Manuel Lima is a designer, researcher and author well known for his work on visualizing and mapping complex networks. A Fellow of the Royal Society of Arts, he was named one of the "50 Most Creative and Influential Minds" by Creativity magazine, and is both the founder of VisualComplexity.com and a Senior UX Manager at Google.
We talked with Lima to find out his thoughts on the connection between data visualization and architecture. The following conversation explores his inspirations and process, as well as his views on how data visualization can help improve the quality of our cities.
Fabian Dejtiar (FD): What inspired you to first turn data into visuals?
Manuel Lima (ML): Sometimes it's hard to explain the specific moment in your life that drove something, but I can recall a very specific moment back in 2004. I was doing a Masters of Fine Arts at Parsons School of Design in New York. I was in the audience attending a lecture from a teacher of mine called Christopher Kirwin. He showed us this "understanding spectrum." It's a diagram that goes by different names, but basically, it shows how data leads into information, information into knowledge, and knowledge into wisdom.
Even though my background is actually industrial design, and is similar to architecture, I wanted to be part of that process. Specifically, building a bridge between information and knowledge, between producers and consumers. I think that's one of the hardest things we can do. Humans are great at collecting information, and that's why we have the big data phenomenon today.
But now we need to make sense of data, converting it into information, and ultimately, knowledge. That's our greatest challenge. For me it was kind of like a calling; here was a moment where I thought, wow, I need to be part of this movement. I wanted to work creating knowledge and wisdom through data.
FD: An important part of knowing what to do with data is knowing what questions to ask. How do you realize what questions to ask? How do you know what is the relevant data to show when there is a lot of data?
ML: I think it always stems from the question that you're trying to answer. In your head, when you put together a visualization, you are either thinking about a question you want to answer or a message you want to convey. Sometimes you go off the data that can serve that purpose. So let's say that you tried to better understand different demographics. You want to try to understand our users, let's say, to navigate through a building or an urban layout. You have that immediate question in your head, and data can provide that answer.
Of course, raw data is rarely enough for you to make sense of a pattern. The answer you need is to visualize it. You need to convert and translate data into visual metaphors and models that we can understand. And therefore, you're going to be able to answer that question through some sort of visualization that can either serve the message you want to convey or answer the question that you want.
FD: How do you do that? How do you work?
ML: There are so many different tools you can use these days. There are different libraries that you can use to make sense of the world. Lately, I'm more interested in the taxonomy, in understanding the system, and I really feel data visualization is a language. In the same way that the written language is made up of building blocks like words that you can combine and create sentences, the same thing happens with data visualization. We have graphics you can combine using color, size, or shape. And through that process, you create messages, meaningful sentences.
For me, I look at the different ways that people have been using this language by understanding the similarities, the differences over time, not just in modern times, but also going back in history and understanding how humans have been using images, symbols and visual communication as a means to convey information.
I'm always fascinated by how long this has been going on. We have been using images for way longer than we have been using written language. The oldest known alphabet is roughly 6,000 years old. The first sort of pictorial representation goes back to 40,000 years ago. We've been using images and symbols way longer than we have been using a written system. So it's somehow embedded in our DNA; we have a strong inclination towards imagery.
FD: Nowadays, people are not only writing and reading more, but also seeing more images - Instagram and Pinterest are full of photography of architecture and cities. That leads me to the next question: how do you think data visualization can help architecture and cities?
That's a great question. I think it can help in three distinct ways:
It can serve as inspiration. One of the great things that I found about visual complexity is when I started putting together the science behind it over fifteen years ago, I noticed a wonderful cross pollination of ideas. I got emails from biologists that were inspired by visual representations that they saw visualizing something that's completely different. And I even remember an e-mail from an architect that I received back then; they were using visualizations that they saw for a social network, and they were applying some of those ideas into a building. Architecture, as any creative field, can be inspired by many different areas, one of them being visualization. It could be a visualization of sound, demographics or whatever, but it can be a strong source of inspiration.
The other way is by the methodology, which can also be inspiring for architecture. Today, many fields are driven by codes. One of the great things that I follow is generative arts and people that are creating art through codes, through algorithms. They are creating interesting shapes and forms that are unpredictable and unique. They're not created by computers, but by machines. So I think that's something that architecture is trying to also understand: how form and shape can be created through code, through algorithm, through machines. I find that very interesting. Especially when when it comes to A.I., and we are talking a lot about A.I. driven design. There's an opportunity for A.I. driven architecture where we can create an element of surprise through this process.
My third point is that visualization can actually bring a lot of insights to architects and urban planners. One of the great projects that comes to mind is called Tracing the Visitor's Eye. This was built in the city of Barcelona 10 years ago or so. They gathered all the images uploaded to Flicker that were taken in the city of Barcelona. Two things were important: the time the image was taken and the location. It was these data points, and they recreated the path that people took through the city. Now imagine the richness of that visualization. All of a sudden, you have this map of where people are actually navigating and which streets they are going on. Where do they start? Where did they end their journey? For a urban planner, or even for the mayor of a city, there's a lot of really interesting knowledge that can come from this process.
The unfortunate thing is, sometimes the people that are doing the really cool, interesting visualizations are not necessarily talking to the people that can really benefit from them, like architects and urban planners. But I think there's a huge opportunity for that type of thinking to be more connected with the people that can benefit from it.