Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
Toward a Framework for Ecologies of Artifacts: How Are
Digital Artifacts Interconnected within a Personal Life?
Heekyoung Jung, Erik Stolterman, Will Ryan, Tonya Thompson, Marty Siegel
Indiana University
School of Informatics (HCI Design)
919 E. 10th Street, Bloomington, IN 47408, USA
{jung5, estolter, wnryan, tstroman, msiegel}@indiana.edu
technical and functional influences. An increasing number of
artifacts communicate with each other, exchange information, and
share content and data [5, 8, 21] creating a network of artifacts.
The reality of being a user of “not one distinct artifact,” but of a
large set of interconnected artifacts is a growing concern that has
not received enough research attention even though these intricate
and complex relationships among artifacts have been addressed by
researchers from different fields [14, 15, 18, 27, 30]. We have
found that most existing approaches in HCI or related fields are
based on the notion of a single well-defined artifact in relation to a
single user or possibly a group of users (for overviews see Carroll
[4] and Rogers [32]) instead of addressing the issues emerging
from considerations of networks of interactive artifacts.
ABSTRACT
Assuming that an interactive artifact cannot be fully understood
by itself due to their increasing number, we explored how
individual artifacts are related to each other and how those
relationships can be investigated for further design and research
implications. This study suggests a concept of ecology of artifacts
to describe any implicit or explicit relationships among interactive
artifacts in one’s personal life. We conducted two types of studies
– personal inventory study and an ecology map study – to explore
multiple dimensions for understanding a personal ecology of
artifacts. We expect the knowledge of artifact ecology would help
designers and researchers in the field of HCI to create and analyze
interactive artifacts considering their dynamic interplays in an
increasingly ubiquitous technology environment.
This study started from an assumption that digital artifacts are
interconnected both in terms of their functions and meanings [6]
surrounding an individual life. As interactive computing
technologies more deeply pervade into everyday objects, we,
designers and researchers in the field of HCI need to improve our
ability to create and analyze these artifacts, especially considering
their relations to other artifacts, our social and physical
environments and human life [10, 25].
Categories and Subject Descriptors
H5.m. Information interfaces and presentation (e.g., HCI):
Miscellaneous.
General Terms
Design, Theory
In this paper, we suggest that a productive way to understand such
a network of artifacts and its influence on human life is to
examine it as an ecology of artifacts. We define a personal
ecology of interactive artifacts as a set of all physical artifacts
with some level of interactivity enabled by digital technology that
a person owns, has access to, and uses (Fig 1). One assumption in
this study is that the notion of ecology, with its biological roots,
serves well both as a metaphor and as a theoretical construct to
support the examination of complex networks of interactive
artifacts. Based on this assumption, this paper aims to explore
appropriate methods to investigate ecology of artifacts focusing
on how to describe and analyze those dynamic relationships.
Keywords
Artifact, personal ecology, interactivity, design
1. INTRODUCTION
Our reality is becoming increasingly interactive and flooded by
new interactive artifacts and traditional artifacts enhanced by
digital technology. A person uses a number of digital artifacts for
a variety of daily activities including work, communication,
entertainment, etc. As interactive artifacts are pervasive and
influential to various aspects of human life, the interconnected use
of multiple interactive artifacts needs to be investigated in order to
fully understand their meanings and values in human life beyond
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that
copies bear this notice and the full citation on the first page. To copy
otherwise, or republish, to post on servers or to redistribute to lists,
requires prior specific permission and/or a fee.
NordiCHI 2008: Using Bridges, 18-22 October, Lund, Sweden.
Copyright 2008 ACM ISBN 978-59593-704-9. $5.00.
Fig 1. A Personal Ecology of Interactive Artifacts
201
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
what Mitcham [22] labels as a phenomenology of artifacts. He
describes this school of thought as dominated by the idea that
technology can be studied as consisting of artifacts with inherent
designed qualities and, once placed in the world, evoke a space of
possibility and limitations to its environment and its users.
2. OBJECTIVES
Every newly designed interactive artifact will inevitably become a
part of someone’s ecology. The challenge for any interaction
designer is to know how to think about a specific single artifact in
relation to its ecology [20]. There already is an intense
“competition” among artifacts for a place within these ecologies,
and only those devices which provide the greatest value in an
ecology will be sustained.
From the works of Verbeek [35], we have been inspired by the
notion that things “act.” This is not an approach that makes
artifacts “alive,” but it does recognize the inherent behavior,
particularly of digitally enhanced artifacts, that make them able to
recognize their environment and act accordingly. From Borgmann
[3], our approach has been influenced by his famous notion of the
device paradigm. According to the device paradigm, there is a
disconnection between the way we design modern artifacts and
what people need to feel grounded in contact with their reality. He
argues that such a development restricts people from having close
relationships with artifacts in a meaningful way. From Latour
[18], we understand the notion of networks as aligned actants
where artifacts in the network work in close relationship with
humans to create combined realities shaping each other.
We are convinced that interaction designers, in most cases, do not
have explicit strategies for dealing with these interconnected
relationships among artifacts except for when it comes to
technical infrastructural aspects or data synchronization across
multiple devices [7, 31, 33]. There are neither theoretical nor
practical approaches, as far as we know, for how a number of
devices should or could be designed to fit in an ecology of
artifacts. As an initial approach for studying an ecology of
artifacts, this paper will introduce a theoretical foundation of
artifact ecologies while considering the influence of
interconnected artifacts in human life. Then, we will investigate
any factors or elements that can be used to describe a personal
ecology of artifacts based on the results from exploratory studies.
It is obvious that these scholars cannot easily be combined into a
coherent theory, and that is not our purpose. However, we fully
recognize that any examination about the relationship between
people and artifacts rests upon a philosophy of technology,
whether conscious or unconscious. As artifacts are becoming
more interactive and even autonomous in increasingly ubiquitous
and pervasive computing environments, a philosophical stance
regarding artifacts offers designers and researchers in HCI
insights and “tools” to reinterpret their meanings and influences
from different perspectives that go beyond the more typical
technology-oriented or user-centered points of views. This
motivated us to investigate inherent properties and values of
artifacts that would influence human life and reality.
