the logic of failure

from the book cover

We’re at the start of the 2010 CHI challenge, another “wicked problem,” and it reminds me of a book I read some years ago: The Logic of Failure: Recognizing and Avoiding Error in Complex Situations, by Dietrich Dörner. The author tells a story of a city council and its mayor trying to fix the volume of traffic, noise, and air pollution in their downtown area. To solve the problem they introduced speed bumps and the speed limit was reduced to 20 miles per hour. But the results were unexpected: 1) the speed bumps forced people to drive in a lower gear, thus increasing the noise and exhaust fumes; 2) but because of the reduced speed, people spent more time shopping, and this actually increased the number of cars in the downtown area; 3) eventually fewer and fewer went downtown because of these reasons; 4) people started shopping in a near-by mall; 5) downtown stores, once thriving, were facing bankruptcy, and 6) tax revenues were significantly down. “The fate of this environment-conscious town demonstrates how human planning and decision-making processes can go awry if we do not pay enough attention to possible side effects and long-term repercussions, if we apply corrective measures too aggressively or too timidly, of if we ignore premises we should have considered.” [p.2]

It seems we’re wired for this kind of ad hoc thinking. Hunting, building a fire, or chasing away a wild animal does not have much significance beyond the act; “…our prehistoric ancestors did not have to think beyond the situation itself. The need to see a problem embedded in the context of other problems rarely arose. For us, however, this is the rule, not the exception. Do our habits of thought measure up to the demands of thinking in systems? What errors are we prone to when we have to take side effects and long-term repercussions into account?” [p. 6]

As we build systems, we need to be aware of short-term, long-term, and unexpected consequences. Smart designers understand the motivations and needs of their target population. Who is your target population? What do you know about them? If you build a tool for them (to help them walk more or to enjoy their walks or…), what are the side effects? Might there be unintended consequences? Are you able to anticipate some of these and design for them?

How is your project avoiding the logic of failure? It may be impossible for us to avoid some of these unintended consequences, but we might mitigate against some of them through careful design.

It is far from clear whether ‘good intentions plus stupidity’ or ‘evil intentions plus intelligence’ have wrought more harm in the world. People with good intentions usually have few qualms about pursuing their goals. As a result, incompetence that would otherwise have remained harmless often becomes dangerous… Failure does not strike like a bolt from the blue; it develops gradually according to its own logic. As we watch individuals attempt to solve problems, we will see that complicated situations seem to elicit habits of thought that set failure in motion from the beginning…

— Dietrich Dörner, The Logic of Failure

design courage

Atul Gawande

I enjoy reading non-fiction, mostly around HCI issues, but occasionally in other areas. One such book was Better: A Surgeon’s Notes on Performance, by Atul Gawande. From the book jacket: “The struggle to perform well is universal: each of us faces fatigue, limited resources, and imperfect abilities in whatever we do. [Sound familiar?] But nowhere is this drive to do better more important than in medicine, where lives may be on the line with any decision.” Gawande describes three core requirements for success in medicine:

1. Diligence — attention to detail.
2. To do right — despite moral obstacles.
3. Ingenuity — arising “from deliberate, even obsessive, reflection on failure and a constant searching for new solutions.”

Medicine is a profession that involves risk and responsibility; and so does human-computer interaction design. As we consider Gawande’s core requirements for medicine, what are the parallels in hci/d?

We are half way through the semester; we are at an important turning point as we engage in the CHI 2010 International Design Competition problem: will we dig deep within ourselves to find our excellence, or will we simply do what’s necessary to complete the task?

Maya Lin

Maya Lin is a design hero. At the age of 21, while still an undergraduate at Yale she submitted her design to a competition: the 1982 Vietnam Veterans Memorial in Washington, D.C. And, of course, we all know that she was the winner. What is less known is the political battle she endured, often ugly and filled with racist innuendos. Lin understood that to do right (in Gawande’s terms) meant to defend her vision of personal and national loss. Her design allowed the memorial visitor to enter a “pain of loss,” not to purge it but to contemplate it.

Get inspired.

Listen to Atul Gawande as he speaks to the Commonwealth Club of California.

Maya Lin’s design document is quoted here:

Walking through this park-like area, the memorial appears as a rift in the earth, a long, polished, black stone wall, emerging from and receding into the earth. Approaching the memorial, the ground slopes gently downward and the low walls emerging on either side, growing out of the earth, extend and converge at a point below and ahead. Walking into this grassy site contained by the walls of the memorial we can barely make out the carved names upon the memorial’s walls. These names, seemingly infinite in number, convey the sense of overwhelming numbers, while unifying these individuals into a whole.

The memorial is composed not as an unchanging monument, but as a moving composition to be understood as we move into and out of it. The passage itself is gradual; the descent to the origin slow, but it is at the origin that the memorial is to be fully understood. At the intersection of these walls, on the right side, is carved the date of the first death. It is followed by the names of those who died in the war, in chronological order. These names continue on this wall appearing to recede into the earth at the wall’s end. The names resume on the left wall as the wall emerges from the earth, continuing back to the origin where the date of the last death is carved at the bottom of this wall. Thus the war’s beginning and end meet; the war is ‘complete,’ coming full- circle, yet broken by the earth that bounds the angle’s open side, and continued within the earth itself. As we turn to leave, we see these walls stretching into the distance, directing us to the Washington Monument, to the left, and the Lincoln Memorial, to the right, thus bringing the Vietnam Memorial into an historical context. We the living are brought to a concrete realization of these deaths.

