ProGram: A Grammar for Progressive Visualization Pipelines

Abstract

Progressive Visual Analytics (PVA) systems have demonstrated the effectiveness of time-bounded, incremental computation for exploring large-scale data. However, PVA implementations are often difficult to reuse and extend because their progressive behaviors are typically implemented through ad hoc, hard-coded mechanisms distributed across servers and clients. To address this limitation, we present ProGram, a grammar for progressive visualization. ProGram is designed not only to support core progressive features while reducing implementation burden, but also to extend the design space by introducing novel mechanisms, such as lockstepping and temporal prioritization to better support diverse usage scenarios. Finally, we demonstrate the utility of ProGram by reproducing progressive visualization systems.