Content uploaded by Pierrick Thébault
Author content
All content in this area was uploaded by Pierrick Thébault on Jan 28, 2014
Content may be subject to copyright.
International Journal of Design and Innovation Research 1
Volume X – n°Y / 2011
IJODIR. Volume X – n° X/2010, pages 1 à X
Designing for the Ubiquitous Computing era:
towards the reinvention of everyday objects and
the creation of new user experiences.
Pierrick Thébault1,2, Henri Samier2, David Bihanic3, Simon
Richir2
1 Alcatel-Lucent Bell Labs France
Route de Villejust
91620 Nozay, France
pierrick.thebault@alcatel-lucent.com
2 Arts et Métiers Paristech, LAMPA
2, Bd du Ronceray
49000 Angers, France
pi.laval@ensam.fr
3 PRES Lille Nord de France, CALHISTE
Le Mont-Houy
59313 Valenciennes Cedex 9, France
david.bihanic@univ-valenciennes.fr
ABSTRACT. Researchers of the Ubiquitous Computing community (Ubicomp) have been
pursuing the vision of a world where technologies and services permeates every object of our
lives for years. With components getting smaller, cheaper and more powerful, it has become
possible to manufacture connected objects capable of interacting with resources of the World
Wide Web. This opens up the possibility for researchers and practitioners to consider
information as a design material and objects as platforms for services. By allowing users to
personalize, complement or repurpose the functions of their objects, such services have a
great impact on the way artifacts are designed. Designing for the Ubiquitous Era requires
modifying our practice and reinforcing collaboration between disciplines at every steps of the
creation process. In this article, we discuss the need to reinvent objects and to investigate the
tools supporting the creation of engaging user experiences.
KEYWORDS: Ubiquitous Computing, design research, interaction design, product design,
user experience design, connected objects, smart objects, object-based services, web of
things.
2 IJODIR. Volume X – n° X/2010
1. Introduction
In the early 90’s, Xerox PARC researcher Mark Weiser envisioned a new computing
paradigm where information services are not processed and delivered by personal computers
but by a variety of devices proliferating in the environment. In a foundational paper entitled
“The computer of the 21th century” [Weiser, 1995], he described the possible future
relationships between people, practice, and technology of the Ubiquitous Computing
(Ubicomp) era. Often referred to as the pervasive penetration of everyday life with computing
technologies, this influential vision led researchers to explore the interaction of computing
through physical objects rather than general-purpose computers (i.e. physical computing), the
ability of environments to sense and respond to the presence of people in an intelligent way
(i.e. ambient intelligence) and the opportunity to digitally identify or augment physical objects
with information or services (i.e. the Internet of things).
If embodying the UbiComp approach in people’s everyday life has been an obstacle for a
long time, the exponential growth and the inexpensiveness of computing capabilities (as
described by Moore’s law), the miniaturization of electronic components and the
democratization of wireless networking infrastructures today allow not only lab researchers
but also manufacturers to embed connectivity chipsets, tiny web servers and microcontrollers
into everyday objects. Products as the Roomba1 robotic vacuum cleaner, the Nabaztag2
smart companion or the Ambient Orb3 light illustrated the feasibility of creating augmented
objects that do not look like traditional computers, as depicted in figure 1. The borders
between the “white goods” (i.e. appliances) and “brown goods” (i.e. consumer electronics)
have then started to shade, while objects’ value tended to shift from mediums to data with the
democratization of “software as a service” (SaaS) [Turner, Budgen, et Brereton, 2003] and
“cloud computing” technologies [Buyya, Yeo, et Venugopal, 2008].
Since the establishment of the World Wide Web as a platform of services with
considerable social and commercial benefits, end-users are moreover looking for new ways
to access information anywhere, anytime. In the same way they embraced the instantiations
of Web services as mobile applications, they are likely to want to consume data on new
connected objects that can be considered as services “avatars” (i.e. a physical artifact
embodying a specific facet of a Web service). According to User Experience designer Mike
Kuniavsky, “information no longer needs to be the purpose of an object, but one of many
qualities that enables it to be useful and desirable in ways that are more directly related to
people’s wants and needs” [Kuniavsky, 2010 44]. The design of objects is therefore no longer
restricted to form, function, material and production, it should considerate objects as
platforms for experiences, services or activities [Buchanan, 2001], whose functionality are
offered or enhanced by the services they instantiate.
In this article, we propose to study the change brought about by the computational and
connectivity capabilities of objects and to highlight the need for defining a new design
practice. In a first part, we discuss the way the Ubicomp community acknowledged the need
of considering user experience design in their research and show that the re-design of
objects constitutes a novel challenge. In a second part, we illustrate the new capabilities of
objects and identify the different types of services that can be built on top of them. In a third
1 Designed by iRobot. http://www.irobot.com
2 Today sold under the name « Karotz ». http://www.karotz.com/
3 Designed by Ambient Devices. http://www.ambientdevices.com/
Designing for the Ubiquitous Computing era 3
part, we highlight the pragmatic issues brought by services and discuss their implications in
the design of connected objects. We finally present the design strategy that led a small team
of product and service designers to collaborate in order to explore the impact of connected
objects on form, users’ perception and service’s experiences.
Figure 1. A market-based map of connected objects.
2. From a technological-driven research to a user-centered research in Ubicomp
As stated by Bell and Dourish, Ubicomp has always been an unusual research area.
Unlike other computer science topics building upon past results, it is not driven by “the
problems of the past but by the possibilities of the future” [Bell et Dourish, 2007]. Although
researchers have been pursuing Weiser’s vision and developed their agenda to anticipate
future trends and meet future needs, their practice is often conceived as being primarily
driven by technology. Based on a literature review, we propose in this part to retrace the way
Ubicomp has been explored by researchers among the past decades and discuss the shift
from a technological-driven research to a user-centered research. We then discuss the role of
designers in the creation of connected objects.
2.1. User needs, user experience and artifact design in Ubicomp literature
By analyzing the thinking that underlies Weiser’s vision from a phenomenological
perspective, Araya first stressed out the need for studying the transformations brought by this
4 IJODIR. Volume X – n° X/2010
“emerging form of technological absolutism” on people’s perception of the world [Araya,
1995]. This critical philosophical analysis led researchers as Edwards and Grinter to open up
the discussion about the technical, social and pragmatic issues of Ubicomp technologies in
homes such as reliability, administration and interoperability with researchers of various
disciplines [Edwards et Grinter, 2001]. Bohn et al. more recently broaden the scope by
exploring the social, economic and ethical implications of living in a world with connected
objects [Bohn et al., 2004]. They especially examined, with regards to the potential business
models, the privacy and control delegation issues that could lead users to reject such
artifacts.
Even if predicting the future is difficult, these discussions reflect the needs for designers
to balance the desire of technological capabilities with existing lifestyles and acceptable
practices. This brought researchers to consider users needs for the creation of Ubicomp
applications. Especially, they stated the difficulty to conduct evaluations with large number of
users due to the considerable technical work to be performed. This brought Scholtz and
Consolvo to propose a framework to facilitate results’ sharing with other researchers [Scholtz
et Consolvo, 2004]. Carter and Mancoff meanwhile discussed the use of summative and
formative evaluations and the need for iterative designs to be adopted [Carter et Mankoff,
2004]. Following a user-centered design process, Davidoff et al.’s ethnographic study on
device control in the domestic context also illustrated the gap between researchers’ focus and
design principles that can be derived from field trips observations [Davidoff et al., 2006]. If
these principles were acknowledged as difficult to embed into working systems, they
constitute valuable insights for the design of user-centered systems.
