ScenariWiki

Principal advantage of fragmented authoring is that fragments can be reused across multiple documents. The anticipated problems raised in C2M can therefore be interpreted as a repurposing and aggregation problem, which is relevant to the research community outside the Scenari context. ScenariWiki is intended to appeal to a public that is unfamiliar with the Scenari authoring tools. The Wiki paradigm was chosen as it is one of the most well know collaboration platforms. However, there are significant differences between a traditional Wiki and ScenariWiki. Most notable is that a Wiki poses as few as possible constraints to an author. In contrast, ScenariWiki exploits the formalized document model used by Scenari. Therefore, an author should respect the imposed document model. To this end he is aided by the Scenari editor (which is part of ScenariWiki). If the functional overlap between ScenariWiki and a traditional Wiki turns out to be too limited, we might decide for an alternative name that is more appropriate.

The research objective we focus on is collaboration and re-purposing using fragmented documents in general. Nevertheless, the proposed solutions should as much as possible align with the Scenari framework that, besides practical advantages from an implementation, has proven to be successful in an industry environment. The objective of our research is to provide a scientific justification of the concepts implemented in Scenari, and, if they exist identify bottlenecks in the current Scenari model.

Therefore, we respect the following design considerations:

• complement existing Scenari solutions (i.e. ScenariWiki should use existing Scenari technology and workflow)

• Lightweight experimentation platform viable to use off-the-shelve technology (i.e. development effort to test ideas should be minimal)

• Provide alternative technology perspective and open new exploitation avenues (e.g. knowledge bases)

The ScenariWiki architecture is based on three components:

• Scenari (client and server), provides one side of the backbone of the system. Notably the graphical authoring environment (including formalized document models), the storage component and the publication generation engine are used by ScenariWiki.

• ClioPatria is a data repository framework specialized in graph computation and web integration. In ScenariWiki it is used to manage the network and metadata associated with the network and items. In addition, we use it as intermediate to incorporate visualizing solutions. As such, it is the he secondary part of the ScenariWiki backbone.

• Firefox extension, which provides a two way interface between the Scenari editor (and consequently the Scenari Server) and the ClioPatria framework. It runs as an extension of scenari client and takes in charge some of the computationally expensive tasks, such as document transformations that otherwise would impact the server.

The Firefox extension captures every "save" event generated by the scenari client. If a save event occurs all actions implemented by Scenari are carried out. In addition to that ScenariWiki archives the fragment in a dedicated atelier for future reference (a unique hash is appended to the name of the fragment to identify it). Furthermore, an RDF record is generated that contains metadata about the respective fragment. This record includes:

• the time and date when the fragment was edited

• the author of the fragment

• A reference to the fragment on the ScenariServer

• the previous version (if it exist) of the fragment

• references to the children of the fragment (if they exist)

• The HTML representation of the fragment

The record is send to the Cliopatria, which indexes it.

ClioPatria is a web framework (comparable to tomcat or cocoon) that stores and indexes RDF and makes it available to other processes using web interfaces. For ScenariWiki we use it to index the fragment network and visualize fragments. (to be completed)