Work Contexts in Design

Context: Interplay between design of computer systems and applications within
social and organizational settings.

Work contexts affect design of interfaces
Application design must consider the organizational setting in which it is used.
e.g.
Past - stand-alone computers exchanged printed documents created with any word processor.
Present: - Now documents are shared electronically requiring standards

Past - engineers and scientists used computers - text-based interfaces
Present - people with a wide variety of skills use computers - wide range of interfaces
 

Interface designs have shifted from generic windows to domain-specific application interfaces
Developers must combine interface design skills with knowledge of specific work domains.

Two Case Studies

1. Office application with ten thousand copies distributed throughout corporation (W.S. Orlikowski)

• Whether and how the use of collaborative tools change the nature of work and pattern of social interactions in the office?
• What are the intended and unintended consequences?

• Peoples' mental models about technology and their work.
• Structural properties of the organization (e.g. policies, norms, and reward systems)

• When people do not understand the point of groupware, they will interpret the system as more familiar, stand-alone software such as a spreadsheet or wordprocessor.
• If corporate culture is antithesis of groupware (e.g. competitive, hierarchical) the technology will not facilitate collective use.

• Consultants - do most of daily work

• Partners responsible for overall management of firm

• Technical team of technical support engineers

• Partners wanted more cooperation among consultants but consultants needed to prove individual worth
• Sharing information over network threatened consultants positions and was in conflict with reward system.

• CIO marketed Notes within organization to high-level managers
• E-mail used enthusiastically, "sharing" feature not.
• Technologists too busy deploying notes to educate populace
• Little education about Notes - bread misunderstanding and skepticism
• Education did not emphasize "collaborative nature"
• Users saw Notes as personal productivity tool
• Corporate structure did not change to support Notes:

• Reward systems based on "billing"
• No new procedures on data access or confidentiality
• Questions about security
• Firms culture of competition not changed

2. Sales support software

• System was never used - complaints about awkward interface
• System was fixed at expense of millions of dollars.
• Still not used
• More interviews showed the software is based on the following model of sales process:

- Customers begin with how much they have to spend!
- Sales force tries to find adequate system for that amount

Result:
- Program did not support backward reasoning
- New interface meant nothing

Shifts in focus:
    - From product to process in research and design
        Fix the process, the product takes of itself.
    - From individuals to groups
        Personal computer islands are disappearing
    - From laboratory to work place
    - From novices to experts
        Fields and users mature
    - From analysis to design
    - More design skills
    - From user-centered to user-involved design
    - From requirements specifications to prototyping and iterative design

Maximize user involvement: Participatory approaches to design

• Intuition - risky
• Consult organization's standard practices or policies manual - unreliable
• Hire domain expert - relies on one person
• Bring users into the team

• Develop applications that enhance workplace skills through increased flexibility
• Focus on interplay between technology and organization of work
• Developers must be immersed in process and culture of workplace
• Users must be educated in technological possibilities
• Must make process interesting and engaging
• Approach create scenarios of future system use - users draw on knowledge of their work practices to see if things are ok

• Quantitative analysis is impossible but can quickly move beyond bad prototypes

• Enid Mumford developed sociotechnical design in England in 1970's
• Gave clerks and shop floor workers a greater voice in reorganization with practices as part of introducing new systems
• Focuses on facilitated group meetings involving developers, managers, and future users
• Like Scandinavian approach, but must consider business mission.

• Used in Europe
• From general systems theory and cybernetics
• Looks at all aspects of system
• Each port a significant whole unto itself and being a component of larger systems that have their "emergent properties" not predictable from knowledge of parts (eg body)
• Participatory at all levels but a simple theoretician pulls everything together

• Add more structure to participatory design

• Research ethnographers try to understand a workplace through immersion and extended contact
• Anthropologists carry out much of the field research
• Given enough time and money this research can provide most detailed understanding of a site

• Method falls between observation and interview
• Core issue: intensely interview people while they are working
• Interrupts user to ask questions
• Goals: support, extend, and positively transform the work of individuals, teams, and businesses
• Contextual interviews are steeped in technology and focus interviewing on understanding work practices.
• Emphasizes the difference between the work that people want to accomplish and work required to use the tool
• Three principles guide the contextual inquiry process:
• Preserving user's work context to obtain a more accurate and concrete understanding
• Creating a partnership with users to reach a shared understanding of the work processes begin examined
• Attending carefully to the focus that the interviewer adopts

• Results of interviews and analysis give five kinds of work models:
• Context models: organizational culture, policies, and procedures that affect work
• Physical models: physical environment as it impacts work
• Flow models: major roles and communication paths at work
• Sequence models: temporal sequences of important activities within a setting
• Abstract flow models: composites used to discuss possible changes in work practices

• IS and HCI communities developed differently and remain apart
• HCI and IS use words in different ways reflecting different priorities
• In the main think about what communities the authors came from IS, HCI,…
• Networks forcing IS and HCI to converge