Finally in this set of three ‘perspectives’ session, Ed Chamberlain from Cambridge University Library.

Why expose bibliographic data?

• Natural follow on from philosophy of ‘meeting reader in their (online) place’
• Already exposing data to others (OCLC, COPAC, SUNCAT etc.) – lots of work to setup each agreement and export – Open data approach might give easier way of approaching this
• Offer value for money (for taxpayer)
• Internal academic pressure – ‘we are being asked for data’

e.g. use Rufus Pollock – wanted to do ‘analysis of size and growth of the public domain using CUL bibliographic data (http://rufuspollock.org/tags/eupd)

The COMET (Cambridge Open METadata) project will be releasing large amounts of bibliographic data under an Open Data Commons License. Formats will include MARC21 and RDF – partnering with OCLC so linking into related services such as FAST and VIAF.

Ed thinks the library sector should have following ambitions around resource discovery:

• Hope to see ‘long tail’ effect – exposing data to large audience
• ‘Out of domain’ discovery
• Multiple points of discovery at multiple levels for multiple audiences
• Services for Undergraduates, for academics AND for developers

Practicalities/Challenges:

• Licensing
  • While individual records may not be protected by copyright, collections of records may be – and often obtained by library from shared catalogue resources/commercial suppliers under contract
  • Ideal is full unrestricted access
  • Better to publish data (as much as you can, even if necessary to have more restrictive licensing attached)
• RDF vocab and mappings – no standard
• Triplestores – for managing RDF – but new technology, seems complex

Opportunities:

• Strong platform for future development
• Linked formats and open licenses are virtuous pairing
• Huge scope for back office benefits

Need to also think beyond bibliography – what about holdings? libraries (physical locations)? librarians as linked data (!) (finding people with specialisms etc.)