Abstract
As more visualization libraries and systems are being developed using new Web technologies, it has become easier to produce visualizations with a plethora of rich interactions. These offer the users, the opportunity to examine details and explore hypotheses that have not been directly encoded by a visualization designer; using dynamic computations and real-time DOM manipulations. The advent in Web technology has made these accessible to a wider audience, which is another reason for its wide spread popularity. However, an analysis made using current Web-based visualizations is often transient: forgotten when the user moves to a new page or reloads the page. Some interactive visualizations allow users to share links or xml scripts that encode the state or save a snapshot of the current view or state on a server, but most work focuses on preserving the end result rather than how that result was achieved. As Web-based visualization libraries and systems become more intricate, the ability to understand how a discovery was made, its provenance, becomes a concern. Shareable Interactive Manipulation Provenance (SIMProv.js) was developed for this purpose: the library can be easily integrated with the user's visualization scaffolding, providing options to capture, manipulate and share provenance. Recognizing both privacy and storage concerns, it provides solutions for users to store and load provenance locally or share it with others through their preferred private channels. It is completely client controlled, the provenance is locally stored with no server services. The library supports modern Web libraries, offering developers the ability to customize how provenance is captured. The library further supports capture of provenance in the context of streaming data, where it records the streaming data along with provenance, which serves a crucial purpose of explaining how the analysis has changed over time. By including information about the state of the streaming data with the provenance information, the library allows users to explore how analyses have evolved with respect to data and to replay past analyses on more recent data. The library also tries to bring in the concept of collaboration to the realm of interactive visual exploration. This augments the current visualization library to be collaborative with provenance capturing capabilities, thus eliminating the need to exchange provenance information after analysis, as everything is done in real-time, with the collaborators viewing all the interactions along with active participation.