Design / Redesign

Its good to see activity-oriented design showing up in discussions, often as a contrast to user-centered or human-centered design. I'm not sure we should treat this as a school of design. It is a context for design, but not a "school" that displaces the decades-old and hard-won position of user-centered design. There are reasons we should encourage both, and many, approaches.

First of all, as a design researcher you find many different contexts for design that call for variations of method, approach, and analysis. A website redesign does not require cognitive work analysis. But to understand the complexity of multiple contexts of use, as in intellectual work in large organizations, theories of activity and distributed cognitive work help us understand the information relationships and structural constraints that trigger tasks. We can design for those tasks in the context of activity. So activity-centered design makes sense in situations where we can really understand the nature of practice.

This is not always as simple as it seems. One of the key issues in Activity Theory (that you hear from the Scandinavians) is that we cannot always know the correct level of activity and context upon which to focus. It may take some time to learn the domain before we know where to locate "activity" vs. context of activity that enables it. And then do we design for activity, or for actions that comprise that activity?

Secondly, designing for objects makes sense in a lot of applications, and is a completely different context of design. And there are information structures where we should design to the object and not activity. For example, searching - search is part of somebody's contextual (and usually unknowable ) activity, but is an action level function - a targeted locating of an information object.

When browsing publications - the activity is the object. IN design terms, it is a type of search of a problem space where we define features for people to find information objects.

In a similar way, think of corporate websites, which are more of an information architecture problem than an activity oriented design. Some firms honestly do have real user activities - such as looking up your account or paying bills at a telecom site. But for the "corporate presence" type of site, who are the "using" customers and what do they want? Do we privilege certain activities at the expense of others? Or, instead, are the information structures themselves the target of "off-screen" activity interests. It does no disservice to say that many firms are right to provide accurate, usable information about their services. Sometimes the only activity on a site is providing a way to get in touch.

Finally, who’s to say that good-old human-centered design, good human factors work, does not include the analysis of activity and the appropriate selection of the design space? I have always considered human-centered a larger context than user-centered, which may encompass activity or object-centered as well. If we do a good job of understanding the functions of the human in a socio-technical system, we will account for activity, task, information structures, and information objects.