Dominikus BaurData Visualization
The W3C introduced a new client-sided way to detect a visitor's location with the HTML5 Geolocation API. For the OECD Regional Well-Being project, we decided to give this new way of learning about a visitor a try and get them started with the visualization in their home region. However, actually providing them with a nice user experience while moving through this somewhat complex process proved to be difficult. In this post, I'll discuss our issues and workarounds.
Well-being exists on a scale from the individual to whole countries. In the OECD Regional Well-Being project - a part of OECD's Better Life Initiative - we dive deeper into this hierarchy: the interactive, web-based visualization shows various aspects contributing to quality of life for over 300 regions all over the world. In this post I'll discuss the project as a whole and its design decisions (with more technical posts coming up).
In the selfiecity project we explored the depths of a seemingly shallow data set: 3200 selfies from five cities around the world. We were wondering what exactly was there in a selfie and how these spontaneous self-expressions would reflect cultural backgrounds. A combination of manual and automatic analyses gave us fascinating insights which we represented visually in a series of charts and graphs. At the project's core there's also the selfiexploratory, a tool to interactively explore all selfies and their metadata in the browser.
Last week I was invited to the 5. Data Science Day in Berlin, organized by The Unbelievable Machine Company. We spend one day with fantastic speakers from various areas, discussing the implications of mobile devices on data science and visualization.
In my talk - 'Big Pictures in the Small' - I focused on a more high-level picture why visualization will go mobile and what form that will take. In addition to our smartphones, we've grown accustomed to a whole zoo of other (portable or stationary) computing devices. I discuss how that's eerily similar to computing environments of the past and how we need more excuses to do data science on couches. Also, be warned that this is only a re-recording of my talk. So if you're into disembodied voices mispronouncing English words - today's your lucky day ;)
The main purpose of most computing interfaces is efficiency. There's nothing more annoying for anybody using a computer than sluggishness, the inability to produce the right results or seemingly unnecessary waiting time. At any point in time during the computing experience, the person in front of the machine has a goal in mind and wants to make steps towards it. We've grown so accustomed to this efficiency that anything that slows us down is seen as a problem and we only have little patience for any distraction. And if we fall into the trap of clicking on a hyperbolic link on the web ('Data analysts hate her! Mom discovers one weird trick for finding insights') or lose ourselves in hour-long TV tropes or Wikipedia binges, we feel the sting of guilt throughout.
Jacob Nielsen has called this very purpose-driven behavior lean-forward: We're deeply engaged and treat computers and the web as an active media. This is in contrast to TV, books or magazines, which is about passive consumption, entertainment and lean-back.