On 2016-02-23 10:18 PM, Stefano Cossu wrote:
Hi there,
1. At the moment we are not exchanging data with other institutions in a
"smart" way. The recent work on our new DAMS will bring more
opportunity in this direction though. The key for this is having adopted
RDF as the lingua franca for our DAMS.
Before we tackle inter-institutional interoperability, we want to
better connect departments within the museum and heterogeneous data
sets such as library, archives and collections. This will have more
obvious and immediate advantages for us.
Harmonizing library, museum and archive catalogue data into RDF (if
that's what you mean) would be a significant achievement. Clearly, it
would be of much wider value than within your own institution. Certainly
it's an aspiration which this group should support you in/help you with.
2. CIDOC CRM seems to be the most comprehensive and flexible ontology
to encompass the widest range possible of cultural heritage items.
However, CIDOC CRM cannot be practically used as a cataloging ontology,
but rather as a harmonization tool. It should sit between the cataloger
and the consumer, and accessed only to machines that can handle its full
complexity.
The approach I have adopted is to generate Linked Data RDF on the fly,
starting in my case from native museum catalogue data encoded in XML
records [1]. The idea is that each object in the [Wordsworth Trust]
collection is published with its own unique, persistent, useful URL. I
would commend this strategy as a good way to get started: stake a claim
for your collection in the Web information space.
I'm working to improve the RDF, and one way of doing that would be to
include more CIDOC-CRM structure in it. The other, bigger, problem I
see is that all the data is just string values. What we desperately
need are more shared Linked Data frameworks (for people, places, events,
...), whose URLs we can begin to embody in our catalogue data.
If you see 'interchange' as a selective publication exercise, rather
than a mass-data-shifting one, then there is no down-side to just
publishing that subset of your catalogue data which is useful for
sharing. Museum records are full of internal management stuff which has
no place being published, just as bibliographic records contain lots of
technical MARC data that won't be of interest to a general public. One
useful thing we can do is to agree what story we are trying to tell, and
work back from that conclusion to a view on what data is required for
sharing.
(Also, dynamically rendering catalogue data on request ensures that it
is always up to date.)
There are several publishing schemata that map to CIDOC-CRM and expose
only the concepts meaningful to a human end user. There is not, as far
as I know, a cataloging schema that is encoded in RDF and maps to CIDOC
CRM. I am aware of ongoing efforts to serialize the Getty's AAT (which
contains CDWA terms) into RDF [1], but having a separate, formalized
cataloging ontology based on CDWA would be a great advancement in this
area.
Note that all three Getty 'vocabularies' (should one now say
'ontologies'?) are fully published as Linked Data, and so are available
as RDF.
Best wishes,
Richard Light
[1] e.g. http://collections.wordsworth.org.uk/Object/WTcoll/id/GRMDC.C144.9
which resolves to
http://collections.wordsworth.org.uk/Object/WTcoll/id/rdf/GRMDC.C144.9
when RDF is requested
3. We use Fedora [2] which is completely content-agnostic and allows to
build any sort of content model. Fedora and its satellite
projects encourage the use of PCDM [3] as a very basic and
broad-scoped ontology on top of which more domain-specific ontologies
can be layered to satisfy any kind of content modeling.
4. My team (5 people) is in charge of designing and implementing our
collection information systems. A separate department, Digital
Experience and Access, acts as a broker for end users' (staff and
public) needs and is in constant dialog with us. I act as an
interpreter who translates semi-technical requirements into
specifications.
[1] http://www.getty.edu/research/tools/vocabularies/lod/index.html
[2] http://fedorarepository.org
[3] https://github.com/duraspace/pcdm/wiki
On Thu, 18 Feb 2016 16:33:51 +0000
"Delmas-Glass, Emmanuelle" <[email protected]> wrote:
Dear all,
In order to get us started with the LAM interoperability SIG, we
would like to get your feedback on a few questions.
1. Use cases: as a museum, library or archive, whenever you tried to
integrate your data with other institutions, what worked, what
didn't, and why?
2. Interoperable metadata schemas and/or ontologies: what is out
there that can help bring collection data and bibliographical data
together? What do you think about them, what are the challenges and
how do you plan or wish to utilize them?
3. Existing interoperability tools: what software platforms do you
use? If you could design your own, what would they be?
4. Staffing: what staff member(s) usually work together on these
questions of data interoperability at your institution (list titles)?
What new staff position(s) would be useful?
If you could give brief replies that would be great. The goal is to
assess the challenges of our community as well as the opportunities.
This common base should lead to some interesting discussions that we
could bring up in an in-person meeting at the next MCN conference in
New Orleans.
Emmanuelle and Stefano
Emmanuelle Delmas-Glass
Collections Data Manager
Collections Information & Access Department
Yale Center for British Art
http://britishart.yale.edu
203-410-4069
--
Stefano Cossu
Director of Application Services, Collections
The Art Institute of Chicago
116 S. Michigan Avenue
Chicago, IL 60603
--
*Richard Light*
_______________________________________________
You are currently subscribed to mcn-l, the listserv of the Museum Computer
Network (http://www.mcn.edu)
To post to this list, send messages to: [email protected]
To unsubscribe or change mcn-l delivery options visit:
http://mcn.edu/mailman/listinfo/mcn-l
The MCN-L archives can be found at:
http://www.mail-archive.com/[email protected]/
