For the last few weeks, a few of us here have been locked away in the highest of the NM Towers, working away frantically scoping out, designing, prototyping and building 3 real-time visualizations to accompany Channel 4′s pre-budget Dispatches special, How to Save £100 Billion LIVE, which was aired on Monday evening. It’s been a privilege to work on, a highlight of my career even, but no small task by any means!
Just over 3 week’s ago, we were tasked with producing 3 visual representations of the conversation around the show, as it unraveled, in real-time, on Twitter, the goal being to give the audience at home a voice in the proceedings, and access to the collective thoughts and opinions of others watching, making it easy to follow the live TV show and online conversations, simultaneously.
We immediately set about defining how we might clearly and accurately represent the conversation visually, settling on slicing and dicing it from a few key angles: who’s talking, what are they saying, how much are they saying, and where are they saying it, complemented by some bite-sized, easily-digestible statistics, such as number of tweets, tweets per minutes, number of ‘tweeters’ and most authoritative tweeter (hat tip: the awesome Infochimps trst.me API).
Teaming up with local data visualization whizz kid, Mr. Ollie Glass, we got to work right away, first prototyping 2 proof of concepts using the Processing environment, and then iteratively testing and re-working until we were (mostly) satisfied. A few last minute deployment and integration issues kept us all on our toes (also, I’m not sure I helped massively by demanding a last minute complete rewrite of 2 out of the 3 visualizations), but after 3 week’s of grueling days and impossibly long nights, I’m incredibly happy with the results (*yay*), both from a technical, and a functional ‘social TV experiment’ perspective! Go team! :)
I will happily follow-up with a technical ‘deep dive’ if anyone shows even the slightest amount of interest (FYI: it involves helicopters, chimpanzees, lots of Sellotape and a jazz guitarist) but for now here’s a quick run-down of the data (visualized in a very neat list):
- Total tweets: 2,020
- Total tweeters: 1,017
- Top tweeter: Sunday Telegraph Business Editor, @ahmedk123
- Conversation velocity: a whopping 23TPM (Tweets Per Minute)