From Edwards’ and Grinter’s point of view, the question of “designing the smartness” has
a great impact on the architectural and implementation tradeoffs of technologies [Edwards et
Grinter, 2001]. As more and more researchers were working towards the creation of proactive
systems acting on behalf of humans, Rogers argued that technologies should be designed
“not to do things for people but to engage them more actively in what they currently do”
[Rogers, 2006]. She told us that users, designers and researchers should collaborate to
address less ambitious challenges and try to “go beyond what is currently possible to do with
our existing bricolage of tools and media”. The focus should be put on designing user
experience supporting specific activities and contexts. To do so, she advised working towards
small “ensembles or ecologies of resources that can be mobile or fixed” rather than complex
context aware environments.
For industrial and interaction designers, connected objects capable of communicating
with each other’s and delivering information and services constitute a tremendous but very
difficult challenge. Hjelm recommended to “use the power of design to visualize and express”
the complex issue of hiding computer technology and to develop the aesthetics of connected
objects in “an organic relation to its own time” [Hjelm, 2001]. Norman’s information appliance
model [Donald A. Norman, 1999], which aims at reducing the functional complexity of objects
to deliver information or services through simple user interfaces, had a important impact
before the rise of smart phones and touch interfaces. It demonstrated the need to design
affordances reflecting the actual capabilities of objects. Another approach developed by
Mavrommati and Kameas consisted in envisioning “hyper-objects” that work in synergy and
“share their capabilities in a communal pool” [Mavrommati et Kameas, 2003]. Interconnected
all together, all these objects therefore share the processing capabilities they individually
have. If designers agreed on saying that objects’ form should be redesigned to communicate
their new capabilities, they pointed out the inappropriateness of adding screens on every
artifact. They instead claimed that conceptual models people have of objects would simply
evolve with the use of new objects, whose redesign is nevertheless rarely addressed.
Designing for the Ubiquitous Computing era 5
2.2. The role of design in Ubicomp research
By augmenting objects with information shadows allowing user to gain knowledge,
instrumenting them with sensors generating data about their use or their immediate context
and hacking them to allow the remote automation of their capabilities, Ubicomp researchers
invented services that bridge the physical and digital worlds together. If these services are
tied with objects, they nevertheless remained invisible to users who are generally forced to
use a terminal to reveal the digital potentialities. A number of graphical user interfaces and
interaction techniques [Rukzio et al., 2006] involving mobile phones equipped with cameras
or Near Field Communication (NFC) modules have therefore been created to balance the
lack of affordances of objects. If the need for transparent user interfaces and intelligible
applications has been highlighted [A. K. Dey, 2009], bringing user interface designers and
human computer interaction designers to investigate these issues, the opportunity to redesign
the form of objects has rarely been explored.
This can be explained by researchers’ concerns to provide a universal and easy-to-
deploy solution (i.e. in most cases, mobile-mediations interactions are chosen to ensure the
adoption of systems) and by the designers’ lack of competencies, or interest, to experiment
with embedded systems. Thanks to the democratization of easy-to-use prototyping platforms
such as Arduino4, designers are nevertheless today able to conduct their own research
regarding the design of connected objects. Rapid prototyping methods combining cheap
hardware pre-assembled boards, easy-to-program software and 3D-printed materials can be
used to explore design concepts much more quickly than traditional engineering methods.
These new possibilities lead product and service designers to strengthen their collaboration in
order to considerate the service as an inseparable way of the product [Schneider, 2011] and
to experiment forms and interactions with regards to these services.
As pointed out by Kuniasvky, designing connected objects not only requires the design of
the physical object, its software interface and its hardware interface, but also of its
interconnections with other objects and its virtual representation on the Web [Kuniavsky,
2010 18]. Instead of considering objects as independent artifacts, designers should work
towards shaping the service delivery mechanisms that enable objects to work as ensembles
[Encarnaçao et Kirste, 2005]. The way users can spontaneously articulate the services
between objects and make them coax should be considered as an opportunity to enhance
and shape the user experience. Places’ scale, objects’ ecosystem and population of course
need to be considered in the design of this experience. It therefore implies “a shift in attitude
to the process of design, to the sequence in which the experience is created, and to whom is
involved at what stage” [Kuniavsky, 2010 46].
We propose to address that challenge from a practice-based research. The creation of
design artifacts and their confrontation with users is indeed needed to leverage new or
substantially improved insights. We argue that it is up to designers to leverage participatory
and iterative design methods to generate and evaluate ideas regarding users’ needs,
services’ cases, interaction techniques and objects’ shapes. Such research cannot therefore
be conducted without an exhaustive understanding of the technical possibilities offered by the
Internet connectivity of objects. In the next parts, we especially focus on envisioning the
opportunities and issues of using services as a design material and discuss the need to
define a new design practice.
4 http://www.arduino.cc/
6 IJODIR. Volume X – n° X/2010
3. When objects meet the World Wide Web: novel capabilities and types of services
Since the original article of Weiser, Ubicomp researchers have conducted a lot of
research on the development of scalable, flexible, reliable and robust networks of devices or
sensors. This led a part of this community to especially focus on the creation of an Internet of
Things [Bassi et al., 2008] and to picture the Internet as an extension of “a seamless fabric of
classic networks and connected objects” whose impact will be intimately linked to the growth
of the Web” [Hourcade et al., 2009 4]. The Web of Things approach [Guinard et al., 2011],
which consists in exposing objects as accessible and addressable resources of the Web,
substantially facilitates the creation of UbiComp applications. Popular Web technologies (e.g.
HTML, Javascript, Ajax, PHP) can indeed be used to digitally enhance connected objects and
allow them to communicate with existing Web resources [Boussard et al., 2011]. In this part,
we present the capabilities brought by the Internet connectivity and the Web exposition of
objects and the possibilities in terms of services.
3.1. Possibilities of connected objects
Thanks to embedded microcontrollers, web servers and connectivity chipsets, connected
objects are capable of processing the same information than computers and to query data
from existing Web resources. In the same way Web platforms are interoperating through
dedicated programming interfaces (i.e. API), a vocabulary for resources or information
requests can be created for objects. This allows objects and Web resources to communicate
with each other’s and facilitate new types of interactions. Each of them can therefore be
considered as a resource for the others and deliver content, share knowledge and shape
behaviors. We describe these mechanisms, also illustrated on figure 2, in the following sub-
sections.
Figure 2. New capabilities of connected objects.
3.1.1. Deliver content
Some objects and Web resources are considered as media-related. They generate or
convey the materials (e.g. text, picture, sound) that compose a document or a recording. In
the physical world, this is typically the case of a radio, a television, a camera or a portable
Designing for the Ubiquitous Computing era 7
music player. In the digital world, user-generated content platforms such as Ubroadcast,
Youtube, Flickr and Spotify5 also aggregates content that can be communicated from a
computer. A new capability of connected object consists in allowing such content to be
delivered from the service to the object, depending on its user interface offering. This would
allow an alarm clock to play streaming music instead of traditional FM/AM signals. Data can
potentially be converted or reformatted in order to be compatible with a specific object (e.g. a
text can be read if no display is available). In the same way, content produced or stored by an
object can easily be sent or publish on a Web platform. Finally, as content can be shared
from one platform to another, files could seamlessly be redirected from one object to another
(e.g. projecting a presentation from a laptop computer without any video cable).
3.1.2. Share knowledge
A part from the content they are able to transmit, objects or Web resources are also
capable of generating information that can be extracted, compiled or analyzed with regards to
a user need or task. Objects can potentially provide data about their capabilities (i.e. what
they can do), their statuses (i.e. how they are functioning at a given moment) or their use (i.e.
