This talk will focus on the data model underlying a general purpose data visualization system in the early stages of development. The goals of the data model include providing a single method for representing spatial data of diverse kinds, supporting extensive maniupulations of such data (such as boolean operations on volumetric data), and providing that the output of such operations be represented in the same form.
The model is based on simplices up to tetrahedra, where the simplices are defined hierarchically (e.g. a face is defined by its three edges). Information on the connectivity of objects within and between layers in the hierarchy is rigorously maintained. A second tier of the model describes more abstract objects consisting of aggregates of simplices, and serves as a repository for attribute information common to many simplices, as well as grouping the geometric data into logical parts.
The talk will include a review of some previously implemented data models with attention to their problems and how the new model addresses these problems. Snacks will be provided.
See Conundrum Talks for more information about this series.