TECHNOMETHODOLOGY

1
TECHNOMETHODOLOGY

• Priyanka Reddy, CS260, April 29, 2009

2
Introduction

• Sociology has increasing influence on design of
  interactive systems
• Ethnomethodology favored in CSCW and HCI
• Goal Determine how to incorporate concepts of
  ethnomethodology into system design

3
Ethnomethodology Brief Review

• Study of how social facts were achieved
• Human social action is reflexively accountable
• E vs. Ethnography

4
Ethnomethodology vs. Ethnography
5
Ethnomethodology in HCI
6
Ethnomethodology in HCI

• Prominent form of sociological analysis in
  HCI/CSCW
• Used to inform design in 2 ways
• Fieldwork investigations to get insider look at
  methods and practices of work activities
• Get an idea of the temporal organization of
  activities and interactions

7
Plans and Situated Actions

• Lucy Suchmans Plans and Situated Actions
• Plans are used as resources to guide situated
  actions
• Based largely on Garfinkels idea of judgemental
  dope
• HCI community influenced by this book, and in
  turn by ethnomethodology

8
Participatory Design

• Largely influenced by Plans and Situated
  Actions
• PD design around work situation of users
• Ethnomethodology looks at rich descriptions of
  work practices
• Analytic allies

9
Ethonography

• Ethnomethodology used to make rich descriptions
  of work
• Uses ethnography in working settings
• Ethnography has become widely used technique
• Ethnomethodology experienced fieldwork?

10
HCI in Transition

• Ethnomethodology looks like a great solution to
  incorporating work settings into HCI design
  understandings

11
Ethnomethodological Studies of Technological Work
12
Ethnomethodological Studies

• Many studies applied ethnomethodological concepts
  to experience of work with technology
• Common focus of studies is sequential
  organization of working activities
• Goal Determine relationship between disciplines
  of ethnomethodology and computer systems design
  from these studies

13
General Practices
14
Ethnomethodology in Process

• Has ethnomethodology entered into the process?
• No, except requirements capture part of process
  is done by an ethnomethodologist

15
Ethnomethodology for Critique and Design
16
Critique

• Ethnomethodology used to critique in the past
• technology, at best, often fails to support the
  work it is designed for
• lack of successderives frominsensitivity to
  the organization of work and communication in
  real work environments

17
Paradoxes Uncovered

• Paradox of System Design
• Paradox of technomethodology
• Technomethodology concerned with detail and
  moment-by-moment organization
• How can it be applied to design of new
  technologies?

18
Design, not Critique

• Goal is not
• Determine how ethnomethodology can be used to
  critique technologies
• How to apply ethnomethodology understandings to
  better understand conditions in which technology
  is developed
• Goal is
• Understand how ethnomethodological understandings
  of human social can be used in designing
  interactive technologies

19
Technomethodology
20
Technomethodology, defined

• Ethnomethodological perspectives on human social
  action concerns the foundational concepts of
  system design
• Interaction should be between the foundational
  elements of each discipline
• Goal draw foundational relationships from
  which to proceed together

21
Drawing Foundational Relationships

• Described through example of
• abstraction (from system design)
• accountability (from ethnomethodology)
• Concerned with design of systems, not a
  particular system
• Learn from ethnomethodology, not from
  ethnomethodologists or their observations

22
Abstraction, Accounts and Accountability

• Re-evaluating use of abstraction in CS using
  Ethnomethodologys concern with accountability of
  practical social action

23
Abstraction

• Allows systems to be considered at different
  levels of design
• Allows engineers to work on one part of system
  without worrying about others (API)
• Helps us manage complexity by selectively hiding
  it
• Allows for
• Systems built with complex components
• Systems with same interfaces are equivalent

24
Abstraction for UI
25
Metaphoric Interaction

• Use equivalences between 2 abstractions
• Using trashcan to throw away or delete files
• Sometimes metaphor breaks down
• Mac floppy desk in trashcan doesnt delete its
  files
• Hidden details become relevant

26
Everyday Interactions

• No abstractions in physical world
• No hidden information
• Can see what is happening and interact with it
• Organize actions around detail of production of
  action, not an abstraction of that behavior

27
Revisit Accountability

• Critical property of ethnomethodology
• Def action being organized as to be observable
  and reportable and rationalized

28
Real World vs. Computational

• ? Problem with improvised action

29
How to fix this?

• Want to make computer systems more understandable
  to users
• Cant just create 1 account for system
• Account created in every circumstance in which
  system is used
• Account is a run-time phenomenon, not design-time
  one

30
Observable-Reportable Abstraction

• Convey certain aspects of the mechanism
• Retain some abstractions to retain clarity,
  consistency and ease of use

