Recommended
More Related Content
What's hot
What's hot (20)
Similar to Context in context: applying a context-driven approach in an academic library
Similar to Context in context: applying a context-driven approach in an academic library (20)
Recently uploaded
Recently uploaded (20)
Context in context: applying a context-driven approach in an academic library
- 1. Context in context: applying a content-driven approach in an academic library Kathleen Fear @kmfear
- 3. • Relatively low usage, with a few exceptions • Electronic theses and dissertations; musical scores • Needed to think about changing software • Usability research never focused on metadata • Metadata form was challenging for users • Data can be deposited to UR Research, but it wasn’t a focus of the repository design process
- 4. DATA & DOCUMENTATION .csv, codebooks, research design, survey, images, images, notes, shape files, specimens, artifacts, etc. DATA DEPOSIT REQUIREMENTS description, creator, title, publisher, date, donor, rights, collector, taxon, documents, subject, coverage, methods, etc.Data producer Repository staff Data Reuser But based on whose needs? A Context-driven Approach to Data Curation for Reuse DATA & DOCUMENTATION data collection, data producer, and repository information, prior reuse, missing data, research objectives, provenance, advise on reuse, etc. The DSpace Digital Repository Model Metadata librarians Data services staff Outreach librarians Project staff HPC staff?
- 5. Scoping down to data… For your purposes, what does a data repository need to do? Producers / Reusers Data curation profiles Walk-throughs of existing systems Staff IR situation analysis Interviews (informal) What systems match those needs?
- 6. What contextual information is needed to effectively support reuse? What resources can we devote to developing and maintaining our repository? …looking back out to library context What use cases / user needs are in and out of scope? Should data be treated the same as other scholarly resources in the IR? What is the library’s strategic vision for data? User context IT service model Library strategic plan
- 8. • Kyle Parry, CLIR fellow for data curation for visual and cultural studies • Nora Dimmock, Asst Dean for IT • Marcy Strong, Head of Cataloging • Kathleen Fear, Data librarian
- 9. Conflict and consensus • What contextual information supports reuse of these images serving multiple goals (artistic, political, documentary, etc.)? • Where are instances of conflict in what repository staff and different users prioritize? How can these conflicts be resolved?
Editor's Notes
- Rochester’s emphasis on user-centered repository design = good fit for context-driven approach to data curation
- Base assumptions Step 1: identify stakeholders and the designated community
- photo credit for the image of the data producer, repository staff, and data reuser: <a href="http://www.flickr.com/photos/32066106@N06/3000043099">Passing time</a> via <a href="http://photopin.com">photopin</a> <a href="https://creativecommons.org/licenses/by-nc-nd/2.0/">(license)</a> photo credit for The DSpace Digital Repository Model image: http://www.ariadne.ac.uk/issue55/vandeventer-pienaar#27, Figure 4
- Trying to find the edges of our designated community. We aren’t and can’t be a disciplinary repository, so whose needs are in scope and whose are out? What we got was huge diversity, and a range of systems that could do some, but not all of the needs – you can do anything, but you can’t do everything.
- Attempting to focus in on user context for data blew us back out to look at the library context for the repository. In order to interpret and implement the things users tell us they need, we need to have clear justification for who we determine are users. When building a repository that necessarily serves a very broad and very diverse set of users, there will always be competing priorities. Previous repository has a gazillion metadata fields, 99% of them optional, and they’re a major barrier.