We are increasingly working with ever richer metadata and working with records that contain different types of metadata designed for different purposes and different audiences. How do we manage this in the cloud?
We often forget what a radical and profound effect migrating from the card catalogue to the online catalogues had on libraries. On line catalogues and web based discovery is so much part of our daily workflows, and so vital to how we make our collections accessible, we forget that in moving online we were able to give our patrons a much richer source of discovery metadata and with this came new ways of doing things.
However, now we are moving from our own siloed online systems to shared cloud based systems there is the potential for another explosion of rich metadata to populate our catalogues. We need to think about the implications for patrons and our own library services.
To know where we are heading it is important to know where we came from. Consider the following image of a traditional library catalogue card. A lot of information had to fit onto a 12.5 cm X 7.5 cm piece of cardboard. As a result, information had to be brief and laid out in such a way as to maximise how the information could be presented to the library patron. There are roughly 560 characters, including spacing, that make up the information on this particular card.
When Henriette Avram developed the MARC standard to enable libraries to share metadata tape storage was limited and expensive by today’s standards. The whole point of the MARC coding was to enable libraries to share metadata more readily, but the actual structure of the leader fields, the MARC field numbers, the first and second indicators, and the sub-field coding, was designed to save space by using numeric and alpha coding to represent concepts, as well as to more efficient data storage by enabling variable field lengths.
Moving our records online has enabled us to offer a much richer source of metadata to give library users better outcomes. For example, the full record on Libraries Australia for the publication listed above contains 3,024 characters, including spacing. This is a 540% increase in discovery metadata. In addition to new content such as RDA, many records have extensive additional information in the 5XX notes fields including formatted notes in the 505.
Yet, while metadata got more varied and richer, for as long as we downloaded data from a National Bibliographic Agency such as Libraries Australia, or from some other source such as SCIS, we could choose what metadata we wanted in our catalogue and strip off what we did not want, especially if the extra metadata could potentially confuse our library patrons.
In a truly cloud based collaborative catalogue environment while we can access very richer metadata our ability to strip off what we don’t want may be limited. For example, as a school library using a cloud based catalogue many of our records offer:
… and this is just the extra rich data in the 6XX subject headings. Furthermore, increasingly this discovery metadata is enhanced with linked data elements. For example, LCSH and FAST. While the extra discovery metadata offers more search terms for library patrons to use, different types of metadata often has specific purposes that goes way beyond supporting keyword, author, title and subject searching. MeSH for example have a specific purpose and structure to suit the needs of medical libraries.
The challenge is: how will the emerging collaborative and cloud based library information management and discovery platforms allow libraries, or indeed individual patrons, work with this increasingly rich source of discovery metadata?
Instead of the old model where libraries stripping out metadata when they download it into their own local system, will libraries be able to choose what type of metadata in the cloud they can suppress, and what discovery metadata they can priorities?
If using more than one type of discovery metadata, can a library choose to give a clear indication to the more advanced library patrons using the discovery platform as to what metadata is what? In a medical library a research may need to know which is the MeSH heading and which is the LCSH. In an Australian school library the SCIS metadata may be used for discovery but the ScOT headings, because it is used as the thesaurus to describe the national curriculum, has uses that go beyond search functionality. And how will all this work in a linked data environment?
Will the emerging collaborative and cloud based library information management and discovery platforms enable libraries to select from the linked metadata on offer and choose what type of linked data can be used to bring in additional, and appropriate, resources on the fly, and in real time? See the following for a pilot study of a catalogue using a couple of lines of coding to search local Australian Curriculum metadata to then do a linked data query to bring in additional external resources on the fly and in real time:
Chadwick. B. (2016), ‘When MARC consumed ScOT: a tale of linked educational metadata’, 18th VALA Biennial Conference and Exhibition, Concurrent Session 2 Data Stuff, Tuesday 9 February 2016, VALA / Libraries, Technology and the Future, Melbourne, Australia. http://www.vala.org.au/vala2016-proceedings/1009-vala2016-session-2-chadwick