ThinkCap comes pre-configured as a WAR file with a set of templates and functions that make it work as a knowledge collaboration portal. Installing ThinkCap in a servlet container creates a web-based collaborative environment where users can construct full conceptualizations in ways that can be tuned to reflect how much technical content they want to see.

The ThinkCap administrator uses the Administration View manages the contents of the knowledge base and the user permissions. As delivered, ThinkCap does not require a login to browse the knowledge base, although this can be changed by the administrator. Only known users with editing permission can propose new concepts to be included. Only users with administrator privileges can edit the ontologies directly.

For users, ThinkCap provides several views that allow interacting with the knowledge base in several ways, ranging from a familiar, Google-like interface to a fully graphical and complete view of the ontologies. Editing is possible in all views if the user has the proper permissions.

The entry view for most users, and the one that requires the least knowledge of ontologies, is the dictionary view that comes in two flavors.

The simplest view presents only a single search field where users can enter search terms. Upon entering of search terms, the knowledge base is searched and all concepts in it are tagged for relevance of the search terms, just like in a reqular web search. Behind the simplicity of this view lies a sophisticated text indexing engine (based on Apache Lucene) which searches not only the concept labels and descriptions from ontology annotation properties, but also any external resources (such as web pages, PDF documents, word documents) that the concepts are connected to. All matching concepts are presented to the user in order of relevance, with short descriptions and a summary of their properties. Buttons allow the user to bring up extensive explanations of the concepts or to go to a different view (e.g. the Thesaurus view discussed later) centered on the concept, to explore its relationships with other concepts in detail.

The Dictionary view has a second, more advanced interface that allows more sophisticated constraints to be specified. This view contains a geographical widget to restrict to concepts that are relevant to a particular area, a similar temporal widget, and several text fields that allow searching in particular fields, load keywords from specific disciplinary areas, or turn search on associated documentation on and off.

The Thesaurus View allows a graphical representation of a concept space where most complexities of an ontology are hidden, and relationships between concepts are substituted by a single link that can be read as "related to". This allows a quick grasping of the conceptual space without being presented with too much detail or unfamiliar notation. Concepts can be seen with their associations and their descriptions pop up when the mouse pointer rests on a concept. The graphical map, based on our GrOWL toolkit, is fully dynamic and interactive: users can move concepts on the screen and browse the full knowledge base by clicking on the concepts and following associations.

to be written

to be written