Online 2D/3D graphic interfaces using XML “repurposable” heritage contents

International audienceIn the field of the architectural heritage, documentation analysis and organisation are vital to the researcher when trying to understand the evolution of edifices and sites. Documentary sources provide partial evidences from which one will infer possible scenarios on how an edifice may have changed throughout the centuries. But the growing mass of documents researchers handle has underlined the necessity to find out solutions for enhanced interfacing and visualisation of this wide and heterogeneous documentation. <br />On the other hand, 2D or 3D representation has historically been at the heart of the way edifices or sites are described, visualised, documented and understood. Why shouldn't it be so today? State of the art in our field shows this is far from being a reality. 2D graphics, with the development of SVG-based applications, have undoubtedly found a role in geo-visualisation, but not yet in architecture. Moreover, 3D models most often remain only in relation with communications goals. Virtual renderings, although presented as visualisations of an edifice, mask the semantics behind the scene. Such seducing results may be of a great use, seen as visualisation of geometrical shapes, but in no way can they be considered as visualisation of the deeply uncertain architectural heritage data. The paper proposes an approach of data visualisation in which graphic codes help in interpretation. Appearance of an object represented in a scene, shows the actual level of the knowledge about the object (included in the ensemble of analysed documentation). The documentation is provides links to architectural concepts with respect to the notion of scale. Dynamically generated 2D/3D graphics are used both as visualisations of the documentation's analysis and as interfaces to the documentation's database. Our experimental set is the historical centre of the city of Kraków (Poland). <br />The paper introduces the recent developments of our research: the handling of multiple scales and consequently of multiple interfaces (2D/3D), the use of each object's “repurposable” XML data in the dynamic generation of graphics, the creation of “timeline” scenes that graphically simulate the city's evolution