Please view the video before reading below. Best enjoyed with headphones!

Music: LOONA - BUTTERFLY XIIX (Intro Demo) by jen orbit


Butterfly is an experimental data visualization and an exploration into how to visualize soft data (qualitative) compared to hard data (quantitative). It is meant to show an approach to visualizing this kind of information that can visualize what is not proven or too theoretical or not measurable. In order to examine visualizing soft data, the visualization evolves based off of the butterfly effect from chaos theory and the word butterfly as the soft data in an ever-changing composition.


Starting Out


At the beginning of this process, I asked myself: How do I merge user experience, user interface, data visualizations and patterns (surface design) - all sectors of design that I’m interested in - within one project? I looked at several reference photos in each area of design I’m interested in and looked at photos that blend the topics together or ways where I can blend ideas together.

I was interested in the question: what is really data? Often associated with numbers, true facts, computers and scientific research.

How might we show data that is not true/vague/ unknown? Not numbers?

How might we make a data visualization that reveals what is not proven or too theoretical or not measurable?

How might we visualize that?

Then thought about how can I relate data to user experience, can I track “un-trackable data” or “vague data”?. This formed my base theory for my project on a proposal to visualize vague data:

Vague Data ☞ Pattern ☞ Output




Starting out with research, I looked at topics that are typically vague in concept:

1. Human behaviour

Human behaviour is always changing as it is unpredictable. We don’t know what people are thinking and this unpredictability can create the data. The possibility of visualizing this is to create pattern recognition to track the unpredictability and would have to be a tactile experience to be affected by human behaviour. I looked into user-generated forms, mostly installation art, where the user creates the composition and is highly interactive.

2. Unproven Science theories

There are tons of science theories still unproven/too theoretical such as dark matter, dark energy, wormholes, black holes, string theory, alien life. The visualization would have to have pattern recognition to be determined by unproven theories. This would be more towards traditional data visualization since this is still in the science field. It could be a data visualizations that make a picture as a whole. I also looked into glitch art as a way of visualizing, as it presents the uncertainty of the theories.

3. Human psyche

The human psyche is defined as “the totality of the human mind, conscious and unconscious.” This can include the stream of thought, dreams, memory and so on. The human psyche is similar to human behaviour where it cannot be predicted in data. To track this, it would have to create patterns by pure conscious or unconscious actions (whatever is first picked by the user) by the neuroscience of free will. This pattern could take any shape or free form patterns. I looked into optical illusions because it is the mind, or the consciousness, that is judging the image and recognizing a shape out of it. Interaction would be in the object as well as it is always moving or breathing.

After research and brainstorming, I concluded on planning a tactile experience (which uses user experience). I looked into the theory of how interactive installations work and different interactive installation examples such as Yayoi Kusama’s Obliteration Room or TheGreenEyl’s Appeel. The initial approach was to involve the participant into making their own creative decisions, which is blending the human behaviour and human psyche topics together as art data, to become a part of an evolving art system. From my research in interactive installation design, the theory in “Approaches to Interactive Art Systems” by Ernest Edmonds, Greg Turner and Linda Candy states that the art data is the components of the installation: the participants, the designer, the installation itself which then makes the whole art system in order to show that data can be outside of numbers/facts. While interacting, it should be able to track it as some sort of data. 

This also connects to a type of design I researched: conditional design. Conditional design is where it is the process that is the design, the main focus. Human behaviour is always changing, it is unpredictable as we don’t know what people are thinking. This unpredictability can create the data and can create the pattern to track the unpredictability. In order to involve human behaviour in data tracking, it would have to be a tactile/interactive experience to be affected by human behaviour and I originally aimed to do a peg board for an interactive installation.


During Design


Once I actually got into designing, my approach soon turned into focusing on just the pattern and output areas of my visualizing theory as I started to design the actual product. The original intent was to visualize interactions from an interactive tactile installation (the butterfly design was going to be a pegboard design then animated digitally), but instead, this is a visualization that shows what could have been that interaction data.

At first I had different abstract iterations of the pattern of the animation but in the end, I chose a more literal pattern of a butterfly. The butterfly pattern was chosen as the pattern design because it is based on the butterfly effect. The butterfly circle pattern and colours are based on the continuous loops in the illustration of the butterfly effect in the Lorenz attractor. The butterfly effect is from chaos theory and it “is the sensitive dependence on initial conditions in which a small change in one state of a deterministic non-linear system can result in large differences in a later state.” In my project, the initial conditions were set by me in choosing the composition and the small changes were the letters interacting with the composition. The large differences are that with each change (each letter showing up) changed the surrounding dots to a different colour, therefore affecting the letter’s surrounding environment. The later state is the final form of the butterfly pattern which resolves to the full pattern being shown and showing the viewer what they were looking at from far away at the end of the video. The song is also chosen with intent as the obscure voice that appears throughout the song says “Fly like a butterfly”, further connecting to the butterfly theme.

Human behaviour is still involved in the piece as the vague data is myself making a visualization out of the word butterfly. Even by planning out where I would place the letters in the visualization triggered pure conscious or unconscious actions, which then turned into the shape of each letter. Therefore, the human psyche is largely tied with my own interaction with the visualization/software and this relationship played into the project.


The animation took a few iterations, as I had to figure out what would be the best workflow for the challenges of animating this video: how to match the animation to the beat of the song, how to animate between the different letter placements and how to animate the different colours in and out. I had to test out different workflows of animating using a small sample of what the video would look like and then apply it to the rest of the animation.  I had to experiment with different techniques, with following video tutorials online on similar animation treatments, to figure out how to get the animation I wanted for the circles.


Final Result


I interacted with the data and the visualization knowing what it will look like but to the viewer, it withholds the full visualization. This makes the viewer see the visualization in parts and think about what is being presented to them. The original visualization is manipulated to become a series, and only at the end of the video, it makes a whole composition. Separately, each part is a composition in itself but all compositions make a whole composition. The bright colours against a dark background are from referencing the computer-generated visuals of the butterfly effect.

In the end, I still accomplished my original goals and objectives that I set out when proposing my idea before designing:

  1. The artwork will evolve itself based off of the choices used

  2. The artwork itself is influenced by the human presence/actions

  3. The output of the artwork is predefined by myself and changed by the participant*

*The word participant can be dualistic. As the designer, I can be considered the participant as I was the one interacting with the composition and changing it and a participant can also be the viewer where they are the one piecing together the visualization, therefore they are changing their perspective of the visualization.