "open", information must be lifted from its original context. The Open Data Handbook published by the Open Knowledge Foundation describes the legal, social, and technical aspects of linked open data, and provides the following definition: " Open data is data that can be freely used, reused and redistributed by anyone - subject only, at most, to the requirement to attribute and sharealike".

Since last year I have been involved in an EU project called Linked Heritage - Coordination of Standards and Technologies for the enrichment of Europeana. The main objective of Linked Heritage is to provide the cultural heritage portal Europeana with new data from both the public and private sector, to improve the quality of the metadata fed into Europeana and to improve the search, access and use of content in Europeana. The work in Linked Heritage is based largely on the results of a previous EU project called Athena. Linked Heritage has recently published "Your terminology as a part of the semantic web recommendations for design and management", (which can be downloaded from the project website). It is aimed primarily for people working in European cultural heritage institutions (especially museums) with recommendations on terminologies and multilingualism).

For improved search and access we need to use established standards, methods and formats. It is important to see the overall picture - although there are cultural heritage materials so peculiar that a custom-made solution for standards or formats at first glance appears to be preferable, it is often not preferable in the long run. The use of accepted standards makes it easier to develop links with data models and to take advantage of existing development of work with standardised data.

The way in which cultural institutions digitise and manage their digital cultural heritage data is the most important factor for whether their data can be a part of the Linked Open Data, both in Europeana and the Semantic Web. With various investments in digitisation already underway, it is urgent to start a broader discussion on the issue.

/ Sanja Halling, Digisam

No comments:

Post a Comment