How? When? By who?). On the other side, Web resources can bring value by obtaining
specific information from multiple trusted sources that often require for a user to engage in a
computing experience to be delivered. From an object perspective, connectedness allow for
external knowledge to be communicated in a transparent way to ease users’ life or augment
their awareness. This would for example enable a digital photo frame to stress out
unexpected event such as public transportations, traffic jams or weather issues by displaying
widgets or alerts in the specific time of the morning. From a service perspective, objects can
be used to gather very accurate information without user intervention (i.e. this is especially
valuable for pollution sensing or goods tracking). Fine-grained statistics can also be published
on a number of professional or social platforms to contribute to create shared representations
of the real world or of people’s activities and to better shape delivery mechanisms (e.g.
suggesting Youtube videos based on users’ television consumption).
3.1.3 Shape behavior
Objects are generally designed to respond to users’ interactions. The manners in which
they function therefore rely on human operations and can be easily predicted. Some of them,
such as the washing machine or video recorders can be programmed. The inner working of
Web platforms is slightly different: their API allows for other resources to control their features
and compose them without any user intervention. In similar approach, the capability of
connected objects to interoperate with other objects therefore makes several type of
automation possible. Object to object interactions can be implemented to support domotic
applications (e.g. setting up the heaters and shutters according to temperature and light
sensors) and commands chained and be sent after a specific user’s interaction with an object
(e.g. automatically reducing the light intensity and declining calls when a user is turning his
DVD player on). Web resources can also develop specific behaviors according to the
information they receive from objects. Publishing rules can for example be implemented in
order to avoid user-sensitive information to be broadcasted on social network platforms (e.g.
posting a message on Facebook only when I lose weight). More importantly, Web resources
can trigger the functions of objects with regards to the knowledge they acquired.
5 These Web resources can be considered as the digital equivalent of the mentioned
objects.
8 IJODIR. Volume X – n° X/2010
3.2. Towards new services built with or for objects
Connected objects allow for new types of services to be instantiated on top of them. By
seamlessly blending with artifacts, such services enable manufacturers and even end-users
to customize objects with new content and augment them by shaping new behaviors from
shared knowledge. We argue that different kind of services can be designed according to the
capabilities they leverage. In this section, we present different used cased scenario motivated
by examples in the literature that illustrate the potential of services aiming at aggregating
objects’ data, augmenting objects’ capabilities and orchestrating an objects’ ecosystem.
3.2.1. Computer or mobile-based services to aggregate objects’ data
“Leila is a young woman really concerned about healthcare. She goes out running twice a
week to keep in shape and like monitoring her effort and measuring her performances. The
connected shoes se bought few years ago do a really good job, but she was also interested
in meeting new people when she practices sports. That is why she subscribed to a service
that helps her getting in touch with runners sharing the same profile. A part from allowing her
to bond with people, this service also offers to monitor her weight and quality of sleep. By
gathering data from her connected scale and by wearing a sensor bracelet at night, it
provides Leila with comprehensive information about her daily life. All she has to do is to
access the service on the laptop computer or to launch the dedicated mobile application. A
few weeks ago, she paid for a premium feature that allows her to get advice from a personal
coach, at a distance. She is also able to navigate through a social network of people using
the same objects than her to compare their personal statistics or share comments, hints or
reviews.”
In this example, we illustrate the opportunity to extend existing Web resources to objects.
Such services aim at facilitating the aggregation, the storage and the presentation of objects’
states and uses. Designed for computer or mobile displays, they provide users with a
comprehensible overview of a small or large-scale objects’ ecosystem and make the creation
of social networks of “things” or smart metering tools possible. If domains as energy
consumption and life logging are most likely to be addressed in homes, such services can be
used in the industry to track and manage assets. Delegation control of connected objects
would then allow to remotely commanding them from a secured service.
3.2.2. Object-based services to augment objects’ capabilities
“It is 7:30 am, Peter is woken by a song that has been randomly chosen from the music on
demand platform he is subscribed to. His connected alarm clock postponed his waking after
querying the local weather and traffic jam platforms. Due to the recent snowfalls that
happened during the night, it would have been impossible to take the usual road and go to
work without wasting several hours in traffic jams. As a precaution, all the meetings Peter
was supposed to attend this morning have automatically been re-scheduled to later in the
day. In accordance with his company’s policy, he will work from home this morning. Peter is
informed of the situation and presses the snooze button, triggering at the same time the
publication of a micro-message on his favorite social network platform. He walks to his
bathroom where the temperature has been perfectly set for his arrival. The house
management service triggered the heater fifteen minutes before he woke up and has just sent
a message to his coffee machine to be sure he will be having a lovely breakfast.”
In this scenario, the service can no longer be considered as a traditional Web or mobile
application, but as a set of new features specifically designed for a type or model of objects.
Such services enhance objects’ inner systems by enabling a new means of interoperability
with other objects and Web resources and allowing users to customize their experience. They
Designing for the Ubiquitous Computing era 9
facilitate the circulation of content between objects and Web resources, offer to add a social
aspect to objects by leveraging existing social networks mechanisms and provide users with
the ability to shape a behavior based on open data or automate certain tasks or operations.
3.2.3. Environment-based services to orchestrate an objects’ ecosystem
“Mike and Terry put a lot of effort into designing their home. They especially like
decorating their living rooms with lamps and ambient displays such as digital photo frames
and are looking towards shaping customized experiences. They create two services for their
home that allows them to modify the ambiance of the room. As they watch a different movie
every night, they like the mood to be changed automatically when their connected television
is running a video on demand. Shutters, lights and speakers are adjusted without any
intervention while their digital frames is turned off. Usually, the latter are displayed pictures
taken from the social network accounts of people that are physically present in the room.
Mike and Terry especially like this feature when friends are coming for dinner, it always start
new discussions.”
While in the last example services were augmenting objects, they here operate at a
bigger scale. By involving multiple interconnections with objects, they allow to orchestrate
users’ environments. Objects’ behaviors can therefore be preprogramed and triggered
according to given contexts or events identified by the service. Data generated by objects can
be used in conjunction with Web resources such as open data to define spatiotemporal
situations or sense the presence of people. This makes automation of certain activity possible
for inhabitants that could use assisted living systems or that are simply looking to save time.
Content can also be delivered in a so-called “intelligent” way according to the social
preferences or presence.
4. From the anticipation of issues to the identification of design implications
Tremendous possibilities are today offered to designers of connected objects. They are
no longer required to define the number of delivered features but can potentially create open
objects that can be charged with new services after being shipped to users. Objects can
therefore be considered as open platforms which purposes are defined by people according
to their needs or situations of use. This opens up the possibility for objects to be reconfigured
and complemented, but also brings a number of issues that are likely to happen if designers
do not consider the implications of new services on the design of objects. In this part, we
discuss these issues and highlight the aspects that need to be taken into account in the
creation of connected objects.
4.1. Pragmatic issues of connected objects
Many questions are left unanswered regarding the manner in which services, and especially
the ones that augment objects or environments, orchestrate the life of objects, and by
extension, ours. From a user-centered perspective, we propose in this section to anticipate,
how users are likely to encounter issues related to the connectivity, dependency, setting and
interoperability of connected objects. We argue that the lack of transparency regarding how
objects allow users to interact with services, use personal data and share information
constitute barriers to the adoption of connected objects.
