Reproduce “OpenRoundup: Multi-Table Data Wrangling Through Interactive Visualization” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Stephen Kasica, Charles Berret, Tamara Munzner

OpenRoundup, an interactive visualization system from the University of British Columbia, enables data journalists to consolidate multiple disparate tables into a single analysis-ready output without requiring programming. The system successfully replicated 17 real-world journalistic programming workflows and was found useful by professional journalists, particularly for visualizing and teaching relational data operations.

Key insights

  • An "eager consolidation" strategy, where a composite table's schema is iteratively built and refined, effectively addresses exploratory data integration when target schemas are initially unknown.
  • Complex relational operations can be effectively simplified into a minimal, declarative, and visually intuitive vocabulary (Stack and Pack) that is accessible to non-expert users, enhancing data literacy.
  • A client-side architecture leveraging WebAssembly for data processing offers crucial data privacy and local execution capabilities, aligning with the sensitive nature of journalistic data workflows.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.