Really like this visualisation of local tweets from the royal wedding. Chico Charlesworth and Tom Martin, both developers in the BBC social team, put this together in three days (read Chico’s post on creating the visualisation).
I asked Chico to run me through the tech side of the project, which used the Twitter API to collect the tweets: the Java back-end loads the local tweets, which were collected on the day, into memory and makes them accessible to the front-end through an Ajax call. The front-end then uses jQuery to playback the tweets, updating the map and the stats.
Chico used a name/gender database to determine the gender of each tweet, based on the user’s first name (I like that).
It’s a neat and simple visualisation which captures a remarkable amount of information – images, Foursquare check-ins, gender of tweeters.
This kind of visualisation could be applied to a range of other events. But in order to make it work there needs to be a mix of visual thinking and awareness of available data. That is key – think visual and data first then work out what you want to do.
Interesting to see that there were just under 3,000 local tweets in a 12-hour period – at an event that brought together hundreds of thousands of spectators. This reflects the fact that most people do not have geo-location enabled when they tweet.