Specifically, this exploratory study consists of two parts, the
personal inventory study and the ecology map study. The
personal inventory study is to investigate multiple dimensions of
properties and values in using interactive artifacts through semistructured interviews. The ecology map study is designed to
examine people’s mental model toward the meaning and relations
among interactive artifacts through visualizing the maps. Results
from both approaches are ultimately aimed to suggest a
conceptual foundation for the study of an ecology of artifacts and
to discover particular issues that would emerge when artifacts are
considered in relation to each other. However, the purpose with
our studies at this stage has not been to generalize patterns of
using digital artifacts nor to solve their specific problems.
3.2 Ecologies of Artifacts
As a core and fundamental concept in our study, we have chosen
the notion of ecology. Some researchers have previously taken the
ecological perspective to study human artifacts, since the
metaphor provides meaningful analogies to explain complex
interactions among artifacts, humans and environments. Gibson
introduced affordance theory to explain how the environment
influences the visual perception of humans providing an
ecological foundation for designing new interactive artifacts in
usable ways [11]. Nardi and O’Day defined the concept of
information ecology as a system of people, practices, values, and
technologies in a particular local environment [24]. Forlizzi
introduced the concept of product ecology to analyze the social
use of products [9]. Based on social ecology theory, she specified
various interacting factors in a product ecology including people,
activities, the built environment, and the social and cultural
contexts, which can be used as a framework of designing social
product. Krippendorff has emphasized the ecological meaning of
an artifact as consisting of its possible interactions with other
artifacts, which successively evolve while guiding particular
users’ choices, driving an increase or decrease of species of
artifacts, and thus transforming everyday life [17]. The message
from these authors is that designers need to consider the
ecological consequences of an artifact, not only its independent
visual or usable qualities, when creating a new artifact.
We expect that understanding the concept of artifact ecologies
would lead to knowledge about how people experience and
strategize the use of interactive artifacts and the development of
their ecologies over time. Accordingly, this knowledge would
support practice and research for analyzing and designing new
interactive artifacts considering the complex relationships of an
artifact to other artifacts and to its external environments.
3. THEORETICAL FOUNDATION
To ground our study, we reviewed several theoretical studies for
our study of artifact ecology. Specifically, a philosophical
foundation about relations among humans, things, and their reality
inspired us to bring the fundamental importance of understanding
the influences and relations among interactive artifacts into focus.
Related studies that employed ecological perspectives in the field
of HCI were also briefly discussed.
3.1 Conceptual Grounds on Artifacts
While the philosophy of technology has extensively explored how
humans relate to their reality, there has been less interest in the
more concrete relationship between humans and their artifacts. In
recent years, however, we have seen several new theoretical
attempts in the philosophy of technology bringing artifacts and
“things” into focus. These attempts create an overall framework or
map of existing philosophy on new technologies and artifacts. All
attempts mentioned below can be viewed as a striving towards
202
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
their artifacts without revealing any intent to investigate
ecological aspects. Instead of using a couple of opposite
adjectives, we asked participants to select particular adjectives to
understand aspects of using digital artifacts from their subjective
perspectives and also to analyze interview data more efficiently.
3.3 Ecological Concepts in HCI Research
As research on mobile and ubiquitous computing is increasing in
the field of HCI, more integrative perspectives are required to
investigate the relations between artifacts and their diverse
contexts of use. Although several studies do not directly mention
the terms of ecology, their approach to analyze interconnections
among multiple devices provides insights for ecological concepts
to this study. Oulasvirta et al. [29] suggested the concept of
mobile kits to solve problems of managing multiple devices
focusing on physical handling of devices and cross-device
synchronization. In another article [28], he also introduced
broader issues of context-awareness, seamlessness and nondisruptiveness when multiple devices work together in ubiquitous
computing environment. Mainwaring et al. [19] investigated the
relationships among people, places and services when using
mobile kits through various urban interfaces. Their study
considered ubicomp as a paradigm of trusted and environmentally
embedded computing emphasizing its increasing influence on
everyday environment. Dearman et al. [7] focuses on the personal
computing with multiple devices suggesting design implications
for task and context-oriented device usage rather than applicationspecific usage. As those studies indicate, mobile and ubiquitous
computing is hard to be investigated without considering its
interconnections with environment and other artifacts as an
ecosystem. Based on previous studies, this study aims at
suggesting a common ground to describe and understand the
artifact ecology with an analytic framework.
In the beginning, the participant was asked to list every interactive
artifact they use on a post-it card and to select adjectives that
describe each artifact the best. Example adjectives were given to
cover various dimensions of properties of interactive product such
as physical appearance, usability, functions, etc., but the study
was also open for participants to come up with their own words.
Table 1 shows the list of example adjectives. In the end, the
participant was asked to describe detailed reasons for selecting
some adjectives as a debrief session. This study provided us with
a quick understanding of people’s overall use and reflective
response about each artifact. More importantly, it provided us
meaningful insights about the influence of digital artifacts in a
personal life and implicit relationships among different artifacts,
which will be discussed later.
4.2 Part 2: Ecology Map Study (EM)
As an extension of the first part, sketching methods were
employed to visualize relationships among personal artifacts as an
ecological map. Sketching has been frequently used to discover
people’s implicit conceptions toward complex topics [34]. As the
concept of ecology is very new and hard to be articulated, we
expect that methods of visualization would be appropriate to
reveal implicit relationships among artifacts, complementing the
interview results from the personal inventory studies. The ecology
map study was done in a simplified way of sketching by using
artifact-labeled post-it cards, which were made in the previous
part of the personal inventory study.
4. METHODS
Based on the theoretical foundation, we conducted exploratory
studies to get more detailed insights about any perceived relations
among digital artifacts. As an exploratory attempt to develop a
preliminary framework for studying artifact ecologies, our study
mainly investigated how people consider their interactive digital
artifacts and any ecological concepts among them.
