Y. A. Bishr1), H. Pundt2) , W. Kuhn2), M. Molenaar1), M. Radwan1)
1)International
Institute for Aerospace Survey and Earth Sciences (ITC), Enschede, The
Netherlands
yaser@itc.nl
2) University
of Muenster, Germany
Institute for Geoinformatics
(IFGI)
pundt@ifgi.uni-muenster.de
Summary
The information communities model is an important part within the framework of the Open GIS interoperability specification (OGIS). The paper discusses the idea of information communities, raising issues like:
The complexity of geodata is one main factor which makes interoperability between different data sources and software systems so difficult to achieve. The complexity is caused by various factors, such as the underlying digital formats imposed by a particular software application or acquisition method and the complexity of higher level descriptions, conventions, and rules imposed by individuals, organizations, and disciplines using the software (Buehler & McKee 1996). The notion of interoperability has different meanings depending on whether it is used by network designers, operating systems designers, or application software engineers. As shown in figure 1, interoperability can be viewed at six different levels, where network protocols are at the lowest level and the information community interoperability is at the highest level. Today's desktop and enterprise environments lack interoperability between geographic information systems. The lower four levels of interoperability provide a distributed computing platform where interoperable GIS can be built on. By GIS interoperability we mean that users can transparently access and share remote geospatial databases and services regardless of their underlying GIS platform. Different applications have different world views, different representations, different schemas and hence different semantics.
Heterogenity at the information community level is a semantic problem, which is due to the differences in the interpretation of the spatial data encoded in the database. The problem of heterogenity at the application level has to be tackled, considering application interoperability and semantic interoperability as synonyms.
In response to the need for interoperability the Open GIS Consortium (OGC) is working on the Open GIS interoperability specification (OGIS). The OGIS framework includes three parts (Buehler et al., 1996), the Open Geodata Model, OGIS Services Model, and the Information Communities Model.
The proposed paper will present a refined model of information communities and semantics mapping. The first chapter will describe the current view of information communities briefly. In addition to this view it will be shown, that problems of semantics mapping can occur both within and accross information communities boundaries. This view is schematically shown in figure 2. The assumption that semantic mapping is essential for data sharing between and within information communities requires case studies of specific applic ation areas.
In the second chapter we will present some examplary problems of geodata sharing which can occur in freshwater ecology, watershed management and transportation. These problems include the use of:
The next section will describe an approach how data sharing between different information communities can be realized without casualities in semantics. In the real world semantic interoperability is achieved by providing metadata about the related data set. Notwithstanding the ability of the metadata to provide an insight in related data sets, users are required to map the retrieved data from the domain of the provider to their own domain. The availability of semantic translators is required to support this task. Semantic translators are middleware components which allow heterogeneous applications to communicate and share data: