OCLC’s Regional Council Meeting for members and customers in EMEA (Europe, Middle East, Africa) took place over the 26 and 27 February in Strasbourg this year.
I had the opportunity to attend as I was asked to contribute to a workshop before the start of the conference proper on CONTENTdm, OCLC’s digital asset management system. We have been using this successfully over the last few years for our My Leicestershire History Website. The workshop was very practical with a number of people like me outlining case studies of using CONTENTdm. We all agreed that it is an easy system to get to grips with but, like any system, has its limitations – one of which is “customisability” (especially if you are using the hosted service).
The theme of the conference was sharing data which – given the context – largely, though not exclusively, meant sharing library metadata in new and more efficient ways to increase the visibility of ‘library resources’ on the Web and provide new services.
The programme opened with an excellent presentation from Jean-Baptiste Michel of Harvard University on how he and colleagues have used a huge dataset derived from Google Books to analyse the changing prevalence of words. The resulting dataset is well known and can be queried online. It vividly illustrates the power of large scale data analysis – in this case, using data not originally created for this purpose. It also illustrates how important re-use of data is to enabling new kinds of research. The team at Harvard are now moving on to use Open Library, JSTOR, the New York Times and arxiv.org as further sources of word occurences for analysis. Michel saw potential for libraries to develop services in this kind of area – providing tools and support to enable researchers to analyse data in new kinds of ways.
Roy Tennant, Senior Program Officer at OCLC, then outlined how OCLC is working to make WorldCat a source of linked data on the Web, thereby opening up access to library resources at ‘Web scale’ level. This presentation demonstrated how important persistent identifiers are in the linked data world and the challenges of creating and maintaining them. How do you identify an author uniquely and persistently, for example? Good old library authority files are being seen in a new light in this regard with OCLC, national libraries and others working together on the OCLC hosted Virtual International Authority file initiative. The Library of Congress is now working on a linked data model as part of its BIBFRAME initiative.
A presentation from Marie-Christine Doffey of the Swiss National Library illustrated that open licensing and harvesting of metadata is now mainstream activity for European national libraries.
Jay Jordan, CEO of OCLC, spoke about OCLC’s new ‘Web scale’ platform for libraries – WorldShare. Continuing the theme of identifiers, his presentation included mention of OCLC’s involvement with development of the International Standard Name Identifier system – ISNI. And, as regards researchers, there is also, of course, ORCID. Most WorldShare sites are currently in the US or Australia. A first site is about to launch in the UK – at Bishop Grosseteste University.
Not everyone was happy with this emerging environment of cloud based services and open sharing of metadata. Some concerns were expressed about the loss of national control of metadata and associated services. Others, however, noted that national authority files were still needed and indeed becoming more important to enable data linking – suggesting that the future of national bibliographic authorities lies in this area. ‘Catalinking’ to re-coin a term from the conference …
(p.s. A first set of presentations has now been made available on YouTube).