Originally posted as a thread on X.

Was re-watching this Jeffrey Heer talk from six years ago. And in the context of data visualization, he brings one point from Edward Tufte which made me think - “Show data variation and not design variation.”

And then he asks, “How might our tools spur us to exercise skepticism and consider new questions?” And to think of it, the asking questions aspect is still extremely dependent on humans and editorial judgement. Or maybe that is just my unawareness of the latest tips/tricks.

With better tools, more focus on communication and research around how people read charts, the design variation is catching up for sure. What about the data variation? Are there tools that spur us to consider new questions?