Information Technology Reference
In-Depth Information
ties that help the stakeholders gain the maximum
benefit of the space.
The main challenge of a cross-domain interop-
erability platform is how to solve the problems
which are encountered due to changes that will
happen in spatial and timing spaces. Due to this,
the smart space is only 'smart' if it is able to
handle these changes in a way that makes the
smart space attractive for its application providers
and end-users.
iii) the inherent dynamism of smart spaces makes
it possible to tolerate changes that happen all the
time in the spatial and timing dimensions. The
stack of the defined ontologies (i.e. foundational,
core, domain, application) helps in the separation
of concerns. Moreover, the separation of concerns
is also made while retrieving a view of the core
ontology for the specific purpose of an application
under development. In this case, the separation
of concerns is supported by the SmartModeller
tool. Aggregation is understood to be an activity
that retrieves information from a smart space,
enhances it and provides the results for the use
of the smart space entities. A set of mechanisms
have been defined and implemented for handling
the dynamism of smart spaces; the adaptable KPs
developed for monitoring, reasoning and adapting
quality, context and behavior of a smart space are
assets that facilitate the managing of the changes
in the spatial and timing spaces.
As a conclusion, although many of the support-
ing software services are still under development
and need to be validated in field tests, most of the
challenges and issues related to the operation and
maintenance of smart spaces have already been
identified or/and partly solved. However, it still
remains open how to collect feedback from smart
space end-users and how to communicate it back
to the smart space developers and owners.
Solutions and Recommendations
In order to improve the understanding of the
stakeholders, i.e. application/information provid-
ers, smart space providers, maintenance providers
and the owners, we have developed and tested a
set of facilities, implemented as KPs and a tool
that is connected to the smart space for monitoring
data, events, quality, structure and the behavior
of the smart space according to the interests of
the respective stakeholders. The tool is able to
record and visualize what is happening in a space
and how perfectly it works. However, it doesn't
record how individual space users behave, what
their interests are, and how they experience the
smart space. Although we don't expect these kinds
of special KPs to bring any technical problems,
there is a privacy issue that has to be considered:
the recording of habits, desires and experienced
quality has to be made anonymously. So far, dem-
onstrators are used for validating the capabilities
of a smart space. However, field tests will also
be carried out when the cross-domain pilots are
refined and ready for the use of real end-users.
Ontology driven software engineering that
enhances the model driven engineering is one step
forward in seeking a comprehensive solution for
managing the evolution of cross-domain architec-
tures, due to the facts that i) abstraction helps the
understanding and sharing of knowledge among
developer teams; ii) aggregation by a means of
shared information makes it easy to develop partial
solutions and integrate them via a smart space; and
FUTURE RESEARCH DIRECTIONS
The development of smart spaces has, at least, two
main trends: the focus of smart space development
is on i) combining the physical, digital and user
contexts in order to provide enriched experiences
for the smart space users; or/and ii) constructing
smart spaces that are able to self-monitor, self-
reason, self-configure, and self-organize their
capabilities and resources based on trade-offs anal-
ysis made at run-time. The 'context-awareness'
trend aims at an increased added value of end user
applications. The 'autonomic smart spaces' trend
Search WWH ::
Custom Search