Hard
Stable
Soft
Unstable
Heavy
Safe
Light
Dangerous
Mechanical
Friendly
Emotional
Unfriendly
Compact
Primary
Large
Secondary
Flexible
Professional
Rigid
Relaxing
Attractive
Remarkable
Ugly
Mundane
Young
Private
Old
Public
Clean
Manageable
Dirty
Scattered
Neat
Duplicable
Shabby
Irreplaceable
Reliable
Secure
Fragile
Vulnerable
Economic
Connected
Luxurious
Isolated
Lovely
Redundant
Poor
Essential
Portable
Relieving
Stationary
Nervous
4.1 Part 1: Personal Inventory Study (PI)
Sensitive
Fast
Unresponsive
Slow
The first part, the personal inventory study, is aimed at
investigating what digital artifacts people own and how they use
and value each of these artifacts. Referring to the Repertory Grid
Technique (RGT) [13], we conducted semi-structured interviews
with individual participant. The RGT is a method of elucidating
the personal construction of her or his environment (i.e. artifacts,
other persons) based on semantic differentials (i.e. bad-good,
isolated-connected). Considering its appropriateness to explore
multiple dimensions of design space from idiosyncratic
perspectives, we modified the method in a simpler way to
understand what people are mainly concerned with about each of
Handy
Useful
Inconvenient
Useless
Simple
Advanced
Complicate
Elementary
Playful
Valuable
Serious
Worthless
Easy
Efficient
Difficult
Boring
Obtrusive
Helpful
Calm
Burdening
Encouraging
Smart
Discouraging
Dumb
Exciting
Timeconsuming
Multi-purpose
Singlepurpose
The exploratory study consists of two parts, the personal
inventory study (PI) and the ecology map study (EM). 5 sessions
of pilot studies were conducted to refine the study process and
methods before the real study. 10 subjects participated in the
actual study and each session took about 90 minutes (60 mins for
the PI and 30 mins for the EM in overall). All participants were
from the department of Human Computer Interaction Design (all
graduate students, 5 men and 5 women) at our university.
Assuming that they are interested in new computing technologies
with diverse backgrounds (i.e. industrial design, computer
science, information systems), we expected this group to be more
appropriate for finding grater dynamic usages of various digital
devices compared to average user groups.
Table 1. List of example adjectives to describe each artifact
203
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
We did not give any specific instructions for drawing the maps,
but asked participants to attach post-its labeled as artifacts on a
whiteboard according to any relations they have conceived, to
draw any graphic elements (e.g. line, box, circle), and to write any
notes to express those relations more clearly. A white board was
used so that subjects could draw or write any comments and edit
them. Many of participants were not sure what or how to draw at
first, while others categorized their artifact-cards on the board
very quickly. Despite these differences, the study results led us to
meaningful insights about personal ecologies of artifacts.
5.1 Results1 (from PI)
A total of 105 items of interactive artifacts were mentioned by 10
participants. The selection of “digital interactive artifacts” was
quite dependent on participants’ subjective decision. We asked
them to include any types of interactive artifacts using digital
technology that they frequently use or consider important to their
life. Portable devices (i.e. laptop, mobile phone, digital camera,
music player) were mainly mentioned considering the lifestyles of
our participants as graduate students. Other items included
headsets, computer monitors, mice, hair dryers, microwaves,
coffee makers, vacuum cleaners, thermostats, etc. Although some
of them may be not directly related to ‘digital technology’, we
included them in the study considering their subjective importance
to participants. During this study, we did not mention any
ecological concepts to participants. Table 2 shows the number of
artifacts that each participant (P1, P2, P3, …, P10) mentioned.
5. RESULTS
In this section, we describe the results of our analysis of the user
study - both from the personal inventory study and the ecology
map study. As mentioned earlier, the purpose of our study was not
to find general patterns or problems of using digital artifacts in
our everyday life. Instead, it was an exploratory study with a
purpose to investigate if any ecological concepts are perceived or
described by users when they describe their use of digital artifacts
or if any emergent patterns could be discovered when considering
the artifacts in terms of ecology.
We analyzed the results from the personal inventory study by
coding every single adjective that participants selected from the
list and all additional comments mentioned to describe each
artifact. First we categorized the adjectives and comments into
similar types regardless of artifact types. Some of them were
about attributes inherent in the artifact - designed properties,
while others were about personal reflections on artifacts subjective values. We are aware of the difficulty of making these
distinctions, both philosophically and practically, but here we use
them as a first way of organizing our results.
The goal of analyzing the results is that it will lead to a
preliminary framework to understand the concept of artifact
ecologies, to explore methods for studying it, and to discover any
important issues that need to be further studied. Therefore, the
findings from the study may not necessarily be something new but
already familiar to us. However, we have looked for signs that
would reveal new design and research issues from an ecological
perspective. Specifically, the results were analyzed by focusing on
two points. First, we investigated all the particular properties or
values of single artifacts from which people construct relations
among artifacts. This is mainly done based on the analysis of the
personal inventory study.
Second, we analyzed different
groupings of artifacts with regard to their purposes or contexts of
use based on the ecology map study.
P1
P2 P3
P4
P5
P6
1
1
1
1
3
1
1
1
1
1
1
Desktop PC
1
1
1
Laptop
1
1
1
Flash driver
Printer
P7 P8
P9
1
1
1
1
1
1
1
1
1
1
7
10
1
P10 SUM
5
1
1
1
1
1
1
1
1
1
2
1
1
1
1
1
1
1
9
Mp3 player
3
2
1
1
1
2
1
11
Voice recorder
1
1
1
1
1
GPS navigator
1
1
Cable box
1
Others
Total
3
12
1
5
1
1
1
1
Home theater
DVD player
5
3
1
Television
1
1
2
1
1
7
1
1
2
1
1
4
1
2
4
8 11
9
5
9
7
Properties are perceived attributes inherent of artifacts. Based on
some earlier preliminary research we categorized them into the
physical, the functional, the informational and the interactive
covering various aspects of user perceptions toward different
types of artifacts. The physical aspect is the artifact as a physical
manifestation, as a thing; for instance, the laptop, the cell phone,
or the mp3 player. Any descriptions about physical size, weight,
appearance or interface are categorized into this property.
Sometimes, aesthetic qualities from artifact appearance are linked
to artifact’s emotional value (i.e. cute, attractive, awesome). The
functional aspect of an artifact refers to what an artifact is used
for. Comments about practical aspects of using artifacts, such as
task, performance, or features are categorized into this functional
property. The informational aspect is similar to the functional
with the difference that it is about data or content, instead of
functional application. Direct comments about data, information,
or contents created or stored in an artifact are categorized into the
informational property. The interactive aspect is the way a person
can interact with a physical artifact. Issues related to usability or
ease of use are categorized into this property.
