We present an approach for the lightweight development of web information systems based on the idea of involving crowds in the underlying engineering and design processes. Our approach is designed to support developers as well as non-technical end-users in composing data-driven web interfaces in a plug-n-play manner. To enable this, we introduce the notion of crowdsourced web site components whose design can gradually evolve as they get associated with more data and functionality contributed by the crowd. Hence, required components must not necessarily pre-exist or be developed by the application designer alone, but can also be created on-demand by publishing an open call to the crowd that may in response provide multiple alternative solutions. The potential of the approach is illustrated based on two initial experiments.

Abstract. We show how tag clouds can be used alongside more traditional query languages and data visualisation techniques as a means for browsing and querying databases. Our approach is based on a general, extensible framework that supports different modes of visualisation as well as different database systems. A number of demonstrator databases and interfaces will be used to show how tag clouds can be used to visualise and browse data or metadata and even a mix of both in object databases and relational databases. Further, we will demonstrate synchronised browsing based on tag clouds as well as ways in which tag clouds can be combined with other forms of querying and data visualisation.

No abstract

There is a vast body of research dealing with the development of contextaware web applications that can adapt to different user, platform and device contexts. However, the range and growing diversity of new devices poses two significant problems to existing approaches. First, many techniques require a number of additional design processes and modelling steps before applications can be adapted. Second, the new generation of platforms and technologies underlying these devices as well as upcoming web standards HTML5 and CSS3 have partly changed the way in which web applications are implemented nowadays and often limit the way in which they can be adapted. In this paper, we present XCML as one example of a domain-specific language that tightly integrates context-aware concepts and adaptivity mechanisms to support developers in the specification and implementation of multi-channel web applications. In contrast to most existing approaches, the objective is to use a more lightweight approach to adaptation that can dynamically evolve and support new requirements as they emerge. Our solution builds on versioning principles in combination with a context matching process based on a declaration of contextdependent variants of content, navigation and presentation in terms of context expressions at different levels of granularity that are specific to the application. To support this, a formally defined context algebra is used to parse and resolve the

Context-awareness is a requirement in many modern web applications. \Vhile most model-driven web engineering approaches have been extended vvith support for adaptivity, state-of-the-art dew lopment platforms generally provide only limited means for the specification of adaptation and often completely lack a notion of context. \Ve propose a domain-specific language for context-aware web applications that builds on a simple context model and powerful context matching expressions.