18
Carlo Combi, Sara Migliorini, Barbara Oliboni, and Alberto Belussi
example, precipitation, temperature, and pollution values. This view of space as a
continuous field is in contrast with the view of the entity-based model that describes
the geographical objects that populate the space.
Any effective language for representing geographical information should be able
to model data in both the above described approaches. The GML [13, 20] satisfies
this requirement. It is an XML grammar written in XML Schema that is used to
model, exchange, and store geographical information, including both the spatial and
non-spatial properties of geographical features.
GML provides different kinds of objects for describing geography including features, coordinate reference systems, geometry, topology, time, units of measure, and
generalized values. GML is the only language we found in the literature that supports
the representation of semistructured data for geographical information.
In GML, there are essentially three types of geographical object: (1) basic GML
feature is the type used to represent a geographical entity, that is, any meaningful
object in the selected domain of discourse such as a Road, River, Person, or Administrative Boundary; (2) coverage is a type of feature that includes a function with a
spatial domain and a value representing a set of homogeneous two- to n-dimensional
tuples (i.e. it describes the spatial distribution of the Earth phenomena); (3) observation is considered to be a GML feature with the time at which the observation took
place and with a value for the observation.
GML is designed to support interoperability through the provision of basic geometry tags, a common data model (features-properties), and a mechanism for creating
and sharing application schema.
2.2.4 An XML-based Representation of Semistructured Spatiotemporal
Information
In [14], the authors explore how information about location and time can be integrated and represented in XML. In particular, they consider how to represent spatiotemporal types through XML Schema. The authors propose several data types in
XML Schema to represent both basic temporal elements (i.e. a sequence of disjoint time intervals) and histories of spatial objects. The VT of a spatial object is
represented by a temporal element and spatial histories are modeled as temporal sequences of spatial objects, through the parametric XML schema type Time-Series.
The XML-STQ spatiotemporal query language allows the user to express queries
containing both temporal and spatial objects; as an example, it is possible to detect when a geographical area has some specific spatial relation with another spatial
object.
This proposal deals with the representation of spatiotemporal information in the
XML context, while the data model we present in this chapter is able to manage
spatiotemporal information in the semistructured data context. Moreover, in [14],
the authors propose the representation of histories by means of sequences (with
respect to time) of spatial objects, while our new data model allows one to represent the VT dimension of each piece of geographical information.
Carlo Combi, Sara Migliorini, Barbara Oliboni, and Alberto Belussi
example, precipitation, temperature, and pollution values. This view of space as a
continuous field is in contrast with the view of the entity-based model that describes
the geographical objects that populate the space.
Any effective language for representing geographical information should be able
to model data in both the above described approaches. The GML [13, 20] satisfies
this requirement. It is an XML grammar written in XML Schema that is used to
model, exchange, and store geographical information, including both the spatial and
non-spatial properties of geographical features.
GML provides different kinds of objects for describing geography including features, coordinate reference systems, geometry, topology, time, units of measure, and
generalized values. GML is the only language we found in the literature that supports
the representation of semistructured data for geographical information.
In GML, there are essentially three types of geographical object: (1) basic GML
feature is the type used to represent a geographical entity, that is, any meaningful
object in the selected domain of discourse such as a Road, River, Person, or Administrative Boundary; (2) coverage is a type of feature that includes a function with a
spatial domain and a value representing a set of homogeneous two- to n-dimensional
tuples (i.e. it describes the spatial distribution of the Earth phenomena); (3) observation is considered to be a GML feature with the time at which the observation took
place and with a value for the observation.
GML is designed to support interoperability through the provision of basic geometry tags, a common data model (features-properties), and a mechanism for creating
and sharing application schema.
2.2.4 An XML-based Representation of Semistructured Spatiotemporal
Information
In [14], the authors explore how information about location and time can be integrated and represented in XML. In particular, they consider how to represent spatiotemporal types through XML Schema. The authors propose several data types in
XML Schema to represent both basic temporal elements (i.e. a sequence of disjoint time intervals) and histories of spatial objects. The VT of a spatial object is
represented by a temporal element and spatial histories are modeled as temporal sequences of spatial objects, through the parametric XML schema type Time-Series.
The XML-STQ spatiotemporal query language allows the user to express queries
containing both temporal and spatial objects; as an example, it is possible to detect when a geographical area has some specific spatial relation with another spatial
object.
This proposal deals with the representation of spatiotemporal information in the
XML context, while the data model we present in this chapter is able to manage
spatiotemporal information in the semistructured data context. Moreover, in [14],
the authors propose the representation of histories by means of sequences (with
respect to time) of spatial objects, while our new data model allows one to represent the VT dimension of each piece of geographical information.