Thursday, January 31, 2008

Bob Horn at VizThink '08

Nancy Duarte and Cliff Atkinson interviewed Bob Horn about his work during the first general session of the conference.

Nancy described his book, Visual Language: Global Communication for the 21st Century as "an encyclopedia of visual language and visual thinking" and said it was a big influence on her. (See my earlier blog post on the recent republication of this book.) Bob claims that visual language is a new language being born, an international auxiliary language. He defines it as the "tight integration of words and visual elements". He uses a familiar and simple example of this tight integration, the "one way" sign:

Neither the words alone, nor the arrow alone, convey the meaning. You need both.

Much of his work lately has been in the form of large murals, which can convey both an overview, or big picture view, and details. He says these can help us "think bigger thoughts" about the complex problems we face, because they provide a way of "keeping the entire strategic context in front of us". He thinks that "large murals should invite people to make their own patterns, associations, and objections." Ideally, they provide information which allows us to form thoughtful opinions, and make reasoned decisions, considering the multitude of factors and perspectives involved.

Here is the leftmost one-quarter of a large timeline mural on management of nuclear waste in England:

Bob pointed out that murals can convey emotion as well as information.

I attended Bob's breakout session. Here is the session description. My mind map notes from the breakout session are at

Wednesday, January 30, 2008

VizThink08 conference

I just got home from the VizThink conference in San Francisco. It was an inspirational and educational 4 days: 2 days in a pre-conference infographics workshop with former Newsweek graphics director Karl Gude, and 2 days of conference sessions with other greats in the world of visual thinking and visual communication. I'll be posting more pictures and notes from it soon.

Tuesday, January 22, 2008

Visual deception 2: 3 colors appear as 2

The second color deception exercise in the Albers/Nelson curriculum is to make 3 colors appear as 2. In this case, there are two ground colors, and the figure color is chosen such that on each ground, it appears to be the same color as the opposite ground. That's pretty abstract - here's what it looks like:

And here's the proof that the two figure colors are the same, and different from the two grounds:

Again, the grays were taken from my array in this post.

For this deception to work, the ground colors need to be closely related, and the figure color must be right in between them.

Monday, January 21, 2008

What's wrong with these arrays?

In week 2 of color class, Dick gave each student a set of color chips and asked them to sort them into an array, discarding any that were not part of the "family". In the two arrays above, there is one imposter in each family. Can you spot them?

Answer: In the top row, the leftmost rectangle is not part of the family. If you think in terms of the pigment primaries (cyan, magenta, and yellow), it has yellow in it, which none of the others in that row do. In the bottom row, the imposter is the second from the left. It leans toward magenta more than any of the others.

Here are the arrays with the imposters removed:

Once you have an array of related hues, you'll notice the fluting I pointed out in a previous post. If you look at an edge of one of the child colors, you'll notice it seems to glow with the color of the rectangle on the opposite side of it. This is a strong indication that you've identified family members.

Something interesting happens with certain color choices for parents. Would you have predicted that this gray was the child of these pink and green parents?

How about in these contexts?

Look for the magenta and green glows along the edges of the children.

What's happening here? Magenta and green are opposites on the color wheel, or complements. When you mix complements, you get gray.

Here is another complementary mix. The child seems to favor the yellow parent a little, rather than being neutral gray, indicating that I didn't quite choose complements for parents. But the glowing effect along the edges, called "halation", is quite nice.

Sunday, January 13, 2008

Visual deception 1: 3 colors appear as 4

This is the first of the "visual deceptions" we are studying in Dick Nelson's "Interaction of color" class. Many of the principles are easiest to recognize and learn from in shades of gray (values), before trying them in color.

I chose my values from the array I created for the last post.

Here is a proof that they are the same:

This one is quite interesting. If you look at the horizontal bar, doesn't it look as if it changes color right at the boundary between light and dark?

