This is an essential truth but there’s a lot of filling in to do before being able to apply the maxim in the world, and Manuel begins with these sub-clauses: start with a question; interactivity is key (discuss); cite your source; [use] the power of narrative; do not glorify aesthetics; look for relevancy; embrace time [cf. interactivity]; aspire for knowledge; avoid gratuitous visualizations.
I have others of my own, but I’ll start by referring back to the origin of the debate which is (I can say with authority, as one of Ian Douglas’ naysayers) that, in half a decade of observing dynamic and interactive data visualisation emerge, there’s a frustration that the critique hasn’t moved beyond cataloging new typologies (of which an excellent job has been done by Manuel’s visualcomplexity.com, the infosthetics blog, and by books such as Data Flow and Else/Where.
I wrote a review of Else/Where in Eye in 2006 (you can read it here) which took up the theme:
“Else/where reveals that, whereas the abstractions we use to represent relationships, dimensions and properties in the physical world are universally understood and ingrained in our consciousness, the visual language of intangible landscapes is immature”.
And In reference to one particular diagram:
No matter how much well-researched data the Map of World Government contains within its frame, the message is unclear, the scale wrong, the detail obscured by arcane pictograms and its visual intensity diminished by over-reliance on text. This ‘map’ leaves its reader powerless. Like…quite a few of the maps of networks and conversations within Else/where the mapmaker has allowed the urge to map everything to suffocate the delivery of meaning.”
Comprehensiveness alone is not a virtue in a map.
The review continues: “A good map does more than say ‘these are all the things that are here’; it enables its reader to plot a critical path or course of action; it scales effortlessly between landscape and detail as the diagrammatic representation and the object fold directly into each other.
This ‘bad’ diagram is contrasted with a good one – an oldy but goody:
A fine example of a good map of a logical landscape is Smartmoney’s Map of the Market; this uses area to signify the size of market capitalisation, and colour (red and green) to signify sector and stock performance. The map provides an instant visual cue to the location, intensity and significance of events and scales through into detailed information on individual stocks. It tells a simple but ever changing story well and it doesn’t try to tell too much. This is a well-designed map.
(Apologies, map and diagram are being used interchangeably, here to mean ‘gathering and arraying data in visual form’).
In the discussion surrounding Manuel’s manifesto there’s been talk about a divide between information art and data visualisation, between beauty and utility and I’m hoping that’s a false dichotomy. The small sins lie in mapping for mapping’s sake (so start with a question), in hiding the wood for the trees, in using the wrong form or typology to express the data, in failing to use the fourth dimension as a drawing plane (and in using the third when you shouldn’t). A revelation can be small or big and should be treated honestly, and so the greater sin is the misrepresentation of inconsequential nothing as portentous; something big is no less profound for being hidden deeply and requiring effort to divine on the part of both producer and consumer. The diagram is a reductive medium, but because there are gaps within it you can add your own stuff into it, creating infinite possibility. That’s why they’re beautiful things, that’s why they’re satisfying.
I worried in the Else/where review about language and the interpretation of meaning in a medium in which shared understanding is undeveloped (think how hugely abstracted a physical map is, how easily we understand its conventions, and how much meaning it can impart). We don’t understand many of the new conventions of interactive visualisation, which is why there’s a critical path of our own to be trodden between, on the one hand, the makers of merely pretty pictures and, on the other, genuinely revelatory representations that we simply don’t get because we haven’t learnt the language. I can’t for the life of me comprehend in print Ben Fry’s visualisations of the human genome, for example, but that doesn’t mean that a scientist versed in both the science, symbology and iconography won’t get it too. Fry’s work – which demonstrates perfectly Manuel’s point that interactivity is key – has enabled geneticists to communicate effectively through a common visual language.
'Fry’s interactive visualizations – which allow data comparisons from multiple viewpoints – have already proved incredibly useful in helping to digest the data. These kinds of visual representations become ‘memes’ – cultural units that spread very rapidly when they click for people’, says geneticist Dr Eric Lander. ‘There’s just no substitute for visualizing data: you see patterns…that you won’t be aware of in any other way’. [quoted from Else/where]
Once the pattern becomes apparent ideas can emerge that articulate into design and inventions. Here’s an extract from a short essay I wrote with Fenella Collingridge. It’s about the use of diagrams in architectural design but there’s lots of crossover in the way that, here at Made by Many, we use diagrams to push through from research into possible solutions.
Diagrams consolidate thinking. They can give visual expression to the multiple layers of material, temporal and human traces that emerge and interact without having to give them the coordinates of a map, so the hierarchical relationships between things remain fluid and can change. Although it is (mainly) a two dimensional form diagrams can express ideas in multiple dimensions. They enable the shaping of data and sensory impressions into a system, allowing patterns to emerge and become real by showing what lies between the visible incidents, artefacts or moments that we otherwise cannot see.
The diagram [then, can be...] a transformative moment in design. It’s produced at the point at which observations are explained and so possibilities become apparent. A diagram is a visual explanation that couples observation and reason within a visual representation. The simple act of making an idea visual is an incredible thing, but when you get it right it’s so simple and obvious: it’s a moment of clarity where you’ve actually managed to connect together the different parts of an idea to produce a discovery or insight, and therein is a visualization of a truth and the essence of a design.
We don’t want any funny business, we want to be clear about the data we collect and how we use it. To find out the details click here