User interface (UI) mashups follow principles and architectures, which should enable even non-technical end users to develop their own widget-based applications. Recently, so-called choreographed solutions have been proposed, which lower the entry barrier even further: mashup designers are supposed to simply place widgets in a shared context and communication among them would emerge automatically. However, existing approaches for enabling self-organization among widgets do not take any measures to prevent undesired or faulty behavior like e.g., self-reinforcing loops or hunting. This paper demonstrates two developments to address this problem in the context of publish-subscribe-based UI mashups. The proposed algorithms are implemented as extensions to the open-source mashup platform Apache Rave.
CITATION STYLE
Tschudnowsky, A., & Gaedke, M. (2015). Loop discovery in publish-subscribe-based user interface mashups. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9114, pp. 683–686). Springer Verlag. https://doi.org/10.1007/978-3-319-19890-3_57
Mendeley helps you to discover research relevant for your work.