Interactive Design and Game Development MFA thesis research

For class wednesday (Theory of Interaction Design), we were tasked with dissecting an existing definition of interaction design. As I had just finished “Where the Action Is”, I used the assignment as an opportunity to start writing about it (and digesting Dourish’s ideas).

Paul Dourish defines interaction design as a new discipline at the center of various converging practices. IxD is concerned with the design of interactive artifacts with an eye towards the context(s) of their use and their expressive potential. He describes the approach as “holistic and expressive” and aimed towards the creation of artifacts that that not only communicate the values of their contexts, but can also be appropriated by users to better suit specific situations, and finally shape the environments of practice they live within. Specifically Dourish states:

Where once interfaces might once have bee created by people whose backgrounds included engineering or psychology, they are most likely now to be created by people who come from backgrounds in areas such as graphic design or product design. The new discipline that has been formed at the intersection of these disciplines is generally called “interaction design” (Winograd 1996) [1]

While he only spends a few paragraphs on interaction design explicitly, his book makes the case for an approach or stance (for use in evaluation) that serves as the basis for goals of interaction design as described. While his statements that IxD is “holistic” and “expressive” seem broad or banal, the great majority of his book explicates, in detail, the complexities and nuances of context an expression as related to the design of interactive artifacts. He does this by organizing a breadth of philosophical, psychological, sociological, and computer science theory under his concept of “embodied interaction”.

Dourish coins the term “embodied interaction” to describe how interactive artifacts exist in — and should be designed for— the world. Building on the philosophy of phenomenology and its emphasis on embedded actors, the interrelatedness of thought and action, and how meaning arises through action. His definition (and its implications) is built on a fair amount of conceptual lead up so I’ll try and grossly simplify. Dourish views embodied interaction as a continuous flow of action in response to a dynamic situation; The focus of a user’s action is also dynamic and will shift to include or ignore all kinds of layered abstractions in order to manipulate them or manipulate through them. Because of this, interactive artifacts must be flexible to situational needs and help communicate the nature of the situation at hand not just potential task based functions.

He goes on to use the idea of embodied interaction to describe the conceptual relatedness of tangible and social computing initiatives. Discussing social computing, in particular, underscores what it means to view interaction design holistically. In showing how a user’s focus moves fluidly between any number of interrelated contexts, and how the new contexts provided by interactive artifacts change the space which they are used in, Dourish shows how interactive systems are not just tools for facilitating the isolated execution of a rationally constructed, content-specific tasks; but material, or a medium, used in simultaneously reflecting on and manipulating the numerous overlapping ontologies of the situation that extends beyond the artifact.

Where traditional approaches formulate interaction in terms of two sets of capabilities —the raw functional capabilities of the software and the raw cognitive capabilities of the user — social computing introduces a new model. This model is based on alternative views of human social behavior, observing that the sequential organization of interaction does not simply result from the “execution” of a formal plan in the user’s head, but instead arises from a process of continual response to the circumstances within which it was being produced— circumstances that include not only a set of prior expectations about likely actions, but also the outcome of earlier actions and the emergence of new concerns an opportunities.[2]

Along with this interconnectedness of action and situation, he holds that “content” and the representation of that “content” are also interrelated and interchangeable. Both “parts” can be worked with or worked through.

What it takes to be a representation is to be used as a representation in the course of some activity. [3]

Later he provides a simple example of how one’s focus can transform an object from content to tool.

Sometimes I want to move the mouse, the mouse itself is the focus of my attention; sometimes I am directed instead toward the cursor that it controls on the screen; at other times, I am directed toward the button I want to push, the e-mail message I want to send; or the lunch engagement I am trying to make.

As meaning is made from acting within any number of intersecting systems, he stresses that digital artifacts should strive less towards an ontologically static description of a work process, and more towards expressing, or reflecting, the situation that the work is taking place within.

Even more straightforwardly, it’s a good idea to build systems that tell you what they’re doing.

I found this odd as it seems to embrace the highly stigmatized “implementation design”. It should be noted, though, that Dourish is implicitly discussing frequently used interfaces that live embedded in a work environment. His case examples include air traffic control and hospitals. In these instances he makes a number of points about how the materials used by workers aren’t just vessels for information, but information in of themselves about the work environment and are often used in unexpected ways as situations demand. He goes on to describe how computer systems that simultaneously simulate and hide the bureaucratic guidelines they’re based on not only loose a lot of this information but can prevent workers from appropriating the artifact in ways that facilitate actually getting shit done. In other words, he argues that interactive systems should expose their workings so that users can understand the situation as is — opposed to as understood by institutional guidelines — and, if necessary, allow users to fudge things as needed. In this sense I fully agree — how this plays out in limited use artifacts (like grocery market self checkout aisles) though is another question. (I suppose the interaction design approach would aim to transcend the checkout kiosk all together).

Jef Raskin in The Humane Interface also makes an argument for “fudgability”. Interestingly, Raskin’s proposal, while squarely in the realm of his generally pragmatic approach, struck me as inconsistent with his anti-modal ideals; If the user is to have the ability to manipulate their data in a non-sequitur way, I don’t see how they won’t be forced to violate the standard, flattened, interactive nomenclature that Raskin argues is so essential to good use. Even if “fudgability” fits within Raskin’s idea of monotony, it’s built on starkly contrasting premise.

Dourish’s definition of interaction design in Action was made in passing to service a larger commentary he wished to make on the ideal of invisibility advocated by designers such as Raskin.

[Invisibility] appears to be at odds with the influence of design…Design communicates…It reflects an attempt to make interaction engaging… Invisibility is not engaging. Invisibility does not communicate… An invisible artifact, one that does not impinge on my world at all, is not a tool I can use effectively for anything.[4]

Here, Dourish emphasizes that expressivity is an essential consideration for interaction designers and situates IxD in stark opposition to ideas like Raskin’s blindly usable interface.

Throughout his book Dourish describes how his ideas, built on long standing philosophical principles, contrasts to the thinking that initially drove HCI and interactive systems design. His concept of embodied interaction is then used to either bridge the gap between various contemporary ideas, like social and ubiquitous computing, which both take as tacit the inter-relatedness of thinking and doing, or refute ideas, like invisibility, that work under the premise of people executing on cognitive processes separated from the world. When he finally addresses interaction design, even if comparatively offhandedly, he positions it as an HCI related discipline infused with communication and holistic thinking skills; A discipline that tacitly understands at least the importance of context and communication.

(Bib page numbers are causing RSS erors (?!?), will return once I figure out what the deal is)
1 Dourish, Paul, “Where the Action Is”, (Massachusetts: The MIT Press, 2001)
2 ibid
3 ibid
4 ibid