Archives and Hydra
What do we think this session is about? Why did you come?
Giarlo - continuation of DLF session re: ArchivesSpace (AS) and integration with repository platforms
Chris Awre - And similarly integration with CALM (UK based, like Archivist Toolkit (AT). Also about ArchivesSpace
Steven Ng - We don't' have any online front end for archives. We would like a unified frontend for workflow
Giulia at UCB - Not sure.
Jim Tuttle - Duke uses AT. Legacy filesystems and variety processes for description, file storage
Lynne, UCB - moving to AS from legacy. Would like a unified system for preservation at the same time.
Jen Green, UMich - Bentley Library would benefit from this discussion
Christian, Denmark - need bit preservation for archival docs. IS everything to be preserved for long term? What about artifacts from exhibit? should those be preserved as well
Matt at GWU - looking at AS, how ti works along Hydra. Are you putting finding aids in repository?
Dermott and Jimmy, Trinity College/Dublin - something we are going to have to face, so what to see what others are doing.
Brian from Unv Utah - what are people doing with archives, finding aids, and how it relates to Hydra
Adam Wead - Want to integrate finding aids with digital stuff in Hydra. Been grappling with this for 2 years. Early attempts were not successful. Let's look for a shared solution.
Erin from Tufts - we use Cider (home grown) for collections, and push it to Fedora. Link to objects from digital library finding aids. Working with DCE on improving processes. Also interested in big files, and MPLP-processed collections.
Kim Schwank, UCSD - Handling ETDs and University records. Listening in.
Cristela, UCSD - Use AT, moving to AS. Maintenance of links to digital objects in finding aids is a concern.
Susan, Harvard/Dance Heritage Coalition - want to include archival collections ultimately.
Karen - WGBH does item level description
---- 30 MINUTE DISCUSSION ON ARCHIVAL COLLECTIONS AS THEY RELATE TO HYDRA ----
What is ArchivesSpace?
Adam explains -- AS is an archival management system. Has database management, role management, also a display function. Uses Ruby and JSON. Adam has tried to look at Hydra for AT and it was unruly.
Giarlo - we spoke with the AS folks. Good architecture. Why not strip out their database and replace with Fedora? Mark M of AS discouraged that. Do you really want to preserve the finding aid? AS has an API, so integration is possible. Now would be a good time to give AS feedback to add necessarily functionality.
But it is all about workflow. Processing workflow matters in this case, since archival processing and digital object creation/management aren't necessarily related.
Adam/Rock Hall points from Hydra objects to finding aid, at item or folder level. Needs to support MLPL practices.
Workflow
Cristela - UCSD did some analysis on database of record. Decision was that description for digital objects - record is in repository. But description must be maintained in AT.
At Tufts - description is handled in central place. Any change is pushed out to Fedora. Works pretty well. Easy to make digital objects available from finding aid.
Adam - the reason to make Hydra interact with AS - being able to bring in all the descriptive context from the official record for the discovery environment (where presumably the digital objects are also available).
this is very hard -- the hierarchy of context/metadata is difficult .
Common Needs for Hydra and AS integration
what would that look like?
Should hydra be fully aware of the hierarchy? Be able to pull it when it needs to, and make it available to user.
See Princeton's approach to finding aids.
If you find a digital object, how do you find more objects in that collection, at other levels and or the same level? Where is that information and context managed?
Does RDF play a role here? isMemberOf …. all in one data stream. BL can take care of the associations based on that. There is a need to build a new vocabulary around description multiple relationships. OAI-ORE could be a replacement for EAD??? and help with this problem.
But what happens when there are changes to the metadata? Pick the system of record and be sure to sync.
For born-digital, the notion of a strict hierarchy is out the window. Files can be associated with different folders, etc.
ACTION ITEMS:
- need a working group in Hydra - to follow this thread both within Hydra and beyond -- Erin, Adam, Steven, Mike
- need user stories
- give feedback on AS API - perhaps a first goal for the working group.
- Karen Wickett - Published dissertation topic related to collection hierarchy, and looking at this from Europeana, and DC data models. Her work should be looked at by the working group.
---- SWITCHING GEARS: 20 MINUTE DISCUSSION ON "Archives" in the sense of preservation master files ---
How to archivally manage Hydra objects?
Karen asked -- how are people keeping track of the archived files? Where are checksums created and managed?
much discussion about practices of file/metadata storage, fixity. Further discussion about TDR, frustration with the high bar it sets. Discussion on Posix as the standard to keep in mind. Object Store.
Hydra is the UI / environment for managing the archival information.