Riley, Jenn. "Implementing Shareable Metadata Practices in a Diverse University Environment" Society for American Archivists Annual Meeting, August 31, 2007, Chicago, IL
2. An vision for shareable metadata
Many robust discovery environments that can
easily exchange diverse metadata
Workflows at Universities that easily bring
together metadata for resources from
libraries, archives, etc.
Large-scale, multi-institution aggregations
that do the same
…and online delivery of content too, but let’s
not get too far ahead of ourselves
8/31/07
SAA 2007
2
3. Challenges for archives
“Archives” not a homogeneous body so a
single workflow likely not possible
Expanding view of mission
Finding the resources
Appropriate description
Technical implementation
8/31/07
SAA 2007
3
4. Challenges for digital libraries
Digital library practice assumes content
digitized
Metadata-only workflows not common
Digital library practice assumes item-level
description
In University DL departments
In metadata aggregations
8/31/07
SAA 2007
4
5. The core descriptive challenge
EAD is not a metadata format.
EAD is a markup language.
It marks up the finding aid as a document
A finding aid is a narrative, not just an inventory
The structure of finding aids can vary among
diverse collections
Let’s assume we have EAD-encoded finding
aids for all collections we want to share
8/31/07
SAA 2007
5
6. The good news
Aggregations not intended to replace
archives-focused discovery mechanisms, but
rather to supplement them
General-purpose aggregations therefore do
not need to make use of all the nuances of
archival description
Your local environment can provide the
robust services you want
Interpretation of resources
Mediation of access
8/31/07
SAA 2007
6
7. Key shareable metadata principles for
archives
Context
Need enough so that the user of an aggregation
can make a decision about whether or not to
follow a link to the local environment
Too much repeated information can pose
challenges for the aggregation and the user
Content
What is the appropriate granularity for shared
records from archives?
8/31/07
SAA 2007
7
8. Some possible strategies
Collection-level records only
Aggregators that understand multi-level
description
Design multi-level description carefully for
future item/file-level view
Link to digital object from the lowest level of
description in finding aid, and use external
system to provide more granular description
Describe at the item level
8/31/07
SAA 2007
8
9. Implications for EAD workflow
Because EAD encoding practice can differ so
widely, decisions about flattening the
structure best made locally rather than by an
aggregator
Adding additional uses for EAD files makes it
all the more important that they’re done well
8/31/07
SAA 2007
9
10. Experiences at IU (1)
New EAD site
<http://www.dlib.indiana.edu/collections/findingaids/>
New system more faithful to encoding, with less
“helpful” fixed presentation
Mutual learning process about archival descriptive
practices
Many decisions made about when encoding
should be changed and when system should be
changed
Results of this process: re-engineering! New
template, report card, better previewing capability
8/31/07
SAA 2007
10
11. Experiences at IU (2)
Some EAD files link to digital objects (more
soon).
Soon, item-level OAI records (DC and
MODS) for digitized items from finding aids
Central DL repository that allows EAD as the
master metadata format
Workflow that allows links from any level of a
multi-level description in EAD
8/31/07
SAA 2007
11
12. What archives in Universities can do
Build archival-specific aggregators
Implement OAI data providers
Talk to DL people about multi-level description
Put more materials online
Put EAD site in your institution’s federated search
Share finding aids with services such as ArchiveGrid
Lobby software vendors for better support of EAD
Implement EAC and other means of sharing
authorities
8/31/07
SAA 2007
12
13. Learning from one another
Item-centric view can be too narrow, but can
help the re-engineering process
More structure in finding aids can be a good
thing
Archives can show libraries why expertise in
descriptive practice is still necessary
8/31/07
SAA 2007
13
14. Thank you!
Jenn Riley, jenlrile@indiana.edu
These presentation slides:
<http://www.dlib.indiana.edu/~jenlrile/presentations/saa2007/>
Metadata For You & Me: A Training Program
for Shareable Metadata
<http://images.library.uiuc.edu/projects/mym/>
8/31/07
SAA 2007
14
Notas del editor
Collection-level records only (UIUC has shown these are useful, but mostly to supplement item-level description)
Aggregators that understand multi-level description (appropriate groupings in search results; hasn’t happened)
Design multi-level description carefully for future item/file-level view (can be part of Greene’s reengineering process described 10 years ago; Prom’s multiple records approach that hasn’t been widely picked up)
Link to digital object from the lowest level of description in finding aid, and use external system to provide more granular description (if you’ve got it, why not put it in EAD, though?)
Describe at the item level (likely the collections that will be digitized are the same that would receive item-level processing per Greene/Meissner)
EAD files linking to digital objects: Mostly item-level description, but some with multiple links from file-level descriptions. Digitization process has forced re-engineering of individual finding aids too