Will,

We had very similar requirements to those that are in your email and on your
blog.  And we developed a web-based administrative interface to our fedora
 repository to meet our own needs.  It allows us to directly manage the
 metadata and content in our repository.  It is scalable, in that we can
 manage all of our different digital object types (image, tei essay, video,
 ead finding aid, book, and collection) through one interface.

 Some of the features of our admin interface:
 * ability to create new digital objects
 * ability to search existing digital objects
 * ability to upload/replace/delete content
 * ability to edit metadata
 * ability to assign digital objects to collection(s)
 * ability to feature digital objects in collection(s)
 * data integrity checking based on digital object type and collection-based 
metadata rules; digital object candidate statuses
 * automated creation of derivatives (at least for images - thumbnails and 
zoomify tilesets)
 * web-based descriptive metadata editing screens
 * ability to manipulate structural and relationship metadata of digital 
objects
 * controlled vocabulary for metadata entry

 Some features being worked on:
 * dynamic lookup during metadata entry for semi-controlled vocabulary lists
 * online help for metadata editing

 In our admin interface, we have developed a tool that meets our needs, but
 is not flexible to meet the needs of the field.  It is currently tied to 
our
 Fedora implementation, our data model, and our choice of metadata schemes.

 Dave


 ----------------------
 David Kennedy
 Manager, Digital Collections and Research
 University of Maryland
 3199 McKeldin Library
 College Park, MD 20742
 [log in to unmask]
 (301) 405-9051
 (301) 314-9408 FAX


 Will Sexton wrote:
> In January of 2007 I sent a post to the Web4lib list titled "Metadata
> tools that scale."  At Duke we were seeking opinions about a software
> platform to capture metadata for digital collections and finding
> databases.  The responses to that inquiry suggested that what we were
> seeking didn't exist.
>
> About a year ago, an OCLC report on a survey of 18 member institutions,
> "RLG Programs Descriptive Metadata Practices Survey Results," supported
> that basic conclusion.  When asked about the tools that they used to
> "create, edit and store metadata descrptions" of digital and physical
> resources, a sizable majority responded "customized" or "homegrown" tool.
>
> Since my initial inquiry, we launched a new installation of our digital
> collections at http://library.duke.edu/digitalcollections/.  Yet we still
> lack a full-featured software platform for capturing descriptive metadata.
>
> We did our own informal survey of peer institutions building digital
> collections, which further reinforced that familiar conclusion -- there
> are lots of Excel spreadsheets, Access and FileMaker databases, etc., out
> there, but no available enterprise-level solution (and we're still happy
> to be wrong on this point).
>
> We also articulated a detailed series of specifications for a metadata
> tool.  The library has committed to hiring two programmers each to a
> two-year appointment for producing a tool that meets these specs.  I just
> posted on this list the job description, for which there are two openings.
>
> I have a longer version of this post on our digital collections blog
> (http://library.duke.edu/blogs/digital-collections/2008/10/10/a-metadata-tool-that-scales/),
> listing our specifications in more detail.  But here are some of the
> basics:
>
> * Digitization:  integrates with, or provides a module for, management of
> digitization workflow.
>
> * Description:  supports a collections-based data model; flexible metadata
> schema (for us, the "Duke Core", derived from qualified Dublin Core);
> authority lists; cardinality and required-field constraints; metametadata
> (i.e., flagging, notations and status indicators for individual items);
> access control; simple and intuitive use.
>
> * Publication:  exports METS documents as well as other common formats
> (CSV, etc.).
>
> * Asset Management:  must be compatible with an asset management policy.
>
> While the Duke specifications are particular to our internal needs, I
> think we captured a lot of what makes the need for a full-featured
> metadata tool felt around the field.  I have some ideas about how to go
> about implementing this set of specifications, but thought I'd see if the
> concept might spur discussion on CODE4LIB.  How would you approach this
> project?  Any thoughts on architecture, platform, data models,
> methodologies?
>
> Will
> --
> Will Sexton
> Metadata Analyst / Programmer
> Duke University Libraries
>