Description:
Collaborative visual analytics (C-VA) is a growing area of research and has been receiving a great deal of attention in recent years. A persistent shortcoming of C-VA systems is that most require custom hardware and software. To ease the development of distributed C-VA systems we created a novel framework and system we call Share.va. Share.va addresses all core requirements of C-VA systems supporting sense-making with data, namely, shared heterogeneous interfaces, provenance, and security. A key element of Share.va is a blockchain mechanism which keeps all shared information in a custom data structure and enables these core requirements naturally. The custom data structure stores the visual properties and underlying data driving the interface as two independent entities. Sense-making and provenance are supported by providing users with annotation, note taking, sandboxing, and history review tools. The blockchain ensures that collaborators have a secure and unmodifiable repository of their analyses. Share.va is one of the first systems to use blockchain as a platform for real-time collaboration, and to best of our knowledge this paper presents the first empirical evaluations of such systems and first use in a visual analytics setting.