Empowering People Rather Than Connecting Them - Human-Computer Interaction and Innovation in Handheld, Mobile and Wearable Technologies - page 274

Spatial interaction, can we provide feedback
gradients as elegant as a Zen garden? Will
well-designed multimodal interaction make
the device feel as rich as a fine musical
instrument?
8. Dancing or commanding? Can we create
metaphors which users can accept which
explicitly support interaction which consists
of a smooth ebb and flow of control between
human and machine, rather than a dialogue
of instructions and responses? Your phone
can actively engage with you at a range of
conceptual levels and timescales. On a scale
from autistic to spookily mindreading, how
alive to your wishes do you want your phone
feel?

Furthermore, when we work with such highly
instrumented devices, we should use the sensors not
just for interaction design, but also for instrumented
usability analysis, when we do our scientific stud-
ies. Standard mobile phones now have sensing and
storage abilities which allow us to log significant
amounts of data, including location, activity levels,
behaviour, and voice characteristics during normal
use, and to use this to better understand the effects
of design decisions on user behaviour, especially if
multiple members of a social network are instru-
mented. This allows users to go into more realistic
environments, but we sense more of the activities
and at a finer grained level than before. We can use
automated behaviour or context detection to initiate
experimental interventions, regaining some level of
experimental control, with the ecologically appro-
priate conditions (Roto et al. 2004, Oulasvirta et al.
2005). Researchers will be faced with demands for
increased levels of objective data and more explicit
behavioural metrics to augment and complement
the more traditional subjective feedback, and the
research community needs to explore, calibrate and
standardize the use of such metrics.

OUTLOOK FOR THE
INTERNATIONAL JOURNAL
OF MOBILE HCI:

How could the International Journal of Mobile
HCI contribute to progress towards the challenges
outlined above?
I think we need theoretical frameworks which
can handle the sensing, modelling and inference
which will be such a vital aspect of modern mo-
bile interaction, and an openness for interaction
design which breaks the 'discrete-event' mould.
We will also need to make sensor interpreta-
tion and signal processing developments which
enable truly engaging context aware interaction,
and we should respect and publish the research
in this area, rather than viewing it as the 'techni-
cal engineering details' which are hidden at the
back of a user study. Without appropriate sensing
and intelligent power management, none of the
futuristic goals can be realised in practice. Many
of the most interesting applications of the infer-
ence technology might end up being used at very
low levels, in order to sense and respond rapidly
to perceived changes in intention.

ACKNOWLEDGMENT

We are grateful for support from: SFI grants 00/
PI.1/C067 and 00/RFP06/CMS052, the EPSRC
project EP/E042740/1, the IST Programme of the
European Commission, under PASCAL Network
of Excellence, IST 2002-506778, and OpenInter-
face Project. Nokia provided a donation of funds
and equipment which further supported this work.

REFERENCES

Clark, A. (2008). Supersizing the mind:
Embodiment . Oxford: Action, and Cogni-
tive Extension. doi:10.1093/acprof:o
so/9780195333213.001.0001