Mixed-Initiative for Situated Interaction with Public Displays
Available from
Jorge Cardoso's profile on Mendeley.
Page 1
Mixed-Initiative for Situated Interaction with Public Displays
Ph.D. Research Proposal
Doctoral Program in Information Systems and Technologies
Mixed-Initiative for Situated
Interaction with Public Displays
- Research Proposal
Jorge C. S. Cardoso
jorgecardoso@ieee.org
Advisor:
Rui Jose´
September 5, 2009
Department of Information Systems
Engineering School
University of Minho
Doctoral Program in Information Systems and Technologies
Mixed-Initiative for Situated
Interaction with Public Displays
- Research Proposal
Jorge C. S. Cardoso
jorgecardoso@ieee.org
Advisor:
Rui Jose´
September 5, 2009
Department of Information Systems
Engineering School
University of Minho
Page 2
Page 3
Abstract
We have become accustomed to seeing digital public displays, but not yet to inter-
acting with them.
Digital displays have an enormous potential for being used as much more than
just information dissemination tools. They have the potential for also being shared
cultural objects in a place, providing a view on the social environment and allowing
a rich and engaging situated interactive experience to its users. A public situated
display should be able to automatically reflect the preferences and expectations
of its present users and also adapting itself over time to the social environment
that characterises the place where it is installed, providing an historic view on that
environment.
Achieving this view requires that display systems become capable of sensing their
environment and adapting to it. Interaction plays a vital role in this. The interactive
features provided be the display system should not only serve the purpose of provid-
ing users with a more engaging experience with the display, but they can also serve
as digital footprints of the activity and interests of the users — the fundamental
way to characterise the social environment surrounding the display.
Providing interactive features in a public display is a big challenge. The display sys-
tem must provide interesting and meaningful interactive features that allow users to
engage with it but at the same time these features must enable the display system
to collect relevant digital footprints that will allow it to better characterise the place
and, ultimately, adapt itself. However, the system must not relinquish full control
to a single person because it must always maintain its global sense of place that is
built over time and — it should not turn itself into an individually controlled display.
Control must be shared by multiple people, possibly with conflicting motivations, so
the display must balance personalisation and public interest. Also, the display must
not wait for user interaction. It should have an active role in its environment, ad-
dressing and enticing users into interaction in order to maintain a dynamic dialogue
with its audience and capturing enough digital footprints of activity.
The objective of this work is to explore mixed-initiative interaction approaches with
public displays that enable this new concept of public situated display: a display
iii
We have become accustomed to seeing digital public displays, but not yet to inter-
acting with them.
Digital displays have an enormous potential for being used as much more than
just information dissemination tools. They have the potential for also being shared
cultural objects in a place, providing a view on the social environment and allowing
a rich and engaging situated interactive experience to its users. A public situated
display should be able to automatically reflect the preferences and expectations
of its present users and also adapting itself over time to the social environment
that characterises the place where it is installed, providing an historic view on that
environment.
Achieving this view requires that display systems become capable of sensing their
environment and adapting to it. Interaction plays a vital role in this. The interactive
features provided be the display system should not only serve the purpose of provid-
ing users with a more engaging experience with the display, but they can also serve
as digital footprints of the activity and interests of the users — the fundamental
way to characterise the social environment surrounding the display.
Providing interactive features in a public display is a big challenge. The display sys-
tem must provide interesting and meaningful interactive features that allow users to
engage with it but at the same time these features must enable the display system
to collect relevant digital footprints that will allow it to better characterise the place
and, ultimately, adapt itself. However, the system must not relinquish full control
to a single person because it must always maintain its global sense of place that is
built over time and — it should not turn itself into an individually controlled display.
Control must be shared by multiple people, possibly with conflicting motivations, so
the display must balance personalisation and public interest. Also, the display must
not wait for user interaction. It should have an active role in its environment, ad-
dressing and enticing users into interaction in order to maintain a dynamic dialogue
with its audience and capturing enough digital footprints of activity.
The objective of this work is to explore mixed-initiative interaction approaches with
public displays that enable this new concept of public situated display: a display
iii
Page 4
that adapts itself to the social environment by leveraging on its interactive features
to engage and entice users into interaction and to collect digital footprints that will,
in turn, enable the display to better adapt to its social environment. To achieve
this goal, we will study interaction mechanisms that can be used to collect digital
footprints and the most effective ways of letting users generate them. We will de-
ploy prototype public display systems in real settings and evaluate the acceptance
of several approaches and determine which are the best at achieving a balance be-
tween giving users control over the display, allowing the display to collect relevant
information and allowing the display to give a coherent image of the place through
a selection of content.
Keywords
Situated displays, Interactive displays, Socially situated, Mixed-initiative interac-
tion
to engage and entice users into interaction and to collect digital footprints that will,
in turn, enable the display to better adapt to its social environment. To achieve
this goal, we will study interaction mechanisms that can be used to collect digital
footprints and the most effective ways of letting users generate them. We will de-
ploy prototype public display systems in real settings and evaluate the acceptance
of several approaches and determine which are the best at achieving a balance be-
tween giving users control over the display, allowing the display to collect relevant
information and allowing the display to give a coherent image of the place through
a selection of content.
Keywords
Situated displays, Interactive displays, Socially situated, Mixed-initiative interac-
tion
Page 5
Acknowledgements
Taking a PhD degree is something that is usually described as a lonely activity. All
things considered, although it might feel like a lonely job, it is in fact a journey
where the contributions of many people come together. Even at this early stage,
this work has already been touched by many people, who I would like to thank.
First, I must acknowledge and thank my supervisor, Professor Rui Jose´. His effort
and dedication in guiding, correcting and supporting me during this year makes the
present work as much his as it is mine.
I would also like to thank my colleagues of the PDTSI doctoral program at the
University of Minho. It is always good to share frustrations, but also solutions.
I must also thank all the doctoral program’s professors who taught us what scientific
research should be — I hope I learned my lesson well.
Finally, I would like to thank my colleagues at both the Mobile and Ubiquitous
Systems (Ubicomp) group at University of Minho and at the Research Center for
Science and Technology in Art (CITAR) at the Portuguese Catholic University for
providing an excelent work environment.
v
Taking a PhD degree is something that is usually described as a lonely activity. All
things considered, although it might feel like a lonely job, it is in fact a journey
where the contributions of many people come together. Even at this early stage,
this work has already been touched by many people, who I would like to thank.
First, I must acknowledge and thank my supervisor, Professor Rui Jose´. His effort
and dedication in guiding, correcting and supporting me during this year makes the
present work as much his as it is mine.
I would also like to thank my colleagues of the PDTSI doctoral program at the
University of Minho. It is always good to share frustrations, but also solutions.
I must also thank all the doctoral program’s professors who taught us what scientific
research should be — I hope I learned my lesson well.
Finally, I would like to thank my colleagues at both the Mobile and Ubiquitous
Systems (Ubicomp) group at University of Minho and at the Research Center for
Science and Technology in Art (CITAR) at the Portuguese Catholic University for
providing an excelent work environment.
v
Page 6
Page 7
Table of Contents
List of Acronyms xv
1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.1 Characterising places . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.2 Adapting to places . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.3 Interacting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.1 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.2 Delimitations . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Reader notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.6 Structure of this document . . . . . . . . . . . . . . . . . . . . . . . . 10
2 Related Work 11
2.1 Content Adaptation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1.1 User-submitted content . . . . . . . . . . . . . . . . . . . . . . 12
vii
List of Acronyms xv
1 Introduction 1
1.1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.2 Challenges . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.1 Characterising places . . . . . . . . . . . . . . . . . . . . . . . 3
1.2.2 Adapting to places . . . . . . . . . . . . . . . . . . . . . . . . 4
1.2.3 Interacting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.3 Objectives . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
1.3.1 Research Questions . . . . . . . . . . . . . . . . . . . . . . . . 8
1.3.2 Delimitations . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.4 Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.5 Reader notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
1.6 Structure of this document . . . . . . . . . . . . . . . . . . . . . . . . 10
2 Related Work 11
2.1 Content Adaptation . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11
2.1.1 User-submitted content . . . . . . . . . . . . . . . . . . . . . . 12
vii
Page 8
2.1.2 Centrally-managed content . . . . . . . . . . . . . . . . . . . . 16
2.1.3 Mixed approaches . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2 Interaction in Public Displays . . . . . . . . . . . . . . . . . . . . . . 21
2.2.1 Individual Interaction . . . . . . . . . . . . . . . . . . . . . . . 22
2.2.2 Group Interaction . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.3 Public Interaction . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3 Human factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3 Methodology 35
3.1 Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2 Work schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.3 Expected results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.3.1 Publication plan . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.4 Resources and access . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4 Current Work 45
4.1 Framework for digital footprints . . . . . . . . . . . . . . . . . . . . . 45
4.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1.2 A framework for digital footprints in public displays . . . . . . 46
4.1.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.2 Interaction guidelines and frameworks . . . . . . . . . . . . . . . . . . 54
4.2.1 Sensing systems challenges . . . . . . . . . . . . . . . . . . . . 55
4.2.2 Interaction Frameworks . . . . . . . . . . . . . . . . . . . . . 57
2.1.3 Mixed approaches . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.2 Interaction in Public Displays . . . . . . . . . . . . . . . . . . . . . . 21
2.2.1 Individual Interaction . . . . . . . . . . . . . . . . . . . . . . . 22
2.2.2 Group Interaction . . . . . . . . . . . . . . . . . . . . . . . . . 24
2.2.3 Public Interaction . . . . . . . . . . . . . . . . . . . . . . . . . 25
2.3 Human factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.4 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34
3 Methodology 35
3.1 Activities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36
3.2 Work schedule . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 37
3.3 Expected results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
3.3.1 Publication plan . . . . . . . . . . . . . . . . . . . . . . . . . 41
3.4 Resources and access . . . . . . . . . . . . . . . . . . . . . . . . . . . 42
4 Current Work 45
4.1 Framework for digital footprints . . . . . . . . . . . . . . . . . . . . . 45
4.1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45
4.1.2 A framework for digital footprints in public displays . . . . . . 46
4.1.3 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . 54
4.2 Interaction guidelines and frameworks . . . . . . . . . . . . . . . . . . 54
4.2.1 Sensing systems challenges . . . . . . . . . . . . . . . . . . . . 55
4.2.2 Interaction Frameworks . . . . . . . . . . . . . . . . . . . . . 57
Page 9
4.3 Planned Experiments . . . . . . . . . . . . . . . . . . . . . . . . . . . 61
4.3.1 A comparison of two approaches for the generation of keywords 62
5 Final Considerations 65
4.3.1 A comparison of two approaches for the generation of keywords 62
5 Final Considerations 65
Page 16
xvi
Page 18
1.1 Motivation
Current public displays have not yet attained their full potential as situated public
displays: an artefact shared by everyone; a cultural and social reference point that
acts also as a mirror of a place, reflecting its social environment; a promoter of
meaningful social interaction.
Public displays have the potential to act as an important reference point in a place,
something that a newcomer can look at to get a sense of the kind of social envi-
ronment that surrounds it. However, public situated displays should not only help
describe a place but also help in defining it. By being more than just a passive
mirror to the social dynamics of a place, displays can also take on an active role
in the definition of what a place is. They can do this by fostering social situated
interactions. By fostering interactions, we mean more than just mediating human–
human interaction. Situated displays should enable new types of social (situated)
human–human and human–machine interaction and be a participative party in those
interactions by suggesting, supporting and accepting them. Even though people may
have personal motivations for interacting with the public display, the result from
these interactions should somehow be publicly meaningful and interesting. Public
situated displays must foster these kind of interactions.
In order to achieve this, public displays need to be able to characterise the social
environment and dynamics of the place they are part of and act as a mirror of that
environment by showing content that reflects an historical view as well as the more
spontaneous and localized events. Displays must also provide interactive features
that support effective public situated interaction between and with people.
Current research around situated displays fails to address this integrated view. Most
developments around this topic, focus on specific display applications rather than
on a concept of generic public display that evolves with its social environment and
that supports meaningful situated interaction.
Achieving this concept of situated display has also practical implications because it
will increase the display efficiency for many applications. A study by Huang et al.
[2008] shows that most public displays fail at captivating people’s attention. Giving
users some control over the contents was one of the recommendations of the study
but presenting more meaningful and interesting content is also an obvious step in
catching the user’s attention. Advertising is an obvious application that can benefit
from this work because this concept of display system naturally provide a means
to characterize and target the current or regular audiences. However, even just
informative or entertainment displays can benefit from a concept of situated display
that takes the specific social environment into account and that allows users to
appropriate the display through its shared control interfacing mechanisms.
2
Current public displays have not yet attained their full potential as situated public
displays: an artefact shared by everyone; a cultural and social reference point that
acts also as a mirror of a place, reflecting its social environment; a promoter of
meaningful social interaction.
Public displays have the potential to act as an important reference point in a place,
something that a newcomer can look at to get a sense of the kind of social envi-
ronment that surrounds it. However, public situated displays should not only help
describe a place but also help in defining it. By being more than just a passive
mirror to the social dynamics of a place, displays can also take on an active role
in the definition of what a place is. They can do this by fostering social situated
interactions. By fostering interactions, we mean more than just mediating human–
human interaction. Situated displays should enable new types of social (situated)
human–human and human–machine interaction and be a participative party in those
interactions by suggesting, supporting and accepting them. Even though people may
have personal motivations for interacting with the public display, the result from
these interactions should somehow be publicly meaningful and interesting. Public
situated displays must foster these kind of interactions.
In order to achieve this, public displays need to be able to characterise the social
environment and dynamics of the place they are part of and act as a mirror of that
environment by showing content that reflects an historical view as well as the more
spontaneous and localized events. Displays must also provide interactive features
that support effective public situated interaction between and with people.
Current research around situated displays fails to address this integrated view. Most
developments around this topic, focus on specific display applications rather than
on a concept of generic public display that evolves with its social environment and
that supports meaningful situated interaction.
Achieving this concept of situated display has also practical implications because it
will increase the display efficiency for many applications. A study by Huang et al.
[2008] shows that most public displays fail at captivating people’s attention. Giving
users some control over the contents was one of the recommendations of the study
but presenting more meaningful and interesting content is also an obvious step in
catching the user’s attention. Advertising is an obvious application that can benefit
from this work because this concept of display system naturally provide a means
to characterize and target the current or regular audiences. However, even just
informative or entertainment displays can benefit from a concept of situated display
that takes the specific social environment into account and that allows users to
appropriate the display through its shared control interfacing mechanisms.
2
Page 20
It means only that that idea will never result in a display perfectly integrated into
its environment, unless some degree of freedom is left for the display to adapt.
The second implication is that a display cannot be designed independently of its
users. A place is, essentially, a social environment strongly associated with a loca-
tion. This means that, ultimately, adapting to a place is adapting to the preferences
and expectations of its users.
The challenge is, then, how to effectively characterize the social environment in a
dynamic way in order to be able to provide a representation of that environment
through the presented content.
1.2.2 Adapting to places
Another challenge, deeply related to the previous, is how the display should adapt to
a place. Given a possible characterization of the place, what is the best adaptation
strategy?
On the one hand, it should always maintain a long-term view that represents the
history of the place. This historical view is what will allow people to identify what
that place is about. It should emerge as something unique to that place, fulfilling
the role of mirror of the social environment. On the other hand, it should allow
people to appropriate it. One of the implications of this is that it should somehow
react to the more spontaneous events that occur and to the currently or recently
present people in order for them to be able to take advantage of it in the moment.
The challenge is combining these two perspectives in a coherent way so that it
contributes to the sense of place.
1.2.3 Interacting
Interactivity is a fundamental aspect and a challenge for public situated displays.
It is a fundamental aspect, because interaction is essential in order for it to achieve
the full potential as a situated display that is in part driven by the preferences
and expectations of people. Interaction allows users to contribute to the display’s
content responding to the ever increasing expectation that people are no longer just
passive viewers but active contributors. A situated display should not be based on
the classic broadcast model in which content is produced and managed centrally;
people now look for more control and ways to contribute to the information flow.
Contributions from people will also, in part, drive the situatedness of the display
— it will not be statically configured to show specific content, but will evolve with
4
its environment, unless some degree of freedom is left for the display to adapt.
The second implication is that a display cannot be designed independently of its
users. A place is, essentially, a social environment strongly associated with a loca-
tion. This means that, ultimately, adapting to a place is adapting to the preferences
and expectations of its users.
The challenge is, then, how to effectively characterize the social environment in a
dynamic way in order to be able to provide a representation of that environment
through the presented content.
