Future Times is a data sculpture consisting of a stack of 100 1-foot square acrylic plates with each plate representing one year from 1906 to 2006. Each plate has 100 holes that correspond to the top 100 most frequent noun phrases associated with the phrase "in the future" for each year according to Google News Archive, Google Scholar, and Google Blog search results. The diameter of each hole varies with the relative frequency of each term. For example, in the plate for 1945, the hole for the term "war" has a large diameter since many people were making predictions about the future of warfare. The plates are each 3/8" thick and are stacked vertically on top of 4 small plastic supports with the oldest plate (1906) at the bottom and the most recent one (2006) on the top. The entire height of the piece (with supports) is 40". A horizontal beam of white light is projected onto one side of the sculpture and randomly walks up and down it to illuminate a single plate at a time. (More on the sculpture here and here and on Vimeo)
The site futureme.org is a web service that allows you to write yourself a letter to be delivered at a later date, up to 50 years in the future. They currently host nearly 400,000 letters, about 30,000 of which are public. By writing a letter to your future self you might, for example, remind yourself of something you are scared of forgetting. or perhaps, you might remind yourself of something you are trying to forget and you want to check on your progress. We scraped some data, calculated word occurrences, filtered out some major stop words and obtained a cloud of recurrent words. Words like "remember" and "forget" show up in the top. From the 30,000 emails sent to your future selves, we selected the ones in which terms like remember, remind, forget, and forgot were highly recurrent. We ended up with about 7,000 emails. The overall concept of the project is to attempt to give viewers a "sense" of the emails from futureme.org without them having to read all 30,000 public messages. The futureme.org dataset is quite rich, so this visualization represents one possibility that we were able to complete within the time constraints of the project. More on the concept and the dataset.
A dark screen offers three white rectangles to solicit a day, a month and a year from the user. There is a moment of anticipation, complete darkness except for the white letters of the date at the very bottom of the screen, then words begin to fill in the space. First the words from the single chosen day of the most recent archived calendar year, 2006. The most frequent terms are the boldest, the least frequent smaller and faded. Then terms from the year before, in a different color fill in more of the dark space. Then terms from the year before that, in another different color, fill in. The cycle continues back till it hits the user-chosen year limit. The cycle stops, the darkness is completely filled with color and words, a different color for each year, a different intensity of color for the different relative frequencies of the words. Some words repeat. Some words stand alone. The final image is a conglomeration of colorful words that fill in a slender column of darkness. It is a pillar of media memory. Intense words, repeated words, stand out. The rest fade back, as they do in our own memories. More on the concept here.
Playing on visualization, our project presents an “audiolization” of the most popular guitar solos for five decades, 1960’s-2000’s. The idea is to create an “average” most popular solo by combining riffs from solos across the years. Since musical styles and tastes change with time (dramatically so, from the golden years of rock n’ roll to the heavy metal, hip-hop and other styles of later years), the “average” solo will comprise a kind of audio history of popular music in the later twentieth century. The effect, we hope, is to explore perceptions of the mean when these are presented aurally rather than visually, expanding the sensory range through which averaging can be perceived and exploring the meanings it has for the users.