Intro To Data Visualization Project 2: Quantified Self

In Dreams I Walk With You

Since early 2019, I’ve been keeping a dream journal on my phone. The original impetus was to better remember my dreams—to be able to hold on to those fleeting fragments I remembered upon waking, and, over time, to be more conscious of them. The practice of logging did in fact help me remember more of my dreams with more frequency, but until now I haven’t spent any time revisiting those entries. 

For this project, In Dreams I Walk With You, I used data visualization to explore one particular aspect of those dreams: who I spend time with on the astral plane. The title for this project is taken from the lyrics of the song “In Dreams,” by Roy Orbison, in which the narrator describes a relationship with the love of his life that exists “only in dreams”—in his waking life they are separated. For myself, and I imagine many others, the relationships in my dreams are not tied to the temporal reality of my day to day. In dreams I spend time with childhood friends, former lovers, family members living and dead, strangers, and people near and far from me in real life. 

My goal was to make visible these dream relationships, and to complement them with information about our waking relationship. Through this combination—and interactive functions that allow the viewer to explore—I hope to paint a picture of my conscious and subconscious relationships, and the overlaps or gulf between them. Ultimately it’s a self-portrait; more art than science. 

In Dreams I Walk With You (full-screen view recommended)

I used a bubble chart to show who I dreamed about the most, what our primary relationship is, and whether or not we are still in contact: Is this a waking relationship, or does it exist only in dreams? What I found is that I dreamed about my parents, my best friends, and my two former longterm girlfriends the most. But I also dreamed about a number of acquaintances, celebrities, and past and current friends. As a group, my friends showed up more than anyone. I also found that, as I suspected, the current status of our relationship didn’t play a huge role in who visited me. I dreamed almost as much about people I’m no longer in contact with or who I’ve never met, as I dreamed about those with whom I’m still in touch. 

I liked the idea of using bubble graphs, pie charts, and color density to convey much of this information, because—like dreams—they are somewhat imprecise. Even the dream journal loses something in translation, and is only based on the dreams I remember. Many more dreams go unremembered, and many details, feelings, and experiences don’t make it into the source material. Meanwhile, the bubble graph makes it easy to see who is on my mind the most, and makes it possible to learn about each individual represented. With regards to color I tried to convey some further relationship between the relationships themselves. For instance an acquaintance is somewhere on the spectrum between a stranger and friend, a classmate is somewhere between friend and coworker. I used a desaturated color scheme to represent the relationships that exist primarily in my dreams, a realm that is hazy, lacking in concrete substance. I used blue throughout because I often dream about bodies of water. I also included a number of filters, so myself and other viewers could make quick slices and dig deeper into my relationships and the ways they are layered with meaning or affected by time. The pie charts themselves also function as broad filters.

I originally intended to include photos of everyone represented. I didn’t have the time or technical ability to pull off that vision here and it also raised some questions about where to draw the line regarding anonymity. In this version I already opted to remove some identifying information. The artist in me wanted to bare it all, but practically, I am also concerned about preserving people’s feelings and not sharing information about their relationship to me they might rather keep private. I wasn’t entirely thrilled with the functionality of the filtering system…it is a bit wonky. I also played with the idea of tying color density to a sliding scale of transparency based on when we were last in contact, but couldn’t yet figure that out while preserving the other color information I wanted to include. I collected geographic information that would be fun to play around with in addition to the visualization here. And in my dream journals themselves, there is more information that could be mined. It would present new challenges, but I could see a similar approach to this project trained on things like environments/locations within the dreams, or any feelings, colors, etc. that I made note of.