Wisconsin Public Television Media Library Online
& CWIS (Collection Workflow Integration System)

Context
PBCore Integration
Contacts

Other Case Examples

Wisconsin Public Television has developed its Media Library Online using an application specification known as the Collection Workflow Integration System (CWIS). CWIS is an open-source software package using mysql and php scripting. It is designed to assemble, organize, and share collections of data about resources, like Yahoo! or Google Directory. However, most importantly, CWIS conforms to international and academic standards for metadata rather than search engine algorithms and folksonomies.

CWIS was specifically created to help build collections of Science, Technology, Engineering, and Math resources and connect them into the National Science Foundation's National Science Digital Library. However, the open-source software can be used for application-specific implementations in other venues. Wisconsin Public Television adapted CWIS for its Media Library Online.

As stated on its Media Library Online website...

Media Library Online is a catalog of media collections held by Wisconsin Public Television. Currently included in the catalog are completed programs from local productions such as "In Wisconsin" and "Here and Now," in addition to field footage and B-Roll. This is a new endeavor to organize and provide access to WPT holdings. The catalog will continue to grow as more items are continually being added. Eventually the catalog will begin to more fully reflect the holdings in the Media Library.

Some of the features of the CWIS/Media Library Online include:

• resource annotations and ratings (a la Amazon)
• keyword searching (with phrase and exclusion support (a la Google)
• fielded searching
• recommender system (a la Amazon)
• OAI 2.0 export (with oai_dc and nsdl_dc schemas)
• RSS feed support
• integrated metadata editing tool
• user-definable schema (comes with full qualified Dublin Core)
• prepackaged taxonomies (includes GEM Subject taxonomy)
• user interface themes
• turnkey installation

The Media Library Online functions as a metadata management tool, not a digital asset management system. Still images can be attached to records for previewing. Moving image files cannot. It is a cataloging tool used by WPT as a specific application for indexing its local, non-digital assets. Production, broadcast, and traffic operations are not integrated into the Media Library. The product is web-based; everything in it is searchable and accessible. Import and export functionalities are under testing and development.

The Summary Page of Media Library Online metadata appears in this illustration...

Data Entry Screens for new records are partially represented in these two illustrations...

And finally, an Asset Search screen can be seen in this illustration...

The WPT Media Library Online was the first metadata management system to use the PBCore metadata dictionary as part of its cataloging system. The Library, as presented by the CWIS software, has 48 metadata elements or fields of descriptions, many based on PBCore v1.0 first published in April of 2005.

The Media Library Online uses a non-hierarchical approach to its data records (flat), not implementing repeatable containers of metadata. Like PBCore v1.0 and Dublin Core, Media Library Online, as presented through CWIS, uses a one-to-one relationship between a complete metadata record and each format or instantiation of a media asset. A single record is devoted to a single format or title.

As observed in the above illustrations, a direct comparison is possible between the metadata dictionary for PBCore and that of WPT's Media Library Online. The PBCore Project is highly supportive of the CWIS implementation. It was the first instance of PBCore, viewable in a real-life application that catalogs media assets.

CONTACTS:

Other Case Examples