Brought to a sharp awareness of such a loss, it is up to each individual to resolve or come to terms with this loss. For death, is in the end a personal and private matter, and the area contained with this memorial is a quiet place, meant for personal reflection and private reckoning. The black granite walls, each two hundred feet long, and ten feet below ground at their lowest point (gradually ascending toward ground level) effectively act as a sound barrier, yet are of such a height and length so as not to appear threatening or enclosing. The actual area is wide and shallow, allowing for a sense of privacy, and the sunlight from the memorial’s southern exposure along with the grassy park surrounding and within its walls, contribute to the serenity of the area. Thus this memorial is for those who have died, and for us to remember them.

The memorial’s origin is located approximately at the center of the site; its legs each extending two hundred feet towards the Washington Monument and the Lincoln Memorial. The walls, contained on one side by the earth, are ten feet below ground at their point of origin, gradually lessening in height, until they finally recede totally into the earth, at their ends. The walls are to be made of a hard, polished black granite, with the names to be carved in a simple Trojan letter. The memorial’s construction involves recontouring the area within the wall’s boundaries, so as to provide for an easily accessible descent, but as much of the site as possible should be left untouched. The area should remain as a park, for all to enjoy.

[For Maya Lin's complete submission, see: http://upload.wikimedia.org/wikipedia/commons/f/fb/MayaLinsubmission.jpg]

think obvious

Many times we get stuck with our designs because we’re trying to “think outside the box.” By this we mean “to think differently, unconventionally or from a new perspective.” [http://en.wikipedia.org/wiki/Thinking_outside_the_box]. The name apparently comes from a famous nine dot puzzle: link all nine dots using four straight lines or less, without lifting the pen.

Try it. The solution can be found here: http://en.wikipedia.org/wiki/File:Ninedots.svg. Most people fail this problem, assuming they’ve not seen it before; and when given permission to draw outside the lines, many continue to fail. What seems to help is practice on simpler dot-puzzle solution training (Weisberg, 1993, Creativity: Beyond the Myth of Genius).

Nevertheless, the dot puzzle haunts us. It’s as if anything other than out-of-the-box thinking is less creative. Even the admonishment to be “computer imaginative” suggests this kind of unconventional thinking. And for us as interactive designers we get stuck. In our quest to be imaginative we forget the problem that’s in front of us. Often the solution is to “think obvious.” So often it’s the simple and straight-forward solution—the obvious solution—that works so well.

I remember when the first graphics browser, Mosaic, was released in 1993 by a team at the University of Illinois’ National Center for Supercomputer Applications (NCSA). I was interested in this for two reasons: it made use of the then new world-wide-web in a powerful way, and it was an invention from my alma mater (and located just two blocks away from where I used to work on the campus). What was so impressive about Mosaic is that it provided simple support for text, graphics, sound, and video in a single application.

At the time I was working as the Director of Research and Development at Indiana University’s Center for Excellence in Education (CEE). My colleague, Gerry Sousa, and I were developing a concept called “the virtual textbook,” suggesting that textbooks as we know them would soon disappear and be replaced by “a hypermedia information environment that combines local and distributed course content, instruction, and powerful support utilities in a single integrated problem-centered delivery environment—an on-demand educational delivery system” (Siegel, M. A., & Sousa, G. A. (1994). Inventing the Virtual Textbook: Changing the Nature of Schooling. Educational Technology, 34(7), 49-54.) Given this description, it’s not difficult to understand why we were interested in Mosaic and web technology; we saw it as a technical solution, in part, for the virtual textbook.

Why am I telling you this story? Not because Mosaic was an obvious solution for world wide web users, and not because the browser would turn out to be an obvious solution to virtual textbook development. No, I’m telling you this story because of what happened next.

While Gerry and I were working in the CEE, one of us turned to the other (I can’t remember if it was Gerry or me who first made the suggestion) and said, “You know, with all of these ‘web sites’ coming out, maybe we should make an index and categorize them.” “No, that’s too simple an idea,” responded the other. “Anyone can do that. We’ll never get known for creating an index.”

Dumb.

In 1994, two Stanford students, David Filo and Jerry Yang, created a web site called “Jerry and David’s Guide to the World Wide Web.” The index grew into categories and subcategories. In 1995 they incorporated and called their business Yahoo!

What’s the point of the story? Gerry and I thought that creating an index of web sites was too simple, too obvious. We wanted to work on the more sophisticated idea of a virtual textbook. We rejected what was before us. We didn’t pay close attention to what people needed at the time—a simple way to categorize and access the growing number of web sites. Filo and Yang delivered the obvious.

Designing the obvious is not simply about the overall goal. It’s also about the detailed solutions to a problem. Sometimes the simple and the straightforward—done well—is the best and most elegant solution to a problem. Sometimes, of course, it helps to think outside the box. But often the solution is right in front of us and we can’t even see it.

Postscript. Fifteen years later, we see glimmers of what Gerry and I conceived for the virtual textbook: the web and browser technology is well-established; electronic whiteboards and portable computing devices—laptops, iPhones, and Kindles—are readily available. What’s still missing are the demonstrated instructional strategies to drive the virtual textbook: “The Virtual Textbook is an environment. Its goal is to move students beyond content mastery to information seeking and problem solving skills: asking appropriate and significant questions of the content, evaluating and synthesizing information from diverse sources (including sources accessed over networks), understanding the difference between (and appropriate applications of) facts and opinion, grasping multiple and diverse perspectives, and drawing insights from these perspectives within the context of one’s own knowledge base and experiences. It provides the student of the future with the power to become effective and creative learners.” An elusive goal? Maybe not. Perhaps the solution is right in front of me and I can’t see the obvious.