10 IJODIR. Volume X – n° X/2010
4.1.1. Connectivity
Connected objects are provided with new access to services and seamless data
circulation. However, do all objects need a continuous connection to the Internet and to the
Web to deliver services and fulfill their functions? There is a growing concern about the
consequences on health of electromagnetic waves emitted by wireless technology [Carvajal,
2007] and an uninterrupted connectivity will slow the adoption of this kind of objects. Some
users already feel like they need to keep “free zones” where technology, considered as
harmful or source of conflict, is less present or absent [Frohlich et Kraut, 2002]. This is
especially true in homes where family members seek to separate personal and professional
spheres. In addition to these social issues, we need to consider the environmental dimension.
The over consumption of energy would nevertheless have less impact for electricity as it
would be delivered in cost-efficient manner by “smart grids” [Mazza, 2002]. That is why
problems of “online <> offline” management [Woodruff et al., 2007] and representation of the
connection status should not only be addressed from a technological but also a social
perspective. It is important that technology remains transparent when users either want to
feel disconnected from the Internet or want to interact with the primary functions of their
products.
4.1.2. Dependency
Once an object is augmented by services, it becomes more or less dependent on the
infrastructure on which it relies. Bohn et Al. tell us that electronic books “appear to be more
error-prone and less autonomous than normal books” because of their loss of autonomy
[Bohn et al., 2005]. In general, appliances that surround us depend on an electrical
infrastructure but at the same time can also be used independently of any other product.
Allowing services to add features to objects and shape their behavior means there is a need
to choose a "referent". In that case, which entity, from the embedded system or the service
will control the other? The act of moving the entire intelligence of objects to the Web has
many advantages (i.e. centralized data, absence of conflicts, apparent transparency) but in
doing so risks altering the relationship that the user has with objects. The latter becomes an
“empty shell” that merely executes the decisions taken more or less consciously by the user
through a global life management service. The issue of control is even more crucial when a
service rely on Web resources to shape the intelligence of objects. We argue that the balance
must be found in the “master <> slave” relationship. It is important connected objects do not
depend on their connectivity to fulfill their functions.
4.1.3. Settings
Because of their ability to process information, new objects can potentially deliver and
interconnect with a multitude of services. Determined by manufacturers at the time of
conception or by users after the purchase, their number and their nature are subject to
change. Whatever the approach is, the issues surrounding the setting up of services seem
inevitable. What kind of input methods will allow the user to provide the login and password
required by the online services? Existing objects do not offer the proper user interfaces for
alphanumerical entries. It would be indeed be very difficult to type text on a clock that has a
small LCD screen using a limited number of unlabelled buttons. A synchronization phase (in
situ or in a short perimeter) and the use of a suitable terminal are necessary. The use of
intermediate artifacts (e.g. RFID tags) or recognition systems (e.g. finger prints, face, voice,
DNA) that help the user to deal with authentication is possible, however, it adds complexity.
Beyond issues of synchronization, associated service representation is also important for the
transparency of experience. Will the user be able to make the difference between two objects
configured with different services or should he leverage augmented reality systems? Since
Designing for the Ubiquitous Computing era 11
the perception of the object extension will affect its use, the notion of affordance [D. A
Norman, 2002] is crucial.
4.1.4. Interoperability
New capabilities of connected objects make “object to object” interactions possible
through the Web. Encarnaçao and Kirste tell us that future environments will be composed of
smart ensembles sharing information to assist users with their tasks [Encarnaçao et Kirste,
2005]. Soon the artifacts will be able to use the semantic description of their functions
[Boussard et al., 2011], to register their status, and a spontaneous creation of ecosystems will
be possible. However, how will several interconnections or links between objects be made
visible to the user? What level of control will the user have on the modeling of each object’s
behaviors? A system that does not integrate the user with the design of the intelligence can
be understood as a form of manipulation or control. On the other hand, a “do-it-yourself”
approach encouraging people to build their own ecosystem may require too much
participation from the users. The interoperability issue described previously deals with
connectivity, dependency and setting problems. Again, it seems crucial to give to the user
every possible means to understand how objects operate together and how data flows
circulate.
4.2. Design implications
By disseminating the technology in objects, we turn them into platforms supporting the
instantiation of services. In doing so, we risk creating a gap between the perceived functions
of objects and the role they are design to accomplish. It is therefore necessary to adapt the
user interfaces of objects so that they can be widely adopted by the general public. We think
there is a need for a transitional stage where intelligence is made visible, so that users have
the feeling of living “with machines, not inside a machine” [Kaplan, 2009 201]. In this part we
give greater consideration to objects that are used collectively and argue that a new means of
control in the design of interactions and a shift in the manner smartness is considered are
required.
4.2.1. Collective use
Most objects work without knowing the identity of users and deliver a generic user
experience. Connected objects are nevertheless likely to deliver a customized experience (as
computers do). Insofar as associated services aggregate personal data (preferences, history
of actions, private information), objects inherited from existing appliances should be able to
“understand” the context of their use so that they remain “user agnostic”. People should be
able to use their connected clock, as it is still an appliance, without having to login or state
their identity. The identification methods chosen by manufacturers must be transparent,
responsive and reliable. It is unlikely that people adopt products that require further actions to
ensure the authentication, or whose execution time will be perceived as too long [Dearman et
Pierce, 2008]. On the other hand, a weak security level may lead to misuse. In the same way
that computer programs can reorganize themselves to better adapt to their environment, the
concept of “behavioral reflexivity” would allow a connected object to suggest appropriate
applications or reconfigure itself according to the circumstances. This can be achieved by
embedding identification systems (e.g. cameras, fingerprints scanners) or rely on external
devices such mobile phone or wearable RFID tags (e.g. bracelet, implants). All the data
related to the use of object (i.e. how, where, when and by whom) and the understanding of
the surrounding environment will then contribute to the creation of service delivery
12 IJODIR. Volume X – n° X/2010
mechanisms and shape the object’s intelligence. The way this information is shared within an
ecosystem and used as inputs by services present major challenges for designers.
4.2.2. Control by the user
Norman tells us that objects should be designed according to the tasks they support [D. A
Norman, 2002]. If users often have a limited understanding of the internal mechanisms of
objects, the consequences of their manipulations are usually made visible by visual or audio
feedback. The hybridization of the digital and the tangible allows new interactions between an
object and services. Commands are executed by systems with the implicit consent of users
and may affect the operation of applications or products, having consequences not only with
the web, but also on reality. User interfaces should be adapted so they can inform, suggest
and warn the users according to the situations of use. We argue that existing displays and
input methods currently integrated into objects are not tailored for this. With the exception of
touch terminals using graphic user interfaces, objects cannot rearrange their buttons or
switches according to the type of services they offer. In order to facilitate the user adoption of
connected objects, there is a need to materialize their capabilities and to give people new
means of control over their data storage and sharing. Objects should not present an obstacle
for the way in which people manage the boundaries between online, offline, personal and
professional spheres. The design of new interfaces that could allow for the simple connection
and disconnection to and from the Internet, activating and deactivating services or to swap
data from one object to another could solve some of these practical problems and enhance
the user experience. Especially connected objects that can potentially support a multitude of
applications, related or not to the primary functions of the product. We argue that combining
physical interfaces and virtual interfaces will address these problems and that it is up to
designers to work on the organization and prioritization of proper interface layers that will
reflect the potential of the digital sphere. Mixed reality technologies involving see-through
head mounted displays and projected interfaces would allow visualizing and manipulating
information in novel ways.
