This post has permanently moved to http://archiveshub.ac.uk/locah/2010/08/27/making-sense-of-modelling-ead/
Please update any links and bookmarks.
We apologise for any inconvenience.
Tags: Archives Hub, barriers, EAD, jiscexpo, linkeddata, locah
This entry was posted
on Friday, August 27th, 2010 at 8:48 am and is filed under Archives, Linked Data.
You can follow any responses to this entry through the RSS 2.0 feed.
Both comments and pings are currently closed.
The example you give of creating a URI for the ‘concept’ of “Sir Ernest Shackleton as the explorer and leader of the British National Antarctic Expedition of 1907-1909″ sounds like it might be a good case for using the new foaf:focus – see http://wiki.foaf-project.org/w/term_focus for some more information (note although at time of posting the wiki says foaf:focus doesn’t exist, it was, in fact, added to FOAF recently in v0.98 http://xmlns.com/foaf/spec/#term_focus)
Funny you should mention that. We are, indeed, planning on using foaf:focus for this. I think we’ll be using it quite a bit, to link concepts to things that are represented in the Hub access points. We can then link several concepts to the same thing.
The issue of modelling the data rather than the record is something I’m struggling with at the moment for our library catalogue records of course materials. These are in MARC format, with quite a bit of local practice dictating where non-standard information (such as the course code) goes.
The MARC records bring together stuff that would be better separated out (especially in terms of ‘carrier’ vs ‘content’). This especially happens where we have several pieces of content in a single carrier (e.g. several items on a single DVD) – as the MARC record tends to focus on cataloguing the ‘carrier’ first, and the information about the ‘content’ finds itself relegated to unstructured fields. I don’t know if EAD gives you a better place to start from this perspective?
Yes, I see what you mean. I think the ‘carrier’ information for us will largely be the properties of the main unit of description – extent and access being two good examples. We’ll have URIs for the finding aid and for the EAD finding aid. After that, its basically modelling the content.
…well, i say that…I’ve just had a conversation with Beth and we’re realising that its not so easy to separate content from carrier all the time. Makes me think of the whole CIDOC Conceptual Reference Model approach where they are very strict about this, but have ended up with something very complex as a result.
We do have the same issue that much of the content is in more unstructured fields of course. But we have archival creator, repository, language and access points which are quite nicely structured.
Playing with the very unstructured biographical history should be fun…
I’d love to come and have a chat with you about it all – I quite fancy a trip to Leam again!
I’m not sure about the assertion:
archiveshub.ac.uk/id/person/sirernesthenryshackleton ‘created’ http://archiveshub.ac.uk/search/record.html?id=gb15sirernesthenryshackleton
He created the material that is described in the resource:
but that’s not quite the same thing…
Yes, you’re right. In reality, the triple will need to refer to the unit of description rather than the finding aid. We will also be thinking about the concept of ‘created’, which is different in an archival sense. In fact, within our data model, we do have the finding aid (and the EAD encoded finding aid) separated out from the unit of description.
[...] this subject. Linked data will generally require some data modeling, and as the Locah project report, this may mean having to change your data model [...]
This blog has moved to http://archiveshub.ac.uk/locah/ is proudly powered by
WordPress MU running on UKOLN Blogs. Create a new blog and join in the fun!
and Comments (RSS).