The Design Principles and Practices of Interoperable Smart Spaces - Advanced Design Approaches to Emerging Software Systems - page 34

with the following attributes: a significance for
architecture/application/space and relations to the
interoperability levels, related scenarios, existing
enablers and required enablers.
Next, the specified requirements were carefully
analyzed and prioritized. For ranking purposes,
we used the same criteria as when ranking the
scenarios: i) the maximum business impact, and
ii) the fast and low-risk realization criteria. As a
conclusion, the IOP requirements of high priority
(16 quality requirements and 12 functional/non-
functional requirements) were used as architec-
tural drivers that guide the definition of the IOP
principles to be followed while architecting the
IOP. Evaluation criteria were also defined for IOP
instances, based on these two sets of requirements.

features should be met at a high quality level.
Among the “qualities” considered to be relevant
to assess the value of the IOP, the following items
should be primarily addressed:

• business generated by the IOP based smart

environments,

• the easiness to develop applications on top

of IOP,

• models that describe interaction among us-

ers, their environment and their history

• the security and dependability of IOP

based applications

• IOP performance, energy efficiency and

scalability.

We claim that an IOP with significantly high
scores on the above properties may originate a
new market for applications that can interoperate
independently from their business/vendor/manu-
facturer origin. This can introduce a radical change
to the traditional application scenario, which is
based on fixed business boundaries. The expected
applications may be cross-domain, may adapt to
the user situation, may spontaneously start when
required and have the potential for significant
market penetration and socio-economic impact. In
order to define the architectural style of the IOP,
sixteen principles were distilled, starting from the
above vision, and summarized in Table 2.

IOP Principles

Prior to defining the IOP principles, several work-
shops were carried out in order to get a consensus
on the focus of the IOP. As a conclusion of these
discussions, the partners involved in the smart
space development agreed on the main objective
of IOP: to provide an infrastructure that assists
users with added-value interoperable information
about objects existing in the environment of the
user. Therefore, the IOP reference model is to be
defined at a high abstraction level and be simple
and agnostic with respect to i) the use-cases, ii)
information and iii) the physical environment
(including legacy equipments). These were ex-
pected to enable the level of extensibility required
to support multi-domain and cross-domain ap-
plications. Thus, IOP combines the information
interoperability solution with existing service
and physical level interoperability solutions. The
interoperability is based on the common ontology
models of information that may originate from
heterogeneous legacy and embedded devices
spread in the environment or may be produced
by the aggregators of IOP information. IOP may
be active and trigger external entities to react to
relevant and selected environmental changes. Such

IOP Enforcement

The IOP herewith is considered to be a simple, por-
table and open source implementation. It is called
Smart-M3 2 and it takes the principles seriously. It
consists of an interoperability component hosting
the shared information space, seen as a service and
accessed through a semantic protocol by engines
interpreting the shared information space through
an ontology; the shared information store is called
SIB (Semantic Information Broker), the agents
are named Knowledge Processors (KPs) and the
semantic protocol, i.e. the interface between the