NDSA:June 4 Meeting Minutes

June 4, 2014 CWG Meeting Minutes

Attendees (38)

• Linda Reib - AZ State Archives
• Chris
• Deborah Rossum
• erin engle
• Edward McCain
• Sara Holladay
• Paul/Sara/Xiaomei, U Iowa
• Jennie Knies, University of Maryland
• Christie Moffatt
• Joel Wurl
• Glen McAninch
• Lori Donovan
• Amy Kirchhoff
• Jennifer Weintraub
• Rebecca Johnson Melvin/U Delaware
• Amy Hodge
• Rachel Howard
• JR
• Midge Coates (Auburn)
• Lisa Schmidt
• Rachel Wise
• Mitchell Brown
• Butch Lazorchak
• Mark Phillips
• Aaisha Haykal, CSU
• Cathy Hartman
• Sara Holladay
• Gail Truman
• Abbie Grotke
• Amy Hodge
• Jaime - NIU
• Kevin McCarthy
• Ricc Ferrante (Smithsonian Institution)
• Alex Duryee (AVPreserve)
• Chad Garrett
• Gail McMillan

AGENDA and MINUTES

This is the fourth and last in our series of talks about the Content Areas outlined in the 2014 National Agenda. This month we focused on Research Data. We had two presentations:

1. Paul Soderdahl (along with his colleauges Sara Scheib and Xiaomei Gu) was up first, discussing Data Management at the University of Iowa and their Data Management Survey. Their slides are here: File:NDSA Content WG presentation on Data Management at UI.pdf

Questions that followed included:

Gail McMillan: Could you say a little more about publishing vs. other dissemination?

Just like with a university press or other traditional academic publisher, a significant part of the publisher’s task is to disseminate the publication as far and wide as possible, so to a large extent these two are closely related. But other dissemination can happen outside of the publisher’s control and/or beyond the lifespan of the publisher’s involvement. For instance, if a data publication is released with CC0 license, others are permitted to redistribute. Another data repository might acquire one of our publications for inclusion in their repository, meaning there are more possibilities for dissemination beyond the publisher’s own “marketing strategy.”

Glen McAninch: Which of the models works best for you or are you working on a hybrid approach

The idea behind the model is to acknowledge that we wear multiple hats with different motivations that have varying lifespans, and they all need to coexist. This model arose out of conversations where we had multiple library staff – e.g., an archivist, a preservationist, a technologist, and a publisher – gathered together and realize we were looking at the same problem through different lenses and each had different expectations about what would need to happen and where the concentration of effort was most needed.

Rebecca Johnson Melvin/U Delaware: For each of these horizons, are you suggesting all be handled throught library repository?

Yes, ideally a generalized repository could handle the same object along its path on all four horizons.

2. Then we heard from Mark Phillips from UNT. His slides are here: http://digital.library.unt.edu/ark:/67531/metadc288009/

Questions that followed:

Chris: Ever run into a retiring professor who discovers his closet full of mainframe tapes? Mark: Not yet, but in future he expects this to be more common.

Rebecca Johnson Melvin/U Delaware: what generic metadata are you using to describe the data deposits versus the readme.txt description? Mark: Duplin core-based UNT-L. It's all documented, and a generic library view of metadata, straightforward. The guide is here: http://www.library.unt.edu/digital-projects-unit/quick-start-metadata-guide . They also have style guide for how readmes should be done.

Aaisha Haykal, CSU: sorry, if I missed this, but is this system home grown or is it open source? Mark: it is "purpose built" locally to manage collections, based on CDL's microservices. Various components should be released open source but works for their specific environment.

Rebecca Johnson Melvin/U Delaware: Do you have any evidence yet of other researchers using this collected data? Mark: Not yet - hard to show true re-use. Have some download stats but hard to interpret.

Cathy Hartman: could you talk a bit about the preservation part of the system. Mark: Preservation repository to manage bag-it bags. A large scale system, with replication of content, fixity info, verifications. Access system (Django) to dsiplay content.

Joel Wurl: Re. the issue of "uptake," are you noticing any patterns in terms of which faculty from which disciplines are most inclined to cooperate/participate? Mark: The faculty cooperating are the ones we work with already, so not much uptake. Haven't strongarmed any others really, haven't made a huge effort to push it. Waiting to see. Need to do more outreach, people don't full understand.

Rebecca Johnson Melvin/U Delaware: Are any of the data sets deposited to meet a federally funded grant requirement? Mark: Not yet, except one library-related IMLS project (written into data management plan).

Rebecca Johnson Melvin/U Delaware: Would you accept data deposits with limited access restrictions or limited size restrictions? Mark: So far we are open to the world. We can limit pretty easily to UNT community, and can provide embargo periods if need be. The system manages/handles that.

Other announcements

Christie Moffatt from NLM will be taking on co-chair role after the summer meeting. Welcome Christie! Many thanks to Cathy!

We're working on details for the July meeting at DC. Will make a decision about holding one early July at our normal time. Abbie is on leave then so not sure... keep an eye on the listserv for more details.