4.2.3. Task support
More fundamental questions regarding the way people’s life revolve around tasks, and by
extension objects, are underlined in the design of connected objects. By serving users and
simplifying their daily life, automation, for example, also takes away many opportunities for
experiences. Borgmann tells us that objects should not procure their commodities
instantaneously but ask for attention and engagement [Borgmann, 1987]. Feelings of
competence, autonomy, routine and familiarity play an in important role in the construction of
pleasurable experiences [Green et Jordan, 2002] and by extension, happiness. We argue
that objects should provide new services, but not deprive users of the experience they used
to deliver. Connected objects can potentially be incredibly smart and fully automated, but they
should let people touch, make or even hear the sounds they make. Instead of trying to work
towards the creation of an ambient intelligence, we argue that designers should leverage the
new capabilities of objects to allow users to reconfigure or repurpose them in a serendipitous
way. Objects could then be used in multiple ways and under different circumstances to
support users’ tasks. This approach can be compared to the system design philosophy called
“recombinant computing”, which dictates that “computing environments can be created from
the bottom up by creating individual entities to be part of an elastic, always changing whole.”
[Newman et al., 2002]. While some objects can be augmented by following a traditional
approach, others can be designed for general purposes and serve multiple situations. Shape
shifting technologies or materials, such as shape memory alloys or polymers, could
hypothetically be integrated into objects to enable tangible reconfigurations of objects. This
Designing for the Ubiquitous Computing era 13
would allow shapes to reflect the content of services, and lead to the creation of a form
grammar.
5. Towards a design strategy for reinventing objects and shaping service’s
experiences
In previous parts, we demonstrated that the augmentation of objects with new features
brought by services has not only a great impact on users’ relations with them but also on the
way they interact with them. By adding smartness to objects, new behaviors can be shaped
and triggered without any user intervention. This raises questions regarding the type of
objects that should be augmented and the way they should be redesigned to reflect the
possibilities of services. While the latter can constantly be recomposed or reconfigured, the
form of objects is most likely to remain the same. The challenge is therefore to balance the
physical constraints of materials with the limitless flexibility of information. Such question
opens up for numerous design tracks to be explored. It also leads designers to consider the
creation of new artifacts instead of envisioning the evolution of existing objects. Since people
makes specific representations of what an object can do, trying to enhance existing objects
may lead to misunderstanding and a slow adoption. Besides the issues of form, which are
inseparable from interaction ones, the question of usage is also crucial. By turning objects
into platforms of services, we offer people to define the features of objects by themselves. It
is therefore needed to identify common needs that would be supported by the service
offerings. The difficulty is now that services can deliver an experience that is no longer
restricted to the scope of the object. Some interactions can happen in another room of the
house or even in another part of the city. In the following sections, we present the first
elements of the design strategy that is currently developed at Bell Labs. After presenting the
context of our work, we discuss the ongoing research conducted on the design of tools and
experiments that allow us to explore the futures of connected objects, investigate users’
mental models, identify needs and shape new user experiences.
5.1. Context of work
Within Bell Labs Applications domain, highly technical components are being developed
by a team of engineers with a strong Ubicomp background to enable the creation of
connected objects and services. Driven by the Web of Things approach, this research
especially aims at modeling objects’ representations, architecting objects’ gateways, and
creating service models. A part from semantic descriptions of objects’ capabilities, service
deployment mechanisms and access control systems, navigation tools and connected objects
have been designed to illustrate the potential use of the overall infrastructure in domestic,
corporate or urban contexts. In order to ensure the user adoption of such technologies, three
members of Application Studio (i.e. Bell Labs user-centered multidisciplinary department)
stepped into that research project: an interaction designer, a product designer and a
usabilist/psychologist. In this context, we aim at developing a design practice where multiple
disciplines intervene at different steps of the projects. By placing users at the center of our
research, we work towards the creation and the evaluation of tools designed to provide us
with insights regarding users’ perception, needs and acceptance of connected objects. This
joint-collaboration with technical researchers led us to identify three research tracks that
compose our design strategy.
14 IJODIR. Volume X – n° X/2010
5.2. Envisioning multiple futures of objects
Observing market trends, companies’ prospective vision of technologies (i.e. often
communicated as videos illustrating innovative interactions or services in various use case
scenarios) and artifacts produced by the research community is a first step in the
understanding the context. An exhaustive collection of pictures and its organization in boards
allowed us to picture the current state of the market and to identify the current evolutions of
objects, as illustrated on figure 1 and 4. This led us to envision different futures involving
different type of modifications regarding the design of objects and explore few of them, as
described in the following sub-sections.
5.2.1 Insights
Our analysis of the state of the art showed that objects can whether:
– remain the same and need a terminal. Most researchers and manufacturers of
connected objects currently adopt this approach. Mobile or computer mediations are
leveraged to allow users to configure or interact with the capabilities that cannot be accessed
from the physical user interfaces.
– become docks for terminals. By making technologies converging into one device,
smart phones have already replaced many objects. Alarm clocks, radios or music players are
not as popular as before. This leads manufacturers to redesign such objects as docks that
are used in combination with terminals.
– become a touch interface. The success of touch interactions leads the industry to
integrate touch screens into objects’ form factors. This tends to reduce the place of physical
user interfaces to the benefit of graphical user interfaces that can be refreshed with a single
tap.
– disappear in the environment. With the ongoing research conducted on flexible and
very thin displays that can be embedded into walls or windows, companies illustrate a future
where information and services are delivered by the environment. This would bring people to
own and use fewer objects than today and interact with services through multiple surfaces.
– be redesigned with physical and virtual users interfaces. Concept products as the
Olinda radio6 demonstrate that services instantiated on top of objects can be materialized
with dedicated buttons, displays or interactors. As mentioned in part 4, projected or mixed
reality interfaces can be used to avoid overloading the physical user interface.
– explode in several modules. Researchers of tangible computing have already
explored the concept of graspable building blocks that can be combined to produce (e.g.
especially in tangible programming) or interact with systems. Such approach can be adopted
to create modules for each core feature of a given object.
– be reinvented to allow shape shifting. A more prospective vision consists in picturing
objects as dynamic entities capable of changing their forms according to the context of use or
to the associated services. This approach can be considered as the most disruptive since it
does not follow any pre-established standards regarding the shape of objects.
6 A radio augmented with a social network designed by Berg.
http://www.berglondon.com/projects/olinda/
Designing for the Ubiquitous Computing era 15
Figure 3. The evolution of an alarm clock.
5.2.2 Exploring the visions with users
From the described approaches, we consider the last three are as the most interesting for
research. While the first three undergo a strong influence of smart phones, changing the
natural qualities of objects, the others aims at creating artifacts that maintain their role in
users’ life and environments. In order to investigate the interaction design and product design
dimensions of connected objects, we have conducted several projects described as followed:
– a workshop with design students. We asked a group of twenty-one fourth-year
students in interaction design and spatial design to study the usage of an object and to
explore its futures. They had five days to come up with new features and services that
augment its capabilities, and to work on a complete re-design based on the selected
approaches. They were especially asked to illustrate the gestural language allowing users to
interact with these new capabilities. This initiative showed that students had difficulties to
project themselves in a world where the role of objects is reinforced by services leveraging
existing Web resources. As such objects are likely to compete with computers and mobiles
that can do more, and “better”, they instead focused on adding smartness to “dumb” things or
creating tangible interfaces. If deriving from a mono-functional object to create new services
can be considered as easier, we argue that students could not really complete the exercise
because of a lack of meaningful objects’ augmentations and pre-established mental models
(i.e. we discuss these issues in the next section). Time constraints certainly also had an
impact on students’ productions.
