Page 451 -

442 CHAPTER 14 Online and ubiquitous HCI research

and interactions can tell us a great deal about the dynamics of group conversation
and the spread of key ideas.
Online and ubiquitous HCI research will likely continue to be shaped by—and to
shape—emerging tools and technological approaches. Improved conferencing tools
with richer integration of simultaneous screen-sharing and webcam feedback will
likely enable richer and more informative online research studies, while new social
applications will enable novel interaction patterns, promote further study, and sug-
gest further innovations in the next generation of tools and applications. Comparable
advances in sensors and ubiquitous tools will facilitate the collection of richer,
higher-resolution, and higher-fidelity data.
As the scope of online HCI research increases, ethical concerns associated with
frequent and often unobserved data collection will expand as well. Although the
comparison of alternative web site designs via A/B testing may be relatively benign,
integrated analyses of social media interactions, health data collected by wearable
devices, and other ubiquitous sensor data may identify insights not possible from any
single dataset, possibly revealing sensitive information that some participants might
prefer to leave undiscussed.

DISCUSSION QUESTIONS

1. The earlier discussion of remote usability testing cited a lack of direct feedback
from participants as a possible drawback. For in-person studies, careful
observation of facial expression and body language during usability tests might
help researchers identify moments of frustration or other emotional responses.
These cues might not be available in synchronous remote usability studies,
as some web conferencing tools might have limited video feedback through
webcams, while others support only screen sharing. However, there might be
some advantages to the lack of direct feedback. For example, in some cases,
participants might be willing to be more frank in providing direct feedback if
they do not have to see the facial expressions of the person administering the
experiment. Are there other types of studies or questions that might provide
better feedback if conducted remotely (as opposed to in-person)? How might
you evaluate the suitability of different questions for remote versus in-person
usability studies?

2. Ubiquitous mobile social computing through smartphones blurs lines between
“traditional” social networks and sensor-based ubiquitous computing. From
location-based apps used for identifying friends who might be physically
nearby to the Pokemon Go game that challenged users to explore and find
Pokemon characters on city streets and in natural environments, these apps
present the possibility of capturing and studying very rich datasets. What are
some of the challenges associated with studying datasets that might combine
geographical locations, social media posts, and detailed interactions with