Since I got involved in the middle of my Sophomore year, I’ve been working as a research assistant in the Interactive Data Lab at the UW, helping to build the high-level Vega-Lite grammar. Previously, the lab was known as the vis lab at Stanford, before it moved to UW. The lab is run by Jeffrey Heer who is one of the original authors and developers of D3.js, Vega, Protovis, and Prefuse in addition to his many other notable contributions to the field of data visualization.
My role on the project involves helping a couple Ph.D. students, Dom and Ham, in the lab with the development of Vega-Lite. Specifically, I’ve been working to introduce a label layout algorithm to Vega and Geo-projections to Vega-Lite.
Vega-Lite is a high-level visualization grammar. It provides a concise JSON syntax for supporting rapid generation of visualizations to support analysis. Vega-Lite specifications can be compiled to Vega specifications.
Vega-Lite specifications describe visualizations as mappings from data to properties of graphical marks (e.g., points or bars). It automatically produces visualization components including axes, legends, and scales. It then determines properties of these components based on a set of carefully designed rules. This approach allows Vega-Lite specifications to be succinct and expressive, but also provide user control. As Vega-Lite is designed for analysis, it supports data transformations such as aggregation, binning, filtering, sorting, and visual transformations including stacking and faceting.’
Vega-Lite is part of the larger Vega project, which aims to provide the greater data vis community (developers, data scientists, journalists, whomever) with a grammar for expressing data visualizations. Such a grammar allows for visualizations to be created declaratively, which affords greater portability, flexibility, and other benefits. Further, a well defined and powerful grammar makes subsequent data analysis and tool development much easier as can be seen by other projects from the lab like Voyager and Lyra as well as papers like: revere-engineering visualizations.