7
Digital camera
2
5.1.1 Properties of artifact
10
Mobile phone
Game system
Our scheme resulted in over 700 codes in seven groupings- four
types of artifact properties (physical, interactive, functional and
informational) and three types of subjective values (experiential,
emotional, and social). Some of codes belong to more than one
property or value considering their interconnected relations. In
that case, they are counted as separate codes. The rationale for
coding the study data to properties and values is to suggest a
holistic perspective to compare and analyze various types of
digital artifacts with general criteria. Those artifact properties and
values form a tentative framework to organize some core aspects
of an artifact ecology.
2
1
3
19
8 19 11
11
105
Table 2. Number of artifact items for each participant
204
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
among artifacts. Those factors are manifested as diverse patterns
of use, which we named as layers of artifact ecology.
5.1.2 Values of artifacts
Although value can be interpreted in many ways [10], in this
study we focused on its subjective and reflective aspects as being
derived from the use of the artifacts, which is closely related to
Norman’s concept of reflective responses [26]. We discovered
three criteria of values from the study data: the experiential, the
emotional and the social. Experiential value is connected to
personal memories or reflective responses toward an artifact.
Emotional value is related to how a participant feels toward an
artifact as in the overall impression or character of the artifact
elicited from all perceived properties of a specific artifact [16].
Social value is about symbolic meaning or economic value of an
artifact in a social context. In most cases, specific values are
evoked from a particular property of artifact (if performance of an
artifact is ‘slow’ or its interface is ‘not responsive’, people feel the
artifact is ‘dumb’, which is reflected as an emotional value toward
the artifact). However, regardless of artifact properties, they may
value an artifact as “valuable” or “memorable” according to their
long-term experience of use.
6.1 Ecological Factors (from PI)
From the personal inventory studies, certain properties and values
of individual artifacts were mentioned especially in relation to
other artifacts of the whole inventory. Although those properties
are small in numbers compared to all the properties and values
that describe characteristics of individual artifacts, they are
noticeable as emerging factors when an artifact is considered as
belonging to an ecology. They transform the use of individual
artifacts when individual artifacts are connected to others. We
define them as ecological factors. These factors “build” an
ecology by making connections between specific artifacts or
commonality throughout a group of artifacts. We will discuss
those ecological factors emerging from each property and value.
6.1.1 Connection through information sharing
Connection through sharing information between devices is one of
the most significant ecological factors emerging from the
informational property of digital artifacts. We found that it is
common that participants use more than one computer (laptop,
home PC, or public PC) and that they use other small devices such
as external driver, music player, or smart phone to transfer the
information around.
Coded data from the personal inventory studies provide a tentative
framework to understand overall aspects of perceiving digital
artifacts. With this framework, we could explore various
dimensions of product properties and values (beyond prevalent
usability issues) that can be used to describe a personal ecology of
artifacts and any of their interconnected relationships. They will
be discussed in the later part of findings section.
As many of digital artifacts are used as a group in connection with
others, individual artifacts sometimes tend to be considered as
parts of a higher system instead of separate independent ones.
Accordingly, their role or importance is perceived comparative to
each other. Adjectives describing individual artifacts such as
“primary,” “complementary,” or “secondary” revealed their
relative importance among the whole inventory items. In many
cases, computer is considered as one of the most important artifact
in personal ecologies described as “focal point” or “core.”
5.2 Results2 (from EM)
We also analyzed all the graphic or text items drawn in the 10
ecology maps and got over 100 codes of three types; labels of
grouping, graphic elements to group or connect artifacts (circles,
lines or arrows) and annotations to explain relations or
connections among artifacts.
Participants grouped artifacts by posting several cards more
closely together and others quite apart according to their
perceived relationships. Graphic elements such as circles, lines, or
boxes were often used to distinguish or clarify those groups but
not always. We analyzed them focusing on participants’ intentions
of using graphic elements to represent different type relations.
Sometime, they labeled each group according to reasons for those
relations. The labels cover purpose of use (“work,”
“entertainment”), physical context of use (“home,” “lab,” “living
room”), similar appearance (“minimal design”), same memories
(“gifts”), technical relations (“computer,” “camera”), etc. Some
groups are left without any labels but still separately positioned
apart from other groups. The types of labels and their
relationships will be analyzed in the findings section. Text
annotations—either descriptive sentences or brief words—were
used to add detailed descriptions about the ecology maps. We also
included verbal annotations, not only written messages on the
board in our analysis. The results of the ecology map studies gave
us more subjective ideas about the whole inventory of artifacts in
relation with owners’ identity or lifestyles.
[P3] “My laptop is a core of my computing system. I have a home
computer, but almost never use that for my work. I have all my
work data in this computer (laptop) and I can do so many things
with it. It is the most important device for me.”
[P4] “I have an iPod touch, but I rarely use it for music because I
have another classic iPod. I use the Touch for checking my emails
and schedules on the move or when I cannot use my laptop. I
consider it as a part of my laptop, just for checking emails.”
For several participants, it was found that web applications
(particularly emails and schedulers) take an important role in
connecting different digital devices.
[P2] “After I purchased my iPhone, I got more interested in
harmonizing my laptop, iPhone, and my desktop through a web
scheduler. I don’t use my paper scheduler any more... My
schedule is always updated through web in any of my devices.”
What is interesting from this factor of connection through
information is that people tend to appropriate digital artifacts
according to their customized needs regardless of the initial
design intents of artifacts. For example, as Participant 3 and 4
mentioned above, only a few features of an artifact can be
selectively used according to its relative roles in its higher group
of artifacts while other features of the artifact are remained
unused. Similarly, it is noticeable that artifacts are separate in
their physical manifestation; they are working like modules of a
whole system linked by sharable and duplicable digital
information.
6. FINDINGS
By analyzing details of the results from two studies (PI and EM),
we observed interesting phenomena emerging from the relations
among a group of personal artifacts. Specifically, some properties
and values from individual artifacts display ecological factors,
which shape the functional network or aesthetic commonality
205
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
phone, portability is considered as one of the most important
physical properties for interactive artifacts in general. At the same
time, due to the versatility of compact digital devices, people tend
to have a couple of different devices that can work for the same
tasks, for examples, digital camera and camera phone. Likewise,
diverse contextual affordances make physical properties (i.e.
weight, size, interface, etc) of artifacts more important to users
when they select one device among similar ones. For example,
participant 2 mentioned that the camera of his mobile phone is
more useful when taking pictures to memorize some information
(car numbers, street signs) just for the record on the walk,
although he prefers to his digital camera in other situations.
