The context is used as the base for a information request. It's sot of part of the request. It contains your preferences used to select the wanted view.
For the RDF::Service API,
every returned object is actually a Context object containing a
node object. The context object holds
the context for the node. All methods in the API are actually
context methods.
The context holds infromation about the session, and the reasoning process itself. This includes a memory of what things already has been asked, in order to recognize thought loops.
The context will complement the requests with implicit criterions. These criterions will come from the context itself, the session and the agent behind the session.
Statements in the current session context will have precedence over contradiciting statements in the general preference for the agent. This will be handled using versioning, saying that the session statement replaces the other contradicting statements, but only for the current session.
The trust model and the other things is part of the preferences in the context.
Not implemented. But we do use a context wrapper.