1.2.2 Adapting to places
Another challenge, deeply related to the previous, is how the display should adapt to
a place. Given a possible characterization of the place, what is the best adaptation
strategy?
On the one hand, it should always maintain a long-term view that represents the
history of the place. This historical view is what will allow people to identify what
that place is about. It should emerge as something unique to that place, fulfilling
the role of mirror of the social environment. On the other hand, it should allow
people to appropriate it. One of the implications of this is that it should somehow
react to the more spontaneous events that occur and to the currently or recently
present people in order for them to be able to take advantage of it in the moment.
The challenge is combining these two perspectives in a coherent way so that it
contributes to the sense of place.
1.2.3 Interacting
Interactivity is a fundamental aspect and a challenge for public situated displays.
It is a fundamental aspect, because interaction is essential in order for it to achieve
the full potential as a situated display that is in part driven by the preferences
and expectations of people. Interaction allows users to contribute to the display’s
content responding to the ever increasing expectation that people are no longer just
passive viewers but active contributors. A situated display should not be based on
the classic broadcast model in which content is produced and managed centrally;
people now look for more control and ways to contribute to the information flow.
Contributions from people will also, in part, drive the situatedness of the display
— it will not be statically configured to show specific content, but will evolve with
4
Page 21
the contributions of the individuals who frequent a place and, thus, with the place
itself.
There are, however, practical problems with this approach. The first is getting peo-
ple to interact with a public display — something that is very different from what
people are used to interact with. Taking the initiative to interact with people is
something the display system must be prepared to do in order to receive sufficient
input. The second problem is how to design and provide interaction models that
people can understand and use effectively. And the third problem is how to pro-
vide public, shared mechanisms that truly support public situated interaction. Even
though users will usually have personal motivations when interacting, the display
system should try to play with personal motivations so that the result is interest-
ing and meaningful to the public. The interaction mechanisms must allow a wide
range of people to contribute to the display without adopting a personal interaction
paradigm that could jeopardize the concept of a public, shared object.
1.3 Objectives
In order to achieve the concept of situated display previously outlined, it is im-
portant to study three interconnected aspects: social sampling methods that allow
the display to automatically gather relevant data to characterize the place; a place
model that is able to integrate both the historical data and event data in order to
create a flow of content that is at the same time representative of the general social
environment of the place but also reactive to the more instant happenings; and pub-
licly shared control mechanisms for interaction with the displays that engage users
into generating, producing or selecting content.
One of the three important aspects for socially situated displays is the ability to
sample the social environment in order to characterize and adapt to it. This social
sampling is not simply a matter of collecting raw sensor data but, somehow, inter-
preting its social significance. The environment of a situated display is a place: a
physical location with an evolving social context. There is no “social” sensor capa-
ble of measuring that phenomenon. However, just as web sites like Amazon.com or
Last.fm are able to learn our tastes in books and music by recording and analysing
our interactions with them, so too can displays learn the preferences and tastes of
their audiences. For this to happen two related things must occur: displays need to
become interactive and ways of collecting and using relevant information from these
interactions need to be devised. If displays are able to collect individual preferences,
they will become much more integrated into their social environment and will be-
come much more meaningful to people — they will truly become situated displays.
Whether online or offline, our daily life generates already generates various digital
traces — digital footprints — of our activities: every time we use our credit cards,
5
itself.
There are, however, practical problems with this approach. The first is getting peo-
ple to interact with a public display — something that is very different from what
people are used to interact with. Taking the initiative to interact with people is
something the display system must be prepared to do in order to receive sufficient
input. The second problem is how to design and provide interaction models that
people can understand and use effectively. And the third problem is how to pro-
vide public, shared mechanisms that truly support public situated interaction. Even
though users will usually have personal motivations when interacting, the display
system should try to play with personal motivations so that the result is interest-
ing and meaningful to the public. The interaction mechanisms must allow a wide
range of people to contribute to the display without adopting a personal interaction
paradigm that could jeopardize the concept of a public, shared object.
1.3 Objectives
In order to achieve the concept of situated display previously outlined, it is im-
portant to study three interconnected aspects: social sampling methods that allow
the display to automatically gather relevant data to characterize the place; a place
model that is able to integrate both the historical data and event data in order to
create a flow of content that is at the same time representative of the general social
environment of the place but also reactive to the more instant happenings; and pub-
licly shared control mechanisms for interaction with the displays that engage users
into generating, producing or selecting content.
One of the three important aspects for socially situated displays is the ability to
sample the social environment in order to characterize and adapt to it. This social
sampling is not simply a matter of collecting raw sensor data but, somehow, inter-
preting its social significance. The environment of a situated display is a place: a
physical location with an evolving social context. There is no “social” sensor capa-
ble of measuring that phenomenon. However, just as web sites like Amazon.com or
Last.fm are able to learn our tastes in books and music by recording and analysing
our interactions with them, so too can displays learn the preferences and tastes of
their audiences. For this to happen two related things must occur: displays need to
become interactive and ways of collecting and using relevant information from these
interactions need to be devised. If displays are able to collect individual preferences,
they will become much more integrated into their social environment and will be-
come much more meaningful to people — they will truly become situated displays.
Whether online or offline, our daily life generates already generates various digital
traces — digital footprints — of our activities: every time we use our credit cards,
5
Page 22
buy something at the supermarket using discount cards, make phone calls, access
the internet, use an ATM2 machine, etc. These digital footprints could be used
to infer our tastes, interests and activities (and sometimes are). In a similar man-
ner, certain digital, footprints resulting from interaction with public digital displays,
could be used by the display system to adjust its content to better suit the interests
and expectations of those who normally use it. The question is, then, what kind
of digital footprints will be relevant for situated displays? How can display systems
use those digital footprints and what interaction/sensing mechanisms are necessary
to allow users to generate them?
Collecting footprints is, however, not sufficient. The display system must have a
model of place that is able to combine several digital footprints to give a high level
representation of place. This model must be able to characterize a place in an
historical way, aggregating long-term digital footprints. However, the place model
must not be simply a data model but also incorporates simple processes that are able
to react to events and more immediate user actions in order to appear responsive
and to give a sense of intelligent behaviour and also to give a sense of appropriation
to users. This model must also integrate more static administrative definitions of
place that allow the display owner to define an overall theme for the display. What
type of models are best for this purpose, what data should they incorporate and
what processes should they implement are questions which we intend to study.
A fundamental way of collecting digital footprints that will serve to characterize
the place where the display system is located is by letting (and enticing) users to
interact with the display. Interaction with public situated displays is fundamentally
different from interaction with a desktop system or even with a collaborative display
system. A situated display must be able to accomplish two things regarding user
interaction: it must let users contribute to the content and also provide some level
of control over that content. These two aspects together contribute both to the
generation of digital footprints that can be incorporated into the place model and
also to the sense of ownership and appropriation necessary to captivate users and
continue to engage them in interacting with the display. Users must be able to
contribute directly to content and to give feedback on the content selected by the
display system. The problem is how to enable a shared control mechanism that
allows multiple users, possibly with different intentions, to simultaneously manifest
themselves. These shared control mechanisms must be able to:
Establish a balance between the reactive and the proactive behaviour. The
display should not be completely reactive to user actions in order to guaran-
tee that the displayed content is based on the characteristics of that place.
However, the display must still be able to react to more immediate events and
requests in order to appear responsive, intelligent and give users some sense
of control.
2Automated Teller Machine.
6
the internet, use an ATM2 machine, etc. These digital footprints could be used
to infer our tastes, interests and activities (and sometimes are). In a similar man-
ner, certain digital, footprints resulting from interaction with public digital displays,
could be used by the display system to adjust its content to better suit the interests
and expectations of those who normally use it. The question is, then, what kind
of digital footprints will be relevant for situated displays? How can display systems
use those digital footprints and what interaction/sensing mechanisms are necessary
to allow users to generate them?
Collecting footprints is, however, not sufficient. The display system must have a
model of place that is able to combine several digital footprints to give a high level
representation of place. This model must be able to characterize a place in an
historical way, aggregating long-term digital footprints. However, the place model
must not be simply a data model but also incorporates simple processes that are able
to react to events and more immediate user actions in order to appear responsive
and to give a sense of intelligent behaviour and also to give a sense of appropriation
to users. This model must also integrate more static administrative definitions of
place that allow the display owner to define an overall theme for the display. What
type of models are best for this purpose, what data should they incorporate and
what processes should they implement are questions which we intend to study.
A fundamental way of collecting digital footprints that will serve to characterize
the place where the display system is located is by letting (and enticing) users to
interact with the display. Interaction with public situated displays is fundamentally
different from interaction with a desktop system or even with a collaborative display
system. A situated display must be able to accomplish two things regarding user
interaction: it must let users contribute to the content and also provide some level
of control over that content. These two aspects together contribute both to the
generation of digital footprints that can be incorporated into the place model and
also to the sense of ownership and appropriation necessary to captivate users and
continue to engage them in interacting with the display. Users must be able to
contribute directly to content and to give feedback on the content selected by the
display system. The problem is how to enable a shared control mechanism that
allows multiple users, possibly with different intentions, to simultaneously manifest
themselves. These shared control mechanisms must be able to:
Establish a balance between the reactive and the proactive behaviour. The
display should not be completely reactive to user actions in order to guaran-
tee that the displayed content is based on the characteristics of that place.
However, the display must still be able to react to more immediate events and
requests in order to appear responsive, intelligent and give users some sense
of control.
2Automated Teller Machine.
6
Page 26
1.4 Methodology
This research will be mainly based on real setting experiments to evaluate differ-
ent adaptation and control mechanisms. These experiments will be measured along
quantitative and qualitative data by analysing interaction logs and performing in-
terviews and questionnaires. The initial phase of this research will, however, be
more focused on the analysis of current work to devise initial interaction footprints
and interaction mechanisms that generate them. These will serve as the basis for
the subsequent experiments. Chapter 3 will present this methodological approach
in more detail.
1.5 Reader notes
In order not to clutter the text too much with the definitions of the several well-
known acronyms used throughout the text, acronyms are defined the first time they
are used using a footnote. For easier reference, a list of all acronyms is also given in
the beginning of this document.
1.6 Structure of this document
This document is structured in the following way: The current Chapter presented the
motivation for the work to be pursued, the main challenges faced when addressing
this problem, the objectives of this work and the general methodological approach.
Chapter 2 presents some work related to this project, more concretely, work related
to some of the challenges posed by the problem that is being addressed. Chapter 3
describes the methodological approach that will be followed in the course of this
research, the main activities to be performed and the expected results. Chapter 4
describes work accomplished so far. Finally, Chapter 5 presents some concluding
remarks.
10
This research will be mainly based on real setting experiments to evaluate differ-
ent adaptation and control mechanisms. These experiments will be measured along
quantitative and qualitative data by analysing interaction logs and performing in-
terviews and questionnaires. The initial phase of this research will, however, be
more focused on the analysis of current work to devise initial interaction footprints
and interaction mechanisms that generate them. These will serve as the basis for
the subsequent experiments. Chapter 3 will present this methodological approach
in more detail.
1.5 Reader notes
In order not to clutter the text too much with the definitions of the several well-
known acronyms used throughout the text, acronyms are defined the first time they
are used using a footnote. For easier reference, a list of all acronyms is also given in
the beginning of this document.
1.6 Structure of this document
This document is structured in the following way: The current Chapter presented the
motivation for the work to be pursued, the main challenges faced when addressing
this problem, the objectives of this work and the general methodological approach.
Chapter 2 presents some work related to this project, more concretely, work related
to some of the challenges posed by the problem that is being addressed. Chapter 3
describes the methodological approach that will be followed in the course of this
research, the main activities to be performed and the expected results. Chapter 4
describes work accomplished so far. Finally, Chapter 5 presents some concluding
remarks.
10
Page 27
Chapter 2
Related Work
The current work can be best described as research in the area of situated displays.
In order to give a better perspective on this area, this chapter presents an overview
of work related to the current research on public situated displays. It begins by
giving a perspective on content adaptation approaches used in earlier systems. It
continues by describing some research around a particular, and fundamental, issue
around the design of situated displays — interaction — and how past display systems
have addressed the question. This chapter ends with some research that has tried to
provide a view on design considerations that are important in order to successfully
create a situated display.
This review of related literature followed a standard methodology of collecting seed
papers (and books) from a keyword search mainly on Google, Google Scholar [Google,
2009], ACM1 Digital Library [ACM, 2009] and IEEE2 Xplore [IEEE, 2009] publica-
tions databases.
2.1 Content Adaptation
The problem of providing content that is relevant to a display’s surrounding envi-
ronment, namely people, has been tackled before in many different display systems,
using different approaches. Two major approaches can be identified in the solu-
tions encountered to deal with this problem: let the users themselves decide what is
relevant content; or, design the display with centrally-managed and defined content.
1Association for Computing Machinery.
2Institute of Electrical and Electronics Engineers.
11
Related Work
The current work can be best described as research in the area of situated displays.
In order to give a better perspective on this area, this chapter presents an overview
of work related to the current research on public situated displays. It begins by
giving a perspective on content adaptation approaches used in earlier systems. It
continues by describing some research around a particular, and fundamental, issue
around the design of situated displays — interaction — and how past display systems
have addressed the question. This chapter ends with some research that has tried to
provide a view on design considerations that are important in order to successfully
create a situated display.
This review of related literature followed a standard methodology of collecting seed
papers (and books) from a keyword search mainly on Google, Google Scholar [Google,
2009], ACM1 Digital Library [ACM, 2009] and IEEE2 Xplore [IEEE, 2009] publica-
tions databases.
2.1 Content Adaptation
The problem of providing content that is relevant to a display’s surrounding envi-
ronment, namely people, has been tackled before in many different display systems,
using different approaches. Two major approaches can be identified in the solu-
tions encountered to deal with this problem: let the users themselves decide what is
relevant content; or, design the display with centrally-managed and defined content.
1Association for Computing Machinery.
2Institute of Electrical and Electronics Engineers.
11
Page 28
2.1.1 User-submitted content
One approach to deal with the problem is to let users themselves define the content
that the display will present. Some situated display applications are well suited for
this approach because the concept they try to reproduce is based on user submitted
content only. This is the case with digital bulletin boards and similar. These situated
displays are designed specifically to allow people (usually within a work group) to
post news, photos, comments, etc., to a public display.
Notification Collage [Greenberg and Rounding, 2001] is one example of a display
where content is completely defined by its users. The system was designed so that
work colleagues could stream their webcam to the display, put photo slide shows,
show web pages and leave notes to each other using a desktop application that was
mirrored in a semi-public display. Figure 2.1 shows an example of the Notification
Collage’s display elements. The Notification Collage was deployed as a means to
increase awareness of co-workers’ activities and was designed in a bulletin board
style. What is relevant content for the Notification Collage is completely defined
and controlled by its users. The system plays only an almost insignificant part in
the presentation of content by laying out the new content items randomly in the left
side of screen, which results naturally in old content being obfuscated by new; the
arrangement of the right side of the screen is managed by users also.
Figure 2.1: The Notification Collage display: examples of media elements that can
be posted [Greenberg and Rounding, 2001].
Another example of a display designed as a type of digital bulletin board is the
12
One approach to deal with the problem is to let users themselves define the content
that the display will present. Some situated display applications are well suited for
this approach because the concept they try to reproduce is based on user submitted
content only. This is the case with digital bulletin boards and similar. These situated
displays are designed specifically to allow people (usually within a work group) to
post news, photos, comments, etc., to a public display.
Notification Collage [Greenberg and Rounding, 2001] is one example of a display
where content is completely defined by its users. The system was designed so that
work colleagues could stream their webcam to the display, put photo slide shows,
show web pages and leave notes to each other using a desktop application that was
mirrored in a semi-public display. Figure 2.1 shows an example of the Notification
Collage’s display elements. The Notification Collage was deployed as a means to
increase awareness of co-workers’ activities and was designed in a bulletin board
style. What is relevant content for the Notification Collage is completely defined
and controlled by its users. The system plays only an almost insignificant part in
the presentation of content by laying out the new content items randomly in the left
side of screen, which results naturally in old content being obfuscated by new; the
arrangement of the right side of the screen is managed by users also.
Figure 2.1: The Notification Collage display: examples of media elements that can
be posted [Greenberg and Rounding, 2001].
Another example of a display designed as a type of digital bulletin board is the
12
Page 29
WebWall [Ferscha et al., 2002]. The WebWall is an infrastructure that can be used
to create large display applications that allow users to post various media elements
and interact with existing ones (these elements are instances of service classes, in
WebWall’s nomenclature). Figure 2.2 shows some types of elements that can be
displayed by the WebWall. Content can be published through various interfaces such
as mobile phone, email and web interface. Contrary to Notification Collage’s random
placement of items, WebWall avoids overlaps by scheduling content according to its
priority, lifetime and type of content.
