Daniel P. Huffman
Was thinking recently about Stuart Brown's video on the "Chicken-o-meter," and wondering if there are examples of this strategy in #datavis. Where quantitative data are communicated w/ a discrete set of iconic/associative symbols, rather than continuously varying size/color/etc. https://youtu.be/B8HT8aUb5q4?si=3JS_cqH9EvbYXaUs
Mar 22 at 8:36pmWeb
Show threadMatthew EdneyMar 22
@pinakographos do you mean Chernov(?) faces?
Show threadDaniel P. HuffmanMar 22
@mhedney Chernoff faces are sort of in line with the idea, though they feel separate to me. Their appearance as faces is often only weakly connected to the individual datasets, and the whole of the symbol is based on multiple datasets instead of communicating one. But, I think there's a connection of some sort.
Show threadBarry RowlingsonMar 23
@pinakographos @mhedney I'm sure I read once that Chernoff invented these as an example of how not to do data viz, but somehow they caught on.
Show threadDaniel P. Huffman6d ago
@geospacedman @mhedney In his initial paper, I believe Chernoff used faces to visualize *geology data*, so I'd believe that.