Charting Past, Present, and Future Research in Ubiquitous Computing

G. D. Abowd, E. D. Mynatt, “Charting Past, Present and Future Research in Ubiquitous Computing“, ACM Transactions on Computer-Human Interaction, 7(1):29-58, March 2000.

This is Abowd et al. seminal paper that addresses the challenges of HCI in ubicomp and presents the notion of “everyday computing” that requires to design for continuous interaction by addressing the needs of interruption and resumption of interaction. Everyday computing focusses on scaling interaction with respect to time.

Requirements for critical-mass acceptance and collaboration imply scaling with respect to people. A final dimension, time, presents new challenges for scaling a system.

It promotes informal and unstructured activities typical of much of our everyday lives.

The objective is to understand how everyday tasks can be better supported, and how they are altered by the introduction of ubiquitous technologies.

The real goal for ubicomp is to provide many single-activity interactions that together promote a unified and continuous interaction between humans and computational services.

Ubicomp research implicit goal in the context of HCI has been to assist everyday life and not overwhelming it. Current accomplishments range in scale and in size from “inch-scale” personal devices to “yard-scale” shared devices. 3 themes of challenges remain:

• Natural interfaces that faciliate a richer variety of communications capabilities between humans and computation. Current interfaces still do not robustly handle the errors that naturally occur with these systems.
• Context-aware applications adapting their behavior based on information sensed form the physical and computational environment.
• Applications that capture of live experience. Tools to remove the burden of doing something humans are not good at (i.e. recording) so that they can focus attention on activities they are good at (i.e. indicating relationships, summarizing, and interpreting)

These three themes moves ubicomp more into the realm of everyday computing characterized by continuously present, integrative and unobtrusive interaction.

Natural interfaces come with the problem of new type of user input mistakes. Systems must assume that errors will occur and provide ways to handle it. Error handling is nothing new. Research areas include error reduction (elimination of errors is probably not achievable), error discovery (techniques include thresholding of confidence measures, historical statistics, and explicit rules specification) and reusable infrastructure for error correction.

Sensing technologies are not 100% reliable and deterministic. However

Combining measure from multiple sources could increase the confidence value for a particular interpretation.
Designing for everyday computing requires addressing these feature of informal, daily actvities:

• They rarely have a clear beginning or end
• Interruption is expected
• Multiple activities operate concurrently
• Time is an important discriminator
• Associative models of information are need

Some HCI research directions in everyday computing include

• Design a confinuously present computer interface
• Presenting information at different levels of the peryphery of human attention
• Connecting event in the physical and virtual worlds
• Modifying traditional HCI methods to support designing for informal peripheral, and opportunistic behavior

We are still in the early premises of evaluating ubicomp systems and the coevolution with us. Social implications of technologies will often come after people invent new, unforeseen, uses of these technologies.

In order to understand the impact of ubiquitous computing on everyday life, we navigate a delicate balance between prediction of how novel technologies will serve a real human need and observation of authentic use and subsequent coevolution of human activitie and novel technologies. Carroll, J. M., and M. B. Rosson. “Deliberated Evolution: Stalking the View Matcher in Design Space“. Human-Computer Interaction, Vol. 6, 281-318, 1991

Main evaluation challenges include:

• Finding a human need (the why? question…)
• Evaluating in the context of authentic use
• Task-centric § techniques are inappropriate

Other more recent article on that subject from the same authors: The Human Experience in Ubicomp

A ppt presentation summarizing this article.

Relation to my thesis: Addressing the need to scale in space, the number of devices and people using a system. Scale is implicit in the definiton of ubicomp research. Mention to the errors of human inputs to the system and how it is close to impossible to avoid them. I guess that part of the uncertainty in collaboration supported by ubicomp system does not uniquely come from technical limitations and failure but also bad user inputs. Errors must be assumed. A high-level goal of my thesis is to understand how everyday tasks can be better supported, and how they are altered by the introduction of ubiquitous technologies. I am aware that in the evaluation of my experiment platforms, there is a need to build a compelling story, form the end-users’ perspective (finding a human need). I plan to deploy the platform in authentic settings according to the scaling dimensions that characterize ubicomp systems, that is device, space, people and time.