Figure 2.2: The WebWall display: types of service classes supported [Ferscha et al.,
2002].
The Community Wall developed by Grasso et al. [2003] also uses the bulletin board
style. The purpose of the Community Wall was to support communication within
a community of practice — a kind of spontaneous and informal workgroup that is
driven by the common work-related interests and practices — and to “create an
environment that fosters social encounters (conversations) using documents or news
and peoples’ opinions on them as triggers” [Grasso et al., 2003]. The display was
designed as a bulletin board (see Figure 2.3) were anyone could post to. Posting
can be done using several methods such as email, web bookmarklets (JavaScript
programs that can be stored as bookmarks and that can be used to automatically
send the current webpage to the display system), paper (using Xerox’s Flowport(tm)
software to scan the document) and a PDA3 application. Although users are com-
pletely responsible for the content database, the Community Wall is responsible for
choosing what to display from that database. A number of rules exist to define the
priority of each item based on its type, age, rating, number of comments, time, etc.
After applying the rules, the system will present the (around 10–15) highest priority
items.
Yet another classic example of a situated display that allow users to post news items
(and other types of content) is the Plasma Poster [Churchill et al., 2004]. This sys-
tem was deployed and studied at the FXPAL — a software research company, based
in Palo Alto, California. The displays consist of touch-sensitive plasma displays with
an interface (see Figure 2.4) that allows people to interact with it to read, browse,
3Personal Digital Assistant.
13
to create large display applications that allow users to post various media elements
and interact with existing ones (these elements are instances of service classes, in
WebWall’s nomenclature). Figure 2.2 shows some types of elements that can be
displayed by the WebWall. Content can be published through various interfaces such
as mobile phone, email and web interface. Contrary to Notification Collage’s random
placement of items, WebWall avoids overlaps by scheduling content according to its
priority, lifetime and type of content.
Figure 2.2: The WebWall display: types of service classes supported [Ferscha et al.,
2002].
The Community Wall developed by Grasso et al. [2003] also uses the bulletin board
style. The purpose of the Community Wall was to support communication within
a community of practice — a kind of spontaneous and informal workgroup that is
driven by the common work-related interests and practices — and to “create an
environment that fosters social encounters (conversations) using documents or news
and peoples’ opinions on them as triggers” [Grasso et al., 2003]. The display was
designed as a bulletin board (see Figure 2.3) were anyone could post to. Posting
can be done using several methods such as email, web bookmarklets (JavaScript
programs that can be stored as bookmarks and that can be used to automatically
send the current webpage to the display system), paper (using Xerox’s Flowport(tm)
software to scan the document) and a PDA3 application. Although users are com-
pletely responsible for the content database, the Community Wall is responsible for
choosing what to display from that database. A number of rules exist to define the
priority of each item based on its type, age, rating, number of comments, time, etc.
After applying the rules, the system will present the (around 10–15) highest priority
items.
Yet another classic example of a situated display that allow users to post news items
(and other types of content) is the Plasma Poster [Churchill et al., 2004]. This sys-
tem was deployed and studied at the FXPAL — a software research company, based
in Palo Alto, California. The displays consist of touch-sensitive plasma displays with
an interface (see Figure 2.4) that allows people to interact with it to read, browse,
3Personal Digital Assistant.
13
Page 30
Figure 2.3: Community Wall’s screen [Grasso et al., 2003]
forward and comment on items. Users can post items to the display by emailing
the content or using a dedicated web interface for posting. The Plasma Poster was
designed to allow more engagement with the content than WebWall or Notification
Collage and so it displays content items in sequence (showing also a list of thumb-
nails of the previously and the next content items) but it allows users to pause the
automatic scheduling for reading an item, scrolling, commenting and even browsing
content.
Figure 2.4: The Plasma Poster display showing a set of images and thumbnails of
other available content [Churchill et al., 2004].
14
forward and comment on items. Users can post items to the display by emailing
the content or using a dedicated web interface for posting. The Plasma Poster was
designed to allow more engagement with the content than WebWall or Notification
Collage and so it displays content items in sequence (showing also a list of thumb-
nails of the previously and the next content items) but it allows users to pause the
automatic scheduling for reading an item, scrolling, commenting and even browsing
content.
Figure 2.4: The Plasma Poster display showing a set of images and thumbnails of
other available content [Churchill et al., 2004].
14
Page 31
Other types of public display applications are also driven solely by user-generated
content. The Dynamo display system [Brignull et al., 2004], for example, is a large
multi-user interactive display for sharing, exchanging and showing multimedia con-
tent in a communal room of a high school. Dynamo provides a GUI4 like interface
(see Figure 2.5) accessible from various interaction points (wireless mice and key-
boards) so that multiple users can interact with it at the same time. Dynamo allows
users to connect external USB5 devices and access its content to display it publicly
on the screen or to share it with other users by dragging it to a public area or by
sending it to specific people (users had to register to be able to access some of the
functionalities). During the deployment time, students used Dynamo to display and
exchange photos, video and music; to create a pool of public media that anyone
could use; to stage performances to the audience in the communal room; to post
notices to other users; to leave content as gifts to specific people; and to engage in
group discussions and interactions. Dynamo provided only the infrastructure for the
content sharing. Users were ultimately responsible for creating meaningful content
and appropriate the system in an adequate way for that place.
Figure 2.5: Dynamo surface [Brignull et al., 2004]
Similarly, the BlueBoard [Russell and Gossweiler, 2001] system (see Figure 2.6) was
designed as a personal content access and sharing tool. Users access their personal
information by authenticating themselves using a personal badge that is read by the
display system. Several users can authenticate at the same time and access their
personal data and share content by dragging it (the display is touch-sensitive) to
the other user’s p-con (the image that represents the user). BlueBoard is meant to
be used as a more casual display, something that is at hand and allows a user to
rapidly access his personal information and share it with others. BlueBoard uses
a custom content authoring tool to allow users to create their content (calendars,
access email, files, etc) which they can later access and share using the display.
4Graphical User Interface.
5Universal Serial Bus.
15
content. The Dynamo display system [Brignull et al., 2004], for example, is a large
multi-user interactive display for sharing, exchanging and showing multimedia con-
tent in a communal room of a high school. Dynamo provides a GUI4 like interface
(see Figure 2.5) accessible from various interaction points (wireless mice and key-
boards) so that multiple users can interact with it at the same time. Dynamo allows
users to connect external USB5 devices and access its content to display it publicly
on the screen or to share it with other users by dragging it to a public area or by
sending it to specific people (users had to register to be able to access some of the
functionalities). During the deployment time, students used Dynamo to display and
exchange photos, video and music; to create a pool of public media that anyone
could use; to stage performances to the audience in the communal room; to post
notices to other users; to leave content as gifts to specific people; and to engage in
group discussions and interactions. Dynamo provided only the infrastructure for the
content sharing. Users were ultimately responsible for creating meaningful content
and appropriate the system in an adequate way for that place.
Figure 2.5: Dynamo surface [Brignull et al., 2004]
Similarly, the BlueBoard [Russell and Gossweiler, 2001] system (see Figure 2.6) was
designed as a personal content access and sharing tool. Users access their personal
information by authenticating themselves using a personal badge that is read by the
display system. Several users can authenticate at the same time and access their
personal data and share content by dragging it (the display is touch-sensitive) to
the other user’s p-con (the image that represents the user). BlueBoard is meant to
be used as a more casual display, something that is at hand and allows a user to
rapidly access his personal information and share it with others. BlueBoard uses
a custom content authoring tool to allow users to create their content (calendars,
access email, files, etc) which they can later access and share using the display.
4Graphical User Interface.
5Universal Serial Bus.
15
Page 33
One example of a display system that uses a pre-defined content source is the MIT
Media Lab’s Aware Community Portals [Sawhney et al., 2001]. This display was
intended to be used in a transitional space of a workplace. Since the community’s
interests, in this case, were well-known, the display was designed to use a popular
technology-related news site — Slashdot7 — as the main content source. Users
had no possibility to contribute directly to the display’s content; they could only
consume the information presented. The display showed several types of content
such as the current time, weather, cartoons and even MP3 audio files, but would
default to the Slashdot news when users stood in front of the display for some time.
Although the final objective was to provide a collaborative way to define the content
that the display would show, the implemented prototype was deployed without this
functionality and users had no control over what was displayed.
Displays with an objective of providing awareness or background information are
also usually designed with centrally-managed content. In the Proactive Displays
[McDonald et al., 2008] project the authors evaluated a set of proactive display
applications designed to augment and extend the social actions and interactions
that usually occur in an academic conference and are usually designated by “social
networking” and which are, in fact, one of the main motivations for attending a
conference. This was accomplished by deploying a set of displays capable of sensing
the proximity of the participants and presenting information about them. In order
to be automatically identified, participants were given RFID8 augmented conference
badges and were asked to fill a personal profile on a web form, before attending
the conference. The displays were deployed in different spaces of the conference,
according to their functionality:
AutoSpeakerID — an application that displays the name, affiliation and photo
(if provided) of the person asking a question during the question and answer
period following a paper or panel presentation (see Figure 2.7).
Ticket2Talk — shows a theme (an image and a caption) that participants
specified as being willing to talk about during the conference. Content is
shown only if the participant is near the display (which was deployed behind
the coffee tables).
NeighborhoodWindow — shows keywords taken from the participants web-
pages and shows shared and unique interests of those near the display.
Although the content displayed was provided by its users, they did not have much
control over it. Participants were asked to fill a web form designed to gather specific
data, such as the name, affiliation, photo and interests and they could only either fill
in the form or leave it blank. Afterwards, there was no easy way to control what the
7http://slashdot.org
8Radio-Frequency Identification.
17
Media Lab’s Aware Community Portals [Sawhney et al., 2001]. This display was
intended to be used in a transitional space of a workplace. Since the community’s
interests, in this case, were well-known, the display was designed to use a popular
technology-related news site — Slashdot7 — as the main content source. Users
had no possibility to contribute directly to the display’s content; they could only
consume the information presented. The display showed several types of content
such as the current time, weather, cartoons and even MP3 audio files, but would
default to the Slashdot news when users stood in front of the display for some time.
Although the final objective was to provide a collaborative way to define the content
that the display would show, the implemented prototype was deployed without this
functionality and users had no control over what was displayed.
Displays with an objective of providing awareness or background information are
also usually designed with centrally-managed content. In the Proactive Displays
[McDonald et al., 2008] project the authors evaluated a set of proactive display
applications designed to augment and extend the social actions and interactions
that usually occur in an academic conference and are usually designated by “social
networking” and which are, in fact, one of the main motivations for attending a
conference. This was accomplished by deploying a set of displays capable of sensing
the proximity of the participants and presenting information about them. In order
to be automatically identified, participants were given RFID8 augmented conference
badges and were asked to fill a personal profile on a web form, before attending
the conference. The displays were deployed in different spaces of the conference,
according to their functionality:
AutoSpeakerID — an application that displays the name, affiliation and photo
(if provided) of the person asking a question during the question and answer
period following a paper or panel presentation (see Figure 2.7).
Ticket2Talk — shows a theme (an image and a caption) that participants
specified as being willing to talk about during the conference. Content is
shown only if the participant is near the display (which was deployed behind
the coffee tables).
NeighborhoodWindow — shows keywords taken from the participants web-
pages and shows shared and unique interests of those near the display.
Although the content displayed was provided by its users, they did not have much
control over it. Participants were asked to fill a web form designed to gather specific
data, such as the name, affiliation, photo and interests and they could only either fill
in the form or leave it blank. Afterwards, there was no easy way to control what the
7http://slashdot.org
8Radio-Frequency Identification.
17
Page 35
Figure 2.8: The Intellibadge visualisation using a garden metaphor to represent
the conference site. Flowers represent locations, coloured petals the professional
interests of users and walking ants represent the rate of people walking by the
tracked areas [Cox et al., 2003].
be accomplished using time-based approaches that define the start and duration
for each content item or event- or context-based approaches that take into account
the preferences or historic data about who is near the display. Storz et al. [2006],
for example, when developing a software infrastructure for (essentially time-based)
content scheduling identified the following requirements:
Support for scheduling content with a wide range of absolute and relative
timing constraints.
Support for scheduling content across multiple displays in an atomic fashion
(e.g. display this video on all the displays or none of them).
Support for the rapid introduction of interactive content.
Support for numerous independently developed domain specific schedulers that
can share the display network.
Provide an abstraction layer to free the scheduler developer from concerning
themselves with the various video and audio sources/ sinks and switching
operations these necessitate.
Fully time-based approaches, however, are incapable of determining which content
is more relevant given the current audience and relies solely on the display’s man-
ager ability to anticipate which content best suits the audience at a given time.
If, however, the display is capable of some kind of sensing of its environment, it
can dynamically decide what to present based on the readings from its sensors.
19
the conference site. Flowers represent locations, coloured petals the professional
interests of users and walking ants represent the rate of people walking by the
tracked areas [Cox et al., 2003].
be accomplished using time-based approaches that define the start and duration
for each content item or event- or context-based approaches that take into account
the preferences or historic data about who is near the display. Storz et al. [2006],
for example, when developing a software infrastructure for (essentially time-based)
content scheduling identified the following requirements:
Support for scheduling content with a wide range of absolute and relative
timing constraints.
Support for scheduling content across multiple displays in an atomic fashion
(e.g. display this video on all the displays or none of them).
Support for the rapid introduction of interactive content.
Support for numerous independently developed domain specific schedulers that
can share the display network.
Provide an abstraction layer to free the scheduler developer from concerning
themselves with the various video and audio sources/ sinks and switching
operations these necessitate.
Fully time-based approaches, however, are incapable of determining which content
is more relevant given the current audience and relies solely on the display’s man-
ager ability to anticipate which content best suits the audience at a given time.
If, however, the display is capable of some kind of sensing of its environment, it
can dynamically decide what to present based on the readings from its sensors.
19
Page 37
the Bluetooth names of their mobile phones. The system uses a pre-defined source
for content, the Flickr9 photo sharing website, but allows users to contribute by
specifying keywords and Flicker user IDs that the system uses to search and display
photos (see Figure 2.9). Apart from the fact that users cannot choose the source for
the content, they have a higher degree of control over what is displayed than in a
completely centrally-managed display system. Control is not as high as in an pure
user-submitted display system, however, because a user is not able to display at will
a specific photo. Instant Places achieves a balance between the two, leveraging on
the advantages of both.
Figure 2.9: Instant Places display [Jose´ et al., 2008].
2.2 Interaction in Public Displays
The ability to interact with a display is a fundamental and challenging aspect of
the design of a situated public display. It is fundamental because it gives users the
possibility to do more with the display than just watch it, but it is, at the same
time, a challenge because there are not yet mature enough models for an effective
interaction with public displays.
The challenges are not just technical but, perhaps even more importantly, also social
ones. The question designers of public displays pose is not only “how to provide
interactive features?” or “what mechanisms should be used?” but also “what kind
of interaction makes sense for the public display?” and “how will people use these
features?”.
Public display systems have been designed to support different interactive features,
using a number of different mechanisms. One of the difference between the various
9http://www.flickr.com
21
for content, the Flickr9 photo sharing website, but allows users to contribute by
specifying keywords and Flicker user IDs that the system uses to search and display
photos (see Figure 2.9). Apart from the fact that users cannot choose the source for
the content, they have a higher degree of control over what is displayed than in a
completely centrally-managed display system. Control is not as high as in an pure
user-submitted display system, however, because a user is not able to display at will
a specific photo. Instant Places achieves a balance between the two, leveraging on
the advantages of both.
Figure 2.9: Instant Places display [Jose´ et al., 2008].
2.2 Interaction in Public Displays
The ability to interact with a display is a fundamental and challenging aspect of
the design of a situated public display. It is fundamental because it gives users the
possibility to do more with the display than just watch it, but it is, at the same
time, a challenge because there are not yet mature enough models for an effective
interaction with public displays.
The challenges are not just technical but, perhaps even more importantly, also social
ones. The question designers of public displays pose is not only “how to provide
interactive features?” or “what mechanisms should be used?” but also “what kind
of interaction makes sense for the public display?” and “how will people use these
features?”.
