by Infogram
Create infographics

100 pieces of flute music

A quantified self project where music and design come together

February 6, 2015

[This is a guest post by Erika von Kelsch*, about her visualization project, 100 pieces of flute music]



The Project

(image: The final infographic | Erika von Kelsch)

The premise of the project was to organize 100 pieces of data into a static print piece. At least 7 metadata categories were to be included within the infographic, as well as a minimum of 3 overall data rollups. I chose 100 pieces of flute music that I have played that have been in my performance repertoire. Music was a potential career path for me, and the people and experiences I had through music influence how I view and explore the world around me to this day. The way I approach design is also influenced by what I learned from studying music, including the technical aspects of both flute and theory, as well as the emotional facets of performance. I decided to use this project as a vehicle to document this experience.

The Data

(image: Collecting and organizing my data into spreadsheets helped me to sort and view the data in different ways | Erika von Kelsch)

I started finding my list of 100 pieces by looking through my sheet music library, looking at the most recent pieces. I decided that even though a total of 7 metadata categories were required, it would most likely benefit me to find more categories that i could potentially represent visually. This would give me more options later in the design process, without the need to re-comb the data. The categories I recorded were: name of piece, name of composer, starting key signature, date of composition, my age when played, type of performance, people/group I performed with, length of piece, time spent practicing, how enjoyable it was (1 to 10), how memorable it was (1 to 10). Even though I didn’t end up using all of these categories in my infographic, they were extremely beneficial for the data summaries that I had to create.

Visual Explorations and Sketches

(image: Sketching by hand was useful to see different forms my data could take. I decided to go with a circular organization, based off of the circle of fifths)

Once I was happy with the overall collection and the preliminary data exploration that I had done, it was time to start sketching how the data would take on visual form. I first looked at two different types of overall form that my data could take: a linear structure placed on a straight line which could include bar graphs, or organization into a circular structure.

I wanted to decide what would be the first level of hierarchy in my data. Through brainstorming and thought mapping, I came to the realization that the circle of fifths could be used as a circular structure to place each piece of music, organized by key signature. This was one point from which I started to structure the rest of my infographic. This led me to sketch simple circular organization methods, thereby allowing me to visualize the product of each. By continually sketching at different scales and levels of detail, I worked out different extensions and modules around which I could organize different levels of my data. This set me up with different visual approaches to the same problem.

(image: Once I had decided on a general visual form, I started looking a structure in more detail)

It made sense to represent each piece of music in a bar, with the length of the piece determining how much space the bar would occupy on the circle. The length of the bar would indicate how much time I spent practicing.

(image: I started testing how one individual piece of data would interact with the rules and structures I had created.)

Because the history of music and the time period of the piece were important aspects that were stressed to me by teachers and conductors, I wanted to include a timeline showing date of composition. Because many of the ways I was looking at the data were concrete, I wanted to create a structure that would allow for abstraction, as well as offer contrast against the straight lines and bars I had made to represent other metadata categories. This would also help to develop distinctly different layers of information for the viewer to navigate.

(image: I also sketched at a size closer to final scale with multiple data points tested on the same system)

Taking feedback from professors and peers on my preliminary sketch, I moved forward to building the infographic in Illustrator.

Production and The Final Print

The first section I created was a skeletal donut chart, and divided the shape up into the appropriate amount of slices for my data. Next came the date of composition timeline as well as the concentric circles for my age. I wanted the final visualization to be quite large, and I set my initial file size to 6’x6’.

(image: Progression of selected draft files.)

Drawing all of the individual threads to connect the bars of music to the composition timeline was time consuming, and special care was taken with the handles and nodes to create flow and movement. I decided to create points on each piece of music to help ease the transition from bar graph to the spiraling lines, many of which hand very steep curves which would have been partially or fully hidden without creating these points.

I chose the Univers family for my type treatments, as it allowed for incredibly condensed text for my titles of the circle graph as well as a visible and eye-catching title at a much larger size. the length of information in these spaces varies greatly and the flexibility of Univers’ different weights was very welcome.

Color was one of the largest challenges I encountered. At first, I tried using saturation and tint levels to indicate an increase in sharps and flats as they reached the bottom of the circle of fifths in the different key signatures. Blues represented keys with flats, as a flat lowers a note by a half step. Yellows were used for the sharps, as they raise a note by a half step. I found that even though there were reasons for these choices, the visual impact and representation of information weren’t effective.

Additional research into synaesthesia was conducted as a potential solution to the connection I was creating between key signature and color, but I did not find a schematic global or consistent enough to be used for my data set. Not being a synaesthete, I do not have my own set of color applications to key signature as it fit into my particular sets of data.

Because I was working with 12 different colors and had already tried systems of dual colors with the result of ineffective communication, I decided to try a 12 hue system. I went through many iterations of color to reach something that I felt was balanced as well as eye-catching.

I ended up taking away some of my original metadata categories to help resolve issues of complexity and navigation. This left 8 final represented categories: name of piece, name of composer, starting key signature, length of piece, time spent practicing, date of composition, if the piece was performed, and my age when played.
The final print was done on a Canon imagePROGRAF iPF9100 and the final print size is 60”x62”.

(image: A large final print allowed for the information to be better understood by the viewer as well as create more visual impact.)
(image: title and key.)
(image: Alternate view of the infographic.)
(image: My final data rollups became a reflection of the extra data I had collected at the start of the process.)
(image: Alternate view of the infographic showing the strands connecting to the timeline)
(image: Alternate view of the infographic showing the bar graph’s type treatment)
(image: Alternate view of the infographic showing the bottom portion)


*Erika von Kelsch is a junior working towards a BFA in Graphic Design at Massachusetts College of Art and Design. When she isn’t working on projects, she enjoys drawing, making prints in MassArt’s letterpress shop, and spirographs.You can find more of her work on Behance.

Written by Tiago Veloso

Tiago Veloso is the founder and editor of Visualoop and Visualoop Brasil . He is Portuguese, currently based in Bonito, Brazil.