I’ve recently received (and accepted) an invitation to join a new JISC taskforce looking at Resource Discovery Infrastructure. The scope of the group (to quote from the draft Terms of Reference) is:
Serials, books, archives/special collections, digital repository content – we acknowledge that the group will have to prioritise areas of work because of their different levels of maturity. For example an approach might be agreed for books and serials and then later down the line repository content might come into play. The Task Force will discuss the scope and how best to deal with it at their first meeting.
- Gain an overview of current activity and library infrastructure.
- Identify the requirements for the UK further and higher education, in terms of current priorities and visions for the future
- Articulate how requirements should be met.
- To consider appropriate business models.
- Oversee related studies and scoping work.
- Identify how to take forward the implementation of an infrastructure to meet the future vision.
- Work with key partners/stakeholders.
- Develop a communications plan.
At the first meeting on the 25th November (which seems to be approaching rather quickly) I’ve agreed to give a presentation on What would we do if we were starting from scratch? I’m slightly nervous, as the others who have been invited to present are Tim Spalding from LibraryThing, and Paul Miller from Talis, and so I’m looking to say something appropriately visionary!
So, rather than rely purely on my own imagination, I thought I might as well ask the world – what if you were starting from scratch? Leave a comment, or trackback with your thoughts.
I have no idea. I’m not a library/metadata expert.
Starting from scratch? I guess there would be two aims: user focused (I just want to find items about X) and efficient (non duplicating of work/data).
How much does resource discovery overlap with ‘resource locating’? i.e. I know which resource i want (book X) so I don’t need to discover it, but i do need to find out (discover?) how to get it, what libraries have it near by, where can i find it online, and for both, which formats.
Starting from scratch would we all have catalogues? or would there be a ‘master’ UK catalogue of books, which libraries could simply add holdings to. The trad library catalogue would just be a ‘view’ of this, limited to those items with local holdings attached by default (but easily to extend search to include holdings of other nearby libraries if not available).
What are the discovery needs of undergrads? of academic researchers? how are they different. It seems to me that ugrads are normally quite easy, they want books/items that will be local. Researchers will sometimes want obscure items, and that is where locating a resource becomes more important. (we STILL don’t have a UKwide universal catalogue).
Perhaps Link resolvers have a greater role with more (psuedo) target services both for online and physical items.
Are books/serials really that different? both are things that can be local and online, though serials have an extra level of holdings complexity. that aside, how differently do they need to be treated? They are both things that will be in multiple libraries (generally), i.e. non-unique. therefore starting again you would perhaps want to avoid the messiness of suncat. again, coming back to a central catalogue idea.
And what about ‘data’? that is to say how will people discover open datasets and research data. And will users get annoyed if searching for something, and the results are full of datasets, images and video, when they just want a bloody book!
Moving away from search… How important is it to be able to navigate from one item to related items (like amazon, aquabrowser, google), how could this be done. to discover by following a trail.
[you can skip this next paragraph, it rambles badly]
Special collections, unlike books/serials, are normally unique. the Mass-Obs archive only exists in one location. I don’t know much about Special collections and archives, but i imagine the potential for define relationships to be huge. The archives of a famous actor in one library could be linked to the records related to items from a ‘old theatres’ collection in another. Imagine if, say, an archive of a person, included metadata of their location/work chronologically. What might have only been possible through chance discovery of letters, might be possible by mashing up geo/time data of different people and entities, to discover relationships and research avenues previously unknown. But perhaps I should leave this to those in archives and special collections who know what they are talking about!
just some randome thoughts…
Chris
how far do we get to go back for a clean slate here? couple of hundred years? before the emergence of bureacratic cataloguing and classification practices? probably not;-)
I agree with the above commenter, that a user-centric approach would be key. In this context-driven world, design should be driven by context of use (and also enabling multiple alternate contexts of use). We need to have a much better understanding about this than we do right now — especially around Archives and Special Collections. If we make the assumption that these collections are largely used by humanities researchers and teachers, we need to understand their workflows much better. Much more emphasis has been around STEM — and there are many good examples of ‘collaboratories’ in these disciplines. There are very few around humanities, because the workflows for research and collaboration are different, perhaps more complex and less easy to systematise. In addition, in a context were humanities/social sci teachers (of all levels) are required to have students engage critically with primary sources, how are our archives/special collections and online digital collections facilitating not only access but use of those objects?
I’ll watch the progress of this taskforce with interest — it’s very relevant for us at the Archives Hub and Copac.