Metadata and Metamodeling

Related Work

Metadata repository systems does not necessarily provide all the services we have described previously. Depending on design goals, some systems focus on middleware services such as authoring (or annotating) while others on modeling and queries. Many systems omit security services, and leave the implementation to underlying protocol layers, i.e. if HTTP is used, simple HTTP/Web server authentication is preferred. We first describe several systems, and then, discuss implementations in following sections.

Annotea [67] is a Web-based shared annotation system based on a general purpose open RDF infrastructure. It aims at simplifying metadata generation and accessibility of online well-structured Web documents. Annotea does not specify any implementation model for client applications, however, a modified version of an open-source Web browser, Amaya [110], is provided as a sample client application.

Sesame [25] is “a Web-based architecture that allows persistent storage of RDF data and schema information and subsequent online querying of that information.” It is basically a middleware application that encompasses services and modeling tiers including a Repository Abstraction Layer. Sesame’s modular architecture allows applications to communicate with users in different protocols in the application tier, and various data store solutions to be deployed in the data store tier seamlessly.

Sesame adopts RDF and RDF Schema for modeling, and focuses on queries in RDF databases. It does not specify any propreitary modeling language, identification or authoring methods.

RDFSuite [4, 69] is a suite of tools for RDF validation, storage and querying. A RDF validator parses RDF descriptions and schemas, and creates internal RDF representation models. A persistent RDF store loads RDF descriptions and schemas into an object-relational DBMS using the models the validator creates. Finally, a declarative language, RQL [69], is used for querying RDF descriptions and schemas. RDFSuite does not propose any authoring and realization system. URIs as defined by RDF are used for resource identification purposes.

Scientific Annotation Middleware (SAM) [83] is an architecture that is intended to form common grounds for scientists in a particular field to share and exchange scientific records and knowledge through collaborative resource annotation tools accessible from variety of computing environments and applications. The Middleware comprises metadata management services, semantic services, and electronic notebook (EN) services. Each service can be accessed through either specialized applications or other third party applications using the programming interfaces.

The Collaboratory Multi-Scale Chemical Science (CMCS) [85, 90, 89] is a general purpose metadata management portal for scientific communities. While SAM is a middleware solution, CMCS provides a suite of tools as interfaces to SAM to manage and share metadata and automate metadata generation and handling.

The Indigenous Knowledge Management (IKM) system [62] is an annotation, management, and accessibility system for multimedia resources whose intellectual property rights belong to communities, i.e. indigenous groups, rather than individuals. IKM focuses on protecting the rights of communities by providing access controlled collaborative annotation systems where access controls are based on available metadata models on individual property rights with extensions for communities. IKM comprises software tools to generate, browse, search, and modify metadata which follows a specific schema. It also requires a DBMS to store the sensitive multimedia content.