HydraConnect 2 HAWG Discussion Group

Owned by Bill Branan
Jun 16, 2017

HAWG (Dampeer)

Facilitator: Adam Wead
Notekeeper: Russell Schelby
Timekeeper: Steven Ng

Review Use Cases, Discuss Deliverables

Stories for Integration with ArchvesSpace(AS) seem to be done, are they sufficient for beginning coding?

  • Hydra ArchivesSpace Integration Features
  • Develop with non-AS systems in mind, ideally anything
  • Can’t connect to an abstraction
  • How can we generalize around the AS api?
  • atom used by several systems
  • “keep EAD at bay”
  • point 5, use AS to arrange Hydra objects and link back
  • may wait to see if ArcLight/BeadL(AB) resolves some
  • Categorize stories OR wait until we have programming drive?
  • Important to have something to work on when we’re ready
  • Digital stuff in Hydra, AS pull in data
  • Do AS collections get pulled into Hydra? No,look it up and link it
  • If you are using AS for search, ok. If using BlackLight (BL) you will need to feed it back into
  • Not intent of story -
  • How to discover content discovered through Administrative organization
  • Tufts - finding aids are objects in repository
  • Can Hydra handle this administrative hierarchy? Yes
  • We are running into AB territory
  • Public UX for AS is lacking, would be good to
  • Would AB present EAD? More like it would search through EAD’s - discovery for archives
  • What information do you want in Hydra? RDF, EAD, ???
  • DAO’s in the EAD - EAD as an aggregating method
Next Steps
  • focus on the contract; EAD, RDF, json, plus others
  • atom based on content standards (selected from a drop down), emits EAD
  • DACS-based EAD? Should be focused on a standard (EAD is the medium)
  • Archival Management System (AMS)
  • discussion of standards is parallel
  • (tick) have a HAWG call with Stanford to coordinate efforts around EAD
  • Rockefeller & Michigan still shopping for workflows (surmised)
  • what are the tools that we need? Archival organization comes first,
  • elevate description to discovery
  • one group uses Islandora & Hydramatica (as a back-end process)
  • two institutions (Yale, Duke) would be interested in developing;
  • Duke looking at ILMS grant for the AS/Hydra space, plus Stanford, plus Ohio State, plus Cinci.
  • group was formed to describe integration - is that still our raison d’eter? we seem to have moved past
  • moved to more broader area: focus on integration with Archives generally AMS
  • Do the interested parties have the same goals? Does look like
  • Digital Preservation WG - DP working group
  • ISAD based vs. DACs based
  • Ingest & Emittance may be separate issues
Hydra Connect 3 goal?
  • Specification for the API, based on current user stories
  • Clear picture of resources & commitments Action Items
  • (tick) make announcements via HydraTech plus HydraCommunity emails
Archivmatica/Hydra derivative integration?

split from general discussion

  • (tick) Yes, let’s do maybe do it later today, tabled
  • (tick) People who work on this should come back with a report
  • Define Use Cases for your own systems
  • Avalon looking to change the way they generate derivatives
  • help from group with Hydra derivatives would accelerate development

How to support ArcLight (formerly BeadL) (AB)

  • Can this project help us table some of our stories?
  • Stanford is developing this, goal is to work on Blacklight to have faceted search of finding aids
  • Corey Harper Ichabod very same thing, 2 year term
  • Artifactual considering using ArcLink, DPLA may send a developer to have BL run against SOLR
  • AB will likely live in the HydraSphere
  • Stanford will make this work with AS, tertiarilly it is about Finding Aids
  • This group will help inform the AB project
  • Is AB based on EAD or more? More; Ichabod mostly about EAD.
  • Faceted Discovery but not just keyword search
  • This group could help with providing stories to Gary Geisler

Archvmatica(AM)/BitCurator/ThingsNotArchivesSpace Integration

  • Pitch: work that happens in Hydra that overlap Archival Information Systems; access derivatives
  • Preservation Planning Tool; what tool is used to generate access derivatives
  • Sufia: there is a Hydra derivative gem that uses command line tools; archivist could enter PEAR db entry on how to manage files, make it available to all other systems
  • treat FPR as a generic characterization tool, not specific to Archamatica;
  • worked on specific FFMPEG commands, this tool could
  • registry of how to to crack and process materials
  • want to do proof of concept, deal with long tail of file handling
  • AM uses pro nom id’s, not MIME type;
  • research data might be a good partner,
  • forensic disk images, code4lib article about presenting this as a finding aid;
  • BitCurator (BC)
    • analyze, scan, report forensic information about a disk,
    • end up with a forensic file format
    • Bulk extract just for quick ingest (basic)
    • just came to 1.0
    • archivists believe that forensic disk images will be another resource
    • deleted files is a political football

Repositories

  • Places to put things
  • Archival collections into Hydra - digitized content
  • Born Digital stuff, Digital surrogates
  • Arrangement? some are before ingested - arrange in Archive System, then send bundle to repo, not
  • putting the object into hydra, arrangement plus object; object knows that it exists (PID) what collection it belongs to; the EAD knows about the digital thing and how it fits
  • Is the arrangement part of the preservation system? is more murky, happens before; could build RDF arrangement, but is time consuming;
  • ArchiveSphere is doing some description into hydra
  • How to deal with master and derivatives relationships? PDF made of 200 images;
    • Stanford uses content metadtata stream, labeling masters & pdf;
    • Others keep RelsX folders with masters, physical description;
    • how does it handle change? export mods file for every level, id for each folder; Archvist Toolkit takes about 5 minutes;
    • IU is using hierarchy of objects