Information Technology Reference
In-Depth Information
several approaches to achieve the automation of Web service discovery are
presented and discussed. They vary in complexity and precision and require
different levels of semantics in the description of Web services and requests.
9.2.1 Discovery Framework
A workable approach to automating service discovery must define precisely
its conceptual model and the particular assumptions underlying the proposed
solution. To describe such an approach, we start by aiming to provide a com-
mon understanding of what a service is and the levels of abstraction in its
description, on the basis of [111], and we also state our assumptions about
the elements involved in the process of locating suitable services for service
requesters. First we recap parts of the discussion from Section 4.1 (on Web
service terminology). We use the following interpretation of the term service ,
as described in the conceptual architecture for Semantic Web services pre-
sented in [111]: the provision of value in some domain . This definition regards
a service as a provision of value in some given domain, independent of how
the supplier and the provider interact. Usually, a service provider does not
provide only one particular service, but a set of coherent and logically related
services. For instance, a hotel usually does not provide only the possibility to
book a particular room on a particular date for a given number of nights, but
instead it will offer (and advertise) the general service of booking rooms at
that hotel. Following the terminology of [111], we call this collection of services
an abstract service offered by a provider. The smallest unit of advertisement
is considered to be an abstract service.
In order to deliver a service, a service provider usually needs certain in-
formation from the requester. For instance, a hotel might require the name of
the person booking the room, the requested features of the room, and a valid
credit card number, as input information in order to book a room. This input
data will determine what has to be provided, a concrete service .
Finally, Web services are computational entities using standardized in-
terfaces that allow clients to interact with a provider to access, explore and
consume concrete services. A service requester needs, technically, to interact
with a Web service to get what he/she is actually aiming at: a concrete service,
or, more generally, a set of concrete services that fulfill the requester's goal,
i.e. an abstract service. Hence, we shall not distinguish between Web services
and abstract services of a provider but treat the two notions as synonymous.
Moreover, we can observe that a concrete service is delivered in the course of
or as a result of a particular execution of a Web service, by invocation with
specific data for the required input parameters.
Dynamics of Abstract Services Over Time
In the real world, we have a further complication in regard to discovery and
Web service description. In general, an abstract service offered by a provider
Search WWH ::




Custom Search