Public display systems have been designed to support different interactive features,
using a number of different mechanisms. One of the difference between the various
9http://www.flickr.com
21
Page 38
mechanisms employed is the number of simultaneous users they support. Some sys-
tem support only single-user interaction, other have been designed to support two,
other support small groups and other still support any number of users simulta-
neously addressing the system. There are also differences on the type and level of
control that is offered to the user. Some systems offer full control over the display,
other support browsing and selecting content, other support submitting content,
voting on displayed content, etc.
2.2.1 Individual Interaction
Some public displays can only be interacted with individually. Usually because the
input device supported is the display itself, for example a touch-sensitive display, or
because it has sensors that are not capable of discriminating between several people
— as some computer vision techniques.
In the Aware Community Portals [Sawhney et al., 2001] the display system is able
to detect movement and the presence of people looking at it. Movement detection is
used to stop the cycling of content (weather, cartoons, etc) and display the current
news item; it also allows the person to read the current item in more detail if he
or she stands for some time. The system is also able to detect faces and link them
to the articles read providing an historic perspective over who read the article and
which articles attracted more attention. The Aware Community Portals gives users
a very low degree of control over the display. Users have no explicit control, they can
only stand and read what the display system determines to show at that moment.
The Community Wall [Grasso et al., 2003] is also able to detect the presence of
people near it. Presence detection is used to prevent the display from passing on
to the next set of items without the user having had the time to inspect the cur-
rent screen. The Community Wall also provides touchscreen interaction and allows
users to read an item in more detail, to rate, comment or print it. Commenting
and rating provide important information to the display because they represent an
important part of the situatedness of the content that the display shows. Plasma
Poster [Churchill et al., 2003] also gave users a high level of control over the contents
of the display by using touch-sensitive screens. As in the Community Wall, users
can read the current item in detail by scrolling or expanding it, print, forward it
to an email address or write a comment about it. Additionally, in Plasma Poster,
users could also browse the current list of content items.
The Hermes office door display by Cheverst et al. [2007] also uses touch-sensitive
displays. Hermes consists of several small digital displays deployed near the office
doors (see Figure 2.10) of the Computing Department at Lancaster University. The
primary purpose of the Hermes office door display was to support coordination of
faculty staff and students by using an enhanced digital equivalent to the “gone
22
tem support only single-user interaction, other have been designed to support two,
other support small groups and other still support any number of users simulta-
neously addressing the system. There are also differences on the type and level of
control that is offered to the user. Some systems offer full control over the display,
other support browsing and selecting content, other support submitting content,
voting on displayed content, etc.
2.2.1 Individual Interaction
Some public displays can only be interacted with individually. Usually because the
input device supported is the display itself, for example a touch-sensitive display, or
because it has sensors that are not capable of discriminating between several people
— as some computer vision techniques.
In the Aware Community Portals [Sawhney et al., 2001] the display system is able
to detect movement and the presence of people looking at it. Movement detection is
used to stop the cycling of content (weather, cartoons, etc) and display the current
news item; it also allows the person to read the current item in more detail if he
or she stands for some time. The system is also able to detect faces and link them
to the articles read providing an historic perspective over who read the article and
which articles attracted more attention. The Aware Community Portals gives users
a very low degree of control over the display. Users have no explicit control, they can
only stand and read what the display system determines to show at that moment.
The Community Wall [Grasso et al., 2003] is also able to detect the presence of
people near it. Presence detection is used to prevent the display from passing on
to the next set of items without the user having had the time to inspect the cur-
rent screen. The Community Wall also provides touchscreen interaction and allows
users to read an item in more detail, to rate, comment or print it. Commenting
and rating provide important information to the display because they represent an
important part of the situatedness of the content that the display shows. Plasma
Poster [Churchill et al., 2003] also gave users a high level of control over the contents
of the display by using touch-sensitive screens. As in the Community Wall, users
can read the current item in detail by scrolling or expanding it, print, forward it
to an email address or write a comment about it. Additionally, in Plasma Poster,
users could also browse the current list of content items.
The Hermes office door display by Cheverst et al. [2007] also uses touch-sensitive
displays. Hermes consists of several small digital displays deployed near the office
doors (see Figure 2.10) of the Computing Department at Lancaster University. The
primary purpose of the Hermes office door display was to support coordination of
faculty staff and students by using an enhanced digital equivalent to the “gone
22
Page 40
Figure 2.11: The Range whiteboard [Ju et al., 2008].
a good solution because they provide a familiar interaction technique and can only
be interacted with at arms length so it is easy to see when the display is being used
(this also naturally helps occlude the personal information being accessed). Blue-
Board [Russell and Gossweiler, 2001], for example, also uses a touchscreen display
to allow access to personal information such as calendar, messages and files and to
allow exchanging this information with other BlueBoard users by dragging items on
the screen. To identify users, BlueBoard uses an electronic badge reader so getting
access to personal information is a matter of swiping the badge on the display.
2.2.2 Group Interaction
Some displays provide group interaction mechanisms, i.e., they allow several users to
collaborate, share or exchange information. These applications are usually designed
for workgroup places as in the case of the Interactive Workspaces [Johanson et al.,
2003] but may be used for more informal collaboration as Brignull et al. [2004] show
in the Dynamo system.
The Stanford University’s Interactive Workspaces project is a classic use of situated
displays as a groupware system. This project consists of a meeting room (iRoom)
augmented with large displays and control software (see Figure 2.12) that allow the
collaboration of different people using different software tools to share the control of
the various displays. The Interactive Workspaces is an example of a display system
that is meant to be used in a workgroup — it is not exactly a public display system
— and so requiring users to use special software and hardware is not a problem and
this means that users can be given a high level of control over the system.
However, even more public and lightweight systems can give users a rich interface.
In Dynamo, for example, students in a school share and exchange multimedia files
using a multi-user display. The system provides a GUI like interface and several
24
a good solution because they provide a familiar interaction technique and can only
be interacted with at arms length so it is easy to see when the display is being used
(this also naturally helps occlude the personal information being accessed). Blue-
Board [Russell and Gossweiler, 2001], for example, also uses a touchscreen display
to allow access to personal information such as calendar, messages and files and to
allow exchanging this information with other BlueBoard users by dragging items on
the screen. To identify users, BlueBoard uses an electronic badge reader so getting
access to personal information is a matter of swiping the badge on the display.
2.2.2 Group Interaction
Some displays provide group interaction mechanisms, i.e., they allow several users to
collaborate, share or exchange information. These applications are usually designed
for workgroup places as in the case of the Interactive Workspaces [Johanson et al.,
2003] but may be used for more informal collaboration as Brignull et al. [2004] show
in the Dynamo system.
The Stanford University’s Interactive Workspaces project is a classic use of situated
displays as a groupware system. This project consists of a meeting room (iRoom)
augmented with large displays and control software (see Figure 2.12) that allow the
collaboration of different people using different software tools to share the control of
the various displays. The Interactive Workspaces is an example of a display system
that is meant to be used in a workgroup — it is not exactly a public display system
— and so requiring users to use special software and hardware is not a problem and
this means that users can be given a high level of control over the system.
However, even more public and lightweight systems can give users a rich interface.
In Dynamo, for example, students in a school share and exchange multimedia files
using a multi-user display. The system provides a GUI like interface and several
24
Page 41
Figure 2.12: The Interactive Workspaces meeting room [Johanson et al., 2002].
Figure 2.13: Dynamo interactive display in use in the communal
room [Brignull et al., 2004].
interaction points composed by wireless keyboards and mice allow several users to
use the same display at the same time (see Figure 2.13).
Both Dynamo and the Interactive Workspaces offer a very high level of control, es-
sentially the same as in a standard desktop interface. However, in many situations
this is not possible nor desirable. Even in the cases were the displays are for public
use, when interaction is for single users or a small group of users, during the interac-
tion duration, the display is perceived as being owned by those that are interacting
with it. As Dix and Sas [2008] note on their design space analysis of public displays
and private devices, interaction with public displays, for public use, is usually done
with individual input devices thus allowing several people to interact with the same
display at the same time. There are many examples of display systems with which
interaction can be performed in this way.
2.2.3 Public Interaction
In some systems, authors have opted to build custom individual devices to interact
with the display. Streitz et al. [2003], for example, followed this approach in the
Hello.Wall ambient display. In order to allow users to be detected by and interact
25
Figure 2.13: Dynamo interactive display in use in the communal
room [Brignull et al., 2004].
interaction points composed by wireless keyboards and mice allow several users to
use the same display at the same time (see Figure 2.13).
Both Dynamo and the Interactive Workspaces offer a very high level of control, es-
sentially the same as in a standard desktop interface. However, in many situations
this is not possible nor desirable. Even in the cases were the displays are for public
use, when interaction is for single users or a small group of users, during the interac-
tion duration, the display is perceived as being owned by those that are interacting
with it. As Dix and Sas [2008] note on their design space analysis of public displays
and private devices, interaction with public displays, for public use, is usually done
with individual input devices thus allowing several people to interact with the same
display at the same time. There are many examples of display systems with which
interaction can be performed in this way.
2.2.3 Public Interaction
In some systems, authors have opted to build custom individual devices to interact
with the display. Streitz et al. [2003], for example, followed this approach in the
Hello.Wall ambient display. In order to allow users to be detected by and interact
25
Page 46
vide such a rich interaction experience as using a custom built mobile application
or device, but it has the advantage of lowering the barriers to interaction to the
minimum. Given that most people now own a mobile phone or other mobile device,
only very few will not be able to use the display system this way.
2.3 Human factors
The idea of a public, socially situated display is a very powerful one. It has the
potential to allow the creation of new display applications and new social practices
and to turn displays into a more efficient medium for advertising and message dis-
semination. However, for this to happen, people must look and interact with the
display, and, as Agamanolis [2003, p. 19] has noticed, “Half the battle in designing
an interactive situated or public display is designing how the display will invite that
interaction.” It is not just a matter of offering interactive features to the public.
Those features must offer an obvious value and, in some cases, the display system
must explicitly or implicitly invite people to interact.
Brignull and Rogers [2003] did a study on how people socialize around large dis-
plays and drew some considerations about how to encourage people to participate.
In their observations of the Opinionizer — a system for posting comments about
a topic, in a social gathering, by typing some words in a regular keyboard — they
noticed a “honey-pot effect”. Because interaction with the Opinionizer could only
be done in a single place — the laptop provided for the effect — people would
gather around that single spot creating an interest area where others would try to
get close to to see what the buzz was about. This increased the number of interac-
tions with the system. They also noted that people would approach the Opinionizer
in phases: first becoming peripherally aware of the display; then focusing and so-
cializing about it; and then actively interacting with it. To move between these
phases people need encouragement to cross the thresholds, principally to active par-
ticipation. They also provide some advice on how to design displays that encourage
participation [Brignull and Rogers, 2003, p. 7]:
For example, a lengthy registration process involving form filling is well
known to put people off taking part. The form of interaction needs to be
very lightweight and visible from the offset; it should be easy to do and
importantly, not embarrassing to recover from mistakes that are made.
Participants need to be able to learn how to interact with the system
vicariously, rather than be told or have to follow a set of instructions.
They need to be able to simply walk up and use it, having watched others
do the same. The interface needs to be clear to the person such that
they are reassured that their interaction with it will be a low commitment
activity, that will be quick to do and enjoyable.
30
or device, but it has the advantage of lowering the barriers to interaction to the
minimum. Given that most people now own a mobile phone or other mobile device,
only very few will not be able to use the display system this way.
2.3 Human factors
The idea of a public, socially situated display is a very powerful one. It has the
potential to allow the creation of new display applications and new social practices
and to turn displays into a more efficient medium for advertising and message dis-
semination. However, for this to happen, people must look and interact with the
display, and, as Agamanolis [2003, p. 19] has noticed, “Half the battle in designing
an interactive situated or public display is designing how the display will invite that
interaction.” It is not just a matter of offering interactive features to the public.
Those features must offer an obvious value and, in some cases, the display system
must explicitly or implicitly invite people to interact.
Brignull and Rogers [2003] did a study on how people socialize around large dis-
plays and drew some considerations about how to encourage people to participate.
In their observations of the Opinionizer — a system for posting comments about
a topic, in a social gathering, by typing some words in a regular keyboard — they
noticed a “honey-pot effect”. Because interaction with the Opinionizer could only
be done in a single place — the laptop provided for the effect — people would
gather around that single spot creating an interest area where others would try to
get close to to see what the buzz was about. This increased the number of interac-
tions with the system. They also noted that people would approach the Opinionizer
in phases: first becoming peripherally aware of the display; then focusing and so-
cializing about it; and then actively interacting with it. To move between these
phases people need encouragement to cross the thresholds, principally to active par-
ticipation. They also provide some advice on how to design displays that encourage
participation [Brignull and Rogers, 2003, p. 7]:
For example, a lengthy registration process involving form filling is well
known to put people off taking part. The form of interaction needs to be
very lightweight and visible from the offset; it should be easy to do and
importantly, not embarrassing to recover from mistakes that are made.
Participants need to be able to learn how to interact with the system
vicariously, rather than be told or have to follow a set of instructions.
They need to be able to simply walk up and use it, having watched others
do the same. The interface needs to be clear to the person such that
they are reassured that their interaction with it will be a low commitment
activity, that will be quick to do and enjoyable.
30
Page 47
The honey-pot effect observed by Brignull and Rogers [2003] was mainly due to
the fact that people had only one input device and so had to gather around it,
creating a “buzz”. As they noticed, giving users the possibility of sending comments
remotely might decrease the social awkwardness in getting in front of the keyboard
and typing while everyone else was watching, but it would also remove the honey-pot
effect. However, there are other ways to achieve a similar effect such as exposing
others’ interactions and having a dedicated core group of users as Huang et al. [2006]
suggests.
Huang et al. [2006] surveyed a set of large-display groupware systems to find the
main factors that lead to the success or failure in usage of the systems. Seven group-
ware systems were analyzed: Notification Collage [Greenberg and Rounding, 2001],
MessyBoard [Fass et al., 2002], Plasma Poster [Churchill et al., 2003], Semi-Public
Displays [Huang and Mynatt, 2003], BlueBoard [Russell and Gossweiler, 2001], MER-
Board [Trimble et al., 2003] and Awareness Module [Huang et al., 2002]. Although
the focus of the study were groupware systems, almost all factors that affect the suc-
cess of the display system can be transposed to generic, public, interactive display
systems.
Huang et al. [2006] propose five recommendations:
1. Task specificity and integration. This suggests that designers should try to
integrate new groupware systems into existing practices rather than trying
to create new practices around the display system. Users should perceive
some obvious value to system and for this to happen they should immediately
see that something they already do can be made faster, or better, using the
groupware system. Although not directly applicable to systems that are not
meant to be used as a tool, other display systems can also benefit from having
a clear focus, even if they are meant to support a wide range of uses.
2. Tool flexibility and generality. Although designed with a specific goal in mind,
display systems should be flexible and allow for multiple practices. Each user
should be able to appropriate the system in a personal manner reinforcing the
perceived value that the system offers.
3. Visibility and exposure to others’ interactions. This and the next recom-
mendation are roughly equivalent to trying to generate the honey-pot ef-
fect described by Brignull and Rogers [2003]. “Users often discovered po-
tential uses for the system after observing other users interacting with the
display.” [Huang et al., 2006]. Adapting this recommendation for public in-
teractive displays requires some care. When interacting with public display,
users may be among strangers. People are much more inhibited to interact in
public places, so interaction with public displays should be designed in a way
that allows users to interact without being exposed. So, while not necessarily
exposing others while interacting, public display systems can still try to expose
31
the fact that people had only one input device and so had to gather around it,
creating a “buzz”. As they noticed, giving users the possibility of sending comments
remotely might decrease the social awkwardness in getting in front of the keyboard
and typing while everyone else was watching, but it would also remove the honey-pot
effect. However, there are other ways to achieve a similar effect such as exposing
others’ interactions and having a dedicated core group of users as Huang et al. [2006]
suggests.
Huang et al. [2006] surveyed a set of large-display groupware systems to find the
main factors that lead to the success or failure in usage of the systems. Seven group-
ware systems were analyzed: Notification Collage [Greenberg and Rounding, 2001],
MessyBoard [Fass et al., 2002], Plasma Poster [Churchill et al., 2003], Semi-Public
Displays [Huang and Mynatt, 2003], BlueBoard [Russell and Gossweiler, 2001], MER-
Board [Trimble et al., 2003] and Awareness Module [Huang et al., 2002]. Although
the focus of the study were groupware systems, almost all factors that affect the suc-
cess of the display system can be transposed to generic, public, interactive display
systems.
