Look and feel is a hot topic once again, thanks mostly to Jony Ive’s polarising redesign of iOS 7 – but in the lifetime of PHP we have seen many new graphical design changes. Sure, gradients will take their place next to rounded-corners, brushed metal effects and the blink tag at digital-archaeology.org, but one of the most common visual design elements predates our digital revolution by over 350 years; the Infographic.
I believe that this much-maligned and often vacuous visual equivalent to the sound-bite shall endure as one of the iconic markers of society’s digital development in exactly the same way that the disk drive does. It is destined to mark the boundary between our data naivety and a much needed data cognizance, where correlated data sets do not prove causal connections. We can plot the demise for our current data journalism (that’s plot as in scheme and conjure rather than chart) because what is needed more than an improved presentation layer is a way of addressing how all this data gets here.
Let’s say you rolled your own user admin system to grant access to services and maintain personal profiles. Would a table or two storing profile information and controlling entry be enough? If a colleague is interested in usage statistics for the site on a personal level then no, this is something Google Analytics will not allow*. It’s necessary to identify users each time they access your services in order to log when and with what frequency they access your services. (*Google’s terms of service strictly forbid such use cases for their analytics modules)
Information Design, the capturing of correct (or at least relevant and complete) information is the first part in the development of how you tell your story. In our example, much more data and effort is needed to prepare our personal usage report than a simple last login feature would require. After all, most of the data you currently have is likely to be a simplified user-behaviour model designed to achieve the original feature requests of your site – usage, maintenance and support criteria rather than reports. Cheaper and cheaper disk storage can let your data accumulate without legacy restrictions on efficient table usage – but more incomplete data is just as incomplete.
Yet, even a lack of relevant data is no arrest to the production of the Infographic. A vacuum now exists in every slide presentation where the Infographic is supposed to go, and that space doesn’t stay empty just because the statistics aren’t there for it. Instead, a cherry picked assemblage of correlations is used to satisfy the presentation’s visual needs and the Infographic becomes a statement of opinion, something less than the result of a study or impartial observation of phenomenon. Big Data doesn’t discover facts or even unique new correlations, it hunts for them.
This situation is quite different from the engagement and transparencies of Open Data, which is more often allied to the task of establishing or revealing new perspectives than to routing out a 0.05% efficiency increase. One example I love is Yelp’s word-map, where the frequency of the word “Hipster” in local reviews is plotted as a heat map over a city. Or equally the “Trees, Cabs, Crime” map from Stamen. Even the poorly named ‘WoMEn index’, an attempt to replace GDP with a musical energy consumption rate calculated from Echo Nest & LastFM data sources, is entertaining and informative. Between them, they prove that challenging assumptions and questioning attitudes is as achievable with data as it is with ink or paint. For Information Artists, the number of available datasets to work from equates to the richness of the palette.
Providing views of data should be precisely such a smart process, one which the user will interact with, not just passively observe. If we build micro or single page applications out of tools like d3.js and Angular or directly onto HTML5’s canvas we can invite users to interrogate them for meaning, and they will discover a great deal more.
So as data developers and architects to the APIs required by such devices, let’s make it possible to produce them when we can. While traditional data visualization lets us see things that were previously invisible, like Galileo’s telescope they are equipped with no focusing method and an almost cripplingly narrow field of view. The alternative we may consider – and I’m arguing we should feel obliged to help create – is the empowered interrogation of data whenever it’s presented. Next time you’re creating tables for an application, keep Media Molecule’s famous tagline in mind: “play, create, share.” You can mitigate message bias and validate curiosity at the same time as engaging your users with a whole new kind of geeky fun.
Ben Greenaway has been a software developer for 18 years. After pioneering internet narrowcasting in the 90’s and collaborations with Cyberia and Virtual Futures on web installation projects he led development on A/V installation projects for Pixar and Sony Entertainment and web applications for SMEs in Southern California before moving to the San Francisco Bay in 2010. He now lives in South London where he balances time writing and developing with the demands of an ever growing collection of retro computers.