Design Crux

Information, Captology, Desirability in Design

Information Work: What Is Context Worth?

What’s it going to take to change? I’ve developed a law of technology adoption, which I modestly call Louderback’s Law: Unless a new technology includes breakthroughs in at least two different dimensions — without adding hardship along the way — it will not supplant and older, established one.

—Jim Louderback

For information work, context should trump technology centric factors like bandwidth, once we establish just what context is.

Coined in the landmark book Beyond Culture, anthropologist Edward Hall describes a continuum of cultural communication style, with America near the low context end, Arab medium and Japanese considered a high context culture. While Hall was concerned with national cultures, this same idea can and has been applied to cross discipline communication, and even between differing corporate cultures.

Persona and scenario based development are used by programmers because they provide a high level context within which to view the low context work tasks of writing code.

Context Does Not Depend On Advancing Technology

Products exist in a vast, often–messy environment of services, brands, cultures and competitors. But successful companies are realizing that deliberately and strategically designing products for the context in which they live can result in more imaginative, better integrated, and ultimately more humane offerings.
Products and their Ecosystems: Understanding the power of context in product innovation

Human communication takes place at the intersection of semantics, interpretive cues, and cultural contexts. Increasing bandwidth is only a kludge, information technology allows the user to cope with the flood of bytes coming through. Data, the lowest contextual level of alphanumeric strings (words, numbers, symbols) and their placement in relation to each other, is completely dependent on bandwidth.

Cultural, interpretive and business context require less from technology but much more from designers. As important as context is, information workers don’t often see the context failures causing symptomatic problems.

Problems can start at the basic level of semantics. Stakeholder groups with different contexts nod in agreement, only to go off in different directions and later wonder what happened. One remedy is for members to compose a dictionary of critical words everyone must explicity agree upon. Defining such a simple word as document will set the limits and power the user sees in the resulting content management system.

Programmers will find direct parallels between context and what The Mythical Man–Month calles conceptual integrity. In essence software that never becomes an integrated whole, remaining a feature list or collection of requirements, fails to gain conceptual integrity. And far too many programs fail this basic test. But conceptual integrity is the real secret design–directed companies like Apple have over the competition.

Conceptual integrity is the result of contextual design, where products, software and services work seamlessly, and individual features blur together in the user’s perception. Consequently most products are constructions of parts, not integrated designs. And where discrete features don’t work together seamlessly, serious failures occur in human systems.

Death By Context Failure

According to management, the users of the existing system were technically sophisticated, tertiary–educated, and used the system frequently — probably daily. …When we got to spend time with a few users, however, I found that they were not tertiary–educated or technically sophisticated. They were primarily in clerical and secretarial roles. We also discovered that most of them could not actually use the existing application — instead, they called technical support whenever they needed to make any but the most basic changes.
Contextual Enquiry – A Primer By Gerry Gaffney

Consider the leading byproduct of information failure in the health care field: Lawsuits. Data compiled from various sources indicate doctor–patient communication is the leading factor in up to seventy percent of suits. Length of visit and other bandwidth–type factors had little to do with how much information was exchanged. The big high–payoff information technology? Seminars teaching the difference between talking and communicating information.

A related factor is what I call stat inflation, where urgency crowds out importance, creating far more serious information failures.

Role of Computerized Physician Order Entry Systems in Facilitating Medication Errors uncovers twenty–two system failures delivering wrong medicine or dose to patients. The symptoms are usability failures, the cause is context failure. The report, published in JAMA, cites “…fragmented CPOE displays that prevent a coherent view of patients’ medications, …separation of functions that facilitate double dosing and incompatible orders” (JAMA Vol. 293 No. 10, March 9, 2005) Some dose calculation errors are common enough to be known by the slang term “death by decimal.”

Part of what Anita has learned is how aspects of the job design and aspects of the organizational context make it pretty hard for these nurses to do their jobs effectively. …And there’s no organizational mechanism to learn from these individual failures.
When problem–solving is a problem: HBS’s Anita Tucker’s troubleshooting addresses root causes

Context failure is a loss of the reference points needed to make decisions within a system. Of creating interaction patterns mediated by technology which separate user tasks from larger objectives. Information work problems concern scope, root causes, perspective, coherency and the rules for connecting data points spit out by computers. The design crux is user attention is the new scarce system resource, not computing power. Fragmented CPOE displays scatter attention rather than focus it. Contextual inquiry techniques bridge silos, revealing the rules of attention crucial for situation awareness for any business.

Just like foreign cultures programmers (called homo logicus by Cooper) and users are separated by a common language, but different contexts of construction and use. It just doesn’t come out in stark relief until you view products, services, and interfaces from that foreign context.

DoCoMo managers were so enraptured with their state–of–the–art Internet service that they failed to notice that the long and intricate menus favored by Japanese consumers didn’t score with foreign customers who were looking for more direct and intuitive interfaces. One reason for the failure to communicate: not a single person in the senior management of the company was non–Japanese.
— Why Apple Isn’t Japanese

All you have to do is watch a busy nurse struggle to interact with a mobile monitor. Clearly designed to be a mobile unit, the wheels could not be locked, making one–handed operation impossible. The nurse had to put down the clipboard she was trying to enter data from the monitor on. With one hand she anchored the unit so enough pressure could be applied to the controls to activate them. Going back and forth between clipboard and monitor interaction continued through several cycles. Somehow the monitor’s developer never anticipated a situation when a nurse would use a clipboard with their product.

Context has been said to be worth 50 I.Q. points, but the payoff from solving the right problem in the right way is often priceless.

Related Articles:

Benefits Are Not Just What Civilians Call Features

Content Management Strategy, How To Develop The Other CMS

Information Work Tactics

Resources

Copyright ©2002–2008 John Soellner. All Rights Reserved.