Hello Erin!
Kept this in my draft folder, as it’s an important topic, here in Quebec. Unfortunately, the timing isn’t ideal to gain high quality insight on the issue from as diverse a group as would
be useful. Maybe after the Francophone Summit and
ACFAS « colloque »?
Please note that I’m speaking in my own name, based on experiences with diverse groups over the years. My level of expertise is low enough that I needed help implementing
ULaval Linked Data subject headings as an ontology in
Omeka S, as a proof of concept.
Having said this, we may find useful leads from people involved in this webinar from last year’s OEWeek.
Webinaire : Arrimage d'une schéma de métadonnées pour les REL au Québec - YouTube
Slides : schema-metadonnes-fabriqueREL.pptx
(live.com)
(Wonder if the automatic translation is enough to get the gist of the presentation and ensuing conversation.)
Participants raised similar issues about indexing and cataloging OERs in repositories meant for research material. Some institutions, like UDS, have a separate collection for OERs.
Ressources éducatives libres (REL) (usherbrooke.ca)
That helps in basic discoverability (indeed, we can use this to harvest from a province-wide catalog). Yet it does nothing to solve the metadata issue.
Besides, as EBSI prof Pascal Martinolli pointed out, code repositories à la Gitlab are often more useful for this kind of collaborative work than DSpace, EPrints, or ContentDM (the three repository platforms used in universities
involved in this discussion).
At any rate…
Participants in that webinar were discussing some (unpublished) guidelines which were an update to a previous document:
Guide des métadonnées REL : recommandations du groupe de travail sur les métadonnées de la fabriqueREL (usherbrooke.ca). That guide’s recommendations
are probably easy to grasp without much knowledge of French (they do mention some DC fields and descriptors). At the same time, what I understand is that they’ve improved on this original document.
(With all due respect to people involved, I’ve heard from metadata experts who had raised some concerns about the original scheme.)
Probably important to note: the workgroup involved in that guide and webinar belongs to a province-wide partnership among university libraries.
Again, yes, it uses DC (dc.contributor.affiliation; dc.creator; dcterms.created; dc.subject.classification, etc.).
Other strategies exist, elsewhere in Quebec’s Higher Education. For instance, our nonprofit inherited a
large catalog (which emerged from the era of “Learning Objects Repositories”) which uses a
application profile for IEEE LOM designed explicitly for Quebec’s peculiar context, including a controlled vocabulary for disciplines in our college system (which is a set list with codes, from the Ministry of Higher Education).
It even integrates competency frameworks such as the
ICT Profile created by ped counsellors and librarians across the network.
The catalog’s backend uses different techniques from the Semantic Web and exposes metadata as Linked Open Data. As such, it’s compatible with MLR which happens to be an ISO standard (and which is somewhat analogous to
RDA in approach). Yes, DC’s involved.
In fact, the former manager for that catalog is part of the ISO board about education, learning, and training. This presentation on metadata is possibly geeky enough that it translates well?
Schémas de métadonnées: pour la description de ressources (eduq.info)
All this to say… There are plenty of recommendations from decades of work in this field. The easiest way to go about it would be to start with the pedagogical intention. Despite that workgroup’s recommendation, learning
objectives prove difficult to encapsulate in
COAR Resource Types. They’re more easily encoded in a vocabulary or ontology meant for learning resources than in plain DC.
Also: don’t forget unique identifiers.
😉
(Otherwise, I find Omeka S to be a better-suited platform for OERs than the typical repositories.)
Hope this helps a bit.
--Alex
|
|
|
ALEXANDRE ENKERLI
|
De : Canadaoer <canadaoer-bounces@lists.bccampus.ca>
De la part de Fields, Erin
Envoyé : 16 mars 2023 15:46
À : Canada OER <canadaoer@mail.bccampus.ca>
Objet : [Canadaoer] OER in Institutional Repositories
[AVERTISSEMENT]
Courriel
EXTERNE. Soyez PRUDENT avec les hyperliens et pièces jointes.
Merci!
Hello,
I have a question related to how people deal with open educational resources in their institutional repositories. We are trying to work through some details at UBC about how to efficiently describe OER in our institutional
repository that prioritizes research materials. We were wondering if folks have suggestions for the following:
Are there any best practices/processes you use for the metadata?
Do you have particular Dublin Core fields and descriptors, keywords you use, or descriptive information that makes it easier to find within your repository?
Thanks for any help you can give!
Erin