6.1.2 Connection through functional compatibility
Regarding the functional property of digital artifacts, technical
compatibility is considered as an important ecological factor that
enables information sharing among different artifacts. Out of
many comments on functional properties of artifact by
participants, “connected” or “isolated” was frequently mentioned
considering technical connections or constraints on data sharing
among different digital devices. Participant 1 had the following to
say that he has more than one portable music players, one of
which he cannot use anymore as he had changed operating
systems of the main computer.
[P1] “I feel like my old mp3 player more and more isolated and
useless. It does not work with my computer. I cannot download or
upload any music. Actually, I do not need that anymore but I just
keep that in my drawer. I don’t know what to do with that.”
In addition, the conceptual boundary between physical artifact and
intangible digital content in it has been largely blurred. Sometimes
digital contents tend to be considered similar with materialized
“thing.” For example, participant 5 uses three flash drivers for
saving different types of data (i.e. personal, work, entertainment).
Her mental model of information structure is quite coupled to the
physical entity of flash drivers.
Also, some people mentioned they are satisfied with their laptops
because they have sufficient number of USB ports that enable the
connections with multiple devices at the same time. Similarly,
display or sound devices are described as “flexible” because they
can be used in connection with different devices.
[P5] (pointing out each flash driver) “this is my personal data,
this is my all my work data, and this is my music and movies. I
have used these three flash drivers more than three years. So I
remember almost of my data structure- which is stored where.”
[P10] “Recently I purchased a new television for watching
movies with my home theater system. But sometimes I use it with
my laptop when working at home. It is better than to work with a
small screen.”
Participant 8 also commented that he usually said, “Here’s my
music” when handing his iPod to his friends, indicating the entire
play list inside the device. Participant 9 mentioned that he uses
different computers to manage different types of data, which
means separating work and leisure activities into different
physical entities and contexts.
[P8] “I think I use my home theater system more with my iPod
than with DVD player. I have all my music there (iPod). I don’t
want to listen to them through earphones even at my home.”
Sometimes, a personal artifact can be even connected to others’
personal ecology through technical connections supporting social
activities.
[P9] “I only save my music files in my home computer. Actually I
use it more like a music player. I use my laptop for more
professional tasks.”
[P7]“I always bring my iPod to parties at friends’. I can just plug
my music to their computers (speakers). It is so convenient.”
The pervasiveness of information embodied in appropriate
physical devices supports people’s ability to easily remember the
structure of their distributed personal files without much cognitive
burden [12, 21, 23]. The evolving definition of the relations
between physical and informational properties opens up new
design opportunities to embody appropriate physical and
interactional properties to digital contents considering various
forms of affordance and interactions to the contexts of use.
Although those physical and interactional properties are not
directly related to the informational or functional connections
among artifacts, they set new common criteria where users’ need
and expectations (such as portability or transparency) for
interactive digital artifacts could be grounded on.
This factor of functional connection augments the use of a single
artifact flexibly through customized combinations with different
devices. On another hand, the compatibility of functional
properties among different devices result in overlapping features
across multiple devices. Many respondents selected adjectives
such as “redundant,” “old” or “useless” as functional property or
social value of an artifact. Most of them mentioned that they do
rarely use all the features of their mobile phones (such as music
player or camera). Several participants mentioned that redundant
features with other artifacts are often make them hesitate when
purchasing a new digital artifact with concerns on waste of
technology (memory or storage space) as well as economic cost.
6.1.4 Ecological values changing behavior
Surely there are trade-offs in device convergence between
portability and quality of functions. In this paper, we do not argue
the advantages and disadvantages of convergent devices or singlepurpose appliances, but we emphasize some complaints about the
lack of choices for users to select a device with features that they
precisely need. Designers need to consider how to support to
develop users’ personal ecologies based on their own needs and
artifacts that they already have.
Value criteria defined in the previous section (emotional,
experiential & social) demonstrate influences of a set of digital
artifacts on users’ behavior and perception towards artifacts and
experience with them in general.
The introduction of a new artifact to an ecology can influence
various aspects of users’ daily behaviors as well as the use of
other artifacts. Considering the pervasive influence of digital
devices that empower users to freely control digital contents (by
accessing, creating and editing them), it was interesting that two
of participants mentioned about spontaneous changes in behavior
due to immediate access to or control of digital information by
using a new interactive artifact.
6.1.3 Contextual affordance and commonality
The physical and interactional properties are related to the
commonality of using digital artifacts or specialized affordance of
individual artifacts in diverse contexts of use. As people use a
number of portable digital devices, such as laptop or mobile
206
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
[P8] “After using a DVR cable box, I feel like I cannot wait for
something I do not want to watch. All TV programs are recorded,
I don’t have to watch all of them. I usually skip most of boring
parts and quickly go through the overall storylines… similar with
watching video clips on Youtube… I have a full control over
playing videos. These days, I really hate going to theater for
watching movies. I cannot stand all the boring parts.”
number of artifacts support similar tasks, they may provide subtly
different emotional or experiential values for very specific
purposes of use. In addition, it is noticeable that layers of artifacts
based on purposes of use can represent what participants’ main
activities are in their everyday life and how those activities are
structured through various artifacts.
At the same time, an artifact belongs to multiple layers of
different purposes of use. It could be said that the more layers are
overlapped for a certain artifact, the more important or essential
they are considered representing more connections to users’
activities and to other artifacts. For this reason, computers were
often labeled as ”core system,” “focal point” or “my heart,” with
many connections and layers overlapped in an ecology.
[P9] “After I got my iPhone, I check my emails every half an hour
or so, wherever around the campus. Sometimes, while talking with
my friends, if we come up to something unclear, for example
movie titles or actor names, I immediately search for the internet.
I feel like having any responsibilities for that by having an access
to the Internet all the time.
We also found that individualization and socialization are
important aspects of using interactive artifacts that blur the
distinction between private and public activities. People can reach
others or intentionally detach themselves from others by flexibly
adjusting the boundary of personal or social space using
interactive artifacts.
The layer that is grouping some artifacts based on their purpose of
use explains how they work together for similar types of tasks.
