Text

Dataphilia: Society x Numbers

image

Welcome to the first in a two-part essay examining the role of quantitative information and data visualization in modern society. This first part offers a critical investigation of the uniquely politicized and privileged role that quantitative information plays in our everyday lives. The second part will discuss what all of that means for those in the community of information visualization.

image

In February 2013, famed U2 frontman and activist Bono stood in front of an audience at TED and delivered a most curious manifesto:

It’s fair to say that I am, by now, sexually aroused by the collating of data.”

Quirky one-liners aside: Bono’s TED talk was essentially about discussing exciting new statistics regarding world poverty. Bono promised to “forget the rock opera, the usual bombast”, instead embracing his “inner nerd” as an “evidence-based activist: a factivist”.

Read More

Tags: essays
Text

The Trouble with Chernoff

Above: From Howard Wainer (1979).

In 1973, applied statistician Herman Chernoff proposed one of the most strange and ingenious ideas in the history of information visualization – symbolizing data using faces.

In the intervening years, the so called ‘Chernoff Face’ has become one of the most curious artifacts in the world of information visualization. Despite the best effort of academics to find a way to make the faces ‘work’, the usual response to Chernoff faces from the cartographic community is a combination of fascination and loathing.

Let’s back up though, and look at why on earth you would want to plaster a statistical map with cartoon faces.

Read More

Text

Isometry and the Local Oblique

I like to call them Lost Cartographic Arts. Techniques, tools, and styles of mapmaking that, for whatever their various reasons, aren’t commonly seen anymore.

The lost cartographic art I’ll talk about today are isometric perspective and local oblique maps.

There may be a more proper term than ‘local oblique map’ to describe what I’m referring to: but the concept is simple. It’s a map with a non-planimetric perspective with a very limited (often city- or neighborhood-level) geographic extent. 

image

image

ABOVE: ‘Campus Map’ - Unkown Author, Unknown Date.

BELOW: ‘Seattle’s Alaska-Yukon-Pacific Exhibition’ - William Caughey, 1909.

There’s a lot of useful things that happen when you have a map with an oblique perspective. Most of all, viewers get to see the shape and height of buildings and other landmarks. With a top-down view, all viewers can see is the color and shape of the roof. Unless you are in a hang glider, this is not information you can use to  identify a structure or navigate through an area. Obliques are great for navigating through a crowded place like a city, or just for getting the general feel for the layout or organization of a small area.

Web map services have recently realized this, and are competing to find ways to best recapture the utility of oblique views. Bing has ‘birds-eye’ aerial photographs that the user can toggle on; Google Maps has started commissioning a massive collection of 3D models of buildings that users can see; and, of course, Google also pioneered street-view capability which is indispensable for local-scale navigation.

If oblique views are so handy, why did they go away in the first place? The answer, I think, is honestly laziness.

Okay, maybe laziness isn’t quite the right word. The more eloquent explanation is that digital mapping technology simplified the production process of maps in general, but wasn’t cut out for the production of oblique perspective maps. From its inception GIS software uses a euclidean data structure; everything has an x and a y, and the z (height) just sorta gets flattened out. There are elevation datasets, but these tend to be for the ground only (no buildings), and the technology to 3D render a map (e.g. ArcScene) is a much more recent advancement (and one that is still rather primitive in the grand scheme of 3D modeling technology).

So, making an oblique-perspective map is one that requires a lot of manual labor, which is the sort of investment that most map producers aren’t willing to put in. So the whole art form unceremoniously faded into disuse.

image

image

ABOVE: Screenshot of ‘Simcity 2000’ - Maxis, 1993

BELOW: ‘Pixeldam’ - various artists, various years

Meanwhile, outside of cartographic realms, one form of oblique local was becoming very commonplace: isometric maps. If you want to get technical, isometric projection is a means of projecting 3D space into a two dimensional image wherein width, depth, and height are all at the same scale, and the three axes are spread 120 degrees from each other. The more understandable way to describe it is “it’s that angle that SimCity 2000 uses”. It’s one of those ‘you know it when you see it’ sorta things.

