Week 10.1 Commentary (Isabel Báez)
Edward Tufte
Tufte discusses the way that color used to display data, along with its benefits and shortcomings. Digitally, color can provide significant advantages: it reduces video glare and, moreover, its edge definition strives away from grid-like representations. Sticking to nature-based color is also better on the eye, as they are more familiar and prevent the garnishing that stronger, non-significant colors would bring.
Tufte evaluates the differences between differentiating scales by values versus by hue. Although recognize the faults in the ocean mapping example he presents, where similar values may be hard to tell apart, I do think its benefits overpass those of the rainbow alternative. The ability to relate the blue to the ocean, and the darker values to more depth, is much more valuable. As Tufte mentions, rainbow colors have no meaning towards the mapping, would be harsh to the eye, and would need some kind of legend to give them meaning.
Steven Few
The findings of Few’s study, comparing information digestion from embelished versus plain graphs surprised me. I hypothesized that overtly embelished graphs would distract from the information displayed, in contrast to their plain counterparts. However, Few expresses that participants were able to recall the information of embelished graphs much better. I suppose his justification makes sense: when information is accompanied and supported by multiple visual elements, it is easier to remember. Moreover, an embelished graph with interesting elements might be more memorable overall.
Nevertheless, although I think some embelishmment is beneficial, the Holmes graphs studied by Few’s certainly over do it. The amount of detail in the graph make them very hard to understand first glance. If the amount of time that participants were allowed to study the graph was reduced, I’m sure teh results would differ. Moreover, maybe there exists some middle ground between the embelished and plain graphs presented by Few that superpass them both in initial value interpretation and in recall.