Huang et al. [2006] propose five recommendations:
1. Task specificity and integration. This suggests that designers should try to
integrate new groupware systems into existing practices rather than trying
to create new practices around the display system. Users should perceive
some obvious value to system and for this to happen they should immediately
see that something they already do can be made faster, or better, using the
groupware system. Although not directly applicable to systems that are not
meant to be used as a tool, other display systems can also benefit from having
a clear focus, even if they are meant to support a wide range of uses.
2. Tool flexibility and generality. Although designed with a specific goal in mind,
display systems should be flexible and allow for multiple practices. Each user
should be able to appropriate the system in a personal manner reinforcing the
perceived value that the system offers.
3. Visibility and exposure to others’ interactions. This and the next recom-
mendation are roughly equivalent to trying to generate the honey-pot ef-
fect described by Brignull and Rogers [2003]. “Users often discovered po-
tential uses for the system after observing other users interacting with the
display.” [Huang et al., 2006]. Adapting this recommendation for public in-
teractive displays requires some care. When interacting with public display,
users may be among strangers. People are much more inhibited to interact in
public places, so interaction with public displays should be designed in a way
that allows users to interact without being exposed. So, while not necessarily
exposing others while interacting, public display systems can still try to expose
31
Page 48
the fact that someone has just interacted and maybe in what way. In both
cases, groupware systems or public systems, the objective to let others know
that the system is in use. In groupware systems it is possible to let other know
how it can be used. In public system, its more sensible just to let others know
that the system has been used (and maybe what was done).
4. Dedicated core group of users. There needs to be a group of people that uses
the system regularly to keep it alive and show others that the system is in use
and that they too can use it. This is related to the “visibility and exposure
to others’ interactions”. In order for this visibility and exposure to occur
in the beginning, there must be a core group of users that uses the system
regularly. This may be less of a problem with public interactive displays due
to the number of users that will potentially use the system, but might still be
something to consider in some cases.
5. Low barriers to use. Interaction should be lightweight and the steps needed to
use the system should not detract users. For public interactive displays this
might mean providing an opportunistic form of interaction — users should be
able to just step up to the display and use it. Other more complex interaction
forms may also exist, for more experienced users, for example, but a first time
user should still be able to accomplish something with the display in an easy
manner.
Some of these recommendations have also been noticed by Brignull et al. [2004] in
their observations of the Dynamo system. As they suggest when discussing the
implications for design learned from the Dynamo experience [Brignull et al., 2004,
p. 9]:
The interactive display should fit in and be able to be integrated with
the other artefacts used in the space. (...)
Provide flexibility both in terms of physical and digital arrangements.
(...)
Design interactive applications that the community can adapt to their
own activities. (...)
Provide an initial set of display-based interactions that are intuitive
and can be easily and comfortably followed. (...)
Another study by Huang et al. [2008] focused on public large public display usage
and their aim was to uncover practices associated with how people look at public
displays in various settings. They reported a field study that took place in 24 Central
European cities, where 46 public displays where observed.
They summarized their findings according to the various factors that affect the way
people look at displays [Huang et al., 2008]:
32
cases, groupware systems or public systems, the objective to let others know
that the system is in use. In groupware systems it is possible to let other know
how it can be used. In public system, its more sensible just to let others know
that the system has been used (and maybe what was done).
4. Dedicated core group of users. There needs to be a group of people that uses
the system regularly to keep it alive and show others that the system is in use
and that they too can use it. This is related to the “visibility and exposure
to others’ interactions”. In order for this visibility and exposure to occur
in the beginning, there must be a core group of users that uses the system
regularly. This may be less of a problem with public interactive displays due
to the number of users that will potentially use the system, but might still be
something to consider in some cases.
5. Low barriers to use. Interaction should be lightweight and the steps needed to
use the system should not detract users. For public interactive displays this
might mean providing an opportunistic form of interaction — users should be
able to just step up to the display and use it. Other more complex interaction
forms may also exist, for more experienced users, for example, but a first time
user should still be able to accomplish something with the display in an easy
manner.
Some of these recommendations have also been noticed by Brignull et al. [2004] in
their observations of the Dynamo system. As they suggest when discussing the
implications for design learned from the Dynamo experience [Brignull et al., 2004,
p. 9]:
The interactive display should fit in and be able to be integrated with
the other artefacts used in the space. (...)
Provide flexibility both in terms of physical and digital arrangements.
(...)
Design interactive applications that the community can adapt to their
own activities. (...)
Provide an initial set of display-based interactions that are intuitive
and can be easily and comfortably followed. (...)
Another study by Huang et al. [2008] focused on public large public display usage
and their aim was to uncover practices associated with how people look at public
displays in various settings. They reported a field study that took place in 24 Central
European cities, where 46 public displays where observed.
They summarized their findings according to the various factors that affect the way
people look at displays [Huang et al., 2008]:
32
Page 49
Brevity of glances
Assume that viewers are not willing to spend more than a few
seconds to determine whether a display is of interest.
If the intent of the content is to be informative, present it in
such a way that the most important information be determined
in 2-3 seconds.
Avoid using more than minimal text; even two or three brief
sentences are not likely to be read.
Positioning of displays
When possible, position displays close to eye-height to encour-
age glances.
If theft or vandalism are concerns, consider other ways to pro-
tect a display or make it inaccessible than putting it above arm’s
reach.
Content format and dynamics
Make content continually dynamic to keep user attention longer.
Avoid abrupt changes in content to encourage continued view-
ing.
Design to give users some degree of control over what informa-
tion to view.
Catching the eye
Consider the direction of people’s movement within a space
when deciding where to situate displays.
When choosing where to situate displays, take advantage of
other objects in the environment to draw attention to displays,
rather than relying on the large display to be the eye-catcher.
When possible, consider ways in which the area surrounding
the large display can be enhanced to maximize attention and
increase the chances of glancing.
Small displays vs. large displays
Design to balance feelings of exposure and privacy within a pub-
lic space by considering multiple display sizes and how they af-
fect the viewer experience, perception, and comfort.
Some of these factors are external to the display itself and may not be controllable
by the designer of display applications. Factors such as positioning, direction and
size may not be easily controlled when designing applications for existing locations.
Other factors, however, are internal and should be considered when designing display
applications. Providing dynamic content, some control over that content and not
assuming people will read lots of text may affect the acceptance of the display
system.
33
Assume that viewers are not willing to spend more than a few
seconds to determine whether a display is of interest.
If the intent of the content is to be informative, present it in
such a way that the most important information be determined
in 2-3 seconds.
Avoid using more than minimal text; even two or three brief
sentences are not likely to be read.
Positioning of displays
When possible, position displays close to eye-height to encour-
age glances.
If theft or vandalism are concerns, consider other ways to pro-
tect a display or make it inaccessible than putting it above arm’s
reach.
Content format and dynamics
Make content continually dynamic to keep user attention longer.
Avoid abrupt changes in content to encourage continued view-
ing.
Design to give users some degree of control over what informa-
tion to view.
Catching the eye
Consider the direction of people’s movement within a space
when deciding where to situate displays.
When choosing where to situate displays, take advantage of
other objects in the environment to draw attention to displays,
rather than relying on the large display to be the eye-catcher.
When possible, consider ways in which the area surrounding
the large display can be enhanced to maximize attention and
increase the chances of glancing.
Small displays vs. large displays
Design to balance feelings of exposure and privacy within a pub-
lic space by considering multiple display sizes and how they af-
fect the viewer experience, perception, and comfort.
Some of these factors are external to the display itself and may not be controllable
by the designer of display applications. Factors such as positioning, direction and
size may not be easily controlled when designing applications for existing locations.
Other factors, however, are internal and should be considered when designing display
applications. Providing dynamic content, some control over that content and not
assuming people will read lots of text may affect the acceptance of the display
system.
33
Page 54
some high-level tasks can be devised:
1. Digital footprints framework: Define a set of digital footprints that the display
system can collect from presence sensing and direct user interaction. Charac-
terise the nature of the data associated with those digital footprints.
(a) Publication: technical report and paper.
2. Study and adapt existing interface principles, guidelines and frameworks to
public display interfaces.
(a) Publication: technical report and paper.
3. Keywords experiment (see detailed description in section 4.3.1).
(a) Develop software support for keywords experiment.
(b) Deploy experiment.
(c) Publication: results from the experiment.
4. Study other experiments.
(a) Develop the experiment.
(b) Develop software support, for the Instant Places platform, for the various
kinds of interactions necessary for the experiment.
(c) Deploy the experiment.
(d) Collect and analyse the resulting data.
(e) Publication: results from each experiment.
5. Write thesis. This will be an ongoing task.
(a) Publication: integration of the various results into a set of guidelines and
principles for mixed-initiative interaction with public displays.
Figure 3.1 shows an approximate task start and duration.
3.3 Expected results
This work is expected to yield results specifically related to the proposed objectives
but other outcomes are also expected to arise.
The mains result is the determination of whether this concept of public situated
display really accomplishes the goal of creating a shared display object that people
38
1. Digital footprints framework: Define a set of digital footprints that the display
system can collect from presence sensing and direct user interaction. Charac-
terise the nature of the data associated with those digital footprints.
(a) Publication: technical report and paper.
2. Study and adapt existing interface principles, guidelines and frameworks to
public display interfaces.
(a) Publication: technical report and paper.
3. Keywords experiment (see detailed description in section 4.3.1).
(a) Develop software support for keywords experiment.
(b) Deploy experiment.
(c) Publication: results from the experiment.
4. Study other experiments.
(a) Develop the experiment.
(b) Develop software support, for the Instant Places platform, for the various
kinds of interactions necessary for the experiment.
(c) Deploy the experiment.
(d) Collect and analyse the resulting data.
(e) Publication: results from each experiment.
5. Write thesis. This will be an ongoing task.
(a) Publication: integration of the various results into a set of guidelines and
principles for mixed-initiative interaction with public displays.
Figure 3.1 shows an approximate task start and duration.
3.3 Expected results
This work is expected to yield results specifically related to the proposed objectives
but other outcomes are also expected to arise.
The mains result is the determination of whether this concept of public situated
display really accomplishes the goal of creating a shared display object that people
38
Page 55
Y
ea
r
1
Y
ea
r
2
Y
ea
r
3
Y
ea
r
4
P
D
ST
I
C
la
ss
es
D
ig
it
al
fo
ot
pr
in
ts
fr
am
ew
or
k
P
ub
:
U
I
fr
am
ew
or
k U
I
fr
am
ew
or
k
P
ub
:
in
te
ra
ct
io
n
fr
am
ew
or
k
D
ev
el
op
so
ft
w
ar
e
su
pp
or
t
fo
r
ke
yw
or
ds
ex
p
er
im
en
t
D
ep
lo
y
ke
yw
or
ds
ex
p
er
im
en
t
P
ub
:
ke
yw
or
ds
ex
p
er
im
en
t
St
ud
y
an
d
im
pl
em
en
t
ot
he
r
ex
p
er
im
en
ts
P
ub
:
ex
p
er
im
en
ts
’
re
su
lt
s
W
ri
te
th
es
is
P
ub
:
in
te
gr
at
io
n
of
th
e
va
ri
ou
s
re
su
lt
s
Figure 3.1: Gantt chart for the planned tasks within the doctoral program.
39
ea
r
1
Y
ea
r
2
Y
ea
r
3
Y
ea
r
4
P
D
ST
I
C
la
ss
es
D
ig
it
al
fo
ot
pr
in
ts
fr
am
ew
or
k
P
ub
:
U
I
fr
am
ew
or
k U
I
fr
am
ew
or
k
P
ub
:
in
te
ra
ct
io
n
fr
am
ew
or
k
D
ev
el
op
so
ft
w
ar
e
su
pp
or
t
fo
r
ke
yw
or
ds
ex
p
er
im
en
t
D
ep
lo
y
ke
yw
or
ds
ex
p
er
im
en
t
P
ub
:
ke
yw
or
ds
ex
p
er
im
en
t
St
ud
y
an
d
im
pl
em
en
t
ot
he
r
ex
p
er
im
en
ts
P
ub
:
ex
p
er
im
en
ts
’
re
su
lt
s
W
ri
te
th
es
is
P
ub
:
in
te
gr
at
io
n
of
th
e
va
ri
ou
s
re
su
lt
s
Figure 3.1: Gantt chart for the planned tasks within the doctoral program.
39
Page 56
recognize as a reference in a place. Also, we expect to determine if the path to achieve
the concept is capable of generating a display that adapts itself and is capable of
displaying content that people recognize as relevant and content that people can
relate to in a meaningful way.
Directly related to the objectives is also the characterisation of a set of digital
footprints and the way they can be generated and used to a enable a dynamic
characterisation of a place. This work will also provide some insight on the type of
place model that these footprints are able to support.
We also expect to contribute some design principles and guidelines for the interactive
features that a public display should support and how they should be implemented
in order to provide an easy user experience. This will help display designers to take
more informed decisions instead of having to think and develop everything from
the start. We do not expect to create the next user interface paradigm for public
displays but give a contribution in that direction by developing knowledge about
what interactive features work best for public displays and how people perceive and
use them.
Another natural result for this work is knowledge about the main factors that affect
the deployment of public interactive displays in real settings and how those factors
should be anticipated and dealt with to obtain a successful public display that is
used and perceived as valuable by people.
The motivation for this work is, in the long run, to develop a new concept of situ-
ated display that not only increases the efficiency of the display as an information
dissemination tool but also that allows new uses for public displays. A potential
result of this work is the creation of new display applications and the development
of new paradigms for existing applications. Advertising, for example, is something
that displays have always been applied to. However, the new contextual advertising
model that prevails in the Web has not been translated successfully to the public
display arena. Even though it might not be possible to take full advantage of the
Web advertising model in public displays, it is certainly possible to go further down
that line than the current state. This work has the potential to shed some light in
that path by studying the characterisation techniques that may allow a display to
automatically assign keywords to a place over time, much like the Web advertisement
systems do for visited web pages.
Another area in which developments may be expected is in entertainment. This
display concept has the potential to promote the development of new kinds of public
games or more ludic applications such as new forms of narrative experiences that
take advantage of the adaptive and interactive features offered by the display system
and the possibility to automatically characterise its audience.
40
the concept is capable of generating a display that adapts itself and is capable of
displaying content that people recognize as relevant and content that people can
relate to in a meaningful way.
Directly related to the objectives is also the characterisation of a set of digital
footprints and the way they can be generated and used to a enable a dynamic
characterisation of a place. This work will also provide some insight on the type of
place model that these footprints are able to support.
We also expect to contribute some design principles and guidelines for the interactive
features that a public display should support and how they should be implemented
in order to provide an easy user experience. This will help display designers to take
more informed decisions instead of having to think and develop everything from
the start. We do not expect to create the next user interface paradigm for public
displays but give a contribution in that direction by developing knowledge about
what interactive features work best for public displays and how people perceive and
use them.
Another natural result for this work is knowledge about the main factors that affect
the deployment of public interactive displays in real settings and how those factors
should be anticipated and dealt with to obtain a successful public display that is
used and perceived as valuable by people.
The motivation for this work is, in the long run, to develop a new concept of situ-
ated display that not only increases the efficiency of the display as an information
dissemination tool but also that allows new uses for public displays. A potential
result of this work is the creation of new display applications and the development
of new paradigms for existing applications. Advertising, for example, is something
that displays have always been applied to. However, the new contextual advertising
model that prevails in the Web has not been translated successfully to the public
display arena. Even though it might not be possible to take full advantage of the
Web advertising model in public displays, it is certainly possible to go further down
that line than the current state. This work has the potential to shed some light in
that path by studying the characterisation techniques that may allow a display to
automatically assign keywords to a place over time, much like the Web advertisement
systems do for visited web pages.
Another area in which developments may be expected is in entertainment. This
display concept has the potential to promote the development of new kinds of public
games or more ludic applications such as new forms of narrative experiences that
take advantage of the adaptive and interactive features offered by the display system
and the possibility to automatically characterise its audience.
40
Page 57
3.3.1 Publication plan
An implicit task is the publication of results from the studies and experiments.
Although it is hard to list the specific potential publications that will result from
this work, we plan on publishing technical reports and papers on the following,
general, topics:
The characterisation of the digital footprints that can result from various in-
teractions and how they can be used by situated displays.
The interaction framework for public displays resulting from the study and
adaptation of existing interface design principles, guidelines and patterns.
Results from the acceptance of the various types of shared control mechanisms
and system initiative approaches.
Results from the acceptance of the place model as generator for situated con-
tent.
Design guidelines that result from the various experiments.
