Charting Past, Present, and Future in Ubiquitous Computing

1
Charting Past, Present, and Future in Ubiquitous
Computing

• by G.D. Abowd, E. D. Mynatt
• Georgia Institute of Technology

Ivo Krka September 18th 2007.
2
Overview

• Ubiquitous computing overview
• Natural interfaces
• Context-awareness
• Automated capture and access
• Everyday computing
• Evaluation of ubiquitous systems
• Conclusions

3
Ubiquitous Computing

• Ubiquitous availability
• Wherever
• Whenever
• New interaction paradigms
• Continuous interaction
• Everyday computing
• Scaling interaction with respect to time
• Informal and unstructured activities
• Wearable computers
• Social implications

4
Ubiquitous Computing

• Distributed computing
• Mobile computing
• Sensor networks
• Human-computer interactions
• Artificial intelligence

5
Vision Weiser

• World
• People
• Environment
• Computing resources
• Assisting everyday life
• New devices
• PDAs, laptops (not so new anymore)
• Mobile phones

6
Vision

• New applications
• Leveraging new devices and infrastructure
• Widespread access to information and resources
• Off-the desktop
• Scale
• Many computational devices
• Physical space
• Many people
• Time

7
(No Transcript)
8
Desirable Properties

• Natural interfaces
• Richer communication capabilities
• Context-awareness
• Adaptive behavior
• Location
• Identity
• Capture and access
• Live experience

9
Natural Interfaces

• Classical interaction not enough
• Keyboards/mouse/display
• Human forms of communication
• Speech (voice dialing)
• Handwriting
• Gestures (Nintendo Wii)
• Properties
• Learnability
• Ease of use

10
Natural interfaces

• Five questions Bellotti
• How does the system know Im addressing it?
• How do I know the system is attending to my
  actions?
• How do I make the system perform an action?
• How do I know whats the system doing?
• How do I avoid mistakes?

11
First-Class Natural Data Types

• Prerequisite for new application development
• Easy input handling
• Speech pause cues, speaker identification
• Pen-based computing pen into text
• Example (Classroom 2000)
• Written annotations linked to audio or video
• Merging pen strokes for timestamps

12
Error-Prone Interaction

• Various new mistakes
• Inability to eliminate all errors
• Less-grained sense than humans
• Error handling of recognition-based interfaces
• Error reduction
• Error discovery
• Reusable infrastructure for error correction

13
Context-Aware Computing

• First demonstration
• Location-aware appliances
• Augmenting general human activity
• Location-Based Services
• Object identification
• Vision-based recognition
• Need for context toolkits
• Design
• Implementation
• Evolution
• Reconfigurable Context-Sensitive Middleware Yau

14
Context Definition

• Time, history, other people
• Five Ws
• Who
• What
• Where
• When
• Why

15
Context Representation

• Too many ad hoc limited schemes
• Evolution will enable separation of sensing and
  acting upon context
• Some standardization efforts
• GML (Geography Markup Language)

16
Context Fusion

• Few truly-ubiquitous context aware apps
• Assembling various related information
• Similarity to sensor fusion
• Negotiation and resolution strategies
• Combining services in parallel
• Reliable ubiquitous context

17
Augmented Reality

• Real-time information
• Tour guide systems
• Car navigation systems
• Modification of real world perception
• Incorporating augmented vision and hearing

18
Automated Capture and Access

• Capturing real life experience
• Augmenting human inefficiencies
• Meeting/rooms, classrooms environments
• DEN (http//den.usc.edu)
• Connecting various aspects
• Interpreting/non-interpreting
• Postproduction phase
• Web systems access

19
Challenges in Capture

• Good acceptance
• Distant education
• Capturing informal and formal meetings
• Usually raw streams of info (non-interpreted)
• Need to derive additional information
• Capture techniques not good enough
• Expenses?

20
Challenges in Access

• Playback capabilities
• Real-time usually inefficient or inappropriate
• Synchronization of multiple streams
• Material static on access phase
• Privacy
• Focus on lecturer

21
Everyday Computing

• Scaling ubiquitous computing with respect to time
• Continuous interaction constant presence
• Need for incorporating in everyday life
• Informal and unstructured activities
• Smart home Edwards
• No clear time borders
• User requirements
• Visibility of current state
• Freedom in dialogue
• Overall simplicity in features

22
Everyday Computing

• Interruptions in interaction
• Continue at the prior state
• Sequence of steps that will resume
• Multiple concurrent activities
• Context-shifting
• Background awareness
• Time as an important issue
• Associative models of information needed
• Hierarchical models not enough
• Activities information usually associative
  (multiple views)
• Continuously changing user context

23
Directions in Everyday Computing Research

• Continuously present computer interfaces
• Wearable computers
• Information presentation to humans
• Levels of users periphery
• Connecting physical and virtual world
• Modifying traditional HCI methods
• Informal, peripheral and opportunistic behavior

24
Evaluating Ubicomp Systems

• Balance between prediction and observation
• Coevolution of human activity and new tech
• Finding a human need
• Cutting edge technologies
• Reliability and robustness issues
• Compelling story from end-user perspective
• User-centric feasibility evaluation
• Designing for a currently impossible interaction
• Xerox PARC Flatland (whiteboard)
• Audio Aura (peripheral awareness)

25
Evaluating Ubicomp Systems

• Evaluation in authentic setting
• Classroom 2000
• Large number of classes
• 60 students modified in-class note taking
• Task-centric eval techniques inappropriate
• Informal everyday situations

26
Social Issues

• Sense, understand, react and record phenomena
• Spying others
• Security, visibility, privacy issues
• Users dont know behind their back actions
• Informing users
• Stopping action

27
Conclusion

• Computers are everywhere and they are staying ?
• We are the ones to make the best use of them
• Questions?

28
References

• Weiser M. Weiser Computer for the 21st
  century. Sci. Am. 265, 3 (Sept.), 94-104, 1991.
• Belloti V. Bellotti, M. Back, W.K. Edwards,
  R.E. Grinter, A. Henderson and C. Lopes Making
  sense of sensing systems five questions for
  designers and researchers. Proceedings of the
  Conference on Human Factors in ComputingSystems
  (CHI 2002), ACM Press, Minneapolis, MN, USA, pp
  415-422, 2002.
• Yau S. S. Yau, F. Karim, Y. Wang, B. Wang, and
  S.K. S. Gupta Reconfigurable context-sensitive
  middleware for pervasive computing. IEEE
  Pervasive Computing, 1 (3), 3340, 2002.
• Edwards W. Edwards and R. Grinter At Home with
  Ubiquitous Computing Seven Challenges, Proc.
  3rd Intl Conf. Ubiquitous Computing, Lecture
  Notes in Computer Science 2201, Springer-Verlag,
  Berlin, 256272, 2001.