Features

Onto Wiki provides sophisticated means for navigating, visualising and authoring of RDF-based Knowledge Bases. It serves and consumes Linked Data and comprises a comprehensive middleware API for building custom Semantic Web applications.

We refer to it as a Wiki, since our focus is on simplicity, adaptability and collaboration. However, other than annotating text-based Wiki pages with a special syntax (as suggested by text-based Semantic Wiki approaches), Onto Wiki uses RDF in the first place to represent information. For human users, Onto Wiki allows to create different views on data, such as tabular representations or maps. For machine consumption it supports various RDF serialisations as well as RDFa, Linked Data and SPARQL interfaces. Since its introduction in 2006, the application has evolved into a framework for building Semantic Web applications and was recently updated to support the collaboration across multiple domains and application via Semantic Pingback and RDFauthor

Navigation and Visualisation

Onto Wiki provides a number of ways for navigating through RDF knowledge bases.
These include taxonomy and hierarchy browsing (e.g. via SKOS taxonomies or the class hierarchy), facet-based browsing (with complex filter conditions and attribute based tag clouds and full-text search. These different navigation facilities can also be used as a combined navigation, for example a user can start with a full-text search and refine the results using faceted-search or restrict them to a certain part of the selected hierarchy.
Onto Wiki translates these different navigation features into a single SPARQL query. In addition to rendering the results of such a query as a resource list, Onto Wiki provides an extension mechanism for implementing dedicated views like maps and calendars. Once a specific entity has been selected, the resource view displays all available information for that entity. Besides rendering this information generically in a tabular way, OntoWiki can be extended with domain specific resource views (i.e. vocabulary).

Authoring

All views in Onto Wiki can be equipped with corresponding RDFa annotations. Onto Wiki employs the RDFauthor mechanism to automatically transform these views into editable forms. As a consequence, all information items displayed in Onto Wiki can be directly edited in place. This allows OntoWiki users to edit the knowledge base even without being acquainted with the RDF, RDF-Schema or OWL data models. Following the Wiki approach all changes are put under version control and can be easily rolled back.

In addition to this, the integration of RDFauthor as our main authoring mechanism led to two additional usage options:

• By using Onto Wiki's RDFauthor bookmarklet, users can collect data from different web pages and import it directly into their personal knowledge bases (e.g. contacts or events).
• Onto Wiki can be used as a service for hosting editable mash-ups. These mash-ups use data from Onto Wiki's SPARQL endpoint and other data sources (which are interpreted as named graphs) and provide a merged view on this data. The bookmarklet is able to distinguish the statements from these named graphs and allow the user to edit these data directly inside the mash-up as well as to propagate the changes back to the different sources.

Knowledge Base Evolution

Onto Wiki tries to support the creation of RDF Knowledge Bases from the scratch (rather than using predefined ontologies only) in a wiki way. This means that our users often follow the learning by doing path, where requirements and priorities change over time. Our evolution framework allow Onto Wiki users to apply predefined evolution pattern from a pattern repository to their data, as well as to create their own pattern in order maintain the data quality of their Knowledge Base with little effort. Examples for such evolution patterns include the splitting of classes by using their property values or the transition from a data property to an object property (e.g. the transition from literal based tags to resource based tags).

Linked Data & Semantic Pingback

Onto Wiki acts as a Linked Data server as well as an Linked Data client. Entities that use identifiers being governed by the Onto Wiki installation are automatically served as Linked Data. Onto Wiki supports access control on Linked Data employing the FOAF+SSL protocol as well as traditional local account based authentication. References to entities using foreign identifiers can be de-referenced (i.e. consumed by Onto Wiki) in order to obtain additional information from the original source. Even non-linked data resources can be consumed by utilising specific data-wrapper, that allow the generation of RDF data (e.g. from images, videos or web services).

Often it is beneficial for the owner of a Linked Data resource to be informed about the usage of their resources. When publishing a FOAF profile, for example, it is useful to get notified when others establish a foaf:knows relation. We therefore developed the Semantic Pingback protocol. By employing Semantic Pingback, Onto Wiki users can communicate RDF statements across domains and applications.