Information Technology Reference
In-Depth Information
provides a mapping from the context metamodel
to the target metamodel. The approach tackles the
structural and static parts of smart space service
creation, i.e. context categories, context sources,
temporal constraints and contextual situation.
However, the dynamic aspects of context-aware
services are not supported. The ontology and the
model driven development approach (Soylu et al.
2009) exploit the model abstractions of MDA and
the commonly used modeling languages UML
and OWL. The approach enhances the software
engineering process with computing independent
domain ontology for modeling platform inde-
pendent applications and a context ontology for
reasoning. The integrated process model of the
approach is however very abstract and does not
increase understanding about how the context
ontology is to be specified, represented, and pro-
cessed at design time and run time, and how the
context ontology is to be transformed for the use
of various architectural elements, i.e. applications,
services and data, in run-time reasoning.
As stated above, the ontologies for defining
quality attributes are rare and incomplete. In
(Ovaska et al. 2010), the ontology orientation
is used for defining quality attribute ontologies,
especially for the defining of their metrics. On
the one hand, the quality and model driven design
methodology which is introduced exploits the on-
tology oriented design for specifying, representing
and managing quality attribute specific knowledge
by ontologies. On the other hand, architectural
knowledge is specified, represented and managed
as styles and patterns. Both of the types of models
can evolve separately. The mapping is made by a
tool chain that supports each development phase
of the model and quality driven service engineer-
ing. Recently, the interest on using ontologies for
describing and managing quality attributes has
increased due to the growing awareness of the
importance of quality characteristics in service
oriented systems. Moreover, the quality aspects
need to be managed not only at design time but also
at run time. In (Kassab et al. 2009), an ontology for
non-functional requirements has been introduced
with three views; an intramodel dependency view
for describing the relations between the software
entities, an intermodal dependency view for de-
scribing the structure of interdependent entities,
and a measurement view for defining measurable
requirements. These views are required for manag-
ing quality properties at run-time; the intermodal
view defines what the quality property is and how
to put it into realization, the intramodel dependen-
cies are used for reasoning purposes; and metrics
are used not only for defining quality goals, but
also for measuring how these goals are achieved.
Agile Smart Space
Development and Evolution
Despite the enriched set of software and service
technologies and development methodologies
which were introduced in the previous section,
there are still some challenges and issues that the
developers of smart spaces are encountering. To
facilitate the agile development of smart spaces, we
propose a novel methodology called Agile Smart
Space Development and Evolution (ASSDE). To
provide agility, ASSDE exploits some of the main
properties of SCRUM (Scrum 2009):
1. It breaks the work down into manageable
chunks, called scenarios, which can be imple-
mented in a few weeks. Scenarios embody
user centered design, which is another stone
base of ASSDE.
2. It enables the project team to proceed sys-
tematically, even when a complete and stable
design cannot be defined.
3. It allows large globally distributed teams to
work like small teams by dividing work into
pieces, proceeding in parallel but synchroniz-
ing continuously, stabilizing in increments,
and continuously finding improvements
through refinements and extensions.
Search WWH ::
Custom Search