|
|
Post by rushil on Apr 12, 2016 7:59:46 GMT
Just to summarize the good discussion points we have had here (and we are gonna discuss them in class tomorrow as well). Encoding is not something that can be achieved by following a set of rules. The framework that the author provides, and the results shown can be used to show really inaccurate graphs, or if used properly can prevent misleading graphics. I believe that is one of the major takeaways of the paper -- they give you the tools, the question is whether you make good on it or not.
|
|
|
|
Post by anhong on Apr 18, 2016 19:06:17 GMT
I think we are still pretty terrible at reading graphs. It's the responsibility of designers to use the most accurate encoding to deliver information. However, because of human's perceptual capabilities in different tasks of graph understanding, different techniques should be used. For example, when two things are close and want to be compared, they should be aligned and in the same scale. However, if two things have to be compared in area, it's best if they are very different, otherwise could introduce confusion.
33 years have passed, while human capabilities didn't change much. However, technologies now allow us to construct dynamic graphs to highlight information the user is focusing on, in order to deliver the most effective information, which was not possible 3 decades ago.
|
|
