Skip to main content

Stanford CS448B 10 VisExplainers

· 4 min read

TLDR

This article contains my notes from Stanford's CS448B (Data Visualization) course, specifically focusing on the tenth lecture about visualization explainers. I'll discuss the importance of storytelling, design space of narrative visualization, interactive documents, chart sequences, and GraphScape.

Original

Download PDF

.

Notes

Forward Thinking

  • How does change blindness apply to interactive charts and how should we design around it?
  • Is it feasible to make creating robust and highly customizable visual explainers easier for less technical users?
  • Why are these kind of "bad" visuals are justified in the talk by placing the context in a more specific community when the accessibility for those communities is not inherently better based on format?
  • When deciding which data should be encoded in which channels: should the most important data be noticed first, or noticed the most accurately?

Topics

  1. Storytelling
  2. Design space of narrative visualization
  3. Interactive documents
  4. Chart sequences

Storytelling

As ancient as mankind

All media tell stories:

  • PEOPLE TELL STORIES
  • WORDS TELL STORIES
  • IMAGES TELL STORIES
  • COMICS TELL STORIES
  • MOVIES TELL STORIES
  • VISUALIZATIONS TELL STORIES

Narrative Storytelling

narrative (n): An account of a series of events, facts, etc., given in order and with the establishing of connections between them.

“... require[s] skills like those familiar to movie directors, beyond a technical expert’s knowledge of computer engineering and science.”

- Gershon & Page 01





Hans Rosling's 200 Countries, 200 Years, 4 Minutes - The Joy of Stats - BBC Four

Design Space of Narrative Visualization

A way of structuring information

  • Easier to understand than lists
  • Uncertainty, conflict, resolution
  • Text and visuals can be complementary


《Narrative Visualization: Telling Stories with Data》PDF




info

Interactive Documents





Chart Sequences

Multiple Charts in Data Analysis


Multiple Charts in Storytelling


Can we automatically identify sequences to recommend to a human designer?

GraphScape

Previously we’ve discussed approaches for automatic design of a single visualization (e.g. Mackinlay’s APT)
GraphScape supports automated design methods for collections of visualizations.
Plenty of future work to do here!

Summary

  • Narrative visualizations blend communication via imagery and text with interaction techniques
  • Specific strategies can be identified by studying what expert designers make
  • Automating construction of effective explainers is an active area of Visualization research