Although they are not always used together at the same time, they
are related through ecological factors such as connections through
information or functions to support similar user goals.
6.2.2 Context of use
[P5] “Sometimes I put my earphones when I want to detach
myself from others in public spaces. I don’t really want to listen to
music but I can pretend that I am listening to music when I don’t
want to talk with others around.”
Other participants grouped artifacts according to their physical
contexts of use- where artifacts are located or where they are
mainly used- such as “at home,” “living room media,” “work,
lab,” “carry around, surrounding,” and “communication” (Fig 5).
[P8] “Working with my laptop isolates me from others by
focusing on what I am doing even in public spaces. However,
ironically the portability of laptop allows me to be around others
who may be working on their own things just like me. I can be just
around others without active socializing, which I prefer.”
Although this factor of changing behaviors is not about direct
connections among artifacts, it is an influential factor that may
change overall patterns of human behaviors transforming their
perceptions about the reality and thus their expectations for other
artifacts around them.
Fig 3 and 4. Highly structured layers of artifacts according to
purpose of use (left), layers divided as subgroups according to
purpose of use (right)
6.2 Layers of Artifact Ecology (from EM)
From the ecology map studies, more explicit relations among
interactive artifacts were visualized. We found several patterns of
those relations– how ecological factors are exemplified in real
contexts of use. We define the concept of layers in an ecology as
different types of relations among artifacts that shape an ecology.
Although they are hard to generalize from this initial study, they
gave us insights to compare our framework of ecological factors
with users’ own description of relationships among artifacts.
6.2.1 Purpose of use
Fig 5 and 6. Layers divided according to physical contexts of
use (left), overlapping layers according to multiple
connections among artifacts (right)
Many participants considered relations among artifacts based on
their purpose of use, for example, by categorizing some into
groups of “work,” “computing,” “gaming,” “entertainment,”
“communication,” “graduate studies,” “individual entertainment,”
“archive storage,” “freelance-web related work,” etc (Fig 3, 4).
Specifically, participant3 grouped laptop, USB driver and digital
voice recorder into work-related artifacts, while grouping digital
camera and music player into entertainment. Participant 4 also
divided artifact groups into laptop, desktop, printer, and external
hard driver into work-related computing devices, and television,
gaming system, and AV system into entertaining group.
It is interesting that some participants labeled each interactive
artifact with very specific purpose of use such as “communication
with family” (for a mobile phone) or “communication for team
appointment” (for a computer). This shows that although a
Fig 7 and 8. Artifacts positioned according to different
categories of subjective meanings (similar appearance, similar
memories, what they care or not, etc.)
207
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
Most of artifacts that our participants had mentioned are portable
and multi-purpose devices (laptop, digital camera, music player,
mobile phone) considering their life styles as graduate students.
Due to their portability, some devices are grouped into more than
one category of contexts (laptop is used at living room and school,
mp3 player is carried around). This tells the importance of
considering contextual affordance when designing interactive
artifacts as discussed in the earlier section about ecological factors
from physical and interactional properties.
Participant 5 grouped her voice recorder and a music player
together because both of them were presented by her father.
Although she has some complaints about the features and designs
of those devices and wants to purchase new devices, she still
keeps using those devices because they remind her of her families
and their love.
Participant 8 grouped his artifacts according to his own categories
of “I don’t care,” “ambivalence” and “I care about” (Fig 8). Those
subjective layers of artifacts represent his emotional feelings to
some extent based on the importance or frequency of using
different artifacts. He additionally added an annotation to his “I
care about” group as “Social, This where I live, work+play”
indicating various individual/social activities and thus emotional
attachment are engaged with interactive artifacts in that group.
It is noticeable that they usually did not use only contextual labels,
but also used labels indicating purpose of use such as
“communication” or “work.” Participant 9 used several different
layers of categorizing artifacts (Fig 6), which show potentials to
relate one artifact to another in multiple ways. He mentioned a
“temporal locality” as a reason for the relation between his TV
and laptop because he usually turns on the TV around when he
uses the laptop in the living room. Similarly, Participant 3
mentioned he feels some relation between his laptop and gaming
device because occasionally he needs to use the Internet
connection for connecting to game help websites. These relations
among different artifacts would have never been mentioned in the
description of using an individual artifact.
This ecological layer of subjective meaning shows how value
criteria for individual artifacts could be applied for the
relationships among different artifacts. Further, it also shows how
those relationships shape the overall user experiences. Especially,
it demonstrates the mutual influence of social norms and personal
experiences, which needs to be studied later as an extension of the
current framework for personal ecology of artifacts.
7. DISCUSSION
The layer that distinguishes artifacts based on their contexts of use
tells that participants are pretty much influenced by physical or
temporal contexts when using an artifact. Especially as the
contexts could be very dynamic due to the portability of digital
artifacts, the support for flexible contextual affordances and
seamless connection with other devices around needs to be further
explored in terms of physical and interactional properties [37].
We suggest insights and implications for further studies based on
the ecological concepts−ecological factors and layers−that we
defined in the previous section.
7.1 Insights from Exploratory Studies
In the findings section, we introduced the concepts of ecological
factors and layers based on the exploratory study results.
Ecological factors mostly emerge from the properties or values
inherent in a single artifact and make connections or
commonalities among all artifacts building an ecology. Ecological
layer is the concept that our study suggests might explain how
artifacts are related to each other through ecological factors in
specific environmental or task-related contexts. Additionally,
connections and commonalities can be described as fundamental
elements that construct an ecology emerging from dynamic
interplays among artifact properties and values in various contexts
of use (Fig 9). In this section, we will discuss insights from those
ecological concepts regarding to the methods (PI and EM) that we
employed for exploratory studies.
6.2.3 Subjective meaning
Another layer that we have found from the study is subjective
meaning, which we named for the value-oriented groupings of
artifacts in an ecology. Several participants grouped a few
artifacts more closely than others without using any clear labels.
However, this relationship may be more implicit than context or
purpose of use and is still worthy of our analysis.
Specifically, some people consider the relations among artifacts
from their similar appearances or memories. For example,
Participant 4 grouped her laptop and digital camera together with
a title of “minimal design” (Fig 7). Although she also uses
technical data transfer between laptop and other digital devices
(such as music players and digital voice recorder), she considered
her laptop and digital camera to be more closely related because
of their similar appearances. Moreover, she mentioned that those
design criteria is one of the most important strategies for her to
select new devices to purchase by selecting new artifacts with
similar appearance that she already have in her ecology.
