Hi Christiaan,

It is true, our archivists only use a subset of the EAD cause they use the ISAD(G) (http://www.ica.org/biblio/cds/isad_g_2e.pdf) has a guideline for making EAD descriptions, but we will probably use the official schema http://www.loc.gov/ead/eadschema.html (still in beta).

The recursion in the EAD (and MODS) schema is the core of interest for the archivists in my perspective, cause the EAD will not describe only  one digital object but many, organizing them in an hierarchy of Fonds / Sub-Fonds / Series / Sub-series /  Files / Items like the Communities / Collections /  Items structure in Fez. So the EAD cannot be kept as a document XSD, cause it describes many documents, and a hierarchy above them. We are trying to use the same approach as Fez had to the Communities / Collections / Items hierarchy, keeping this hierarchy in Fedora has a group of records related by RDF, and maybe replacing the Fez structure, or make a Fez for our own.


On Sep 12, 2006, at 8:33 AM, Christiaan Kortekaas wrote:

Hi Luis,
I have experimented with an EAD XSD as a datastream in a Fez content model once before. It was the Princeton EAD XSD. The problect with that XSD for EAD
was that it was extremely large and very recursive. It kept on referening back to an element in the XSD back up the tree so that basically an XSD parser gets stuck in an endless loop if
it doesnt decide stop refering back up the tree at some stage.
I am working on making MODS part of the defautl content model for the next version of Fez right now (for many reasons) and it also refers recursively back up the XSD tree
with the MODS:relatedItem element. I am currently getting around this by removing relatedItem from the MODS XSD schema but the real solution is to only along one recursion back to above the current
XSD tree element which would also solve the problem for the Princeton EAD XSD.
Out of curiousity which EAD XSD have you looked at? Perhaps there is on that is not so infinitely recursive as the Princeton example. Eg an EAD lite or similar would be better as there appeared
to be only a small subset of the entire schema that our archivists really used.
We would like to support EAD natively at some stage in the future as our archivists are very interested in it. I have written an archival system supporting EAD for our archivists before (internally here at UQ about a year ago) so I understand they differing needs to "normal" librarians. We used the archival system "Concord" before this internal PHP/MySQL internal archival system so replicated most of its functionality. We would like to migrate to Fez + Fedora from it at some stage in the future once we can support all the archivists specific needs eg with native EAD and series/provenance/accession structure.

From: fez-users-bounces@lists.sourceforge.net [mailto:fez-users-bounces@lists.sourceforge.net] On Behalf Of Luís Faria
Sent: Friday, 8 September 2006 2:10 AM
To: fez-users@lists.sourceforge.net
Subject: [Fez-users] EAD in Fez

I'm currently involved in a project of implementing a Authentic Digital Object Repository (RODA) using Fedora. This project is aimed for archival institutions, with different needs from the libraries. 
For one the object model cannot be the one implemented by Fez (Community -> Collection -> Record). The chosen object model is a descriptive metadata schema called EAD.

My questions are:

1 - How can I change the Fez Object Architecture (http://espace.library.uq.edu.au/documentation/templates/design.html)?
2 - Are the core of Fez (workflow system, search and browse engines, item visualization - dissemination) and the object model interdependent?
3 - Which is the relation between FezACML and Fedora's XACML? Why not use Fedora's user and object permissions instead?
4 - Can I use other schema instead of the FezMD?
5 - For images (as an example) Fez creates thumbnails and previews while the Ingest Image workflow. Is it possible to do this with Fedora's disseminators?
6 - Could we use pluggable Viewers over the fedora's disseminators to present information  in Fez?

Thank you,
Luis Faria
National Archive Institute of Portugal

Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
Fez-users mailing list