"Supporting Data Journalism through Compilers for Visual Inputs" by Parker Ziegler

Supporting data journalism with Cartokit's compilers for visual inputs, enabling interactive graphics with minimal programming expertise.

Key takeaways

Compilers can open up new interaction models for building interactive graphics, allowing for greater design flexibility and reducing the need for manual reverse engineering.
Data journalists often face a “gulf” between visual input and textual representation, making it difficult to reuse and retarget existing visualizations at new datasets.
The Cartokit system is being developed to address this challenge, enabling data journalists to interact with visual inputs and see those interactions propagated to output programs.
The Cartokit compiler translates visual inputs into intermediate representations, which are then used to generate output programs.
Data journalists can use Cartokit to retarget existing visualizations at new datasets, reducing the need for manual reverse engineering and increasing the efficiency of graphics development.
The concept of “visual inputs” encompasses a range of formats, including Figma files, SVG subtrees, and other visual design representations.
Compilers can help reduce the cognitive load on data journalists, allowing them to focus on higher-level design decisions rather than low-level programming details.
The Cartokit system is being developed with a focus on geospatial data and visualization, but has the potential to be applied to other domains as well.
The goal of Cartokit is to empower data journalists to create high-quality, interactive graphics with minimal programming expertise.

More talks