[foaf-dev] Extending FOAF for modeling organizations
Kris Van den Bergh
kris.vandenbergh at gmail.com
Sat Jun 19 13:26:04 CEST 2010
I would like to discuss an idea I have with regards to using /extending FOAF
for modeling organizations.
FOAF is usually used for saying things about people. But FOAF descriptions
may also contain information about organizations, for example to which
organizations that people belong to. Since FOAF provides a framework that is
extensible, I was wondering to what extent FOAF could be used to model
When looking at the FOAF Vocabulary Specification, the following is stated
about Organizations: “The
represents a kind of
Agent <http://xmlns.com/foaf/spec/#term_Agent> corresponding to social
institutions such as companies, societies etc.” This leads to the following
foaf:Organization rdfs:subClassOf foaf:Agent
But does it end there? I am aware that FOAF is very lightweight and uses
fairly simple modeling constructs. It is rather a means to maintain
consistency and to implement an information merging solution. It relies on
the inferencing structure of RDFS to add completeness to their information
structure. However, I think that its mechanisms could be used capture the
hierarchy within an organization.
This got me thinking where to start. First of all, if there already exist
vocabularies that you know of and tackle the organization modeling problems,
please inform me.
An approach would be the following. I think that we could learn a lot from
the big sociology thinkers like Mintzberg. For example, consider the 7 C’s
for organization characterisitics: organisational design may differ
according to configuration, complexity, compliance, centralisation,
coordination, communication and compensation. I strongly believe that not
all organizations are flat or will be anytime soon. Therefore, I suggest to
build a vocabulary and to develop several patterns that support different
but typical organisational plans/configurations going from a simple
structure to machine bureaucracy, professional bureaucracy, divisional
structure and adhocracy. The patterns will build on differences in
structural, people and environmental characteristics Organizations could
then adopt these patterns and make their changes accordingly.
The resulting ontologies could then form the basis for developing an
enterprise ontology that uses the organisational components as a top-down
approach, followed by a bottom-up approach using a variety of clustered
I know that this involves a lot of unverified assumptions, but as I said,
this is just to structure my mere thoughts. I am certainly no expert in this
matter, that’s why I’d like to discuss this here in the first place, where
there are people that have something useful to add.
If I need to clarify something, feel free to say so!
Eagerly awaiting your responses!
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the foaf-dev