How can we liberate visualization from chronology? Should we?
Time is a metric convention and as such it can or cannot be considered a valid approach to organize data, there could be many other ways to explore an apparently time sensitive subject. What if we group items by the questions they suggest rather than by the time they were produced in? Let’s think of the invention of the light bulb. How different were people’s lives before and after this invention? Is the answer more relevant than knowing the date it was invented? At what point does the time distance make the notion of the date irrelevant or too abstract to understand and to contextualize? As we have been debating about humanities and science, maybe timelines are the missing link: is it the closest we can get to the precision of math? Does it matter?
The open source tool called TimeFlow (more info here) is just an example of a timeline that allows one to zoom in and out of its contents so as to visualize info by the year/month/day selected. I think this is a good way to use a timeline because of the opportunity to consider single events in relation to a longer period of time. Although timeline remains a genre of historical writing that employs rhetorical figures (Zuern, p. 3), the third dimension added by the computer screen -which not only lets us play with zooming, but can send us to relative links- can improve the use of timeline drastically. By implementing and diversifying information and offering an interactive component through which each user can create an independent path, timeline could make significant steps towards that objectivity which has been the focus of the debate around this tool (Zuern, p.4).
Speaking of bar charts, according to Nathan Yau, temporal data can be categorized as discrete or continuous. Discrete are values that are from specific points or blocks of time; continuous are values that can be measured at anytime, during any interval and it is constantly changing (p. 93). What if our timeline would show both, discrete and continuous values? Here is an example.