Database Server Models for WMS and WFS Geographic Web Services - Geographical Information Systems: Trends and Technologies - page 144

As a consequence, end-users must be aware of the distant database schema
and its semantics (with conventional problems such as aggregative/metrics
units, clear defi nition of the scope, and semantics of the attributes). Some
data available in the database schema are relevant for their needs, some
are not.
GetFeatureInfo and GetFeature operations of the WMS/WFS services
will query the database schema. They provide the set of data associated
with a map or a geographical object. They present some limitations since
they cannot query an external schema adapted to a specifi c user's need
and no information about the time required to obtain data is available.
They require the analysis of distant database schemas. Furthermore, people
involved in this task are not the distant database designers since they are
distant application designers. The results of WMS/WFS operations may
therefore not be adapted to decision making from a specifi c application
point of view (e.g., tourist, social, economical, transports).
In order to adapt available attributes to end-users' needs, we
propose a solution that is independent of GetFeatureInfo and GetFeature
operations of WMS and WFS. We still respect the normalized interface of
these operations within these services. We propose an extension of the
database schema in distant databases in order to precise answers that these
operations may provide. The answers may be alphanumerical attributes or
results of methods (i.e., functions) viewed as alphanumerical or graphical
attributes. The proposed solution is to enhance the data model of distant
databases with typed links on a conceptual lattice and information about the
operational complexity required to obtain data. These types are defi ned by
distant database administrators (i.e., people who have the best knowledge
of available information and the required time to get these data). This
solution provides a dynamic data model as a result of querying and is a
semi-automatic process to defi ne relevant data for these operations.
In the second section, we present the state of the art in the spatial
enhancement of data models. In the third section, we present data model
extensions suggested by us. In the fourth section, we present pre-defi ned
rules to choose relevant attributes for these operations. In the fi fth section,
we present a conclusion and some perspectives.
State of the Art
We present in this section, research works devoted to enhance the semantic
in geographical Web services. Two main approaches are possible: the
extension of the norm or to defi ne a service that is closer to end-users'
needs.
In the fi rst part, we present solutions promoted by the Open Geospatial
Consortium (OGC 2005). In the second part, we present a model-oriented