Data Management Plans Meeting Mar 19 2015

From CASRAI

Attendance

   Chris Brown, Jisc
   Anna Clements, St Andrews (co-Chair)
   Joy Davidson, University of Glasgow
   Catherine Grout, Jisc
   Catherine Jones, Science & Technology Facilities Council
   Rachel Proudfoot, University of Leeds
   Hardy Schwamm, Lancaster University
   Veerle Van Den Eynden, University of Essex
   Thomas Vestdam, Elsevier Inc
   David Baker, Casrai
   Sheri Belisle, Casrai

Agenda

   Call to order
   Review action items from last meeting
   New profile format –
       New items proposed from Use Cases – agree whether to add
       Items not referenced by Use Cases – agree whether to keep
   Term definitions
       Are definitions clear
   Upcoming meetings
   AOB

Supporting Materials

   Previous Minutes
   New Draft Profile – Use Case Spreadsheet Info Incorporated
   DMP Draft Charter/Work Plan
   Use Case Spreadsheets

Agreed Actions

    Continue to refine/review term definitions in profile
       Action owner: WG Members
       Completed by: Ongoing
   Reformat profile to dictionary standard, including element action items listed in discussion.
       Action owner: Casrai
       Completed by: Mar 30

Discussion

Meeting brought to order at 1:03 PM GMT

Minutes/actions last meeting. There were many actions; some completed, some still outstanding but gaps in the use case specific elements can be captured in the next stages. Note new use case numbers.

Agenda Item 3 – New Format

Profile reformatted: all elements moved into master list, questions/comments moved into terms list. When imported into Casrai dictionary as draft, we will put a large master profile into the dictionary with all sub-elements and definitions; will also have accompanying sub-profiles that map elements to use cases. Does not duplicate elements but provides specific views based on use case profiles.

Agenda Item 4 – Review New Elements

Use case 6 – need to reflect collaborative, potentially international, element. Things like data selection, IP ownership/licensing are already there. Collaborative – sharing and storage sections – do these cover sharing data during the lifetime of the project? Distributed project across inst. (partners) – resourcing and roles – do these need to be reflected? If DMP at application stage, may not apply but citation and crediting (DOI) might be relevant. Most of these things can be included in definitions of existing terms, but don’t necessarily warrant new elements being added.

Collaborative projects, include work package? Down to dataset level as well. Depends on what granularity is needed. Action: Add work package (text/title/name) to profile. Entity is the project; granularity can be refering to work package or dataset, not just project. What are the dataset attributes? Might be captured in other parts, but roles and responsibility are important. Hardware would also come in this section – element would be “equipment”.

Suggest editing definitions in the Google doc or in the draft once its uploaded. Right now we need to determine if the elements list is complete. We’ll review (highlighted) elements moved up from use cases.

ORCID. Should be an identifier for a person. Pattern in the dictionary is to include two elements to properly cover use case. One is to identify an ID Type, then based on that, second is ID string. Do we want to force a single ID? There are a few in use. The group can propose a list of types the profile will support, such as ORCID, ISNI, etc. Becomes Unique ID Type, and the ID field. List would be ORCID, ISNI Action: Sheri will get list from PRS to propose to this group.

Project ID – is this institutional ID or funder ID? We can have a project ID type, could include a grant number or an inst project ID, but forms of local IDs. Over time, if the community evolves an ORCID-equivalent to identifying projects in a neutral way it can be added. Over time, as IDs come into greater use, they can come in, others that fall away can be removed. 1 & 2 is internal inst. ID, funder project (10) would be external. If intent is to uniquely identify with an internal and an external ID, we can address it. Avoid using “internal” and “external” due to perspective. Institutional and Funder project IDs. Scheme’s can evolve into the profile.

Research funder – Action: refer to current dictionary. Propose to group.

Project dates – is this a start and end date? Action: Yes so this becomes Project Start Date and Project End Date.

Funder ID – goes with researcher funder per current dictionary format.

Call reference – maybe also need Program? Is this a grant ID or reference to call for proposals? Ties project to a program. Would be free text and optional. “Program” would sit between funder and call reference. Depends if machine readability is needed. Action: Add Call Reference – would follow Program. Or could be a URL. Keep as text for the moment.

Bid Deadline – is this an aspect of the call? Deadline would support DMP writer. Action: Flag for review to clarify which use it case it belongs to.

Risk Assessment and Risk Treatment Plan – very generic labels that sound like they’d be a paragraph. Are they just text fields? This is for data centres rather than institutions. Maybe a yes/no binary – is there a plan? A URL – if there’s an existing plan or policy, provide a link. Assessment is perhaps a levelling? High, medium, low. Entry can be adjusted in future.

Expected size of final data set – is this the dataset that would be archived?

Costs of Data Archive – these two belong with storage and backup and/or resources. Wait for list that shows object/attribute organisation to determine if elements are grouped correctly.

Re terminology – why dataset in this element? Its a legacy from the various policies that came together to create this list. As we move to the next iteration, terms will be normalised. For now, calling it datasets solves the issue.

Given time constraints, the next view might facilitate finalising the list. Reformat the list and complete definitions. Are there any that warrant discussion now?

Data publication date – would it be known at this stage? Presumably not. Because we’ve added so many use cases, what started as application stage has evolved in scope. So it fits now.

Format needs revision to what it will be in the final dictionary; we’ll be proposing changes to labels to clarify things and supplying definitions. Come back with a fresh view before import into the dictionary. If possible, the group will be able to review before the next meeting. Will advise.

Members can continue to work on definitions in existing profile.

Meetings going forward will be booked and the group advised of dates and times.

No AOB.

Adjourned at 2:04 PM GMT.