Exclusive conversations about infographics and data-vizualization
The name is well-known in the information design field. Considered by many one of the most forward-thinking and creative internet developers around, Santiago’s quest for understanding the possibilities hidden behind the massive amounts of data we all deal with begun back in 2005 – and it’s a quest with no end at sight.
Since his early days in Bogotá, where he was born, his work has been featured all over the world, he has taught in several universities and speaks regularly at major events like VISWEEK, FutureEverything, VizEurope, STRATA, SocialMediaWeek, NYViz, OFFF and ARS ELECTRONICA.
Yet, the co-founder of the creative ‘visual think-tank’ Bestiaro.org seems to be far from done, when it comes to experiment, invent and develop better ways of communicating through data.
His multidimensional background, his ‘world traveler’ mindset and the unstoppable urge to keep pushing forward the limits of visual interaction make his view of the data visualization scenario rather unique, and he was kind enough to share some of those insights with us:
Visual Loop (VL) – One of the most impressive things about data visualization it’s the range of fields that can benefit from it, don’t you think?
Santiago Ortiz (SO) – That’s for me the most attractive thing about information visualization, it’s perhaps the main reason I decided to work on this field. Back in 2005 I was working at the same time in two collaborative projects: Quiasma, who was based on a set of tagged media files: photos, audio and videos from several trips trough different and conflicting locations in Colombia, and GNOM, a research project aimed to explore interactive visualization techniques applied to a genetic network. Datasets were clearly different in terms of meaning and in terms of the reality they were associated. From the point of view of the structure they were also distinct: one being a list of elements with tags, the other describing a network. I took the first one and by a classic procedure built a network, associating elements with similar metadata. Then I realized that both networks, the genetic and the ‘cultural’ one, had similar number of nodes and relations. So it was only one step to have both datasets depicted with exactly the same method.
And that astonished me. If information visualization is a language, what does it mean that you can use the same vocabulary and grammar to express such different realities?
That year we presented both Quiasma and GNOM in Ars Electronica – that year’s theme was ‘hybrid identities’–, placing both visualizations (that is: the same visualization conveying two extremely different contents) one next to the other.
The question remains unanswered, the astonishment continues. Working in this field allows you (obligates you) to transit through extremely different realities, cultural, technical and science contexts. Reality is everything that is a source of information. And our culture is becoming extremely obsessed in converting every source of information in a source of data. With more and more people and companies producing data-visualization, reality is being mapped by its representations!
Our culture is becoming extremely obsessed in converting every source of information in a source of data. With more and more people and companies producing data-visualization, reality is being mapped by its representations!
VL – You’ve focused a good part of your work around the concept of ‘digital space’, creating and experimenting with interfaces and interactive visualizations. Can you explain us that concept?
SO – I use the very vague concept of space to denote the idea that an interface could be experienced as a place, in the sense people can feel the experience of movement, displacement, the experience of being changing position within a structure that offers a point of view according to a position.
In the case of information visualization the concept of space apply mainly on exploratory interfaces.
VL – In your opinion, how could we transform the ongoing criticism around the spread of bad data visualizations into a more open and positive creative environment for those new to the field?
SO – One story I repeatedly tell to my kids, is the one about an old man and his son that undertake a journey to the nearest town to sell a donkey. First the old man rides the donkey and the kid walks. Someone criticized them because of the small kid having to walk (child labour!). They interchange positions. Then, someone gets outraged by the old man having to walk (disrespect to an elder!). They both climb on the donkey, wich results in new criticism (animal abusers!). Finally, they arrive in town, both walking, and everyone’s laughing at them for not taking advantage of a transportation mean.
I believe information visualization is just a language with everything to be discovered, and we won’t discover new awesomeness without failing. The future of information visualization is being shaped by projects that fail in many senses and that are being criticized.
Every time I read something like ‘Tufte wouldn’t approve…’ I recall Einstein never approved quantum mechanics and for many decades insisted the universe was static, Gandhi was a misogynist, in 2004 Bill Gates predicted the death of Spam in the following two years, and a long etcetera…
I also recall Tufte’s sentence: “Most principles of design should be greeted with some skepticism… we may come to see only through the lenses of word authority rather than with our own eyes.” - Edward Tufte, The Visual Display of Quantitative Information.
Every time I read something like ‘Tufte wouldn’t approve…’ I recall Einstein never approved quantum mechanics and for many decades insisted the universe was static
VL – The volume of data being generated by Social Media activities its certainly exciting and opens a lot of possibilities for those in the field of data visualization. What are the challenges in this area?
SO – Data coming from social media is fantastic. In terms of semantics it contains all types of language uses, intentions and textures, and of course irony in all its possible shapes. That makes social media contents analysis very challenging. In terms of structure it’s also extremely rich: millions of people interacting with text at extreme speeds. The most successful projects that visualize social media data are, so far and with some exceptions, quite static. The recently published emoto is one of the exceptions. Other very recent project that I think is extremely good is Twiplomacy that focus on specific type of twitter users. But in this field, everything is to be done.
VL – What about new projects? Anything you’re working on that you’d like to share?
SO – I’m involved in several projects at the time, most of them being small researches, and a couple of big ones.
The next one to be released is in fact about social media: a set of different interactive exploratory tools, the first one oriented to communities in twitter. By community I mean groups of accounts that often interact: networks of conversations. There are many of them and they are extremely important. Twitter is not just a stack of messages, a huge collection of timelines, there’s a lot of rich structure based on living relations between people. Networks in twitter are not defined only by followers/following but also – and I believe that’s more important– by activity. Twitter is extremely organic, much more than any other social network.
On Facebook you are someone’s friend, or you’re not. And you don’t have any way to seduce someone, to persuade him to be your friend. On Twitter, a relation is something that you can build and grow. This kind of approach requires a lot of social intelligence and a new type of marketing science. And it’s really a big thing, because a community could be the group of people with whom maintain a rich dialogue, learn, be visible and obtain opportunities. A community could be the niche for your business.
Sentiment analysis and, in general, the common analytics in social media gives you results in which the information about ‘who-say-what’ is lost: you get info about who from one side, and info about what in the other. It’s like burning the aliments and losing all the vitamins and nutrients. Community oriented analytics are based in different approaches that try to give people insight without loosing social context. I think the next big think in social media is community analytics.
There’s another project that I will soon release. I don’t want to tell much about it except that it is very related to this amazing research project published very recently: Movie Galaxies.
All projects I’m involved now can be framed into what I call ‘conversations and dialogues visualization’, which would be a field that mixes text analysis and network analysis.
Community oriented analytics are based in different approaches that try to give people insight without loosing social context. I think the next big think in social media is community analytics.
VL – Thank you, Santiago! Looking forward for all the new releases!
SO – Thanks!
We thank Santiago for taking his time to answer these questions, and again for including our Tumblr on the Data Visualization References Network. You can connect with him on LinkedIn and follow his updates on Twitter (@moebio). His full portfolio his available at moebio.com.