for the Journalists of Southern Africa
In the latest instalment of her “Backstory" series, Gill Moodie writes exclusively for journalism.co.za:
Gill Moodie: What is so interesting about data visualisation is that it is so multidisciplinary. You need to be a journalist and to know how to mine data and have an understanding of stats. You also need a visual sense and you need to know how to code. (Click here to read a blog post by Kirk on the different skills required: “The 8 hats of data visualisation design”.)
Andy Kirk: Yes. It is essentially a unique convergence of art and science. For me, personally, when I was at school I was that rare kid who did art and maths and so the discovery of this field – which was only five or six years ago now –was a real “eureka” moment…
But you’re right. It’s something that has a lot of facets and a lot of challenges you get from stats, mathematics and the sciences around compu-science and visual perception and psychology – how people relate to shape and form – and then right through to the artistic side and not just computer graphics but also a design instinct…
You may not have all of those skills but you’ve definitely got a role to play if you’ve got some of those skills. And then it’s about how you find a way to address those gaps – skilling up yourself or collaborating with others who can bring those different perspectives.
Moodie: But even with a design sense, you can still make rubbish infographics. Often you see very counter-intuitive ones. They can be very pretty but you can’t work out how they fit together.
Kirk: Yeah. There’s a real tight rope to walk – and this is something I really focus on when I teach. Good visualisation is not just about how quickly you get something – that’s the first thing. Some things are complex and they need a little bit of time from the reader to invest, to learn how to decode something. But the key skill of the designer is to give them a chance – that when they’ve learned that, they still get some insight.
So many times there’s something that’s intriguing but you never get over that barrier of the confusion, the clutter, the unusual representation. So my key piece of advice is that you don’t need to dumb things down. You don’t need to dilute the essence of a problem but you need to give people a chance that when they invest time to learn something, that they will get there and they’ll get to dig out the insides. And that’s a real difficult thing. A lot of data visualisation you get these days looks very nice and very clever. They’re very intriguing but you never cross into the functional side of things – which is learning something about a subject.
Moodie: And really good visualisation is not just about illustrating the copy but finding entirely new stories and looking at things in a new way.
Kirk: Yeah, absolutely. It’s finding stories and telling stories in data visualisation. And sometimes… you’re trying to create or provide an interface to let readers find their own stories. You’ll frame a subject within a visualisation and it’s all about trying to find what’s interesting to them, combinations of variables…
Moodie: Let me get my head around that. Are you saying you’re giving the reader the opportunity to find what interests them and different combinations of being able to read the information?
Kirk: Yeah, for example, there’s a project called Wind Map, which is a very intelligent portrayal of wind speeds and directions in the US.
Moodie: Golly, that would work well in Cape Town, where the wind is such a big factor in our lives but it varies so much from area to area.
Kirk: Yeah. What this (Wind Map) does is that it creates a single interface of millions and millions of data points. But then it allows you, as user, to drill down into the locations that interest you or into the patterns of strong winds or direction and what it means for you.
Moodie: OK, like if you’re a surfer, it could tell you a lot about swell.
Kirk: That’s right. It allows you to take control and find what’s interesting to you.
Moodie: Andy, you travel the world doing training and speaking at conferences. Are their particular countries that do this better than others, do you think?
Kirk: …Up until about two years ago you’d definitely say the States was definitely well ahead of most others and that’s in terms of prevalence of data visualisation and also just the tech and the awareness of organisations that this is an important thing to embrace.
But certainly over the last three years, Europe in particular had definitely caught up. There’s some great work being done in Germany and Holland and Spain and France… Although it’s still quite fragmented. So we’ve still got fairly small pockets of people doing this in terms of training and design practitioners. It’s still very much the start of the journey…
Moodie: Are there any journalism schools that teach this?