Possible conferences and journals where results from this research work could be
published include:
Conferences
MobiQuitous — Annual International Conference on Mobile and Ubiqui-
tous Systems
Ubicomp — International Conference on Ubiquitous Computing
ICUT — International Conference on Ubiquitous Information Technolo-
gies & Applications
MUM — International Conference on Mobile and Ubiquitous Multimedia
PICom — IEEE International Conference on Pervasive Intelligence and
Computing
IMIS — International Workshop on Intelligent, Mobile and Internet Ser-
vices in Ubiquitous Computing
CHI — Computer Human Interaction Conference
PerCom — IEEE International Conference on Pervasive Computing and
Communications
Pervasive — International Conference on Pervasive Computing
41
An implicit task is the publication of results from the studies and experiments.
Although it is hard to list the specific potential publications that will result from
this work, we plan on publishing technical reports and papers on the following,
general, topics:
The characterisation of the digital footprints that can result from various in-
teractions and how they can be used by situated displays.
The interaction framework for public displays resulting from the study and
adaptation of existing interface design principles, guidelines and patterns.
Results from the acceptance of the various types of shared control mechanisms
and system initiative approaches.
Results from the acceptance of the place model as generator for situated con-
tent.
Design guidelines that result from the various experiments.
Possible conferences and journals where results from this research work could be
published include:
Conferences
MobiQuitous — Annual International Conference on Mobile and Ubiqui-
tous Systems
Ubicomp — International Conference on Ubiquitous Computing
ICUT — International Conference on Ubiquitous Information Technolo-
gies & Applications
MUM — International Conference on Mobile and Ubiquitous Multimedia
PICom — IEEE International Conference on Pervasive Intelligence and
Computing
IMIS — International Workshop on Intelligent, Mobile and Internet Ser-
vices in Ubiquitous Computing
CHI — Computer Human Interaction Conference
PerCom — IEEE International Conference on Pervasive Computing and
Communications
Pervasive — International Conference on Pervasive Computing
41
Page 59
resources (computers and Bluetooth dongles) is partially available (computers for
the UCP site are not yet available but should be easy to acquire either through the
research project’s fundings or through the UCP’s budget).
43
the UCP site are not yet available but should be easy to acquire either through the
research project’s fundings or through the UCP’s budget).
43
Page 64
Presence characterisation The second level of presence information is the abil-
ity to characterise presence. This may involve determining how many people are
near the display or inferring some type of characteristic about viewers, such as age
or gender. Periods of high activity or low activity in a place, or the presence of
people with specific characteristics, can all be used to trigger specific content in the
display.
Commercial people counters [Wikipedia, 2009] that count the number of people en-
tering/exiting a room can be used by a display system to estimate the number of
people nearby. Computer vision techniques such as face detection, gender classifi-
cation [Verschae et al., 2008] and age classification [Kwon and Vitoria Lobo, 1999],
used by some audience metering products [Quividi, 2009], can also be used to char-
acterise and estimate the number of people in front of a display. These audience
metering products can deliver reports about the number, attention span, gender and
age of the viewers of a particular display.
Presence characterisation generates a richer description of people flow. The display
system is able not only to determine periods of presence/absence, it also becomes
able to characterise the changes in the number and type of viewers.
Presence Identification Presence identification corresponds to the ability to de-
tect unique identities in the presences. Determining who is present, in the sense that
the display system is able to determine that the same person is present in different
occasions, gives the display system, not only the possibility to determine how many
people are present, but also to establish a correlation between different people or
groups of people. This may be achieved through face recognition techniques, but the
most common approach is by far the use of some personal device (with Bluetooth
or RFID capabilities, for example) as a proxy for the person.
Bluetooth has been used extensively as presence detection mechanism since many
people already own a Bluetooth enabled mobile phone. The BluScreen system [Sharifi et al.,
2006] uses Bluetooth detection to avoid showing advertisements to users more than
once. The Cityware project [Kostakos and O’Neill, 2008b] explored several ways
in which to analyse Bluetooth mobility traces, including a set of in situ visualiza-
tions about Bluetooth presences [Kostakos and O’Neill, 2008a]. These visualisations
provide people with information about current or recent Bluetooth presences.
Radio Frequency Identification (RFID) tags can also be used for presence identifica-
tion. In the IntelliBadge project [Cox et al., 2003], users participating in a confer-
ence were given RFID augmented badges that were used to track them through the
conference rooms. A display at the conference cycled through several visualizations
of the resulting data. RFID tags have the advantage that they are small and can
be incorporated into many existing artifacts. In situations such as a conference, as
in the IntelliBadge system, where people are already obliged to wear a badge, this
48
ity to characterise presence. This may involve determining how many people are
near the display or inferring some type of characteristic about viewers, such as age
or gender. Periods of high activity or low activity in a place, or the presence of
people with specific characteristics, can all be used to trigger specific content in the
display.
Commercial people counters [Wikipedia, 2009] that count the number of people en-
tering/exiting a room can be used by a display system to estimate the number of
people nearby. Computer vision techniques such as face detection, gender classifi-
cation [Verschae et al., 2008] and age classification [Kwon and Vitoria Lobo, 1999],
used by some audience metering products [Quividi, 2009], can also be used to char-
acterise and estimate the number of people in front of a display. These audience
metering products can deliver reports about the number, attention span, gender and
age of the viewers of a particular display.
Presence characterisation generates a richer description of people flow. The display
system is able not only to determine periods of presence/absence, it also becomes
able to characterise the changes in the number and type of viewers.
Presence Identification Presence identification corresponds to the ability to de-
tect unique identities in the presences. Determining who is present, in the sense that
the display system is able to determine that the same person is present in different
occasions, gives the display system, not only the possibility to determine how many
people are present, but also to establish a correlation between different people or
groups of people. This may be achieved through face recognition techniques, but the
most common approach is by far the use of some personal device (with Bluetooth
or RFID capabilities, for example) as a proxy for the person.
Bluetooth has been used extensively as presence detection mechanism since many
people already own a Bluetooth enabled mobile phone. The BluScreen system [Sharifi et al.,
2006] uses Bluetooth detection to avoid showing advertisements to users more than
once. The Cityware project [Kostakos and O’Neill, 2008b] explored several ways
in which to analyse Bluetooth mobility traces, including a set of in situ visualiza-
tions about Bluetooth presences [Kostakos and O’Neill, 2008a]. These visualisations
provide people with information about current or recent Bluetooth presences.
Radio Frequency Identification (RFID) tags can also be used for presence identifica-
tion. In the IntelliBadge project [Cox et al., 2003], users participating in a confer-
ence were given RFID augmented badges that were used to track them through the
conference rooms. A display at the conference cycled through several visualizations
of the resulting data. RFID tags have the advantage that they are small and can
be incorporated into many existing artifacts. In situations such as a conference, as
in the IntelliBadge system, where people are already obliged to wear a badge, this
48
Page 65
may be a good choice. Bluetooth, on the other hand, is a very widely deployed
technology and many mobile-phones are already Bluetooth enabled. This means
that it is possible to use the Bluetooth discovery features to detect presence without
requiring the user to carry any additional object (as with the RFID tags), as most
people already carry a mobile phone regularly. Also, Bluetooth allows the user to
manage his presence by turning it on or off at will.
Presence self-exposure
Self-exposure corresponds to the ability of the display to obtain information about
the interests, preferences or activities of nearby people. This type of knowledge
about the people that use a place may enable the display to adapt itself to their
expectations and preferences. For this to happen, users must be willing to let the
display system know something about them. This personal information can take
many forms: it may be a reference to a user’s personal webpage, a set of user associ-
ated tags, the username for some social sharing website, a set of interest categories
or even personal information, such as age and gender.
The most common approach for supporting presence self-exposure combines presence
identification with the a priori definition of a user profile that becomes associated
with the identity. This approach was used in the Proactive Displays [McDonald et al.,
2008], were users attending a conference registered their affiliation, interests and per-
sonal webpage before the conference day and were given RFID augmented conference
badges at the conference site. In this system, the user does not have easy access to
their information in order to update it which means that they have less control over
what information the display system uses in a given moment.
Another way to achieve this self-exposure is to use an information device (e.g. mo-
bile phone) with a custom application that allows users to register a profile. This
application can connect automatically, or on demand, to the display system and
communicate users’ preferences. One example of this is Camera-Phone [Toye et al.,
2004], where a custom mobile application is used to interact with public displays.
This application may be configured with personal information that is made auto-
matically available to the display system when a user interacts with the display. One
advantage of this approach is that the information is always available to be updated
by its owner.
Bluetooth naming, as described in [Jose´ et al., 2008], is yet another alternative for
managing self-exposure. Naming is explored to allow users to enter predefined com-
mands in their mobile phone Bluetooth name. Since these names can be read by
any other Bluetooth device, this can be used to provide an opportunistic interac-
tion mechanism to any user since there is no need to install an application. This
approach, however, is less suited for private information since anybody can read the
49
technology and many mobile-phones are already Bluetooth enabled. This means
that it is possible to use the Bluetooth discovery features to detect presence without
requiring the user to carry any additional object (as with the RFID tags), as most
people already carry a mobile phone regularly. Also, Bluetooth allows the user to
manage his presence by turning it on or off at will.
Presence self-exposure
Self-exposure corresponds to the ability of the display to obtain information about
the interests, preferences or activities of nearby people. This type of knowledge
about the people that use a place may enable the display to adapt itself to their
expectations and preferences. For this to happen, users must be willing to let the
display system know something about them. This personal information can take
many forms: it may be a reference to a user’s personal webpage, a set of user associ-
ated tags, the username for some social sharing website, a set of interest categories
or even personal information, such as age and gender.
The most common approach for supporting presence self-exposure combines presence
identification with the a priori definition of a user profile that becomes associated
with the identity. This approach was used in the Proactive Displays [McDonald et al.,
2008], were users attending a conference registered their affiliation, interests and per-
sonal webpage before the conference day and were given RFID augmented conference
badges at the conference site. In this system, the user does not have easy access to
their information in order to update it which means that they have less control over
what information the display system uses in a given moment.
Another way to achieve this self-exposure is to use an information device (e.g. mo-
bile phone) with a custom application that allows users to register a profile. This
application can connect automatically, or on demand, to the display system and
communicate users’ preferences. One example of this is Camera-Phone [Toye et al.,
2004], where a custom mobile application is used to interact with public displays.
This application may be configured with personal information that is made auto-
matically available to the display system when a user interacts with the display. One
advantage of this approach is that the information is always available to be updated
by its owner.
Bluetooth naming, as described in [Jose´ et al., 2008], is yet another alternative for
managing self-exposure. Naming is explored to allow users to enter predefined com-
mands in their mobile phone Bluetooth name. Since these names can be read by
any other Bluetooth device, this can be used to provide an opportunistic interac-
tion mechanism to any user since there is no need to install an application. This
approach, however, is less suited for private information since anybody can read the
49
Page 68
not being displayed. If the user asks to skip the current item, the display system
can infer that the user does not find that item particularly interesting and instead
wants to see what is next. If the display shows a list of scheduled items to appear
next and the user is able to skip to a particular item the display system can infer
interest on that particular item. Content control can be achieved by any selection
mechanism. A touch screen is an obvious choice for this. Both Hermes Photo Dis-
play and Plasma Poster [Churchill et al., 2004] use a touch screen interface to let
users navigate their content. Jukola [O’Hara et al., 2004] also uses a touch screen,
but in this case content is indirectly controlled through voting: users of a bar have
the possiblity to vote on the next music to be played by selecting a music, among a
list. Other selection mechanisms such as the one used by Snap and Grab or Content
Cascade could be used for this purpose. Text commands sent by SMS, email or IM,
as in the In WebGlance [Paek et al., 2004] system where users send an email IM
message to the display system with a number corresponding to an option on the
could also be used.
Content control is also a way to collect users’ implicit interest on an item, similarly
to what happens with content download.
Rating By rating an item, the user is explicitly saying the he likes or dislikes that
item, depending on the value of the rating. This is a way for the display system
to allow a user to explicitly indicate his preferences. Rating is found on many
websites such as Youtube, Lastfm, Amazon, etc. On public displays, rating can be
implemented using any selection mechanism or through text commands.
Voting Displays can also collect users’ preferences by crafting polls which allow it
to extract information directly or indirectly from an individual. For example, sports
preferences of a user can be estimated by asking him to vote on his preferred athlete
from a list of athletes from different sports.
As with rating, voting can be accomplished through many different interaction mech-
anisms. As an example, Bluevote [Bortenschlager and Rehrl, 2007] uses images push
via Bluetooth. In this case the selection command is a picture sent previously by the
display system (by pushing the images to all discoverable Bluetooth devices). Users
send back to the system the picture that corresponds to their vote. Bluevote was
used in a conference setting to allow participants to vote on the best paper award.
Classification Classification is of a different nature than the previous categories
because the result is not a preference but the association of a description or keywords,
for example, with a given content item. This can be a less natural action for a user,
especially for public items, but it can be provided by displays in a more ludic per-
spective following the approach of Games With a Purpose [von Ahn and Dabbish,
52
can infer that the user does not find that item particularly interesting and instead
wants to see what is next. If the display shows a list of scheduled items to appear
next and the user is able to skip to a particular item the display system can infer
interest on that particular item. Content control can be achieved by any selection
mechanism. A touch screen is an obvious choice for this. Both Hermes Photo Dis-
play and Plasma Poster [Churchill et al., 2004] use a touch screen interface to let
users navigate their content. Jukola [O’Hara et al., 2004] also uses a touch screen,
but in this case content is indirectly controlled through voting: users of a bar have
the possiblity to vote on the next music to be played by selecting a music, among a
list. Other selection mechanisms such as the one used by Snap and Grab or Content
Cascade could be used for this purpose. Text commands sent by SMS, email or IM,
as in the In WebGlance [Paek et al., 2004] system where users send an email IM
message to the display system with a number corresponding to an option on the
could also be used.
Content control is also a way to collect users’ implicit interest on an item, similarly
to what happens with content download.
Rating By rating an item, the user is explicitly saying the he likes or dislikes that
item, depending on the value of the rating. This is a way for the display system
to allow a user to explicitly indicate his preferences. Rating is found on many
websites such as Youtube, Lastfm, Amazon, etc. On public displays, rating can be
implemented using any selection mechanism or through text commands.
Voting Displays can also collect users’ preferences by crafting polls which allow it
to extract information directly or indirectly from an individual. For example, sports
preferences of a user can be estimated by asking him to vote on his preferred athlete
from a list of athletes from different sports.
As with rating, voting can be accomplished through many different interaction mech-
anisms. As an example, Bluevote [Bortenschlager and Rehrl, 2007] uses images push
via Bluetooth. In this case the selection command is a picture sent previously by the
display system (by pushing the images to all discoverable Bluetooth devices). Users
send back to the system the picture that corresponds to their vote. Bluevote was
used in a conference setting to allow participants to vote on the best paper award.
Classification Classification is of a different nature than the previous categories
because the result is not a preference but the association of a description or keywords,
for example, with a given content item. This can be a less natural action for a user,
especially for public items, but it can be provided by displays in a more ludic per-
spective following the approach of Games With a Purpose [von Ahn and Dabbish,
52
Page 69
Footprint Interaction Mechanism
Presence Detection .Movement detection (proximity sensor; computer-vision)
Presence Characterisation .Face detection with age or gender classification, people counters
Presence Identification .Bluetooth
.RFID
Presence Self-exposure
.Bluetooth (profile on device name; a priori profile
definition)
.RFID (a priori profile definition)
Suggest Content
.Email/IM
.SMS/MMS
.Bluetooth (OBEX; BT Name)
Actionables
.Touch screen (Standard GUI controls)
.Email/IM (Text commands)
.SMS/MMS (Text commands)
.Bluetooth (Text commands, e.g. BT naming;
Standard GUI mobile application)
.RFID (Proximity activation, e.g. Touch & Interact)
Table 4.1: Mapping between interaction mechanisms and digital footprints.
2008]. Classification requires that the user is able to send free text to the display sys-
tem and so requires a text submission mechanism such as SMS, email, IM, Bluetooth
names, etc.
Footprints for Socially-Aware Interactive Displays
The previous sections have highlighted the types of interaction mechanisms that
we may need if we want to gather a particular type of footprint. This section will
now analyse how those multiple footprints can be used to support various types of
context-aware adaptation processes.
Table 1 presents a mapping between the digital footprints and the most widely used
interaction or presence mechanisms that generate those footprints. This mapping
can be used by situated display designers to help choose the interaction mechanisms
provided by the display in order to be able to collect a given set of footprints.
Overall, the entire set of digital footprints constitutes a collection of data which can
be used to characterise the place profile, enabling the display system to adapt its
behaviour to that particular social setting. Regardless of their generic contribution
to this broad adaptation, specific types of footprint can support specific types of
adaptive behaviour. Figure 1 summarizes the relationships that can be established
between the different footprints and possible adaptation processes.