7.1.1 Conception of ecological factors
Most of the ecological connections were mentioned in relation to
functional and informational properties of artifact. Technical
“connection” or “isolation” among devices is often critically
considered in terms of usefulness of digital artifacts. Also, the role
of informational property is becoming more essential as
information connects different artifacts by virtually flowing
through them. While connections through functional or
informational properties were prevalent in the personal inventory
studies, they were comparatively less visualized in the ecology
map studies. Some participants explained that it might be because
they do not significantly realize those technical issues: they are
usually seamlessly or automatically done without serious
difficulties or they do not remember them in details.
Specific connections or relationships between the physical or
interactive properties of a single artifact with other artifacts were
rarely articulated. We assumed this is because those properties are
more static inherent of an artifact unlike functional or
Fig 9. The compositions of artifact ecology
208
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
informational properties that can be easily connected and shared
among a number of artifacts. However, we could at least find
similar patterns or commonalities from described physical and
interactive properties of digital artifacts from this study. For
example, increasing concerns on the size of an artifact, such as
“portable” or “handy,” were often mentioned as critical physical
properties for individual digital devices. Similarly, interactive
properties such as “responsive,” “easy” or “complicated,” were
often mentioned as main concerns for each device. Although those
properties are not directly related to elements of ecologies
(connections or layers), they imply considerable design issueshow to materialize complicated informational or functional
connections among artifacts with appropriate physical or
interactional properties of each artifact.
in their current ecologies. Reflective values such as emotional or
experiential ones cause a user to feel more attached to some of
their artifacts, even though they are not competitive to others in
terms of objective functional properties or not suitable to their
aesthetic preferences. The value aspects of artifact ecologies need
to be more studied for sustainable and meaningful use of artifacts
[1]. As digital artifacts become more pervasive and influential
surrounding a personal life, it is expected that understanding
someone’s artifact ecology would reveal rich personal information
including their activities, behaviors, reflective values or identities.
7.2 Implications for Further Studies
Although the concept of artifact ecology is yet new and not
significantly considered by many people, at least we could find
some emerging factors that describe the increasing technical,
emotional, or contextual interconnections among artifacts from
the exploratory studies. We expect that understanding those
ecological factors will support designers to create new artifacts
and users to consciously develop personal ecologies while
considering the dynamic relationships among artifacts. At the
same time, as there is an ever increasing emphasis in HCI research
on everyday artifacts that support and improve personal life [2],
we believe that refined ecological concepts can help to predict
how new artifacts can be integrated into already existing network
of artifacts. Based on the results from our exploratory studies, we
suggest the following implications to be further investigated.
7.1.2 Conception of ecological layers
From the ecology map study, we found how people consider the
relationships between artifacts in real use contexts by applying
our initial framework of ecological factors. The purpose and
physical context of use creates layers of artifacts by grouping
relevant artifacts together. It was noticeable that many participants
visualized their own way of appropriating digital artifacts through
multiple connections and layers. For example, people often
grouped some artifacts by selectively using specific features from
individual artifacts or creating different connections of artifact
features depending on diverse contexts of use. In many cases,
those connections were found to be more flexible in artifacts such
as display or speaker systems, which support immersive and
comfortable environments by connecting compact digital devices
with specific features or contents to them.
• Two different methods of our initial exploratory studies, the PI
and EM studies, resulted in meaningful findings in our attempt to
understand and describe dynamic relations among digital artifacts
from an ecological perspective. These studies complement each
other in the way they explore multiple dimensions of the use of
interactive artifacts with their properties and values. Further
investigation with diverse user groups would help to discover
various patterns and phenomena of artifact ecologies based on the
initial framework of artifact ecology from this study.
Due to the multiple layers of an ecology, certain artifacts can
work as a hub in a whole system. For example, laptop belongs to
more than one group of artifacts, which means multiple
connections to other artifacts. It mainly works with home
computer or PDA for work-related tasks or checking schedules,
but also has a relation with a television in physical and temporal
connectivity because a user said that he usually turns on a TV
around while working on with his laptop in the living room. Such
examples shows possibilities of dynamic interconnections among
digital artifacts, which need to be further studied.
• Multi-dimensional properties and values of digital artifacts are
expected to help designers and researchers consider and predict
the overall influences of digital artifacts in human life. As digital
artifacts pervade human life, those considerations would be more
critical in terms of sustainable development of human-artifact
relationships. Considering the mutual influences between artifacts
and users, the perspective of the artifact ecology would provide a
more comprehensive design perspective expanding user-centered
approaches. Moreover, the current framework of a personal
ecology needs to incorporate mutual influences between artifacts
and macro-environments in terms of social/cultural ecosystems.
7.1.3 Ecological values and strategies
The results from exploratory studies can also reveal how people
use artifacts and how they have developed their ecologies to some
extents. For instance, some people showed very structured
hierarchy of artifacts reflecting their strategy of archiving
information through different devices (Fig 3). Others divided subgroups of artifacts considering more local relations according to
specific types of tasks or contexts of use (Fig 4). We also found
one participant displays a flexible attitude to manage a number of
artifacts without organizing any of the information scattered in
different devices in an attempt to reduce any efforts to manage
them (Fig 8). Those people usually mentioned that they do not
need to organize their information in a structured way, because
they remember where all of their data (files) are saved. Although
their artifact ecologies may seem unorganized to others, they are
surely customized under the personal strategies of their owners.
• Ecologies evolve according to individual users’ personal
strategies and appropriation of artifacts. Designers and researchers
should consider how to support users to develop their personal
ecologies flexibly and creatively over time [1, 36]. Beyond
technical connections through informational or functional
properties, contextual affordance through physical or interactive
properties and respect for values need to be critically considered
through a systematic and sustainable design approach.