31
Abstraction and User

• User wont understand everything
• He doesnt need to
• Understands enough to manage relationship between
  their work and systems actions

32
Open Implementation

• Approach to system design that uses idea of less
  abstraction
• Idea abstraction sometimes hides design
  decisions that are critical to effectively using
  abstraction
• Ie. memory overhead to create new window
• Key principle computational reflection

33
Computational Reflection

• Have representation of their own structure and
  behavior
• Causally connected to behavior that it describes

34
Computational Reflection goals

• Flexibility
• Allow user to reconfigure mechanism behind
  traditional computational abstraction

35
OI Strategy
36
OI Strategy

• Value allows programmers to make distinctions
  between what they want to do with system
  abstractions and how they want the system to do
  it
• Feature allows us to articulate relationship
  between
• What is done (implementation behavior)
• What is done by what is done (achievement of
  application ends)

37
Accounts and Accountability

• Goal develop approach to design of interactive
  systems
• Use OI model
• Use reflective self-representations

38
Reflective Self-Representations

• Define as accounts that systems offer of their
  own activity
• Account explanation of systems behavior
• Accountability how the explanation arises
• Concerned with accountability want account to
  arise in the course of action
• Allow users to rationalize activity of the system
  and organize their behavior around it

39
Example File Copying

• Folder abstraction hides details which could help
  understanding

40
Possible Solution A

• Offer different abstractions for each operation
• Drawbacks
• Accounts created at run-time
• Difference between 2 kinds of copying not
  important specifics of an operation is
  important

41
Possible Solution B Accounts Model

• Provide a mechanism for dynamically relating
  users actions to what is actually happening
• Can distinguish different file copying
  circumstances
• Can understand the role of the percentage-done bar

42
Accounts and Ordinary Operation

• Interface accounts offered from within the system
• Provide constant monitoring of action, not just
  recovery from failure
• By enriching info given about systems
  circumstances, we enrich resources for users
  moment-by-moment decision-making

43
Accounts and Mental Models
44
Accounts Example of Technomethodology

• Accounts was an example of Technomethodology

45
Conclusion

• Goal Determine how ethnomethodological practices
  can reconceptualize foundational elements of
  system design
• 1. Developing basic analytic posito

46
Current and Future Work
47
IMPLICATIONS FOR DESIGN
48
Introduction

• Idea Implications for design not best metric
  for evaluation of ethnographic study
• Focus is misplaced
• Misses chance to provide major insight and
  benefit for HCI research

49
History of Ethnography

• Arose within discipline of anthropology
• Marked major transition point ? started focus on
  the user
• Use participation to understand what users
  experienced through their actions
• Adopted into HCI and CSCW

50
Problems of Ethnography

• Marginalization of Theory
• Power Relations
• Technology and Practice

51
Marginalization of Theory

• Dominant view corpus of field techniques to
  collect and organize data
• Used for the requirements gathering phase of
  software development
• View as a toolkit marginalizes the theoretical
  components of ethnography
• Ethnography is not just observations its the
  relationships between those observations

52
Power Relations

• Engineering demands tend to overpower social ones
• Ethnographys natural end-point is not
  implications for design
• Ethnography not the tool for creations of new
  technology
• Analysis of any cultural or social organization
  is best explained independent of any technology

53
Assumptions to Consider

• Who is doing the design in these scenarios?
• What does the design look like?
• At what point can ethnographic contributions have
  greatest impact upon technology development and
  deployment?
• Is success or value of an ethnographic
  investigation best determined by what design
  decisions it can support?

54
Technology and Practice

• Implicit Assumptions in Implications for Design
  approach
• Ethnography is the point of mediation between
  everyday practices and technological design
• People will use technology as something that is
  used as it was designed

55
Social-Technical Gap

• Large gap between
• Technological reach in design process
• Realities of technology in practice
• HCI tries to narrow this gap
• Ethnography not the right tool
• Gap is not the fundamental problem to solve
• Ethnography refigures users
• Design active process of incorporation of
  technologies, practices and settings

56
Moments and Models

• Scenic fieldwork moments
• Models help understand social settings
  organizes the moments into a narrative

57
To be done

• Multi-sited ethnography
• A study is not limited to one site
• Understand culture within broader web of
  relationships
• Transnational flows of people, capital and
  culture
• Rather than doing independent observations,
  contribute to broader ethnographic corpus

58
Implications for Design

• Discount ethnography techniques
• Contextual Inquiry
• Cultural Probes
• Proposed as alternatives to full ethnographic
  methods
• Locate topics outside of relationship between
  subject and ethnographer

59
Conclusions

• Ethnography provides valuable insight into social
  settings
• Can support different ways of thinking about
  those settings
• Important role in interactive system design
• Shaping research strategy
• Discover constraints faced in a design exercise