Tufte articles... My comments

I really can't stand when choropleth maps use all sorts of crazy techniques to parse the data "representative" groups, which really skew the impression a map gives a viewer. Tufte acknowledges,
"Notice how quickly and naturally our attention has been directed toward exploring the substantive content of the data rather than toward questions of methodology and technique." I concur!

Here are a few points he made that I found insightful:
* TIME SERIES displays are best for big datasets with REAL variability; for example weather can be hot one day, cold the next, but over time one can see seasonal patterns.
* AESTHETICS are incredibly difficult keep up while trying to incorporate multiple variables into a map; though it seems now with "faux-GIS" style mapping, layers allow the cartographer to stuff as much in as She wants.
* Charts/Graph are more attractive when WIDER than TALLER. He relates this to the golden ratio, while I'd like to think of it as dependant vs. independant variables (x and y axes); by widening a map, it implies that the pattern can persist indefinately. A tall graph implies that any results are possible.
* A friendly chart is one that does not necessitate a legend. Cartographic convention says One Must Always Create A Legend, but I agree with Tufte that a good map is obvious in its intent and data.

Things I liked:
* Minard is the best, as any Eastern European scholar would agree.
* Dr. Snow's cholera map is pretty awesome too.