– a participatory design session with intermediary artifacts. A different methodology
has been developed with a group of six people (i.e. researchers or interns with a design or
usability background) in order to explore the potential of shape shifting objects. Participants
were given six mockups allowing different types of manipulations (e.g. inflatable, combinable,
retractable, rotatable, foldable, etc.) for short periods of ten minutes. They had to come up
with any kind of ideas related to the type of services that could be communicated by such
moving shapes. Proposals were written on sticky notes and discussed by the group. Each of
them has been categorized and subjectively evaluated. Ideas considered as most relevant in
the context of connected objects have been illustrated as shown on figure 4. This session
unexpectedly produced more insights regarding the way people would like to manipulate such
open-ended shapes than concrete service cases scenarios. While functions remains
undefined or fuzzy (e.g. sending a message to friends, accessing the news, playing music,
etc.), some users had a clear view on the type of interactions they could performs according
16 IJODIR. Volume X – n° X/2010
to the different artifacts. We argue that such intermediary representations facilitate the
projection of users in a specific future of objects and could be used to support sessions on
usage exploration.
– a case study of the alarm clock. We decided to experience the same approach than
the one developed during the student workshop in a much longer period. An interaction
designer and a product designer chose to explore the different futures of the alarm clock. We
picked this object for its collective dimension and its role regarding the preparation of the day.
After conducting short interviews with users and non-users of alarm clocks in order to gather
insights about its actual use, we managed to identify opportunities to enhance their
capabilities (e.g. especially in the way the alarm is set up or awake people) and add new
ones (e.g. supporting users in their morning tasks). Three concepts have been developed
from the different approaches (i.e. materializing services, creating modules, allowing shape
shifting). Each of them results from complex trade-offs between form, function and interaction
and led us to diverge from the original alarm clock. We argue that this methodology forced us
to consider services as a design material of objects, and to reason in terms of tasks that go
beyond the pre-existing mental models of what an object can do. We look towards to
prototyping theses concepts in order to confront them with users in real life situations or
participatory design sessions. We are interested in evaluating their value as objects and
creation tools.
Figure 4. Interactions or service cases generated during a participatory design
session.
5.3. Changing established mental models
People make their own representations of everyday things, called mental models, in order
to interact with the world. Defined by Rouse et al. as “mechanisms whereby humans are able
to generate descriptions of system purpose and form, explanations of system functioning and
observed system states, and predictions of future system states” [Rouse et Morris, 1986 351],
these models evolve naturally through interactions with systems and are therefore affected by
users’ prior knowledge and experiences [Donald A Norman et Draper, 1986 45]. As
mentioned in previous part, these preexisting representations are likely to be impacted in the
use of connected objects whose capabilities have been augmented and whose inner working
Designing for the Ubiquitous Computing era 17
is more difficult to apprehend. If mental models are subject to change, it is needed that
designers investigate the way people perceive the interconnections between objects and
services in order to ensure the intelligibility of the overall system. In the following, we present
an experiment aiming at capturing user’s mental models of an objects’ ecosystem and
discuss some insights.
5.3.1 Investigating mental models with users
As connected objects have not yet been introduced in users’ life (i.e. except from
Nabaztags, so-called smart meters, televisions or radios have not hit the French market yet),
we built a protocol allowing participants to project themselves in a simulated reality involving
well-known objects and services. By asking them to draw a schematic representation of a
non-existing but functioning objects’ ecosystem (i.e. depicted in a provided use case scenario
very similar to the ones proposed in sub-sections 3.2.2 and 3.2.3), we tried to extract their
mental models. Since people generally do not develop the same mental model of a system
[Payne, 2003], we focused on measuring users’ understanding of the different
representations or schemas with a questionnaire. During this qualitative evaluation, we
introduced an anonymous schema designed by our team (i.e. illustrated on figure 5) in order
to get feedback from users. Interviews were also conducted in order to gather more insights
and better understand users’ representations. After the experiment, we finally tried to make
connections between the schemas in order to identify patterns or remarkable strategies.
This 45-minutes experiment has been iterated two times with groups of 6 design students
from different schools and with different background (i.e. the first panel was mainly composed
of product designers interested in the design of connected objects while the second was
mixing interaction, spatial and product designers with no specific affinity with the domain).
Twelve drawings representing a system of height objects (e.g. an alarm clock, lamps,
shutters, mail box, heater, coffee machine, photo frame) leveraging information or knowledge
from Web resources (e.g. several social networks, music on demand platforms, online
agenda, traffic jams platform, etc.) were produced. Besides our interest in knowing which
drawings were best perceived by participants, we were concerned by the validity of the
overall concept of object-based services and environment-based services.
A grid, which incorporate our research questions, was finally created to interpret the
productions with regards to the:
– understanding of object-based and environment-based services. Are services
represented? Can we count them? How many relates to an object? How many relates to the
environment?
– strategies of categorization. Are objects and Web resources organized in clusters?
Can we count them? Are they labeled?
– type of interconnections between objects. How are links directed? How many
different types can we count? Are they captioned?
– representation and application logic. Does the schema follow a specific structure? Is
the logic of the service represented? How?
18 IJODIR. Volume X – n° X/2010
Figure 5. A refined version of the objects’ ecosystem designed by our team.
5.3.2 Insights
Surprisingly, this exercise has been considered as difficult by most of the participants.
Many of them had trouble envisioning a world where connected objects are able to process
information an services. This is reflected by the different strategies adopted by users to
represent the ecosystem. While the clearest drawings can be compared to network maps
where objects and Web resources are indifferently represented as nodes, others use
alternative metaphors, as shown on figure 6. For example, users created personal timelines
showing a sequence of tasks, as well as representations where objects revolve around a
“computer brain” or an “internet bubble” making the link between the physical and digital
worlds. If users did not use the terms object-based and environment-based services, the way
they grouped objects and labeled some of them indicates that such concepts are underlying
in their representations. This analysis validates the taxonomy described in section 3.2 and
brought us to argue that designers and researchers need to:
– Explain the different type of services. Computer-based, object-based and
environment-based services involve different mechanisms and interconnections. Each of
them should be communicated in a way that allows users to clearly distinguish and
understand them. This therefore requires a learning phase where people are introduced to
the capabilities of services.
– Considerate object-based services as single features. If objects can be considered
as platforms of services, the model of multiple services clustering lots of different features
does not seem to make sense for users. In most cases, participants appear to consider
object-based services as a range of simple augmentations leveraging the functional DNA or
the user interfaces of objects. This would require avoiding creating complex services that
combines too many Web resources and propose another term.
– Enable reallocation of services. Results of the experiment showed that people
sometime arbitrarily distribute the intelligence as they see it. While some users may
considerate some objects as querying another, others can see the latter as controlling the
Designing for the Ubiquitous Computing era 19
first one. Representations provided by designers should therefore be flexible enough to allow
users to move a service from an object to another in order to make the system fit their own
mental models.
– Design multi-scale representation modes. We learned that users do not always care
about the functional logic of services. If a graphic language would make some parts of the
inner working of applications visible and allow people to change parameters, exposing the full
logic of services would cause an information overload. That is why visualizing tools allowing
users to “zoom” in their objects’ ecosystem and switch from global views to detailed
representations are required.
We argue that such insights about users’ mental models are required to properly design
connected objects and services. Since services constitute a new design material, it is need to
understand how users perceive them as part of an object or the environment. By collecting
such observations, we aim at driving the creation of systems that supports users’ models and
are therefore easier to understand and to use. In future work, we plan to re-conduct this
experiment with different panels to validate our findings and confront users with real
connected objects. We are interested in observing how mental models can evolve and what
kind of tools or artifacts can support this change.
Figure 6. Different user-defined metaphors of the same objects’ ecosystem.
