Sustainable Data Management/20160720 ESIP summer mtg

ESIP summer meeting

landscape: http://commons.esipfed.org/node/9139

Agenda

Abstract and uploaded talks: http://commons.esipfed.org/node/9139

Speakers, from 3 points of view (<10 min each):

1. Kerstin Lehnert - Repository Registries (COPDESS-Re3Data)
2. Matt Jones - Repositories aggregators (Data ONE)
3. Margaret O'Brien - Contributors (liaison, perspective from Researchers)

Also to discuss, Shelley Stall's visualization at ...link...

For each talk and following, we are discussing the following questions.

• How open are each of these repositories to new or outside data?
• What will it take so can we get more data in repositories?
• How to guide people to the right repository?
• Additional fields to add to the registry to help? E.g. Certifications. (http://gfzpublic.gfz-potsdam.de/pubman/item/escidoc:1397899:6/component/escidoc:1398549/re3data_schema_documentation_v3_0.pdf)
• Any obvious gaps in services that we know of?

Outcomes of today's discussions

Recommendations for registries

• send out reminder to update the registry record regularly
• promote the business card standard
• work to standardize the APIs that journals use to communicate with repositories
• consider opening up correction process to anybody with confirmation by recordholder
• note that some metadata about a repository may live in different registries (e.g. COPDESS doesn't have info about technical services but Council for Data Federation might and so does DataONE -- so registries should ensure linkage across the relevant records
• work with Shelley to improve the currency of the data in re3data
• make sure that the fields we would need for the services

Recommendations for repositories

• take an active role in bringing COPDESS messages to your institution
• create a standards compliant business card
• use re3data as the authoritative source
• validate your record in COPDESS (where you can own your record)
• adopt an existing API rather than developing one of your own
• use loose coupling so submission or authentication systems or other subsystems could be used across groups
• start with small steps to participate in an aggregated network, incrementally improve
• leverage added value from aggregators for things like data citation

Recommendations for liaisons

• develop a set of questions to help narrow the list of options based on Margaret's questions
• hope for consolidation to streamline the options in the longer term
• help registries to track the right kind of info

Next steps for landscape analysis

• will need to get metadata from re3data & COPDESS and CDF (sounds like they are just starting to develop their registry which may have technical info)
• List the predominant standards eg APIs, ORCID, etc.
• help to comment on Erin's and Shelley's flower diagram to make it better for understanding roles and their touchpoints
• look for opportunities to streamline the landscape and fulfill the needs of different perspectives
• summarize surveys about measuring these outcomes and the gaps (work with the ROI group
• Shelley's graphic: could use libraries and business venn bubbles.
• consider the collective approach -- sustain the network of landscape of repositories

Common technical vision

• concern that lots of funding goes to developing bells and whistles (map viewers, authentication systems) and not to new added value
• try to minimize overlaps
• ensure that investment really increases the amount of data available and the ability to access it no matter what repository
• streamline the following for economy of scale:
  • data quality
  • authentication
  • standards for metadata exchange & web services
  • author identifiers
  • helpdesk support (see ScienceGateways.org set up to help set up computational gateways and web-based resource <--nancy wilkins-deer)
  • data preservation (connect with the data stewardship group)
  • citation
  • compliance mechanisms
  • search interfaces
  • ROI framework
  • business models
  • A common place to communicate (this cluster) including stories of how people are using data