We had mid-semester reviews this week at MassArt DMI so I figured I’d finally share what I’ve been working on here. The assignment was to map the experience of “You Are Here.” Extremely vague, as usual (though that can be a good thing). My mind has been focused on politics a lot lately (what with the mid-terms and such) and it’s an area I’m considering exploring for my MFA thesis. So I jumped at this first major assignment with that in mind.
Today’s media does a fantastic job of labeling everything either red or blue, right or left. Anyone with the time to think for a few minutes, however, can identify that political thinking is not that simple. There are so many levels of complications and considerations and we have very little time to actually think about it.
Underlying image by David McCandless & Stefanie Posavec
So my question was: as a dynamic media designer, can I help someone understand where they sit on the political spectrum? And by spectrum I don’t mean a binary decision of red or blue – I mean where they sit along a political continuum. Can I harness the nuance of an individual’s opinion? And better yet, how many people out there are not voting at all, simply because they don’t know which of these two polarizing sides to pick? Is there a way to compare someone’s results with a politician’s and provide that help?
System
I immediately began compiling a list of potential inputs:
Surveys
Questionnaires presented to individuals
Personal voting records
Live Activity
Periodic updates regarding timely issues
Regional polling data
Historical Data
Public voting records
Albeit very high level, I moved forward form this list of inputs and started thinking about the next two important aspects of a system: the algorithm and output.
Algorithm & Output
Early on in the project a classmate of mine, Mr. Kaiser, sent me a link to Vote Smart’s Vote Easy app, a tool put together by a non-partisan group to help individuals decide which candidate to vote for. My immediate reaction was that I was just too late to the game – this had already been done.
Vote Smart's Vote Easy Project
After further inspection, however, I realized this project was much too narrow in scope. Whereas Vote Easy’s system is only 10 yes or no questions, I wanted significantly more detail in my system. I wanted to make sure people weren’t being forced to check a box they didn’t feel comfortable with. As another classmate, Mr. Buckley put it, I was at risk of forcing an objective algorithm on a subjective opinion. At the same time, some people just want an answer so they can go and check the box.
Vote Easy's algorithm
This led me to two important decisions in the project. Firstly, I was going to plot all of my answers on a continuous scale:
Continuous scale
Secondly, I needed to give users the opportunity to decide how much time to dedicate to the project. Some may want to only spend five minutes answers questions. Personally, I’d probably need five months. So I needed users to choose a “resolution” of data: the higher the resolution, the more questions asked.
Users can dig deeper on specific issues
But this is where I hit a real wall: when trying to visualize my output.
My original concept was to plot a user’s opinion on a 2D plane. Each plane would represent a new issue. The position of the “opinion” dot would be determined by metrics that had yet to be formalized. You could then compare lines in a 3D space, connecting dots on each plane.
Plotting users answers on 2D planes in a 3D space.
This didn’t feel right, however. What metrics would I use to plot on the planes? And how would I accommodate the different resolutions of data people would be providing? The planes felt restrictive.
That’s when Mr. Kaiser introduced me to The Shape.
The Shape
Zach pointed me towards the branding of a town in Norway (Nordkyn, to be exact) which changes based on the current conditions in the town (temperature, wind speed, wind direction). The result is a dynamic shape that encompasses multiple levels of data points.
Nordkyn, Norway's everchanging brand
Turns out that was exactly what I needed. Rather than restricting myself to the dimensions of a plane, I could now work inside out from the core of a 3D form. And instead of forcing a great deal of data into a limited number of axes, I could break out into new directions as needed.
The Shape
So here‘s how the system works in full (link to Vimeo – having trouble embedding).
You can see in the video above an extremely rough overview of the user flow for the questionnaire functionality. Some key points;
As the user answers questions from various categories (foreign, domestic, fiscal, social) the corresponding plane on the shape changes colors.
and
When a user decides that a question is not specific enough and chooses to dive deeper, the resolution updates, and the current category’s plane splits, changing the entire form of the shape.
Nuance of Opinion
By mapping political beliefs on in an abstract way, the shape is able to provide value to an individual whether it sits alone or in comparison to others. A user can track shapes over time to see how opinions have evolved. She can also hold it up to others’ shapes, helping her make decisions on which candidate, media outlet, or institution her political thought is most like. And as a shape breaks down into a higher resolution of data, the question of “why” can perhaps be answered. One can realize what nuances of opinion her political choices are based on.
Comparing your shape over time
Comparing your shape to others
Another friend of mine pointed out that his father has very conservative opinions about fiscal policy but rather liberal social beliefs (same-sex marriage, for example). He gets frustrated, however, when his father votes based solely on what he reads and sees in the conservative media. When I showed him this project he was intrigued. Perhaps his father could now see where his political views as a whole exist. In the end, his father may decide that fiscal policy is simply more important than social – after all, this is partly about checking box A or box B eventually. But providing individuals with the ability to truly visualize their political opinion could definitely enlighten.
Further Considerations
This was only a five week project – and it shows. There is a great deal more to explore here.
Specifically, the development of the question set is extremely important. The sheer number of topics and questions that can be compiled is rather daunting. I would need a team of individuals with an extremely varied political background – in terms of experience, party affiliation, institutional affiliation, and category expertise. I’d also like to give users of the application the opportunity to submit questions. After all, how can I embrace the nuance of opinion without making room for opinions I haven’t considered.
Plotting results on the shape itself is not flushed out either. I worked with many different forms before finally going with a tetrahedron – mostly because of its simplicity as a 3D object, upon which I could break out more complex shapes. But maybe this isn’t the right one. And forcing every issue into a four basic category set could also be problematic. Plus, how do I integrate the other data inputs besides user answered question onto the shape?
Finally, how can this shape be used in other applications? Do we share our shapes or keep them private? Do we create a collective of opinion based on every individual’s shape? Should we?