Abstract. In this paper we describe some of the personal information management features found in the PSEW prototype of the NEPO-MUK Semantic Desktop project. Focussing on importing and adding new knowledge to the system as well as navigating, browsing, and searching the existing knowledge we show how a user can use NEPOMUK to integrate her information across application boundaries and create and discover new structures in her information space.
The NEPOMUK Semantic DesktopThe paper presents parts of the outcome of the now completed NEPOMUK project 1 . NEPOMUK aimed to create a specification for a Social Semantic Desktop and a prototype implementation of this specification. This paper will only describe the details around the prototype implementation, for information on the Semantic Desktop specifications such as the ontologies, the architecture blueprint and service descriptions we refer the reader to the project deliverables available from the project website as well as [1,2].The NEPOMUK project resulted in two parallele implementations of the Semantic Desktop, one written in C++ which is tightly integrated with the KDE4 desktop environment, and a cross platform version written in Java. This paper deals only with the Java version. Furthermore, we focus only the Personal Information Management (PIM) components of the NEPOMUK Semantic Desktop, that is, only the components used on a single desktop machine. NEPOMUK also includes social features allowing various kinds of distributed search and information sharing, but they will not be covered here.The NEPOMUK approach to personal information management aims to overcome the limitation of having a separate applications for each type of data, such as email, photo-manager, publication-managers, etc. Instead all the information is integrated into one information-space and the user is free to relate and annotate information elements across application boundaries. The central hub for working with this information-space is called the NEPOMUK Personal Semantic Workbench (PSEW) and a typical view is show in Figure 1. Here we see a range of different views on the user's data, and a person concept, "Claudia", 1 http://nepomuk.semanticdesktop.org L.