Isometry has been around for a long time- for over 500 years in Chinese art and  over a century in western technical drawing. Where isometric perspective really entered the public consciousness, though, is in early computer graphics. It’s relatively easy to render using both vector graphics or good ol’ fashioned pixels. Early entertainment software in particular really took to isometric perspective, because it allowed the illusion of 3D space while being rendered using 2D tiles. Owing to this, isometry is one of the most common forms of representation in the pixel art community. (fun fact: before I was a cartographer or an illustrator, I was a pixel artist!)

There’s some curious wrinkles to the isometric view: first off, the height of objects obscures what’s behind them, potentially obscuring them entirely. There’s also some famously wacky illusions that happen when all 3 axes are all on the same scale: this is how we get things such as the Penrose stairs or Escher’s waterfall illusion.

Nevertheless, isometric perspective is a pretty attractive way to render a local-scale oblique; one that once again remains uncommon in Cartography largely because it’s less conventional and requires a bit more effort and knowledge to produce. Fortunately, “unconventional and labor-intensive” is this blog’s unspoken credo, so check back two weeks from now when I try my hand at an isometric map.

But first, next Sunday, it will be time for another map review. I’ll be tackling one from the world of Fine Art, so things might get a bit heady. It’ll be fun though! Tune in!

Text

The Cartographic Aesthetic

Here’s a statement for you to evaluate.

As it stands now, maps are made to look strongly alike.

Even if you have no map design training, you’re familiar with ‘cartographic conventions’; those little things that give maps a cohesive visual language. North is up, water bodies are blue, text curves around rivers, et cetera et cetera. When I say maps look alike, I am not actually referring to these sorts of things, although they do play a role in making maps a visually consistent medium. Rather, I’m talking to the most structural, fundamental aspects forming the look of modern-day maps. For example, think of how well the maps you look at everyday conform to these descriptions:

  • The view is generally planimetric (i.e. ‘top down’), rather than oblique.
  • Linework is thin and precise-looking.
  • Aside from the inherent distortion of map projection, geographic distortion is minimized as much as possible.
  • The typography has a legibility-focused, non-decorative personality.
  • It is produced digitally, and opts for vector data formats when possible.
  • Map symbols are abstracted to their most geometric representation feasible. Cities become dots, points of interest become stars, airports become tiny airplane silhouettes, etc.
  • Visual parsimony: the elimination or minimization of graphic elements that serve purely as flourish. The only things on the map are landscape features, quantitative data, and/or functional elements such as scale bars and gridlines.
  • A color palette that is ‘even’: not too bold, contrasting, or saturated.

I’m sure there are more examples I haven’t touched upon, but essentially, these are some the universally accepted, yet unspoken rules that dictate how a map ‘should’ look. In other words, they are the components of a certain aesthetic now ubiquitous in map creation – the Cartographic aesthetic.

Though maps have almost always exhibited some of these qualities, to some degree, it’s obvious that the general look of maps have evolved over time. Colonial-era maps were drafted in pencil and abound with ornate graphical trappings. These gradually gave way to the maps of today, which have the clinical, technical, minimalist appearance described above.

It’s obviously impossible to capture a truly representative set of examples on how map aesthetics have changed over time, but here’s my best attempt.

TOP: Adriaan Reland - ‘Het Westerdeel van het Eyland Groot Java, 1718’ (1718). This Dutch map of Java has many illustrative elements, including very pictorial map symbols. Yep, I’m counting those drawings of elephants and stone workers as map symbols- they’re a visual icon representing geographic features at that point.

MIDDLE: Author(s) unknown - ‘Great Britain. Her natural and industrial resources’ (c. 1940). This map contains almost identical information to the map above: cities, economic activities, and land cover. What’s different is fewer graphic embellishments, a more drafted appearance, more legible text, and a more formal symbol scheme based more on geometric shapes, rather than pictorial symbols.

BOTTOM: Philippe Rekacewicz - ‘Utopian Africa’ (2011). Once again, we have a map showing natural resources. This time, the map is completely devoid of illustrative elements, and the symbols are wholly abstracted into geometric shapes. The most distinctive aesthetic choices for maps in the digital age.

