The database structures of the archive systems of the partners differ as as can be observed by comparing their → databases. A new metadata scheme was needed, which includes the present contents and processes them for use in the web context.
The OASIS Metadata scheme can be labeled as integrative. It considers the existing fields of metadata of the partner databases. In addition, the basic metadata scheme was produced for the use and circulation of media art content in the context of the internet.
After extensive investigations, the partners agreed on an expansion of the → Dublin Core metadata standards. The expansions consider the specific attributes of media artworks. Alongside, there was implemented a metadata category, which considers the online transfer of multimedia data.
Some of the net specific metadata provide transparency of data sources during transfer. Others serve the accurate online processing in the → OASIS Middleware. Those operate invisibly behind the user interface.
As a reference for the shared → OASIS Metadata model, the
→ DCMI | Dublin Core Metadata Initiative was used, because its references (recommendations) could be adapted to the compilation of electronic content.
In addition, the → W3C–Consortium recommends the DC for the compilation of Web content, and even in the container format → MPEG 7, the included metadata are filled with the help of DCMI categories. → read more
Fifteen predefined core elements characterize the standards in three types of metadata: content based, technical description and rights properties.
OASIS Specific Metadata
During the development of the communication model for the OASIS interface, the problem arose, that common metadata provided by the partners was not sufficient for the networked search: the origin, possible rights and various other information were supposed to be stated, which were to document the systemic processing mechanisms during the search.
Thus, an entire section was arranged through OML-labeled metadata (OML = Online Media Resource) → read more
Connected to these, other tools were utilised, which assist the
→ automatic indexing & systemic processing of metadata: language, text and facial recognition tools were researched and implemented.