But Dick says, "If you can push it, why not SHOVE it?" Meaning, really emphasize the effect. To do that, you have to recognize what causes the illusion, and what variables you can manipulate to heighten it. In this case, making the ground values as different as possible from each other is what does the trick. This is what he terms "exploitation".

Here, if asked to order these values from lightest to darkest, I would say white, the right gray strip, the left gray strip, then black. But in reality, both gray strips are the same middle gray.

What are the principles involved in this illusion? The figure color must be related to both ground colors, in other words, it is a child of the two parents. The ground colors should differ from each other in both hue and value. (Here, they differ only in value. I will explore this illusion in hue, or color, in a later post.)

So what's the point? The point is, a color looks different depending on its background. So don't use one color on two different backgrounds and expect the viewer to "read" it as the same - they won't.

Eye training: discriminating value

Dick Nelson has started a new round of his 10-week color class, "Interaction of Color", based on his training with Josef Albers. The students are Maui artists in a variety of media. I volunteered to be his assistant, so I "drive" the computer during demonstrations, and he has me review and comment on his assignment handouts to make sure they're as clear as possible.

In the first class, one of the "discovery" exercises was to sort a set of black, white, and gray strips into an array of equal steps. (See this post on arrays.) The idea is for the value change between pairs to be even, as the distance between stair steps is even.

We are using Adobe Illustrator for this class. It has a function called "blend" which will create intermediate colors between two objects. This can be a smooth gradient blend, or discrete steps, which is what we are using. Illustrator appears to use a linear interpolation to create its blends. We don't perceive the resulting steps as being visually equal. (We'll go into the technical explanation of that later in the course.) He wanted to make a point of this, so we will learn to trust our own eyes, and make our own judgments, not just trust the computer results "blindly".

The arrays at the top of this post illustrate the difference. The top array was produced by the blend function in Illustrator. I built the values in the second one by repeatedly finding the middle value between two extremes. The differences are most pronounced at either end. Looking at the top row, at the light end, the jump from white to the first gray produced by Illustrator is too great. We can imagine a large number of intermediate grays. Then the difference from the second to third does not "feel" as large as from the first to the second. Similarly, at the dark end of the Illustrator row, the final gray is almost black, so that feels like a small step, while the step leading up to it feels larger. In my array, on the bottom, the steps from one value to the next seem more even, so if you were walking up them, you wouldn't stumble.

If you click on the image to see the large version, you may notice an interesting phenomenon when you look at individual gray strips. They appear to have a fluted appearance, as if each rectangle had a gradient, from left to right, of darker to lighter gray. But each is a single, solid shade of gray, which you can prove to yourself if you block off its neighbors. This phenomenon has to do with how we perceive. What we see depends on the context.

The whole point of the class is to increase our awareness of, and sensitivity to, the facts of our visual perception, so we can use that knowledge to create the effects we want.

Thursday, January 3, 2008

The meeting game

What game are we playing? Maybe we need some ground rules.

This evening, I was reflecting on a meeting I participated in recently. It was reasonably productive, as meetings go, but I perceived a lack of focus, and thought we could have accomplished more if we'd had some ground rules. I started thinking about ground rules, and how to introduce them. Why does that notion seem so foreign? It's as if we assume that just because we all speak English, we assume we know how to hold a meeting and get something done. Individually, we each know how to get something done. But our instinctive or habitual approach may not work for everyone in our group.

What if we all decided to play our favorite game at the same time? That's how I felt at this meeting:

One person mentions a topic or raises a question: A basketball is in play. Someone answers, kicking the ball. Oops, are we playing soccer? Someone else lobs in a football. Next, a tennis ball. Another, a baseball. Now - what's that - it's a whiffleball! Oh, now there's a shuttlecock in the air! Wait, a bocci ball! A bowling ball! Juggling pins! Oranges! Apples! Knives! Flaming batons! Wow, this is exciting! Oh - were we supposed to be having a meeting?!?!? Hmmm....