Designed for Good: Community Well-Being Oriented Online Databases for Youth - Community-Built Databases: Research and Development - page 309

different databases. Consequently, only parts of the data can be stored in one
database. Although persistence layers, for instance in service-oriented architec-
tures, do allow distributed data repositories [ 38 ], these usually are within the
boundary of one organisation or business network.
The diversity of the social collaboration layer means that any attempt to work
with this data in a structured manner can be difficult. Therefore, our conceptual
architecture proposes that the social collaboration layer is designed in a way which
allows that parts of the data on this layer are projected upon a semantic integration
layer.
The semantic integration layer allows the addition of a semantic meta-data to the
data entered in the process of social collaboration. A technology that is capable of
integrating data from different sources in different formats is the semantic web. The
use of semantic web technologies is a common approach to integrate data [ 20 , 32 ,
80 , 95 ]. The semantic web principles are based on the open world assumption,
meaning it is always assumed that more perspectives are added to a certain set of
data and the principle Anyone can say Anything about Any Topic meaning that there
is no ownership of information and that everybody can contribute to any topic go in
line with youth' desire to freely express themselves in ways that, due to the rapidly
evolving nature of the Internet, cannot be anticipated. A community database for
youth must develop these concepts further not just to semantically enable the
content stored in the community database but to be prepared to provide a common
platform to meaningful integrate content from other web sources.
The community governance layer is enabled by the semantically enriched data of
the semantic integration layer. Youth users contribute content by composition,
orchestration and choreography. Governance must address all of these in order to
create a youth space that enhances well-being. While composition is relatively easy
to govern as all of the content resides in one database, orchestration is more
problematic as the data entered by the users is dynamically generated from data
that is stored elsewhere. The choreography of data poses even greater challenges as
the meaning of data may depend on what users have contributed to other platforms.
To fulfil the challenging task of governing a community database, which is
conversation driven rather than ontology driven, advanced intelligence mechanisms
must be employed. These mechanisms can be driven by visualisation, automatic
reasoning or semantic graph-based queries.

12.10 Implementation

We have conducted an explorative implementation to illustrate the key aspects of
our architecture and framework. However, our focus was not on the delivery of a
complete horizontal solution to all of the issues relevant for youth-oriented com-
munity-driven databases. Our focus was on a vertical prototype, which can help in
discovering difficulties in the implementation of projects or reveal issues that
cannot be derived from a theoretical discussion.