Improving Intelligibility and Control in Ubicomp Environments
•Download as PPTX, PDF•
1 like•676 views
The document discusses improving intelligibility and control in ubiquitous computing environments. It provides examples where systems trained inhabitants instead of the other way around, causing reluctance to correct systems and "fights" over control. The authors propose making invisible computers more visible by answering "why" and "why not" questions from users through graphical behavior notation, cancel commands, and ambient projection systems to provide real-time feedback and increase intelligibility and user control.
Report
Share
Recommended for you
Recommended for you
Recommended for you
Recommended for you
Recommended for you
Recommended for you
Recommended for you
Recommended for you
Recommended for you
Recommended for you
More Related Content
Similar to Improving Intelligibility and Control in Ubicomp Environments
Similar to Improving Intelligibility and Control in Ubicomp Environments (20)
Recently uploaded
Recently uploaded (20)
Improving Intelligibility and Control in Ubicomp Environments
- 1. Improving Intelligibility and Control in Ubicomp EnvironmentsJo Vermeulen, Kris Luyten and Karin Coninxfirstname.lastname@uhasselt.beHasselt University – tUL – IBBTExpertise Centre for Digital Media
- 2. in⋅tel⋅li⋅gi⋅bil⋅i⋅ty:the quality or condition of being intelligible; capability of being understoodcon⋅trol:to exercise restraint or direction over; dominate; command
- 4. Why are intelligibility and control important?HAL 9000 from “2001: A Space Odyssey” (1968)
- 5. Example: smart lighting applicationSee also: http://ailab.wsu.edu/mavhome/
- 6. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 7. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 8. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 9. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 10. This observation is not new V. Bellotti and W. K. Edwards. Intelligibility and accountability: human considerations in context-aware systems. Hum.-Comput. Interact., 16(2):193–212, 2001.
- 11. W. K. Edwards and R. E. Grinter. At home with ubiquitous computing: Seven challenges. In Proc. UbiComp ’01, pages 256–272. Springer-Verlag, 2001
- 12. K. Rehman, F. Stajano, and G. Coulouris. Interfacing with the invisible computer. In Proc. NordiCHI ’02, pp. 213–216. ACM, 2002.
- 13. T. Erickson. Some problems with the notion of context-aware computing. Commun. ACM, 45(2):102–104, 2002
- 14. L. Barkhuus and A. K. Dey. Is context-aware computing taking control away from the user? Three levels of interactivity examined. In Proc. Ubicomp ’03, pp. 149–156. Springer, 2003.…
- 15. Lack of intelligibility and control leads to …?
- 16. Frustrated users
- 19. UnderstandingWhat actually happens here?
- 21. UnderstandingMost common: Why & Why not
- 22. Relatedwork[Ko et al., CHI’04][Ko et al., CHI’09][Myers et al., CHI’04][Lim et al., CHI’09][Lim et al., Ubicomp’09]
- 23. No implementationforUbicompyet[Ko et al., CHI’04][Ko et al., CHI’09][Myers et al., CHI’04][Lim et al., CHI’09][Lim et al., Ubicomp’09]
- 28. Whynot: intelligibility and controlDo
- 30. Our solutionMaking the invisible computer visible
- 31. Our solutionMaking the invisible computer visible
- 32. Our solutionIntelligibility: Graphical Behavior Notation
- 33. Our solutionControl: cancel command
- 34. Our solutionAmbient projection system
- 35. Our solutionAmbient projection systemprojector withwide-angle lenssteerable projectors
- 36. Relatedwork[Rehman et al., Ubicomp’05][Dey et al., CHI’03]
- 38. Our solutionVisible Computer: user study
- 39. AcknowledgementsSketches: Daniel TeunkensCo-author: GeertVanderhulstDesign:Karel RobertCo-author: Jonathan SlendersMovie: MiekeHaesenPhotography:stock.xchng
- 40. ConclusionsJo Vermeulen, GeertVanderhulst, Kris Luyten, and Karin Coninx. Answering Why and Why Not Questions in Ubiquitous Computing. In Ubicomp ‘09 Supplemental Proceedings (Poster), pp. 210-213.
- 41. Jo Vermeulen, Jonathan Slenders, Kris Luyten, and Karin Coninx. I Bet You Look Good on the Wall: Making the Invisible Computer Visible. To appear in Proc. of AmI '09, Springer LNCS, 10 pages.Undohttp://www.jozilla.net/http://www.edm.uhasselt.be/
- 42. Backup slides
- 43. Example: smart lighting applicationSee also: http://ailab.wsu.edu/mavhome/
- 44. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 45. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 46. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 47. Example: MavHome“With inhabitant three, we noticed a new phenomenon in the course of our experimentation — the system did more training of the inhabitant than the inhabitant did to the system. There seemed to be a reluctance to give prompt feedback on the inhabitant end. On interview, the inhabitant said that they were learning to live in the dark because it was too bothersome to correct the system. This is probably human nature. We also observed a few fights between the system and the inhabitant over control that ultimately was won by the inhabitant when feedback caused the system to change behavior, but for a short duration the system caused some duress to the inhabitant—not a desired effect.”[Youngblood et al., PERCOM’05]
- 48. This observation is not new V. Bellotti and W. K. Edwards. Intelligibility and accountability: human considerations in context-aware systems. Hum.-Comput. Interact., 16(2):193–212, 2001.
- 49. W. K. Edwards and R. E. Grinter. At home with ubiquitous computing: Seven challenges. In Proc. UbiComp ’01, pages 256–272. Springer-Verlag, 2001
- 50. K. Rehman, F. Stajano, and G. Coulouris. Interfacing with the invisible computer. In Proc. NordiCHI ’02, pp. 213–216. ACM, 2002.
- 51. T. Erickson. Some problems with the notion of context-aware computing. Commun. ACM, 45(2):102–104, 2002
- 52. L. Barkhuus and A. K. Dey. Is context-aware computing taking control away from the user? Three levels of interactivity examined. In Proc. Ubicomp ’03, pp. 149–156. Springer, 2003.…
- 54. Our solutionAmbient projection system
- 55. Our solutionVisible Computer: user study
- 56. Our solutionRelation to Stages of Action model
- 57. Our solutionRelation to Stages of Action model
- 58. Our solutionRelation to Stages of Action modelfuture work