I’m technically on vacation this week, but I cannot take a vacation from the Interwebs. While browsing feeds yesterday, I saw this post from Gizmodo titled “Foursquare, As Seen by the Predator”.
My crush on foursquare is well-documented, and I’ve always liked “The Predator“, the original one, not all the sequels. So, I was intrigued.
Turns out Steven Lehrburger (@lehrblogger) built a mashup of foursquare checkins displayed as a heat map overlaid on a Google Map. He calls it the Where Do You Go project, and he put it together for NYU’s ITP Winter Show.
He offers a couple different visualizations, several of which look like the Predator’s thermal vision. I guess that makes sense for a heat map. My heat map for Portland is above, and here’s the one from my visit to San Francisco for Open World.
Steven took his inspiration in part from 4mapper, which I covered recently when foursquare launched their API, and his implementation is sweet. A couple functional differences aside from the visualization, Steven’s mashup is dynamic, so as you checkin it updates, and he is pulling all my checkins, more than the 250 limit reported by John of 4mapper.
Maybe the API has been updated, or maybe foursquare, like Twitter, has a firehose API. That would make sense, since these checkin data are a goldmine to advertisers.
The dynamic bit is interesting to me, especially this time of year, as I’ve ventured out to places I don’t normally go to buy gifts, as depicted on the map. Yeah, I usually shop online, but this year I waited a bit too long.
From the beginning with foursquare, you can see the value of aggregated checkin data to local businesses. Now that Steven and John have put these checkins on a map, you can literally see how valuable geo-data can be.
I got you another gift, an early prediction for 2010. Here it comes:
Get used to geo. It’s coming soon to Facebook and will be the story of 2010. The main story lines will be mainstream acceptance or rejection of geo-tagging update, and advertisers falling over themselves and paying big bucks to get access to geo-data.
Anyway, for those interested, Steven built his mashup thusly:
Written in Python and hosted on Google App Engine, uses the Google Maps and Foursquare APIs, relies heavily on code adapted from the gheat-ae Google Code project, uses Mike Knapp’s OAuth library from github, uses jQuery and BlueprintCSS for the front-end.
His final conclusion is pretty funny.
Even simple web applications are hard to build to be robust, scalable, and ready for a larger audience.
So, enjoy the holidays. I’ll be back next week to finish the epic WebCenter VM journey, review 2009 and look ahead to 2010.
Find the comments if you feel motivated.