There are two explanations you could offer for why these aesthetic features rose to prominence. The first is utilitarian. Planimetric, undistorted views allow for the easy and sensible measurement of distances between map features. Non-decorative typefaces are easier to read. Digital production makes mapmaking and data sharing much easier, and the Cartesian concept of vectors are highly appropriate for representing real-world space. The other explanation of the cartographic aesthetic is rhetoric. Several of the features described above, such as inoffensive palettes and geometric symbols, do not serve to make the map any more accurate in terms of portraying geographic reality, or serve any other readily apparent functional purpose. What they do, however, is to make the map look objective and authoritarian. With their tidy, technical appearance, maps have an aura of precision, of professionalism, of intelligence, of trustworthiness. Mapmakers have long known these qualities aren’t totally true, of course. All maps are inherently inaccurate abstractions of reality; all maps contain uncertain and/or incomplete information; all maps reflect some sort of bias from their author. But by employing the cartographic aesthetic, these subtle and unfortunate implications aren’t communicated to the reader. The mapmaker does everything in their power to make their map appear as geographic gospel.

TOP: Jean Klare & Louise van Swaaij - ‘World of Experience’ (2000). This ‘map’ so perfectly co-opts the aesthetic and visual language of cartography, it’s almost easy not to notice that it isn’t portraying any sort of geographic reality, but rather a subjectively organized set of concepts meaningful to the human condition. Non-cartographic subject matter, done in a cartographic style.

BOTTOM: Paula Scher - ‘The World’ (1998). Do you trust this map? This acrylic painting contains all the same information as a more ‘legitimate’ world map: the only difference is in aesthetics. Its most standout features are the medium used, its dense visual hierarchy, and its ‘distorted’ text and geography. Cartographic subject matter, done in a non-cartographic style.

This isn’t to say that the cartographic aesthetic is inherently malicious, or that it should be abandoned. Imperfect as they may be, a properly produced map is the culmination of many sophisticated and reliable components. Our topography is produced by space-age satellite measurements; roads and other features are meticulously geocoded; socioeconomic data comes from thorough, well-funded censuses- just to cite a few examples. If a map is built upon adequately current, functional, and trustworthy information, then it just makes sense to design the map look authoritative, because it is authoritative.

There is a curious side effect to having all maps adopt this same formal aesthetic, though. The cartographic aesthetic is so effective at appearing ‘objective’ because it rejects the inclusion of any appearance of human touch. The human touch is imperfect, you see, and imperfections damage the credibility of a map. A human hand can’t draw a perfectly straight line. A human hand can’t reproduce the highly precise forms of a typeface. A human hand can’t create a perfectly matte field of color. A professional-looking map, ergo, shouldn’t look like it had an actual designer, but rather that it somehow sorta sprang straight from the ether. This creates a unique situation for map designers, relative to their other peers in the commercial arts. If you are a painter, or an illustrator, or a graphic designer, you’ll generally want your work to be distinctive and have a personality. Offering something that no one else can is how you elevate your craft, and it’s how you distinguish yourself as a creative (and therefore earn a paycheck). Cartographers, however, are essentially encouraged to make their work as personality-neutral as possible. Or, at the very least, they may explore their own artistic voice only within the confines created by the cartographic aesthetic and the nature of their subject matter. One cartographer may like sans-serif fonts and drop shadows, while another may always work in darker backgrounds and rounded rectangles. Are these tiny stylistic choices really enough to distinguish the works of these two cartographers, the same way we can instantly tell the difference between, say, Picasso and Dali? Probably not. So then, is it a worthwhile goal, in and of itself, to create a field where each map designer’s works are highly distinct from each other? Where with enough familiarity, you can identify who made a map from a mile away? I honestly don’t have an argument one way or the other. But what I am pretty sure of, and I hope I’ve made the case of with this whole writing, is: mapmaking has rules, and if you want to create a truly distinctive map, you can’t be afraid to break a few of them.

Tune in two weeks from now, when I make a map that goes ahead and breaks a few rules. Next Sunday, though, we’re going to have the first of MapHugger’s in depth map critiques. You’ll get to learn what my favorite map of all time is, and why I like it so much! Stay tuned.