Information Design for the Web

Research focus

The Peer review has evaluated this group as Excellent

The growth of Web Applications as the fundamental infrastructure for business and social activities has generated a strong interest in methods, environments, and tools supporting their design and deployment. The continuous evolution of technologies calls for a foundational, technology-independent approach, rooted in the tradition of information modelling. The main focus of this research is centred upon a conceptual modelling language, called WebML (Web Modelling Language), that was invented in 1999, patented in 2003, and has given rise to significant research results, most of which have been translated into applied technology through a spin-off of Politecnico. WebML describes a conceptual model of the Web application in which the various aspects of the specification (respectively the content, the hypertext, and the presentation) are orthogonally combined. The most innovative aspect of WebML is the modelling of hypertexts as collections of elementary units and links, where the units describe both the visualization of elementary elements of a Web page and the operations performed by the application, and links between units capture the user behaviour. The model was initially focused on the display and management of contents, but it has been progressively extended to incorporate other features of modern Web applications, including: process management, Web service invocation and publication, management of adaptive and reactive computations; management of collaborative applications; methods for improving the accessibility, and more in general the quality of Web applications; support for new media, technologies and architectures, including rich Internet applications, VOIP, and Web architectures for embedded systems. All of these aspects have given rise to publications and to the development of prototypes and have been experienced in test-beds or applications as part of funded projects. Some of these aspects are reported in papers currently in print on major journals, such as ACM Communications and ACM-TOIT. A WebML specification is a graph, therefore WebML specifications are supported by a visual design tools with extensible components; such tool, called WebRatio, has been initially developed at Politecnico as result of EU-funded projects in the fourth and fifth framework, then has been the core of the spin-off company Web Models, participated by the Politecnico. The company has developed WebRatio, a web development tool suite (see www.webratio.com) and has now about 20 employees, subdivided in the two locations of Milano – for commercial development – and Como – the software factory. Future directions of research include the adaptation of the WebML approach to the evolution and new challenges of Web Technology. Recent work has addressed: - Model-Driven Development of Rich Internet Applications: WebML conceptual model and WebRatio code generation technology are extended to transfer more application logic from the server to the client. - Model transformations in the MDA context: WebML has been generalized using UML Meta-Object Facility (MOF), to make it consistent with OMG’s Model-Driven Architecture (MDA). The code generation techniques will be then generalized into a more abstract model-transformation framework, capable of addressing such tasks as: the generation of metric models for evaluating different size measures of a project (e.g., for automatically producing the Function Point count from a conceptual application model); or the generation of models for driving the automatic testing of applications. 39 - Automatic front-end synthesis from process models: the aim is to bridge the process modelling phase to the hypertext generation phase, by automatically generating (low-level) hypertext skeletons from (high-level) process models, according to different styles of process enactment. In the above context, special emphasis is given to reverse engineering, i.e. the ability to reconstruct the process from the generated hypertext when the latter is subject to modifications and evolution. - Conceptual modelling for the Semantic Web: an extension of the WebML approach towards the semi-automatic production of semantic descriptions of applications and services is investigated, so as to turn conventional software products (produced by using WebML-related technologies) into components which fit with Semantic web service architectures (such as WSMO). Through this method, semantics which is normally contained in high level models of applications and services gets exposed in a format which is compatible with Semantic Web Architectures. - Embedded Web: this line or research aims at down-scaling Web architectures and applications to embedded systems, adapting the conceptual modelling primitives, runtime architectures, and code generation techniques to the space and time constraints of embedded architectures. Applications are envisioned in domotics, intelligent buildings, cultural heritage, and industrial automation.

Dipartimento di afferenza

Dipartimento di Elettronica e Informazione (DEI)

Docenti afferenti

Stefano Ceri (full professor)
Piero Fraternali (full professor)
Giuseppe Pozzi (associate professor)
Daniele Braga (assistant professor)
Marco Brambilla (assistant professor)
Alessandro Campi (assistant professor)
Sara Comai (assistant professor)
Marco Masseroli (assistant professor)
Maristella Matera (assistant professor)