Reaction: Effectively communicating numbers

This article talks about different ways that any person can convey numerical information to the user in a more informative way. The article focuses on the technique of building different types of graphs. The author starts by introducing the problem of presenting numerical data which is normally seen as list of numbers related to any different values. This data can be effectively put into a graph which summarizes the information and helps recognizing patterns.

The article explains the different guidelines which are necessary in order to build graphs. The most important guideline is the type of data present. Normally, there is a relation between the categories present in the data and the data itself; the appropriate way to display such data-sets differs from normal. The article cites seven common relationships and ways to represent them.The article also explains the ways to encode data by giving examples about each type; i.e. points, lines, bars, boxes.

The author also lists an excellent design process to build an effective graph by eliminating the distractions, placing objects where they belong, displaying appropriate messages and most importantly understand the data and decide about the best way to encode values.

I found this paper to be very useful as an information resource as it contains so many subtle points that we do not normally consider while drawing common diagrams such as graphs. This paper is very useful to beginner and expert alike because it focuses on making the basics stronger and concentrates on what matters the most. I do wonder whether these guidelines apply to all cases or are there any exceptions where you need to step out of the design process to make the visualization more appealing.