Visual Analytics: Thoughts at Week 3 and Links

I’m a few weeks into my class on Visually Enabled Reasoning, and it’s a deeply interesting class so far. We are starting with some lectures and readings on human cognition, reasoning, decision-making and mental modelling. Taken with the topics we covered in my Decision Insights for Public Policy, I’m really inspired by the idea that it’s not a debate about whether we’ll “submit to the logic of the machine”, but that we’ll nuance our understanding of what machines do best and what we as humans do best, and find ways to support effective dialogue between what people do best and what machines do best. Values come up as a topic a lot, because those lie at the heart of policy decisions.

It takes all my might sometimes not to laugh outloud in class, because much of the work in the visual analytics field originates in intelligence analysis and law enforcement, and I think the layman’s understanding of what this looks like probably comes from shows like CSI, Criminal Minds or Lie To Me. (Fun fact: the work of Paul Ekman, the inspiration for the lead character in Lie to Me played by Tim Roth, is excerpted heavily in my textbook. Small world.) The vision put forth through those dramatized procedural formats, stocked with genius-level characters constantly pulling associations out of thin air, speaks to the dream that this stuff isn’t just for playing Bejeweled at the drop of a hat, or immersing us in trivialities — that it can, in fact, empower us to tackle issues in new ways that we were never previously able to.

But I’m admittedly in this class for much more mundane, much less glamourous reasons (unless that TV procedural about urban planners ever pans out). As I stated in my presentation at Open Gov West BC, I’m really interested in how Big Data — the kind being collected by the kinds of systems we are putting into place under the auspices of smarter or more sustainable cities — can help us to understand our collective interdependency. That’s my goal for the class: to understand what helps or hurts conveying that message, as well as how to nuance our understanding of when that holds true, and when it does not.

Since we’re fairly early on in my class so far, I’m just starting to get my feet wet a bit with the tools, and also seeking plenty of inspiration. So far, I’m pretty bowled over by VizCandy. Kelly shares her perspective being educated in sociology and demography, picks great topics and shares her process of making visualizations in Tableau Public. A trip through her blog’s archives is on my calendar…

I’ve also started to play around with Compendium.  I wonder what it would look like, for instance, to visually represent insights around things like the challenges for municipalities in participating in regional planning, or funding transportation.

There’s also some really interesting threads in my class around the concept of “pair analytics” — similar to pair programming, except instead of having two programmers, you have one domain expert and one person adept in using visual analytics tool, working together to understand what the value of representation is. I find I’m halfway in between both of these roles — still learning about the domain, and interested in, but not yet proficient in, the tools.

Post a Comment

Your email is never published nor shared. Required fields are marked *

*
*