53
Presence Detection .Movement detection (proximity sensor; computer-vision)
Presence Characterisation .Face detection with age or gender classification, people counters
Presence Identification .Bluetooth
.RFID
Presence Self-exposure
.Bluetooth (profile on device name; a priori profile
definition)
.RFID (a priori profile definition)
Suggest Content
.Email/IM
.SMS/MMS
.Bluetooth (OBEX; BT Name)
Actionables
.Touch screen (Standard GUI controls)
.Email/IM (Text commands)
.SMS/MMS (Text commands)
.Bluetooth (Text commands, e.g. BT naming;
Standard GUI mobile application)
.RFID (Proximity activation, e.g. Touch & Interact)
Table 4.1: Mapping between interaction mechanisms and digital footprints.
2008]. Classification requires that the user is able to send free text to the display sys-
tem and so requires a text submission mechanism such as SMS, email, IM, Bluetooth
names, etc.
Footprints for Socially-Aware Interactive Displays
The previous sections have highlighted the types of interaction mechanisms that
we may need if we want to gather a particular type of footprint. This section will
now analyse how those multiple footprints can be used to support various types of
context-aware adaptation processes.
Table 1 presents a mapping between the digital footprints and the most widely used
interaction or presence mechanisms that generate those footprints. This mapping
can be used by situated display designers to help choose the interaction mechanisms
provided by the display in order to be able to collect a given set of footprints.
Overall, the entire set of digital footprints constitutes a collection of data which can
be used to characterise the place profile, enabling the display system to adapt its
behaviour to that particular social setting. Regardless of their generic contribution
to this broad adaptation, specific types of footprint can support specific types of
adaptive behaviour. Figure 1 summarizes the relationships that can be established
between the different footprints and possible adaptation processes.
53
Page 71
aspects to consider when designing (in a global sense) a public display system.
The following is an elicitation of some of those aspects and, although it may not be
possible to include all of them in a single design, they are, at least, important to
consider in order to take informed design decisions.
4.2.1 Sensing systems challenges
Interacting with any systems requires more of both user and system than just being
able to give or receive input.
Norman [2002] defines a set of seven stages of action when interacting with any
system:
Forming the goal.
Forming the intention.
Specifying an action.
Executing the action.
Perceiving the state of the world.
Interpreting the state of the world.
Evaluating the outcome.
Norman defines two main problems users face when interacting with a system: the
Gulf of Execution and the Gulf of Evaluation.
The Gulf of Execution is the difference between the intentions and the allowable
actions. If the system does not provide clear affordances of the available actions
a user may not be able to translate intent into action. In this case the Gulf of
Execution would be large and hard to cross. Similarly, the Gulf of Evaluation
corresponds to how well the system provides a physical representation of its state
and how well the user can tell if the intentions have been met.
Norman rephrases the seven stages of action into questions the designer should try
to answer when designing an interactive system:
How easily can one:
Determine the function of the device?
55
The following is an elicitation of some of those aspects and, although it may not be
possible to include all of them in a single design, they are, at least, important to
consider in order to take informed design decisions.
4.2.1 Sensing systems challenges
Interacting with any systems requires more of both user and system than just being
able to give or receive input.
Norman [2002] defines a set of seven stages of action when interacting with any
system:
Forming the goal.
Forming the intention.
Specifying an action.
Executing the action.
Perceiving the state of the world.
Interpreting the state of the world.
Evaluating the outcome.
Norman defines two main problems users face when interacting with a system: the
Gulf of Execution and the Gulf of Evaluation.
The Gulf of Execution is the difference between the intentions and the allowable
actions. If the system does not provide clear affordances of the available actions
a user may not be able to translate intent into action. In this case the Gulf of
Execution would be large and hard to cross. Similarly, the Gulf of Evaluation
corresponds to how well the system provides a physical representation of its state
and how well the user can tell if the intentions have been met.
Norman rephrases the seven stages of action into questions the designer should try
to answer when designing an interactive system:
How easily can one:
Determine the function of the device?
55
Page 72
Tell what actions are possible?
Determine mapping from intention to physical movement?
Perform the action?
Tell if the system is in desired state?
Determine mapping from system state to interpretation?
Tell what state the system is in? — [Norman, 2002, p. 53]
Bellotti et al. [2002] argue that emerging sensing systems — systems “where input
is sensed be means other than keys, mouse or stylus (e.g., gesture, voice, or loca-
tion)” [Bellotti et al., 2002, p. 1] — face challenges different from traditional GUI
systems which have, to some degree, solved many of Norman’s questions. Although
not necessarily the same, interactive digital displays can share many of the proper-
ties of sensing systems and, clearly, to be effective need to be designed using a very
different paradigm than that of the GUI desktops.
Building on Norman’s stages of action but adopting a more communicative perspec-
tive to interaction, as opposed to Norman’s more cognitive one, Bellotti et al. [2002]
define a set of issues that sensing systems face:
Address Directing communication to a system.
Attention Establishing that the system is attending.
Action Defining what is to be done with the system (roughly equivalent to Nor-
man’s ’Gulf of Execution’).
Alignment Monitoring system response (roughly equivalent to Norman’s ’Gulf of
Evaluation’).
Accident Avoiding or recovering from errors of misunderstandings.
Analogously to Norman’s stages, these issues can be rewritten as questions that a
user (and system’s designer) must answer to be able to interact with a system:
“How do I address one (or more) of many possible devices?”
“How do I know the system is ready and attending to my actions?”
“How do I effect a meaningful action, control its extent and possibly specify a target
or targets for my action?”
“How do I know the system is doing (has done) the right thing?”
“How do I avoid mistakes?”
56
Determine mapping from intention to physical movement?
Perform the action?
Tell if the system is in desired state?
Determine mapping from system state to interpretation?
Tell what state the system is in? — [Norman, 2002, p. 53]
Bellotti et al. [2002] argue that emerging sensing systems — systems “where input
is sensed be means other than keys, mouse or stylus (e.g., gesture, voice, or loca-
tion)” [Bellotti et al., 2002, p. 1] — face challenges different from traditional GUI
systems which have, to some degree, solved many of Norman’s questions. Although
not necessarily the same, interactive digital displays can share many of the proper-
ties of sensing systems and, clearly, to be effective need to be designed using a very
different paradigm than that of the GUI desktops.
Building on Norman’s stages of action but adopting a more communicative perspec-
tive to interaction, as opposed to Norman’s more cognitive one, Bellotti et al. [2002]
define a set of issues that sensing systems face:
Address Directing communication to a system.
Attention Establishing that the system is attending.
Action Defining what is to be done with the system (roughly equivalent to Nor-
man’s ’Gulf of Execution’).
Alignment Monitoring system response (roughly equivalent to Norman’s ’Gulf of
Evaluation’).
Accident Avoiding or recovering from errors of misunderstandings.
Analogously to Norman’s stages, these issues can be rewritten as questions that a
user (and system’s designer) must answer to be able to interact with a system:
“How do I address one (or more) of many possible devices?”
“How do I know the system is ready and attending to my actions?”
“How do I effect a meaningful action, control its extent and possibly specify a target
or targets for my action?”
“How do I know the system is doing (has done) the right thing?”
“How do I avoid mistakes?”
56
Page 75
Cell interaction zone is the closest zone to the display. In this zone users are
in touch distance and can interact with the individual cells of the display.
Figure 4.2: Interaction zones for the Hello.Wall [Streitz et al., 2003].
This framework was used as a way to infer attention or engagement of the user
towards the display system and, consequently, to define different kinds and levels of
information to display to a user in different zones.
A proximity-based framework has also been used by Vogel and Balakrishnan [2004]
which, building upon the two previous frameworks, divided interaction into four
phases (Figure 4.3):
Ambient display phase: the ambient display phase is a neutral state where the
display shows only overall public information.
Implicit interaction phase: this phase is triggered when a user passes by the
display and appears “to be open to communication” [Vogel and Balakrishnan,
2004, p. 3]. In this phase the display shows an abstract representation of the
user and notifies the user if there is an urgent information item that needs
attention.
Subtle interaction phase: in this phase the user gives a clear indication that
he is interested in the display, by pausing and looking at it, for example. In
this phase the display can show more detailed information.
Personal interaction: the user interacts directly, touching the display to see
more details about personal information.
This framework is very similar to the interaction zones framework, but the authors
assume a more precise proximity/attention sensing mechanism that is able to dis-
tinguish one more level. This was accomplished by using proximity and gesture
recognition systems.
59
in touch distance and can interact with the individual cells of the display.
Figure 4.2: Interaction zones for the Hello.Wall [Streitz et al., 2003].
This framework was used as a way to infer attention or engagement of the user
towards the display system and, consequently, to define different kinds and levels of
information to display to a user in different zones.
A proximity-based framework has also been used by Vogel and Balakrishnan [2004]
which, building upon the two previous frameworks, divided interaction into four
phases (Figure 4.3):
Ambient display phase: the ambient display phase is a neutral state where the
display shows only overall public information.
Implicit interaction phase: this phase is triggered when a user passes by the
display and appears “to be open to communication” [Vogel and Balakrishnan,
2004, p. 3]. In this phase the display shows an abstract representation of the
user and notifies the user if there is an urgent information item that needs
attention.
Subtle interaction phase: in this phase the user gives a clear indication that
he is interested in the display, by pausing and looking at it, for example. In
this phase the display can show more detailed information.
Personal interaction: the user interacts directly, touching the display to see
more details about personal information.
This framework is very similar to the interaction zones framework, but the authors
assume a more precise proximity/attention sensing mechanism that is able to dis-
tinguish one more level. This was accomplished by using proximity and gesture
recognition systems.
59
Page 76
Figure 4.3: Framework for interaction phases [Vogel and Balakrishnan, 2004].
These three frameworks are very similar between themselves in the sense that they
all use a discrete division of attention or proximity to the display. The most salient
difference is that Vogel and Balakrishnan [2004] divided the intermediate phase in
two levels. Table 4.2 shows a comparison of the three frameworks.
Streitz et al. [2003] Brignull and Rogers [2003] Vogel and Balakrishnan [2004]
Ambient Zone Peripheral Awareness Ambient Display Phase
Notification Zone Focal Awareness Implicit Interaction Phase
Subtle Interaction Phase
Cell Interaction Zone Direct Interaction Personal Interaction
Table 4.2: Comparison of interaction frameworks.
A different interaction framework has been proposed by Ju et al. [2008] — an im-
plicit interaction framework — based on two axes: the level of attentional demand
and the balance of initiative taken by the system. Attentional demand is the cog-
nitive load that the system imposes on the user. Foreground processes require a
lot of attention from the user, as is usually the case with standard GUI interfaces;
background processes, like monitoring or ambient displays, impose a low demand
on attention. Initiative can vary between two extremes: reactive and proactive. Re-
active interfaces only respond to explicit user inputs; proactive interfaces act before
the user expresses any need.
This framework defines four main types of interactions that result from the quad-
rants defined by the two axes — reactive/foreground, reactive/background, proac-
tive/foreground and proactive/background (see Figure 4.4):
Reactive/foreground interactions are the ones that take place as result of an
explicit request by the user and that require the user’s full attention. Conven-
tional GUI interfaces fall on this category.
Reactive/background interactions are the result of an explicit request by the
user but take place in the background. An example of this is the “auto-save
60
These three frameworks are very similar between themselves in the sense that they
all use a discrete division of attention or proximity to the display. The most salient
difference is that Vogel and Balakrishnan [2004] divided the intermediate phase in
two levels. Table 4.2 shows a comparison of the three frameworks.
Streitz et al. [2003] Brignull and Rogers [2003] Vogel and Balakrishnan [2004]
Ambient Zone Peripheral Awareness Ambient Display Phase
Notification Zone Focal Awareness Implicit Interaction Phase
Subtle Interaction Phase
Cell Interaction Zone Direct Interaction Personal Interaction
Table 4.2: Comparison of interaction frameworks.
A different interaction framework has been proposed by Ju et al. [2008] — an im-
plicit interaction framework — based on two axes: the level of attentional demand
and the balance of initiative taken by the system. Attentional demand is the cog-
nitive load that the system imposes on the user. Foreground processes require a
lot of attention from the user, as is usually the case with standard GUI interfaces;
background processes, like monitoring or ambient displays, impose a low demand
on attention. Initiative can vary between two extremes: reactive and proactive. Re-
active interfaces only respond to explicit user inputs; proactive interfaces act before
the user expresses any need.
This framework defines four main types of interactions that result from the quad-
rants defined by the two axes — reactive/foreground, reactive/background, proac-
tive/foreground and proactive/background (see Figure 4.4):
Reactive/foreground interactions are the ones that take place as result of an
explicit request by the user and that require the user’s full attention. Conven-
tional GUI interfaces fall on this category.
Reactive/background interactions are the result of an explicit request by the
user but take place in the background. An example of this is the “auto-save
60
Page 79
Measures
To compare the two approaches, the following measures will be used:
The number of detected BT devices during the experiment - indication of
audience size.
The number of times a BT device was detected - indication of frequency/duration
of visits.
The number of BT name changes (interactions) per device during the experi-
ment.
The number of different keywords per device.
The individual weight of each keywords generated.
Procedures
In order to perform this experiment, a display application will be developed for
showing Youtube videos.
The display system will maintain a “Global Tag Cloud” with all tags send by users
since the beginning with popularity increasing every time the tag is detected. From
this global tag cloud a subset with the N most popular tags will be displayed.
In order to select content, the display will use a “Search Tags” set of T tags to
search for content items in Youtube videos. The display will choose a tag from this
set using a weighted random function that gives a higher probability of selecting
some tags. This “Search Tags” set is built by including up to T tags from the
“Present Tags” first and then from the “Global Tag Cloud”. In both cases, the
most popular tags are selected first. In order to guarantee that present users have
high influence over the selected items, the weights attributed to the “Search Tags”
set will be calculated in order to give the subset of user tags a total of 80% of the
total weight, with individual weights relative to the global popularity of the tag.
The rest of the tags will have equal probability of being selected.
Videos will be selected according to the “Search Tags”. : the display system will use
a weighted random tag in the tag cloud to search for videos and schedule N videos
for viewing. After the N videos are shown the process will be repeated. When
selecting videos for scheduling, the display system will try to select videos that were
not previously shown already.
63
To compare the two approaches, the following measures will be used:
The number of detected BT devices during the experiment - indication of
audience size.
The number of times a BT device was detected - indication of frequency/duration
of visits.
The number of BT name changes (interactions) per device during the experi-
ment.
The number of different keywords per device.
The individual weight of each keywords generated.
Procedures
In order to perform this experiment, a display application will be developed for
showing Youtube videos.
The display system will maintain a “Global Tag Cloud” with all tags send by users
since the beginning with popularity increasing every time the tag is detected. From
this global tag cloud a subset with the N most popular tags will be displayed.
In order to select content, the display will use a “Search Tags” set of T tags to
search for content items in Youtube videos. The display will choose a tag from this
set using a weighted random function that gives a higher probability of selecting
some tags. This “Search Tags” set is built by including up to T tags from the
“Present Tags” first and then from the “Global Tag Cloud”. In both cases, the
most popular tags are selected first. In order to guarantee that present users have
high influence over the selected items, the weights attributed to the “Search Tags”
set will be calculated in order to give the subset of user tags a total of 80% of the
total weight, with individual weights relative to the global popularity of the tag.
The rest of the tags will have equal probability of being selected.
Videos will be selected according to the “Search Tags”. : the display system will use
a weighted random tag in the tag cloud to search for videos and schedule N videos
for viewing. After the N videos are shown the process will be repeated. When
selecting videos for scheduling, the display system will try to select videos that were
not previously shown already.
63
Page 80
Users will be able to send keywords to the display system, thus affecting the current
tag cloud, by changing their mobile device’s Bluetooth name according to some
simple rules defined by the display system, e.g., “usual bt name k:soccer,ronaldo”.
In Approach #1, the display system will accept any keyword sent by users. In
Approach #2, users will only be able to define keywords from a list of keywords
presented by the display system. These keywords will be selected from the list of
keywords associated with the scheduled videos.
Evaluation
Evaluation will consist of performing standard statistical tests to detect significant
differences in the distributions of number of interactions and different keywords for
the two approaches.
64
tag cloud, by changing their mobile device’s Bluetooth name according to some
simple rules defined by the display system, e.g., “usual bt name k:soccer,ronaldo”.
