
Goal \| Application \| Challenge '04 \| Examples \| Registration \| Sponsors/Contact \| Former Challenges
Flink Flink is a unique entrant to this year's Challenge. As opposed to products of large EU projects or DARPA funds, it is the result of the effort of a single Ph.D. student driven by the desire to see for himself the state-of-the-art in Semantic Web application development. In particular, the simple question this application intends to answer is whether it is possible today to develop with the minimal effort possible an engaging, cutting-edge SW application from the LEGO blocks available as open source. In this sense, Flink serves as a showcase of the achievements of SW development and highlights the challenges lying ahead. Compared to the entrants to last year�s challenge, advances have been made in the use of OWL, custom inferencing, Web Services integration, the latest features in query languages (SeRQL), contextualization and support for handheld devices. Flink itself is also likely to be unique as a crossover between a social experiment and a semantic application. Flink brings together a number of different knowledge sources and use them to learn about the social structures of the community that created them -- in this case, the community of Semantic Web researchers. The resulting application is a Who is Who of the Semantic Web, which can be of interest to this community as a reflection of their social organization, but is also valuable as an input for Social Network Analysis, a branch of sociology concerned with relational data. (Export to popular network analysis packages is directly supported by the application.) Lastly, Flink is a true a Web citizen, integrating the information sources of the traditional web (HTML pages) with those of the Semantic Web (FOAF profiles). Although scalability is not an issue addressed by the application, extensibility is: due to its modular design, Flink is easily extended with additional features and information sources while preserving a minimal, lightweight and dynamic profile. URL: http://prauw.cs.vu.nl:8080/flink/ Contact: Peter Mika Free University, Amsterdam Email: pmika@cs.vu.nl
Last modified on 7 November 2004

Flink

Flink is a unique entrant to this year's Challenge. As opposed to products of large EU projects or DARPA funds, it is the result of the effort of a single Ph.D. student driven by the desire to see for himself the state-of-the-art in Semantic Web application development. In particular, the simple question this application intends to answer is whether it is possible today to develop with the minimal effort possible an engaging, cutting-edge SW application from the LEGO blocks available as open source. In this sense, Flink serves as a showcase of the achievements of SW development and highlights the challenges lying ahead. Compared to the entrants to last year�s challenge, advances have been made in the use of OWL, custom inferencing, Web Services integration, the latest features in query languages (SeRQL), contextualization and support for handheld devices.

Flink itself is also likely to be unique as a crossover between a social experiment and a semantic application. Flink brings together a number of different knowledge sources and use them to learn about the social structures of the community that created them -- in this case, the community of Semantic Web researchers. The resulting application is a Who is Who of the Semantic Web, which can be of interest to this community as a reflection of their social organization, but is also valuable as an input for Social Network Analysis, a branch of sociology concerned with relational data. (Export to popular network analysis packages is directly supported by the application.)

Lastly, Flink is a true a Web citizen, integrating the information sources of the traditional web (HTML pages) with those of the Semantic Web (FOAF profiles). Although scalability is not an issue addressed by the application, extensibility is: due to its modular design, Flink is easily extended with additional features and information sources while preserving a minimal, lightweight and dynamic profile.

URL:

http://prauw.cs.vu.nl:8080/flink/

Contact:

Peter Mika
Free University, Amsterdam
Email: pmika@cs.vu.nl