Last spring, in Nicholas Felton’s information visualization class, we were tasked to visualize a personal collection for our final assignment. Realizing that the final was due on the exact anniversary of my move to New York City, I decided to visualize my collection of foursquare checkins for the first year. This view is an overview of the entire year, providing a high level view of I moved through the city and how often I was at venues. Further explorations, which I’ll hopefully be getting up this week, dive into the specifics in more detail.
Using the NYC.gov OpenData dataset “Maps of Basketball Courts”, I was able to quickly generate a heatmap of all the public basketball courts maintained by the City of New York, 1,129 in all. Not surprised at all by the noticeable absence of courts in midtown.
2:08 AM Edit :: Added figures 3 & 4 to show the 697 baseball & softball diamonds maintained by the City of New York. A merged map is in order for tomorrow
I love music, and I love data. So I decided to go back and examine my music trends in 2011 to see what I was listening to ::
- Finals at SVA, April and December, generated large listening spikes.
- You can see when I started at Nike in mid June and was too new to be comfortable putting headphones on at work. Plus, I hadn’t found a way to scrobble my music on my work computer yet.
- My trip to Colorado on the 4th of July sees a spike in music consumption, as I spent many hours on airplanes and trains, as well as DJing by the pool at Jakes.
- A lot of my late summer was spent on TurnTable.fm, which lead to a higher variety of songs being played, rather than full albums. This is represented by many thin layers, rather than fewer, thicker layers
- Early September was my breakup with Claire, and I remember not really listening to music during this time. I’m not entirely sure why, but its incredibly apparent.
- I purchased Spotify premium in mid September and saw my music consumption continue to rise from there on out.
- Mid October was the Jay-Z / Kanye West concert, in which I listened to The Throne, as well as their individual albums, pretty much nonstop for 2 weeks before and a month after.
This visualization was created over at lastgraph via my last.fm data, which has consistently logged my music listening from turntable, itunes, and spotify. You can view a high resolution version here.
I’ve been slowly writing some Processing code to create some additional visualizations in the pattern of my listening, such as time of day, and weather compared to genre. Unfortunately, with thesis scheduled to take up all of my time between now and May, I might not get back to it until the summer.
And if you’re interested in what I thought the best music of 2011 was, check out my playlist.
Eric Fischer, of “See Something or Say Something” and “Locals vs Tourists” fame (not to mention an artist in MoMA’s “Talk to Me” exhibit), has taken inspiration from my Nike+ project and created his own using San Francisco MapMyRun data. He went to great lengths to acquire a data set that could help tell this story, and also turned my attention to a blog post that brings up a great criticism of my project. I love seeing the similar patterns emerge between New York, London, and San Francisco :: people gravitating towards water, parks, and bridges. And again, no map of San Francisco is needed beneath it, the runners have created their own map of SF, from the landmass all the way down to individual streets.
After seeing Cooper Smith’s visualizations of data from runners in New York City, I wanted to see what similar data sets would look like for other cities. Nike+ doesn’t have public GPS logs, but MapMyRundoes, if you are willing to spend several hours clicking through search results to hit the “Download” buttons, so that’s what I did to get the tracks for these 771 runs (from June 13 through August 9) in San Francisco.
As Open Source Planning has pointed out, uploaded runs come from a fairly small, self-selected group of people, the most obvious result of which is the total absence of the southeastern corner of the city from this map. It is also a very self-conscious process, so it is biased toward intentional, and often intentionally difficult, trips made for their own sake, and away from the repetitive patterns of everyday life.
Unfortunately the MapMyRun tracklogs do not have date and time stamps, so it is not possible to do the time of day, pace, and interruption analyses that Cooper Smith did. I should have done direction of travel, though.
Last year, in Nicholas Felton’s information visualization class, we were given 1,000 runs of Nike+ data from the NYC area and worked to visualize it in a way that told a compelling story. You can view that project here.
Over the summer, Wired UK contacted me and asked if I’d be interested in doing a followup to the project, this time with 10,000 runners (some 6 million+ lines of data) in London. It was a great opportunity to further explore this running data, this time with a city I’m far less familiar with, which led to some really fun discoveries in the data. In the end, we decided on this heat map direction.
I also worked with them to concept around what the iPad version should show. We decided to render 24 separate heat maps, one for each hour of the day, allowing the user the ability to scrub back and forth through the day. Lots of interesting insights here, but definitely the most interesting is how people stop running through parks at 7 pm. My hypothesis is that runners no longer feel safe in these parks once it becomes dark.
I hope to do a more thorough writeup on my portfolio site this week. In the meantime, UK friends keep an eye out for the November issue. My Fellow Americans, worry not, you can still grab the iPad version here
…and another look at runners’ directions. The Brooklyn Bridge is being used almost exclusively to funnel runners into Brooklyn, while the Manhattan Bridge is used to cross from Brooklyn to Manhattan. It appears these bridges were appropriately named. The Williamsburg Bridge is a perfect mix of both directions, with the majority of runner’s taking the side of the bridge going with traffic.
Last weekend I was running the bridges loop in downtown Portland, and I realized that I had crossed paths with several people twice in my run. It got me thinking about how I had picked to run the route counter-clockwise, while others ran it clockwise. Personally, I deal with it this question a lot - the first day at the Nike berm I was convinced I was disobeying some unwritten rule to run clockwise. Other times, I pick loops based on where I can start uphill and end downhill.
But what would this data look like collectively? Are all people running one direction, or are they randomly picking a direction?
If a runner’s position changes towards the south, his path is drawn yellow. If he’s running north, blue. Here, using the loops of Central Park, we can see that most runners choose to run the loops clockwise, regardless of the direction.
Taking a few steps back in visual complexity to teach myself how to access XML data feeds. Sure, I was able to make this graphic in week two of Nick’s class…but from a static, nicely formatted excel doc. Next week, without any further effort, this chart will look entirely different. That’s pretty rad.
Tufte
This past week, Nike was kind enough to send me to Edward Tufte’s lecture on Presenting Data and Information. Though the lecture was catered to the larger crowd of powerpoint power users from accounting departments, Tufte’s principles towards presenting information have universal application, especially to those of us interested in the ever growing field of dataviz and infographics. Tufte’s lecture was the epitome of his definition of great information presentation; dry, straight forward, absent of superfluities, and extremely informative.
Below, I’ve attempted to type up my notes in a way that makes sense. And where applicable & available, I’ve included the reference that accompanied it.