In Approach #1, the display system will accept any keyword sent by users. In
Approach #2, users will only be able to define keywords from a list of keywords
presented by the display system. These keywords will be selected from the list of
keywords associated with the scheduled videos.
Evaluation
Evaluation will consist of performing standard statistical tests to detect significant
differences in the distributions of number of interactions and different keywords for
the two approaches.
64
Page 82
66
Page 83
References
ACM (2009). Acm digital library. Website. [Visited July 2009]
http://portal.acm.org/dl.cfm.
Agamanolis, S. (2003). Designing displays for human connectedness. In O’Hara, K.,
Perry, E., Churchill, E., and Russel, D. M., editors, Public and Situated Displays
- Social and Interactional Aspects of Shared Display Technologies, pages 309–334.
Kluwer, Dordrecht.
Bellotti, V., Back, M., Edwards, W. K., Grinter, R. E., Henderson, A., and Lopes,
C. (2002). Making sense of sensing systems: five questions for designers and
researchers. In CHI ’02: Proceedings of the SIGCHI conference on Human factors
in computing systems, pages 415–422, New York, NY, USA. ACM.
Bortenschlager, M. and Rehrl, K. (2007). Bluevote - a ubiquitous audience voting
service. In Adjunct Proceedings of the 9th International Conference on Ubiquitous
Computing UbiComp’07, Innsbruck, Austria.
Brignull, H., Izadi, S., Fitzpatrick, G., Rogers, Y., and Rodden, T. (2004). The
introduction of a shared interactive surface into a communal space. In CSCW
’04: Proceedings of the 2004 ACM conference on Computer supported cooperative
work, pages 49–58, New York, NY, USA. ACM.
Brignull, H. and Rogers, Y. (2003). Enticing people to interact with large public
displays in public spaces. In Rauterberg, M., Menozzi, M., and Wesson, J., editors,
INTERACT’03, pages 17–24. IOS Press.
Cheverst, K., Dix, A., Fitton, D., Kray, C., Rouncefield, M., Sas, C., Saslis-
Lagoudakis, G., and Sheridan, J. G. (2005). Exploring bluetooth based mobile
phone interaction with the hermes photo display. In MobileHCI ’05: Proceedings
of the 7th international conference on Human computer interaction with mobile
devices & services, pages 47–54, New York, NY, USA. ACM.
Cheverst, K., Dix, A., Fitton, D., Rouncefield, M., and Graham, C. (2007). Ex-
ploring awareness related messaging through two situated-display-based systems.
Human-Computer Interaction, 22:173 – 220.
67
ACM (2009). Acm digital library. Website. [Visited July 2009]
http://portal.acm.org/dl.cfm.
Agamanolis, S. (2003). Designing displays for human connectedness. In O’Hara, K.,
Perry, E., Churchill, E., and Russel, D. M., editors, Public and Situated Displays
- Social and Interactional Aspects of Shared Display Technologies, pages 309–334.
Kluwer, Dordrecht.
Bellotti, V., Back, M., Edwards, W. K., Grinter, R. E., Henderson, A., and Lopes,
C. (2002). Making sense of sensing systems: five questions for designers and
researchers. In CHI ’02: Proceedings of the SIGCHI conference on Human factors
in computing systems, pages 415–422, New York, NY, USA. ACM.
Bortenschlager, M. and Rehrl, K. (2007). Bluevote - a ubiquitous audience voting
service. In Adjunct Proceedings of the 9th International Conference on Ubiquitous
Computing UbiComp’07, Innsbruck, Austria.
Brignull, H., Izadi, S., Fitzpatrick, G., Rogers, Y., and Rodden, T. (2004). The
introduction of a shared interactive surface into a communal space. In CSCW
’04: Proceedings of the 2004 ACM conference on Computer supported cooperative
work, pages 49–58, New York, NY, USA. ACM.
Brignull, H. and Rogers, Y. (2003). Enticing people to interact with large public
displays in public spaces. In Rauterberg, M., Menozzi, M., and Wesson, J., editors,
INTERACT’03, pages 17–24. IOS Press.
Cheverst, K., Dix, A., Fitton, D., Kray, C., Rouncefield, M., Sas, C., Saslis-
Lagoudakis, G., and Sheridan, J. G. (2005). Exploring bluetooth based mobile
phone interaction with the hermes photo display. In MobileHCI ’05: Proceedings
of the 7th international conference on Human computer interaction with mobile
devices & services, pages 47–54, New York, NY, USA. ACM.
Cheverst, K., Dix, A., Fitton, D., Rouncefield, M., and Graham, C. (2007). Ex-
ploring awareness related messaging through two situated-display-based systems.
Human-Computer Interaction, 22:173 – 220.
67
Page 84
Churchill, E. F., Nelson, L., Denoue, L., and Girgensohn, A. (2003). The plasma
poster network: Posting multimedia content in public places. In Rauterberg, M.,
Menozzi, M., and Wesson, J., editors, Human-Computer Interaction INTERACT
’03, pages 599–606. IOS Press.
Churchill, E. F., Nelson, L., Denoue, L., Helfman, J., and Murphy, P. (2004). Sharing
multimedia content with interactive public displays: a case study. In DIS ’04:
Proceedings of the 5th conference on Designing interactive systems, pages 7–16,
New York, NY, USA. ACM.
Cox, D., Kindratenko, V., and Pointer, D. (2003). Intellibadge : Towards providing
location-aware value-added services at academic conferences. pages 264–280.
Creswell, J. W. (2002). Research Design: Qualitative, Quantitative, and Mixed
Methods Approaches. SAGE Publications.
Dix, A. and Sas, C. (2008). Public displays and private devices: A design space
analysis. In Workshop on Designing and evaluating mobile phone-based interaction
with public displays. CHI2008, Florence.
Electronics, A. (2009). Pressure mats. Website. [Visited April 2009]
http://www.arun-electronics.co.uk/pressure_mat.htm.
Fass, A., Forlizzi, J., and Pausch, R. (2002). Messydesk and messyboard: two designs
inspired by the goal of improving human memory. In DIS ’02: Proceedings of the
4th conference on Designing interactive systems, pages 303–311, New York, NY,
USA. ACM.
Ferscha, A., Kathan, G., and Vogl, S. (2002). Webwall - an architecture for public
display www services. In The Eleventh International World Wide Web Conference,
Honolulu, Hawaii, USA.
Google (2009). Google scholar. Website. [Visited July 2009]
http://scholar.google.com.
Grasso, A., Muehlenbrock, M., Roulland, F., and Snowdon, D. (2003). Supporting
communities of practice with large screen displays. In O’Hara, K., Perry, E.,
Churchill, E., and Russel, D. M., editors, Public and Situated Displays - Social
and Interactional Aspects of Shared Display Technologies, pages 261–282. Kluwer.
Greenberg, S. and Rounding, M. (2001). The notification collage: Posting informa-
tion to public and personal displays. In CHI Letters, volume 3, pages 515–521.
Horvitz, E. (1999). Uncertainty, action, and interaction: In pursuit of mixed-
initiative computing. IEEE Intelligent Systems, 14:17–20.
Huang, E. M., Koster, A., and Borchers, J. (2008). Overcoming assumptions and
uncovering practices: When does the public really look at public displays? In
Pervasive, pages 228–243.
68
poster network: Posting multimedia content in public places. In Rauterberg, M.,
Menozzi, M., and Wesson, J., editors, Human-Computer Interaction INTERACT
’03, pages 599–606. IOS Press.
Churchill, E. F., Nelson, L., Denoue, L., Helfman, J., and Murphy, P. (2004). Sharing
multimedia content with interactive public displays: a case study. In DIS ’04:
Proceedings of the 5th conference on Designing interactive systems, pages 7–16,
New York, NY, USA. ACM.
Cox, D., Kindratenko, V., and Pointer, D. (2003). Intellibadge : Towards providing
location-aware value-added services at academic conferences. pages 264–280.
Creswell, J. W. (2002). Research Design: Qualitative, Quantitative, and Mixed
Methods Approaches. SAGE Publications.
Dix, A. and Sas, C. (2008). Public displays and private devices: A design space
analysis. In Workshop on Designing and evaluating mobile phone-based interaction
with public displays. CHI2008, Florence.
Electronics, A. (2009). Pressure mats. Website. [Visited April 2009]
http://www.arun-electronics.co.uk/pressure_mat.htm.
Fass, A., Forlizzi, J., and Pausch, R. (2002). Messydesk and messyboard: two designs
inspired by the goal of improving human memory. In DIS ’02: Proceedings of the
4th conference on Designing interactive systems, pages 303–311, New York, NY,
USA. ACM.
Ferscha, A., Kathan, G., and Vogl, S. (2002). Webwall - an architecture for public
display www services. In The Eleventh International World Wide Web Conference,
Honolulu, Hawaii, USA.
Google (2009). Google scholar. Website. [Visited July 2009]
http://scholar.google.com.
Grasso, A., Muehlenbrock, M., Roulland, F., and Snowdon, D. (2003). Supporting
communities of practice with large screen displays. In O’Hara, K., Perry, E.,
Churchill, E., and Russel, D. M., editors, Public and Situated Displays - Social
and Interactional Aspects of Shared Display Technologies, pages 261–282. Kluwer.
Greenberg, S. and Rounding, M. (2001). The notification collage: Posting informa-
tion to public and personal displays. In CHI Letters, volume 3, pages 515–521.
Horvitz, E. (1999). Uncertainty, action, and interaction: In pursuit of mixed-
initiative computing. IEEE Intelligent Systems, 14:17–20.
Huang, E. M., Koster, A., and Borchers, J. (2008). Overcoming assumptions and
uncovering practices: When does the public really look at public displays? In
Pervasive, pages 228–243.
68
Page 87
Quividi (2009). Quividi - automated audience measurement of billboards and out of
home digital media. Website. [Visited April 2009] http://www.quividi.com/.
Raj, H., Gossweiler, R., and Milojicic, D. (2004). Contentcascade incremental con-
tent exchange between public displays and personal devices. The First Annual
International Conference on Mobile and Ubiquitous Systems: Networking and Ser-
vices, 2004. MOBIQUITOUS 2004., pages 374–381.
Russell, D. M. and Gossweiler, R. (2001). On the design of personal & commu-
nal large information scale appliances. In UbiComp ’01: Proceedings of the 3rd
international conference on Ubiquitous Computing, pages 354–361, London, UK.
Springer-Verlag.
Sawhney, N., Wheeler, S., and Schmandt, C. (2001). Aware community portals:
Shared information appliances for transitional spaces. Personal Ubiquitous Com-
puting, 5(1):66–70.
Scheible, J. and Ojala, T. (2005). Mobilenin combining a multi-track music video,
personal mobile phones and a public display into multi-user interactive entertain-
ment. In MULTIMEDIA ’05: Proceedings of the 13th annual ACM international
conference on Multimedia, pages 199–208, New York, NY, USA. ACM.
Sharifi, M., Payne, T., and David, E. (2006). Public display advertising based on
bluetooth device presence.
Storz, O., Friday, A., and Davies, N. (2006). Supporting content scheduling on
situated public displays. Computers & Graphics, 30(5):681 – 691.
Streitz, N., Prante, T., Ro¨cker, C., Alphen, D. V., Magerkurth, C., Stenzel, R., and
Plewe, D. (2003). Ambient displays and mobile devices for the creation of social
architectural spaces: Supporting informal communication and social awareness in
organizations. In O’Hara, K., Perry, M., Churchill, E., and Russell, D., editors,
Public and Situated Displays: Social and Interactional Aspects of Shared Display
Technologies, chapter 16, pages 387–409. Kluwer Publishers.
Toye, E., Madhavapeddy, A., Sharp, R., Scott, D., Blackwell, A., and Upton, E.
(2004). Using camera-phones to interact with context-aware mobile services.
Technical report, University of Cambridge, Computer Laboratory, 15 JJ Thomson
Avenue Cambridge CB3 0FD United Kingdom phone +44 1223 763500.
Trimble, J., Wales, R., and Gossweiler, R. (2003). NASA’s MERBoard: An Interac-
tive Collaborative Workplace Platform, pages 18–44. Kluwer Academic Publishers.
Ventura, P., Sousa, H., and Jorge, J. (2008). Mobile phone interaction with outdoor
advertisements. In Workshop on Designing and evaluating mobile phone-based
interaction with public displays. CHI2008, Florence.
Verschae, R., Ruiz-del Solar, J., and Correa, M. (2008). A unified learning framework
for object detection and classification using nested cascades of boosted classifiers.
Mach. Vision Appl., 19(2):85–103.
71
home digital media. Website. [Visited April 2009] http://www.quividi.com/.
Raj, H., Gossweiler, R., and Milojicic, D. (2004). Contentcascade incremental con-
tent exchange between public displays and personal devices. The First Annual
International Conference on Mobile and Ubiquitous Systems: Networking and Ser-
vices, 2004. MOBIQUITOUS 2004., pages 374–381.
Russell, D. M. and Gossweiler, R. (2001). On the design of personal & commu-
nal large information scale appliances. In UbiComp ’01: Proceedings of the 3rd
international conference on Ubiquitous Computing, pages 354–361, London, UK.
Springer-Verlag.
Sawhney, N., Wheeler, S., and Schmandt, C. (2001). Aware community portals:
Shared information appliances for transitional spaces. Personal Ubiquitous Com-
puting, 5(1):66–70.
Scheible, J. and Ojala, T. (2005). Mobilenin combining a multi-track music video,
personal mobile phones and a public display into multi-user interactive entertain-
ment. In MULTIMEDIA ’05: Proceedings of the 13th annual ACM international
conference on Multimedia, pages 199–208, New York, NY, USA. ACM.
Sharifi, M., Payne, T., and David, E. (2006). Public display advertising based on
bluetooth device presence.
Storz, O., Friday, A., and Davies, N. (2006). Supporting content scheduling on
situated public displays. Computers & Graphics, 30(5):681 – 691.
Streitz, N., Prante, T., Ro¨cker, C., Alphen, D. V., Magerkurth, C., Stenzel, R., and
Plewe, D. (2003). Ambient displays and mobile devices for the creation of social
architectural spaces: Supporting informal communication and social awareness in
organizations. In O’Hara, K., Perry, M., Churchill, E., and Russell, D., editors,
Public and Situated Displays: Social and Interactional Aspects of Shared Display
Technologies, chapter 16, pages 387–409. Kluwer Publishers.
Toye, E., Madhavapeddy, A., Sharp, R., Scott, D., Blackwell, A., and Upton, E.
(2004). Using camera-phones to interact with context-aware mobile services.
Technical report, University of Cambridge, Computer Laboratory, 15 JJ Thomson
Avenue Cambridge CB3 0FD United Kingdom phone +44 1223 763500.
Trimble, J., Wales, R., and Gossweiler, R. (2003). NASA’s MERBoard: An Interac-
tive Collaborative Workplace Platform, pages 18–44. Kluwer Academic Publishers.
Ventura, P., Sousa, H., and Jorge, J. (2008). Mobile phone interaction with outdoor
advertisements. In Workshop on Designing and evaluating mobile phone-based
interaction with public displays. CHI2008, Florence.
Verschae, R., Ruiz-del Solar, J., and Correa, M. (2008). A unified learning framework
for object detection and classification using nested cascades of boosted classifiers.
Mach. Vision Appl., 19(2):85–103.
71
Page 88
Vogel, D. and Balakrishnan, R. (2004). Interactive public ambient displays: tran-
sitioning from implicit to explicit, public to personal, interaction with multiple
users. In UIST ’04: Proceedings of the 17th annual ACM symposium on User
interface software and technology, pages 137–146, New York, NY, USA. ACM.
von Ahn, L. and Dabbish, L. (2008). Designing games with a purpose. Commun.
ACM, 51(8):58–67.
Wikipedia (2009). People counter — wikipedia, the free encyclopedia. [Online;
accessed 6-April-2009].
72
sitioning from implicit to explicit, public to personal, interaction with multiple
users. In UIST ’04: Proceedings of the 17th annual ACM symposium on User
interface software and technology, pages 137–146, New York, NY, USA. ACM.
von Ahn, L. and Dabbish, L. (2008). Designing games with a purpose. Commun.
ACM, 51(8):58–67.
Wikipedia (2009). People counter — wikipedia, the free encyclopedia. [Online;
accessed 6-April-2009].
72
Sign up today - FREE
Mendeley saves you time finding and organizing research. Learn more
- All your research in one place
- Add and import papers easily
- Access it anywhere, anytime
Start using Mendeley in seconds!
Readership Statistics
1 Reader on Mendeley
by Discipline
by Academic Status
100% Researcher (at an Academic Institution)
by Country
100% Portugal