While the Sequence-based encoding scheme is robust and faithful in recording a user’s interactions with a visualization, it is a static representation that does not afford future modifications and therefore reuse. In cases where a user’s provenance information needs to be examined and re-applied to automate future analyses, researchers have developed techniques for recording the user’s actions using rules and grammars. An early example of this approach is the HomeFinder system byWilliamson and Shneiderman: a user’s interactions with the Home- Finder visualization result in the generation of SQL queries that are then executed by a back-end database. Outside of the visualization community, one popular example of a Grammar-based encoding scheme is Excel’s AutoFill and Flash Fill techniques. In Excel, a user can provide a few example values in cells and “drag” those values to other cells that are then automatically populated. Under the hood, Excel uses the few examples to learn regular expression rules that are then applied to the empty cells – a technique in the programming languages community known as program synthesis.