Résumé
We are entering an era of ubiquitous geospatial information and massive amounts of heterogeneous data. Typical users take digital data for granted, assuming their quality is high and fits the intended usage. This attitude increasingly leads to mistakes since digital data contain uncertainty, their characteristics vary geographically and the needs differ. Many subtleties in spatial referencing methods, measurement techniques, technologies and database design end up being hidden in the data and may lead to results not fitting the intended uses. The common practice of providing metadata has proven to be insufficient. As a result of the increasing number of accidents and Court decisions regarding spatial data quality problems, our Society becomes more sensitive to properly warn users about the quality of spatial data with regards to their usages. This paper presents an approach to reduce the risks of misuse of geospatial data by improving database design methods with a risk management strategy. It integrates concepts from geomatics engineering, software engineering, risk management, professional ethics and quality-related standards such as ISO 191XX, ISO 3864-2, ANSI X535.4. This paper also presents how this works fits within a more global project funded by GEOIDE in Canada which involves engineers, geographers and lawyers from academia, industry and key government agencies from Canada and abroad. Finally, an overview of the impacts on education and regular courses is also given.