Note that as of 2020, this group has ended.
Introduction: What is Semantic Interoperability?
This group is producing guidelines, best practices and tools for opening library data, in accordance with the linked open data principles and with a special emphasis on semantic interoperability.
There are various initiatives afoot so bringing them together and making sure that interoperability is working in the same way all along the chain is crucial right now.
The concept of semantic interoperability refers to the ability of two systems to exchange data with a shared meaning. To accomplish this, the meaning of the data (or other resources) is encoded as metadata that is in turn tied to shared controlled vocabularies.
The first part of the equation is handled well in the library sector. We produce quality metadata and make use of controlled vocabularies and shared schemas and formats. So the main challenge that should be tackled is making use of shared vocabularies.
It is worth noting that the use of shared vocabularies does not necessarily mean that everyone should use the same vocabularies. A more practical alternative is that the various vocabularies used are linked together. This naturally means that the vocabularies need to be built using linked data principles utilizing dereferenceable identifiers and explicit relations. This sort of work is already under way in many places (see, for example, the various library thesauri that have been linked to the Library of Congress Subject Headings or to Wikidata). Furthermore, IFLA and RSC are also tackling these challenges from certain angles.
There are, however, various approaches to linking and to producing linked data. And even though they are generally compatible and produce results which can be used together, it takes effort to hook up into each different implementation.
The strive for semantic interoperability should not be limited to traditional publications, however. Research data is increasingly being published in reusable form and semantically interoperable metadata tying together research data, scientific publications, and library collections is an important and ultimately necessary step to take. There are undertakings, for example under the banner of the Research Data Alliance, which aim towards semantic interoperability within the scientific world. These should be considered and utilized when building interoperability for the library sector.
A library cannot be interoperable on its own. Individual decisions made locally need to be well informed. A set of guidelines, best practices, and tools for improving semantic interoperability in the library sector should be produced under LIBER’s umbrella.
There are various initiatives afoot so bringing them together and making sure that interoperability is working in the same way all along the chain is crucial. The interoperability should, if possible, encompass not only the library sector but also research data, other memory organization sectors and even more widely the public administration.