Kirk:Yeah, let me point you in the direction of a great podcast. It’s called “Data Stories” and it’s facilitated by two very prominent names: firstly an Italian guy, who is currently based in Germany called Enrico Bertini, and he’s assisted on the podcast by a very prominent German designer called Moritz Stefaner, who is essentially one of the very bestdesigners in the field. He’s one of those unique guys who’s got every hat (i.e. set of skills). He’s a very skilled fellow and they do an informative, very entertaining podcast every couple of weeks…
One of the most recent episodes was an interview with a guy called Alberto Cairo, who is a Spanish fellow currently working at the University of Miami and this episode was really about his most recent book called “The Functional Art”… But Alberto is essentially a data-journalism professor and in this podcast he mentions a number of different schools that are teaching variations of this journalism…
Moodie: You know, we have pockets of innovation in SA but I think one of the reasons this hasn’t gained more momentum here is that laying your hands on the data – especially from the government – is not easy.
Kirk: That’s one of the things that came across in the training (I did for Media24 in SA recently). Certainly in the UK and the US the culture of data openness and transparency has really got some momentum now.
Moodie: Do you think information is generally more accessible to people if it’s visual?
Kirk: Sometimes you don’t need to visualise it, you know, so a table might be fine if you’re just working on stats… The real value of visualisation is to see the patterns and relationships that you don’t see in the raw form…
Moodie: From a journalist’s perspective, you seem to be saying that amid all the different skills needed to visualise data, the journalist’s role is to think about what the people out there would like to know about.
Kirk: Yeah, and I think that that single skill is important – if not the most important – because this is a person who is able to reason about what are the important stories that I think we should tell and that I think people want to hear. (With data sets) you’ve got millions of dimensions of stories – that you can slice – and a lot of the weaker visualisation practitioners may be very good designers but they don’t have that critical sense of what’s the important thing to tell and what to leave out.
And that is something (you find) when you visit a place like The New York Times or The Wall Street Journal. They’ve got such a critical sense of what’s an angle and what’s an interesting angle – and the discipline to eliminate everything else because you can’t tell every story… It’s really about saying: ‘I can step away from the temptation of a certain design technique’ or ‘I can get away from the noise of the data and separate myself from that and say I think we should be looking to tell this story’.
Those journalists out there who are looking to slightly modify their skills sets – if they can pick up these data-handling and statistical capabilities as well as deploying the many free (visualisation) tools – they’ve got a real strength in terms of doing effective work in this field.
Moodie: Is there a British paper which does this well?
Kirk: I would say that the strongest is The Guardian. It’s got a very strong culture of data journalism. Now what they’ve been doing in the last two or three years is really pioneering this data blog on the website, where every day they will publish a data set that goes with a main story. They’ve got a number of data journalists, who will very quickly – once a story breaks – publish a data set and also do some visualisation.
The reason why I distinguish them slightly from others is they don’t always spend too much time developing very sophisticated, polished visualisations. They spend more time gathering the data, cleaning it, publishing it and then doing some quick visualisation. So they use a lot of the free tools out there such as Tableau Public and Google’s visualisation tools like Fusion Tables, for example. If you compare this to The New York Times, for example, (The New York Times) has got the resources, the time, the capacity to create very advanced, very clever successful visualisations and graphics to go with their stories.
So I don’t think at the moment that there’s another UK paper that stands out very strongly on the data visualisation and design front… The Guardian’s (data blog) is a brilliant resource and I would recommend it to people who want to do this.
Moodie: I know you can do really cool things online but surely you can employ data visualisation in print too?
Kirk: I personally think the most elegant work is a static, that conveys motion and emotion. Perhaps I’m biased in this as I don’t have the advanced programming skills to create all the interactives that I’d like to build in my head. But I personally think it’s amazing what can be done (in print) and is being done.
Moodie: Is that because the choices are so stark in print?
Kirk: You’ve got a limited static real estate to work in…
Moodie:So it hones the thinking?
Kirk: Yeah, absolutely. As I said before, about the difference between telling a story and allowing people to find their own stories, what we see in an interactive is that it’s much easier to give people a chance to find their own stories – which I think, in many respects, increases the responsibility of the designer (doing a static visualisation) to find that analysisand find that story. The best graphic artist and the best visualisation people are really deft when it comes to doing a static portrayal.
‘How to start doing data journalism’, Editors’ Weblog, April 2012
‘Design challenges and opportunities behind award-winning online work’, Journalism.co.za, May 2011