Monday, October 30, 2006
Researching Activity, Designing for Activity
I've been thinking about Don Norman's articles (Activity Centered Design) a lot recently, because of his influence on a lot of non-designers who read his pieces and start thinking along these lines. That is a lot of leverage, and we can all learn and benefit from that. He picks up on the right emerging trends and gives them a good push into broader awareness. Now it's up to all of us to make the memes sing! Which gets me into some follow-up on activity.
If we are to design to activity, then what IS activity? Is it a type of task, such as setting up and completing a stock purchase? Is it an ongoing series of inter-related work activities, such as the "activity" of software development? (Coding a single program?) Who should define what the activity is that we are designing to? The designer? Or the person engaged in action?
In activity theory research the identification of activity is not straightforward - the specification of an "activity" is not activity-centered itself, as the coinage of activity-centered design suggest. The components of activity - action and operation - seem better defined, and they are observable. Activity and its context, not so much.
Activity is based on purpose and motive, what the Soviet school calls "object." Read "object" as "end result" plus "motive" and that's what activity design should be centered on.
This requires pre-design research into the activities and context of action. It is not user-centered, because there is no "thing" yet to be a "user" of. This pre-design research into activity conflicts with Don's other recent and influential essay, "Why doing user observations first is wrong." If you are doing activity-based design, based on activity theory, you have to conduct "user research" before designing to activity. I put that in quotes, because, more specifically, this is research into the unit of analysis called activity from the a person's perspective, a person that we hope to recruit as a user once done with our design.
These "levels of activity" are confusing, especially to the instrumental thinking of American industrial tradition, from which user experience is derived more than we might imagine. But they are valuable in identifying the most critical components of a user's practice and their information use cycles related to that practice.
The activity of "centered" should be based on the person engaged in work, not on a task or function we believe we are designing to. In product design, we always already have a set of resources in mind - a product concept, a labor-saving tool, a body of valuable content. These resources are typically the starting point for what WE consider "activity" to be, and our prior work with these artifacts may bias our ability to formulate the big picture, the activity and its purpose and outcomes.
If we are to design to activity, then what IS activity? Is it a type of task, such as setting up and completing a stock purchase? Is it an ongoing series of inter-related work activities, such as the "activity" of software development? (Coding a single program?) Who should define what the activity is that we are designing to? The designer? Or the person engaged in action?
In activity theory research the identification of activity is not straightforward - the specification of an "activity" is not activity-centered itself, as the coinage of activity-centered design suggest. The components of activity - action and operation - seem better defined, and they are observable. Activity and its context, not so much.
Activity is based on purpose and motive, what the Soviet school calls "object." Read "object" as "end result" plus "motive" and that's what activity design should be centered on.
This requires pre-design research into the activities and context of action. It is not user-centered, because there is no "thing" yet to be a "user" of. This pre-design research into activity conflicts with Don's other recent and influential essay, "Why doing user observations first is wrong." If you are doing activity-based design, based on activity theory, you have to conduct "user research" before designing to activity. I put that in quotes, because, more specifically, this is research into the unit of analysis called activity from the a person's perspective, a person that we hope to recruit as a user once done with our design.
These "levels of activity" are confusing, especially to the instrumental thinking of American industrial tradition, from which user experience is derived more than we might imagine. But they are valuable in identifying the most critical components of a user's practice and their information use cycles related to that practice.
The activity of "centered" should be based on the person engaged in work, not on a task or function we believe we are designing to. In product design, we always already have a set of resources in mind - a product concept, a labor-saving tool, a body of valuable content. These resources are typically the starting point for what WE consider "activity" to be, and our prior work with these artifacts may bias our ability to formulate the big picture, the activity and its purpose and outcomes.
