Born Digital

Nancy Enneking - Institutional Archives and Records at the Getty Research Institute

Only born digital not digitized

People don't know what to do with it
file extensions aren't useful in determining what type of item

run through FITS/JHOVE/Droid

Only have free version of FTK

How to identify and get some level of minimal metadata - only have folder structure and filenames

possible systems
hypatia - done for AIMS
curator's workbench
Sufia for WGBH - needs bulk and to be able to handle large files

Hypatia - grant funded with funding ended - major breakthroughs but not a finished project
1. has collection objects with relationship in fedora to all the items in the collection
2. has datastreams with things such as images of the floppy, FTK image and HTML plain text
rendering of the items on the image
3. access controls - same as all hydra applications - rights datastream for each object
4. can do arrangement in it
5. creating sets and editing relationships
6. bulk loading through hydra - set default values for the bulk
7. input mechanism a little peculiar
8. indexing directly from fedora into solr
9. create derivatives for future use without format migration/emulation

Possibly use Tika plug-in for solr to see how well that can be managed

Possibly set up an automated system where you attach each file to a fedora object

what is the relationship with ArchivesSpace? archival management - Not a repository - a shortcoming?
Mike Durbin has been working on the connections but doesn't have all the pieces together

Beginning workflow
1. scripted analysis of the files
2. creation of objects in fedora
3. discovery in blacklight

Community needs to get back to ArchivesSpace about the management of the born digital
Out of the box ArchivesSpace discovery layer looks like blacklight

what does an integrated system look like? Flow from modification of managed finding aids
making it into discovery layer- how to complete the loop?

Mike Durbin described possibly creating a proxy back end to push some data to the system and some to fedora

end-user and curator will both need to be able to play the digital items such as audio/video

What about format migration?
tika and FTK can create derivatives that people can use

a lot of old formats there is no extractor/emulator for

most urgent questions:
1. are the digital items safe - backed up?
2. staff can access when they need to on their own through authentication
3. transfer some of it to the public in the future

Should test out tika - java api - does it change the modification date when data is pulled?

plans for ArchivesSpace at Stanford - inconsistent currently going to be a problem going from AT to AS probably for the Getty as well

There are key connections that need to be made between ArchivesSpace and Blacklight/Hydra

Mike Durbin should have something together in the next couple of months - talk about it at OR - going through user testing now

Move to ArchivesSpace presentation of finding aids in solr/blacklight - not an application issue - an indexing issue
need grant to put together the best way to present finding aids in blacklight
Maybe to start with, put together everyone's solr schemas and talk from there

need item level discovery for digitized items - with context - use case of exhibits with data being pulled from finding aids

« Back