5.4. Supporting latent needs
Besides goods tracking, energy consumption and assisted living matters (i.e. often
discussed in European projects such as SENSEI [Wilson et al., 2009]), researchers and
experts seem to have difficulty finding Ubicomp applications involving domestic objects.
Leveraging Web resources to create new services delivered by objects opens up for
numerous possibilities, but do not revolutionize people’s life. Looking for disruptive
innovations, researchers usually depict use case scenarios that often suffer from a
comparison with computers and smart phones applications, considered as easier to deploy or
20 IJODIR. Volume X – n° X/2010
sometimes use. For Weiser, Ubiquitous computing enables “nothing fundamentally new, but
by making everything faster and easier to do, with less strain and mental gymnastics, will
transform what is apparently possible.” [Weiser, 1995]. Web tablets such as Apple iPad
resonate with the vision of Weiser and especially the Parc Pad device. While tablets did not
offer more features than laptop computers (i.e. on the contrary, the iPad did not initially
supported multitask and still not offer to play Flash animations), they can be grabbed and
used anywhere. One of the strong aspects of Ubicomp is to mitigate the importance of
traditional computers, it makes computing immediate, trivial and easy to use. We do not need
to connect objects to the Internet per se, we simply have the opportunity to access
information and interact with Web resources differently, in more natural ways. It is therefore
up to designers to investigate users daily life and identify what tasks or situations can be
better supported.
5.4.1. Generating service cases with users
We conducted two studies at the beginning of our research project, which illustrate the lacks
of existing methodologies applied to the Ubicomp domain. They were conducted with the
same panel than the one who participated in the mental models investigation and aimed at:
– identifying potential objects to be augmented with a contextual investigation. The six
fourth-year students agreed to participate in a study about their objects’ use. On a daily basis,
they had to fill out a diary with a list of all objects they interacted with (i.e. the following
categories where created: at home, at school, in public transportations as illustrated on figure
8). Ten pages corresponding to ten consecutive days where proposed in the booklet, which
was also composed of another section inviting people to select five of their favorite objects.
For each of them, they had to answer questions about their use (i.e. what do they do with it)
and the motivations that led us to buy them. They also had to provide information about their
degree of collectiveness (i.e. personal, used by few member of the family or shared with
many different people), their mobility (i.e. fixed, sometimes moved, brought outside the home)
and their perception (i.e. functional, aesthetical and affective values). They finally were asked
to describe the features that can be performed by each object and to compare them with
ideas of new capabilities.
– combining objects and Web resources during a participatory design session. We
asked the panel to create service cases involving connected objects. In order to facilitate the
creation process, we provided them with cards representing daily life objects (e.g. television,
photo frame, camera, washing machine, fridge, hair dryer, etc.) and Web resources (e.g.
social network news feed or messages, metro timetable, cultural agendas, traffic or air quality
information, etc.). They had one hour to come up with combinations that would benefit
someone and to fill out a “service card” for each of their ideas. Describing the way objects
and Web resources operate and the added value of this composition, these cards were
shared with other participants at the end of the session and evaluated with sticky notes. Each
student could only vote for the three cases that he found the most interesting. Figure 8 shows
a picture of the session.
Designing for the Ubiquitous Computing era 21
Figure 7. Two pages of the diary used by participants to report their objects’ use.
5.4.2 Insights
These studies produced unexpected insights. While we were looking towards ideas of
objects to augment and new services to instantiate on objects, we found out that users could
not project themselves in the future or predict the long-term value of connected objects.
The contextual investigation confirmed the dominant use of computers and smart phones,
which were picked by every participant, and the reduced number of objects possessed by
students, whose temporary lodging and relatively low income do not allow for much variety.
Users had globally very few ideas on how to augment their most-preferred or used objects,
maybe due to a low motivation and a lack of incentive. If structured diaries are easier to fill
out for participants, they require investing a precious time. We might have improved the
number and quality of contributions by using sampling methods, which consist in asking
participants to answer questions on specific signals (e.g. a text message or e-mail). As
mentioned before, participants did not clearly see the opportunity to enhance objects with
Web resources but mentioned general problems that they encounter in objects’ use.
The participatory design session generated around thirty concepts of object-based
services. While interesting ideas related to alternative ways to communicate information
through ambient displays or lights, most of them suggested functionalities that were already
provided by existing terminals or relayed in science fiction movies. This reveals the difficulty
for participants to evaluate a concept. One that sounds uninteresting might be considered as
indispensable when experienced, while another that seems exciting can be disappointing
afterwards. Many service cases were especially inspired from dreamt visions of the
automated homes. This indicates the influence of what people know in their creation process.
When interviewing Rafi Haldjian7, Kuniavsky learned that focus groups conducted during the
7 Former CEO of Violet, the company that designed the Nabaztag.
22 IJODIR. Volume X – n° X/2010
development of the Nabaztag to brainstorm potential uses for the connected rabbit did not
brought valuable insights [Kuniavsky, 2010 248]. We argue that it is therefore needed to
explore alternative research methods that would allow users to experience few services with
specifically designed interactions. In future work, we plan to investigate the creation of
intermediary objects and probes.
Figure 7. A participatory design session aiming at generating service cases.
6. Conclusion
In this article, we addressed the research domain of Ubiquitous Computing from a design
perspective. We stipulated that advances in technology, changing practices and market
trends made possible the realization of a world were computational capabilities are
disseminated in everyday objects. The extension of the Internet to physical artifacts today
represents a logical evolution for manufacturers whose products are challenged by the digital
convergence of medias and telecommunications. This context brought the research
community to shift from a technology-driven approach to a user-driven one. If not all
technical barriers have been lifted, it is now possible to leverage Web of Things architectures
and electronic platforms embedding Web servers to easily create services mixing objects with
Web resources.
This led designers to step in and to collaborate with engineers to focus on the design of
user experiences and to acknowledge the need to explore the implications of using services
as a core component of objects. The use of information as design material opened up the
possibilities for objects to share content delivery mechanisms, knowledge and behaviors with
other objects, but also existing Web resources. Such capabilities allow for new types of
services that monitor, augment or orchestrate objects to be created. We argued that each of
them revolve around a different medium, respectively a computer platform, a physical artifact
and an environment. We then demonstrated that a number of issues regarding the
connectivity, dependency, settings and interoperability of connected objects could be
anticipated from basic use case scenarios. This led us to identify and discuss the need of
considering the collective dimension of objects, as well as the means of control given to users
and the way services supports their daily tasks.
Designing for the Ubiquitous Computing era 23
In the final part of the article, we discussed the methods and tools developed or adopted
to address three main research tracks on the exploration of possible future of objects, the
capture of users’ mental models and the production of use case scenarios involving objects
and services. A literature review and a benchmark of existing connected objects brought us to
work toward the redesign and the reinvention of objects as functional modules that can be
composed and shape-shifting entities defined by services. We argued that making services
more visible would free users from the mobile mediated interactions that are often leveraged.
We then presented the first results of an experiment aiming at better understanding the
perception of connected objects’ ecosystems by users. We finally discussed the low utility of
conducting diary-based contextual investigations and participatory design sessions in the
identification and creation of new concepts of services.
The first findings of these ongoing experiments today allow us to work towards a user-
centered design strategy. If researchers should keep exploring the outcome of Ubicomp and
its applications, we argue that it belongs to designers to contribute to the research community
by:
• exploring the possibilities and issues of using services as a design material
• studying the impact of services on objects’ shape, interaction and relations
• defining services models that matches the mental models of users
• developing tools that allow to investigate, validate and create with users
As pointed out by Kuniavsky, there is a need to design “multi-scale user experiences” where
digitally mediated activities move between scales and involve the participations of users at
different scales [Kuniavsky, 2010 174]. The major challenge of Ubicomp is to maintain a
perceived continuity of experiences and we hope to contribute to both design and computer
science communities by working towards a better definition of such practice.
