Metadata Workflow/Management UnConference

Owned by Bill Branan
Jun 16, 2017
2 min read

Metadata Workflow/Mgmt (Dampeer)

Moderator: Chris Powell
Notes: Russell Schelby
Convener: Anusha Ranganathan

Chris Powell: What do people use to work on metadata and view it?

  • Oxygen
  • forms, excel ingest, content dm,
  • built-in spreadsheet-like editor

Anusha Ranganathan,

  • in repository, not visible to people
  • people whose job it is just to review, mainly copyright checks, embargo checks
  • few metadata enhancements
  • very simple project, once it is submitted,
  • review process has several steps
  • review, escalated, rejected (rare: duplicate), approved
  • once approved gets migrated
  • software to track? manual XML
  • built hydra admin to move things between buckets: archive to Open Archive
  • has ties to request tracker, which handles email trail, email ticket tracking
  • metadata as a service, only Northwestern is doing that
  • Declan Fleming: Research Data program, meet with faculty, look at data sets, sometimes it is cosmos scale

Chris Awres: we do metadata modeling

  • no real ontologies for most stuff, can’t validate
  • moderator: does like looking at data sets and cleaning up say spreadsheet column names
  • lots of data is someone’s spreadsheets
  • “look at the red column, it means something”
  • USC Karma - lets you link headers to ontology
  • basic metadata - data management plan required for grants, we need to know what the files are

Alicia Morris, Tufts

  • basic recommendations, i.e. text files describing data set
  • student scholarship - staging environment, clean up metadata, push to production
  • identify high-value content
  • image library - working with faculty in art history department, having students fill out spreadsheet to describe course images; library cleans up this metadata with scripts to correct punctuation, capitalization, then ingested into repository

Penn State: Would like to get away from spreadsheets

  • linked data/active triples might alleviate this from a coding exercise
  • tough to validate a spreadsheet
  • how do collections go from selection to digital collection? lots of people involved, lots of technologies
  • Declan: we’ve embraced spreadsheets: provide templates to staff
  • Moderator: getting vendor data back is frustrating: mac/windows artifacts
  • would it be worth it to write a head for a generic metadata editor?
  • Mike Giarlo: There is a gem that does that

Andrew Rouner, Wash U

  • surprised that self-submission as default - wouldn’t expect to have any pre-prints: faculty are busy
  • would be interested in having a discussion about that model
  • note: Oxygen can integrate with some systems, maybe hydra?

Michael, Penn State

  • we agreed on a workflow, pushing forward with a pilot project: a workflow around a Hydra project
  • thinking about HydraDAM workflow, describe the same object through different
  • Declan: we use JIRA - everyone is in JIRA, can attach spreadsheets
  • Oregon has taught their catalogers in GIT
  • making this a Gem would be worthwhile, watch the

Question

  • What tools are you using for creating RDF metadata?
  • Aaron Colburn has a unique way, is building fedora classes not the fly, uses OWL
  • Tom Johnson was also creating a similar system
  • BibFrame editor would create an RDF, doesn’t necessarily need bibFrame backend