E3: Translating from Users to Developers

1
E3 Translating from Users to Developers
Laura Faulkner, Ph.D. Research Associate /
Principal Investigator Applied Research
Laboratories The University of Texas at Austin

• 10 August 2006
• Presentation to the Austin Software Process
  Improvement Network

2
E3 Lifecycle Tools
Ethnography
3
E3 Lifecycle Tools
Ethnography
Evaluation
4
E3 Lifecycle Tools
Ethnography
Evaluation
Empiricism
5
E3 Lifecycle Tools
Ethnography
Evaluation
Empiricism
6
E3 Lifecycle Tools
Ethnography
Pictures
Evaluation
Empiricism
7
E3 Lifecycle Tools
Ethnography
Pictures
Evaluation
Interaction
Empiricism
8
E3 Lifecycle Tools
Ethnography
Pictures
Evaluation
Interaction
Empiricism
Data
9
User Observation
Ethnography

• Process
• Participant-observer watching writing
• Individual user interviews analysis
• Products
• Ethnographic report
• User personas

10
Technical Report Site Observations, Interviews
and Analysis of the Alpha-Beta Evaluation
Office by Laura Faulkner, Research Scientist
Associate Applied Research Laboratories, The
University of Texas at Austin 15 May 2006
Executive Summary The Alpha-Beta team is an
energetic group with a bold and innovative
director. The staff is close-knit, and consists
of mentor and novice personnel who work in pairs.
Tasking is generally divided between evaluation
and analysis, with specific individuals (and
types of individuals) assigned to each of those
approaches per project, with much team
interaction and some overlap. Evaluation is
generally concerned with the overall picture of
product readiness for test, release and
marketing and analysis is concerned with more
detailed looks into products which will be the
appropriate level for any metrics we propose. At
each level there are evaluation and analysis
questions in two
11
Physical Spaces and Other Significant Notes The
physical working space in which Alpha-Beta is
housed is small and tightly utilized. Cubicles
with 2/3 to ¾ walls and no doors compose most of
the individual work spaces, with those of higher
rank or supervision having full, to-the-ceiling
walls and doors that close. Carpet wear patterns
in the hallways between cubicles and up to each
of the doorways support previous indications of a
high level of collaborative efforts. Far from
simply exchanging emails, even the most
introspective of the Alpha-Beta team makes
extensive direct contact with each other in
working project issues. Many of these contacts
are brief touch and go, asking a brief but
significant question at the doorway, then
hurrying back to implement the answer, as
occurred during interviews and interim
observation periods. When back in their own
workspaces, concentration was intense, quiet and
private. One individual sat up with straight
back, close to his monitor, typing quickly, while
another sat in her space, back to the cubicle
doorway, the side of her head resting in her
hand, and hair shielding her from interruption,
writing furiously with the other hand. Each of
them, and all others observed at some point,
demonstrated this
12
Researcher Dr. Gerald Shibusawa Originally from
Chicago, Illinois, Dr. Shibusawa, 37, received
his PhD from Purdue in engineering acoustics. He
works at the University of Washington Applied
Physics Laboratory. On his door are two 8.5 x 11
photos one of his two-month old daughter, and
one of a an experimental he built and
tested. While preferring Linux and MacOSX, he
uses Windows-based systems for interoperability
and compatability. He says, It should work the
way you think it should, not waste my time. He
has a single computer and does who gets credit
for ideas and research. He emphasizes that
ARPANET was for research before it was turned
into the Internet. Geralds Goals Gerry needs
to ensure a steady supply of funding and a
favorable rate of publication about his work and
research so that he can support his family and
maintain his job security. He also wants to
remain able to play with and develop technical
components.
13
Usability Expertise
Evaluation

• Process
• Training for developers
• One-on-one interactions with developers
• Design meeting participation
• Iterative design / prototype reviews
• Products
• Design prototypes, testable interim or end
  product

14
User Testing
Empiricism

• Process
• Optimal Path Test Method
• Repeatable test across multiple users
• Products
• Numeric data of user interactions with current
  design
• Data tied to specific design elements

15
Test Sheet (go to Excel example)
16
Summary

• Ethnography Pictures
• Communicate user characteristics, environments,
  and goals
• Evaluation Interaction
• Communicative relationship throughout design and
  implementation process
• Empiricism Data
• Demonstrate, convincingly, where user needs can
  be better fulfilled

17
Contact information

• Dr. Laura Faulkner
• Signal and Information Sciences Laboratory
• Applied Research Laboratories
• The University of Texas at Austin
• P.O. Box 8029, Austin, Texas 78758-5542
• laura_at_arlut.utexas.edu
• [phone]