Proper conceptual instruments and operative tools are now needed to support the design of database models in the production of digital detailed geological maps by GIS applications. In these activities, problems related to uncertainty, usefulness and quality of geological data should be taken into account, while use-constraints of web-disseminated data are to be determined. In this paper, an analysis is carried out on the phases that characterize the acquisition of structural-geological knowledge and related different levels of uncertainty. Methodologies to represent geological uncertainty in GIS databases are then proposed: tools (ontologies, metadata, relational database tools) implemented in DBMS (Database Management Systems) are planned to help users to understand the origin and representativity of data, as well as the interpretations that lie beneath the geological conceptualization. In this view, geological database models should be flexible enough to describe the plurality of actual geological situations, other than consistent with sharing and interoperability needs. The representation of geological uncertainty, and the possibility to re-trace the knowledge path of data makers, allows evaluation of the quality of input-data with respect to a given purpose. Since these processes of giving information in an explicit format require more objectivity in producing input-data, an improvement of original data quality is expected.