8. CONCLUSION
In this study, we conducted exploratory studies to investigate
relationships between personal digital artifacts with an assumption
that the experience with an interactive artifact cannot be fully
understood by itself, but needs to be considered in relation with
other artifacts—namely in terms of an artifact ecology. Aiming to
Sometimes, reflective values of artifacts also influence how
people develop their ecologies. The social values such as “cheap,”
“expensive,” “old,” or “advanced” were considered critical when
people change or purchase a new artifact, which would be added
209
Proceedings: NordiCHI 2008, October 20-22, 2008
Full Papers
find any implicit or explicit relations within a personal ecology of
artifacts, we employed two study approaches – the personal
inventory study and the ecology map study. Based on the study
results, we introduced the notion of ecological factors. They form
connections between artifacts or commonalities across the overall
artifacts in an ecology, which structures various layers of
relationships among artifacts. We expect that the conception of
those ecological factors would support different strategies to
develop personal ecologies of artifacts for users, designers and
researchers in the field of HCI. We will continue to develop this
foundational framework through further empirical studies with
diverse user groups in our future work.
[19] Mainwaring, S.D., Anderson, K., and Chang, M.F. Living for
the global city: Mobile kits, urban interfaces, and ubicomp.
In Proc. Ubicomp 2005, Springer. 268-286.
[20] Margolin, V.: Expanding the boundaries of design: The
product environment and the new user. Design Issues, 4(1/2),
59-64. (1998)
[21] McCullogh, M.: Digital Ground Architecture, Pervasive
Computing, and Environmental Knowing. The MIT Press.
(2004)
[22] Mitcham, C.: Thinking through Technology: The Path
between Engineering and Philosophy. The University of
Chicago Press. (1994)
9. REFERENCES
[1] Blevis, E. & Stolterman, E.: Ensoulment and Sustainable
Interaction Design. IASDR 2007.
[23] Nardi, B.: Studying context: A comparison of activity theory,
situated action models and distributed cognition. In Proc.
East-West HCI Conference, 352–359. (1992)
[2] Bødker, S.: When second wave HCI meets third wave
challenges. In Proc. NordiCHI 2008, ACM (2006). 1-8.
[24] Nardi, B. A., O’Day, V. L.: Information Ecologies: Using
Technology with Heart. MIT Press (2000)
[3] Borgmann, A.: Technology and the Character of
Contemporary Life: A philosophical inquiry. University of
Chicago Press. (1984)
[25] Nathan, L.P., Friedman, B., Klasjna, P.V., Kane, S.K., &
Miller, J.K.: Envisioning systemic effects on Persons and
Society Throughout Interactive System Design. In Proc. DIS
2008, ACM.
[4] Carroll, J.: HCI Models, Theories and Frameworks. Elsevier
Publishing. (2003)
[26] Norman, D. A.: Emotional Design: Why We Love (or Hate)
Everyday Things. Basic Books, New York, NY (2005)
[5] Castells, M.: Rise of the Network Society, Blackwell
Publishers, Inc. Cambridge, MA, USA (1996)
[27] Norman, D.: The Design of Future Things. Basic Books.
(2007)
[6] Csíkszentmihályi, M.: The Meaning of Things: Domestic
Symbols and the Self. Cambridge University Press. (1981)
[28] Oulasvirta, A.: When Users “Do” Ubicomp. Interactions
15(2), 6-9. ACM (2008)
[7] Dearman, D., and Pierce, J.S.: It's on my other computer!:
computing with multiple devices. In Proc. CHI 2008, ACM.
[8] Floridi, L.: A look into the future impact of ICT on our lives.
The Information Society (23)1. Taylor & Francis (2007)
[29] Oulasvirta, A. and Sumari, L.: Mobile kits and laptop trays:
Managing multiple devices in mobile information work. In
Proc. CHI 2007, ACM. 1127-1136.
[9] Forlizzi, J.: How Robotic Products Become Social Products:
An Ethnographic Study of Robotic Products in the Home. In
Proc. HRI 2007. ACM.
[30] Pantzar, M.: Domestication of everyday life technology:
Dynamic view on the social histories of artifacts. Design
Issues, 13(3), 52-65. (1997)
[10] Friedman, B.: Value-sensitive design. Interactions 3(6), 1623. (1996)
[31] Poole, E. S., Chetty, M., Grinter, R. E., Edwards, K., More
Than Meets the Eye: Transforming the User Experience of
Home Network Management, In Proc. DIS 2008, ACM.
[11] Gibson, J. J.: The Ecological Approach to Visual Perception.
Houghton Mifflin, Boston. (1979)
[32] Rogers, Y.: New Theoretical approaches for HumanComputer Interaction. Annual Review of Information,
Science and Technology, 38, 87-143. (2004)
[12] Greenfield, A.: Everyware: The Dawning Age of Ubiquitous
Computing. New Riders Publishing. (2006)
[33] Schilit, B.N., Sengupta, U.: Device Ensembles. IEEE
Computer 37(12), 56-64. (2004)
[13] Hassenzahl, M., Wessier, R.: Capturing Design Space from a
User Perspective: The Repertory Grid Technique Revisited.
Interaction Journal of Human-Computer Interaction 12(3),
441-459. (2000)
[34] Tohidi, M., Buxton, W., Baecker, R and Sellen, A. User
Sketches: A Quick, Inexpensive, and Effective Way to Elicit
More Reflective User Feedback, In Proc. NordiCHI 2006,
ACM.
[14] Hayles, N. K.: How We Became Posthuman: Virtual Bodies
in Cybernetics, Literature, and Informatics. University of
Chicago Press. (1999)
[35] Verbeek, P.P.: What Things Do –Philosophical Reflections
on Technology, Agency, and Design. Penn State University
Press. (2005)
[15] Hutchins, E. Cognition in the Wild. MIT Press. (1995)
[16] Janlert, L-E. & Stolterman, E. The character of things.
Design Studies 18(3), 297-314. (1997)
[36] Wakkary, R., & Maestri, L.: The resourcefulness of everyday
design. Proceedings of the 6th ACM SIGCHI conference on
Creativity & cognition, 163-172. (2007)
[17] Krippendorff, K.: The Semantic Turn – a new foundation for
design. Taylor & Francis. (2006)
[37] Woodruff, A., Anderson, K., Mainwaring, S. and
Aipperspach, R.: Portable, But Not Mobile: A Study of
Wireless Laptops in the Home. In Proc. Pervasive 2007,
Toronto, Ontario, 216-23.
[18] Latour, B.: Pandora's Hope. Essays on the reality of science
studies. Harvard University Press. (1999)
210
View publication stats