7. Bibliography
Araya, A. A. 1995. Questioning ubiquitous computing. Dans Proceedings of the 1995 ACM 23rd annual
conference on Computer science, 230–237.
Bassi, Alessandro, Geir Horn, Gerald Santucci, Peter Friess, Thomas Sommer, Sebastian Lange,
Gereon Meyer, Ovidiu Vermesan, et Jesper Holmberg. 2008. Internet of Things in 2020: a
roadmap to the future. European Commission: information society and media.
Bell, G., et P. Dourish. 2007. « Yesterday’s tomorrows: notes on ubiquitous computing’s dominant
vision ». Personal and Ubiquitous Computing 11 (2): 133–143.
Bohn, J., V. Coroama, M. Langheinrich, F. Mattern, et M. Rohs. 2005. « Social, economic, and ethical
implications of ambient intelligence and ubiquitous computing ». Ambient intelligence: 5–29.
Bohn, J., V. Coroamă, M. Langheinrich, F. Mattern, et M. Rohs. 2004. « Living in a world of smart
everyday objects—Social, economic, and ethical implications ». Human and Ecological Risk
Assessment: An International Journal 10 (5): 763–785.
Borgmann, Albert. 1987. Technology and the character of contemporary life: a philosophical inquiry.
University of Chicago Press, mars 15.
Boussard, Mathieu, Benoit Christophe, Olivier Le Berre, et Vincent Toubiana. 2011. Providing user
support in Web-of-Things enabled smart spaces. Dans Proceedings of the Second
International Workshop on Web of Things, 11:1–11:6. WoT ’11. San Francisco, California:
ACM.
Buchanan, R. 2001. « Design research and the new learning ». Design issues 17 (4): 3–23.
Buyya, R., C. S Yeo, et S. Venugopal. 2008. Market-oriented cloud computing: Vision, hype, and reality
for delivering it services as computing utilities. Dans Proceedings of the 10th IEEE
24 IJODIR. Volume X – n° X/2010
International Conference on High Performance Computing and Communications (HPCC-08,
IEEE CS Press, Los Alamitos, CA, USA).
Carter, S. A, et J. Mankoff. 2004. Challenges for ubicomp evaluation. Computer Science Division,
University of California.
Carvajal, Doreen. 2007. Cloud of worry gathers over wireless health risks - The New York Times.
http://www.nytimes.com/2007/09/23/business/worldbusiness/23iht-
wireless24.1.7608114.html.
Davidoff, S., M. Lee, C. Yiu, J. Zimmerman, et A. Dey. 2006. « Principles of smart home control ».
UbiComp 2006: Ubiquitous Computing: 19–34.
Dearman, D., et J. S Pierce. 2008. « It’s on my other computer!: computing with multiple devices ».
Dey, A. K. 2009. « Modeling and intelligibility in ambient environments ». Journal of Ambient Intelligence
and smart environments 1 (1): 57–62.
Edwards, W., et R. Grinter. 2001. At home with ubiquitous computing: Seven challenges. Dans
Ubicomp 2001: Ubiquitous Computing, 256–272.
Encarnaçao, J. L, et T. Kirste. 2005. « Ambient intelligence: Towards smart appliance ensembles ».
From Integrated Publication and Information Systems to Information and Knowledge
Environments: 261–270.
Frohlich, D., et R. Kraut. 2002. « The social context of home computing ». Inside the Smart Home: 127–
155.
Green, William S., et Patrick W. Jordan. 2002. Pleasure with products: beyond usability. CRC Press.
Guinard, Dominique, Vlad Trifa, Friedemann Mattern, et Erik Wilde. 2011. From the Internet of Things to
the Web of Things: Resource Oriented Architecture and Best Practices. Dans Architecting
the Internet of Things, éd par. Dieter Uckelmann, Mark Harrison, et Florian Michahelles, 97–
129. Springer, avril.
Hjelm, S. I. 2001. Designing the invisible computer–from radio-clock to screenfridge.
Hourcade, Jean-Charles, Yrjo Neuvo, Reinhard Posch, Roberto Saracco, Wolfgang Wahlster, et Mike
Sharpe. 2009. Future Internet 2020: visions of an industry expert group. European
Commission: information society and media.
Kaplan, Frédéric. 2009. La métamorphose des objets. FYP editions, juin 12.
Kuniavsky, M. 2010. Smart things: ubiquitous computing user experience design. Morgan Kaufmann
Publisher.
Mavrommati, I., et A. Kameas. 2003. « The evolution of objects into hyper-objects: will it be mostly
harmless? » Personal and Ubiquitous Computing 7 (3-4): 176–181.
Mazza, P. 2002. « The smart energy network: Electrical power for the 21st century ». Climate Solutions.
Newman, M. W, J. Z Sedivy, C. M Neuwirth, W. K Edwards, J. I Hong, S. Izadi, K. Marcelo, et T. F
Smith. 2002. Designing for serendipity: supporting end-user configuration of ubiquitous
computing environments. Dans Proceedings of the 4th conference on Designing interactive
systems: processes, practices, methods, and techniques, 147–156.
Norman, D. A. 2002. The design of everyday things. Basic Books New York.
Norman, Donald A, et Stephen W Draper. 1986. User Centered System Design; New Perspectives on
Human-Computer Interaction. L. Erlbaum Associates Inc.
Norman, Donald A. 1999. The invisible computer. MIT Press, août 20.
Payne, S. J. 2003. « Users’ mental models: The very ideas ». HCI Models, theories and frameworks:
Toward a multidisciplinary science: 135–156.
Rogers, Y. 2006. « Moving on from weiser’s vision of calm computing: Engaging ubicomp
experiences ». UbiComp 2006: Ubiquitous Computing: 404–421.
Rouse, W. B, et N. M Morris. 1986. « On looking into the black box: Prospects and limits in the search
for mental models ». Psychological bulletin 100 (3): 349.
Rukzio, E., K. Leichtenstern, V. Callaghan, P. Holleis, A. Schmidt, et J. Chin. 2006. « An experimental
comparison of physical mobile interaction techniques: Touching, pointing and scanning ».
UbiComp 2006: Ubiquitous Computing: 87–104.
Designing for the Ubiquitous Computing era 25
Schneider, Jakob. 2011. This Is Service Design Thinking: Basics - Tools - Cases. BIS Publishers B.V.,
avril 19.
Scholtz, J., et S. Consolvo. 2004. « Towards a discipline for evaluating ubiquitous computing
applications ». Intel Research.
Turner, M., D. Budgen, et P. Brereton. 2003. « Turning software into a service ». Computer: 38–44.
Weiser, M. 1995. « The computer for the 21st century ». Scientific American 272 (3): 78–89.
Wilson, Duncan, Francesca Birsk, Richard Gold, Jan Holler, Mohsen Darianian, Jilles Van Gurp, Markus
Eurich, et al. 2009. SENSEI Scenario Portfolio, User and Context Requirements. SENSEI
Integrating the Physical with the Digital World of the Network of the Future. European
Commission: Information and Communication Technologies. http://www.ict-sensei.org/.
Woodruff, A., K. Anderson, S. D Mainwaring, et R. Aipperspach. 2007. Portable, but not mobile: A study
of wireless laptops in the home. Dans Pervasive computing: 5th international conference,
PERVASIVE 2007, Toronto, Canada, May 13-16, 2007: proceedings, 216.