16
ECOCLUB, Issue 92
OWL builds on RDF and RDF Schema and adds more vocabulary to describe properties and categories: among others, relations
between classes (e.g. disjointedness), cardinality (e.g. "exactly one"), equality, richer typing of properties, characteristics of
properties (e.g. symmetry), and enumerated classes.
Both OWL (Ontology Web Language) and RDF (Resource Description Framework) have yet to prove their appropriateness within
the collaborative network domain. RDF is the basis for expressing information about all things that can be addressed by a URI basis.
In any case we are confronted with a lot of such things, which is in some sense not a reification of natural things. Often we consider
only real world things (see (RMSIG, 2006), (Werthner, 2001)), but we have also to take into account domain specific processes (as
for instance reservation, booking or canceling are not really reifications of a real world thing). This is evidence of proof that RDF is
appropriate .
However, this must first be evaluated. Experiences described in RMSIG, (2006) and Werthner, (2001) would be very useful in this
context. Although it seems that booking, reservation and canceling are very well understood and defined, the situation with value
proposition for example, is less understood.
As mentioned before, RDF has only been applied to very similar/exact Ontology so far, in tourism for instance to make reservations,
cancellations etc. The Business Model Ontology for tourism will be implemented by either basing it on the standardized RDF/OWL
vocabulary, or by enhancing this vocabulary according to the requirements of the given Ontology.
Collaboration Agents
We already introduced Agents in Withalm, 2000. Next, the most essential behaviors of Agents shall be briefly described. Agents will
roam through the portals of participating organizations, looking for ideal partners. Agents will be programmed by a leading partner in
a way that allows him to decide which partner will be the most suitable regarding Business Strategies, Business Models and Business
Processes. To do so, Agents must look for Semantic Web portals that are characterized, that whose contents are presented in
RDF/OWL. So, partners that are interested in co-operating in a breeding environment must implement their BS, BM, BP in
RDF/OWL according to the established outcomes of ECOLEAD and using the proposed ICT-I of ECOLEAD. In applying this, ICT-
I should contain RDF/OWL generators facilitating the preparation of this collaborative semantic WEB. On the other hand the
required RDF/OWL generators for implementing the Agents are also contained in the ECOLEAD ICT-I. Agents must also be
modified according the concrete requirements of the respective domain (including BS, BM, and BP). These modifications could be
done via attribute-based programming. Agents will then be in the position to roughly look for appropriate organisations.
In ENTER 2004 (Withalm, 2004) the crucial problems (see (FIPA, 2006)) of negotiation issues of agents were discussed. As already
outlined, it seems feasible that the negotiation process of agents on the e-marketplace may be substituted by Semantic Web.
Findings
The proposed approach will use actual projects. The project team will be involved in preparation endeavours, and in the first phase
all results on how to build a breeding environment will be collated. In that phase many of the already developed guidelines,
checklists and tools will be applied in the sense that especially all questions concerning the business strategies – above all trust
buildings and legal issues – and business models will accelerate the decision process regarding which organisations should/could
participate.
In particular, applying rules for trust building and business models (particularly those concerning pillar 2 ‘customer interface’, see
Figure 3) will clarify that some organisations are direct competitors and it would not make sense to continue negotiations via
collaboration agents between them. It will also become clearer, which organisations are most appropriate to join the network.
Especially pillar 1 and pillar 3 (see Figure 3) - value proposition and infrastructure management - will be useful to select the right
organisation.
Value proposition and infrastructure management indicate if organisations may prepare a common value proposition and
infrastructure management in a virtual organisation. For instance, if infrastructure management does not agree, all efforts to automate
these business processes are likely to fail.
Further questions to be answered:
Are RDF/OWL ready for Sociological Ontology?
Are SMTEs able to enhance their web portals with Collaboration and Tourism Ontology by applying RDF/OWL?
Have Collaborative Agents the ability to select appropriate partners based to their business- oriented properties?
Is the programming of Collaborative Agents feasible for service providers?
Has Semantic Web the potential to substitute the crucial negotiation issues of Agents on e-marketplaces (FIPA, 2006)
Conclusions
If the application of all these checklists, templates and tools improves and accelerates the negotiation process, and if all these
concepts for finding appropriate partners for a breeding environment are implemented by means of Semantic Web (i.e. Ontology,
RDF/OWL, Agents), they will become a very powerful tool.
