REP(AIR): An Olfactory
Interface For Bike
Maintenance and Care
In this paper we present Rep(AIR), a research-
through-design olfactory interface for communicat-
ing moments of wear and tear on a bicycle. Rep(AIR)
was designed as a probe in an autobiographical
design inquiry to uncover qualities of the relation-
ship between humans and objects as they relate to
breakage and repair. Key used Rep(AIR), along with a
personal probe notebook, on a 12 day international
cycling trip to reect on and document repairs and
maintenance. We share ndings from our analysis
which extend the discourse on repair to include
moments of wear, maintenance, and care as a part
of the ongoing process of everyday use; recenter
functionality to the human-object team, rather
than the object alone; and highlight teamwork and
collaboration as a way to challenge the hierarchical
human-object narrative. We conclude by noting the
role of Rep(AIR) as a tool which gave the bicycle a
voice—revealing the often uncommunicated experi-
ence of wear and tear on an object.
Cayla Key1, Audrey Desjardins2
1 School of Art + Art History +
Design, University of Washington,
Seattle Washington, United States
2 School of Art + Art History +
Design, University of Washington,
Seattle Washington, United States
Keywords: olfactory interface;
bicycle; autobiographical design;
maintenance; repair; care
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 2
Key & Desjardins | Rep(AIR)
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 4
Wear, care, breakage, and repair are all part of the long (or short) lives
of things. Design and human-computer interaction (HCI) scholars
have built a strong theoretical and empirical corpus of work on repair,
care and maintenance in relation to everyday objects, technological
artifacts, and infrastructures. Repair has been positioned in relation
to planned obsolescence and sustainability (e.g., Blevis 2007), as a
practice that negotiates endurance in a sociomaterial system (e.g.
Rosner and Ames 2014; Houston et al. 2016), as well as an everyday
creative practice (e.g., Maestri and Wakkary 2011). In this paper, we
extend this work by turning our attention to the intimate and ongo-
ing practice of maintaining one’s things. Previous works have made
clear that repair should be considered a ‘normal’ part of our life with
objects (Jackson 2014), and that repair and reuse have important
implications towards sustainability and personal empowerment. In
response, we designed an interface to prompt and encourage care
and maintenance interventions in and around an everyday practice—
that of riding a bicycle.
In this paper, we present Rep(AIR)—an impact-triggered scent dif-
fuser mounted to the handlebars of a bicycle—and the ndings of a
12 day autobiographical design inquiry. Rep(AIR) was designed to
interrogate how repair, care, and maintenance might contribute to
the intimate relationship between humans and objects, or in this case
between Key and her bicycle. To do so, Rep(AIR) brings attention to
breakage and wear by establishing a new mode of communication
(through scent) between the bicycle and the rider. Feedback chan-
nels conveying damage to the frame exist inherently on a bicycle;
there might be sounds, vibrations, or visual signs that damage has
occurred. However, all too often these warnings are hidden in the
ubiquity of repetitive and habitual haptic feedback being transmitted
from the bicycle to the rider on a bumpy path, often eluding con-
sciousness and receding to the back of the mind. Rep(AIR) counter-
acts this familiarity by diffusing three scents depending on the type
of impact or wear the bicycle is experiencing. Our design rationale is
inspired by defamiliarization where the habitual is made strange and
unfamiliar (Shklovsky and Reis 1965; Bell, Blythe and Sengers 2005).
This use of strangeness to challenge usual thinking opens a critical
space for people to “interpret situations for themselves, it encourages them
to start grappling conceptually with systems and their contexts, and thus
establish deeper and more personal relations with meaning offered by those
systems” (Gaver, Beaver and Benford 2003). We aimed at making the
minor dings and dents on the frame a strange, delayed, and curious
feedback like scent in order to provoke more thoughtful reections.
Key used Rep(AIR) during two cycling touring trips: 6 days in
Mongolia, and 6 days in Slovenia, in an autobiographical design in-
quiry. During those trips, she gathered data on her own experience in
the form of a personal probe notebook. Data included repair or main-
tenance acts performed, as well as reections on her relationship
to the bicycle. Both the designing and making of Rep(AIR) and the
deployment allowed us to investigate the following research question:
What qualities of a human-object relationship might emerge by bring-
ing more attention to acts of repair and maintenance?
We make two contributions in this paper. First, we describe in detail
the design of Rep(AIR), an olfactory interface responding to impact.
This contributes to the growing corpus of works in HCI and design
in sensory interfaces by offering a fully functional prototy. Second,
in our analysis, we articulate three themes around elevating acts of
repair, maintenance, and care and repair that we observed in the
deployment of Rep(AIR).
In the remainder of this paper, we present related literature around
repair, peripheral interactions, scent as interface and bicycles in HCI.
We then offer details around the methodological approach (auto-
biographical design), design process and Rep(AIR)’s specications,
as well as the results of the 12-day deployment. We conclude with a
discussion on repair as part of a longer ongoing process and on the
multifaceted relationship that emerged through acts of maintenance.
Repair in design
Design and human-computer interaction (HCI) scholars have long
paid attention to repair and maintenance as an important, but often
overlooked, area of interest in the life of (technological) artifacts. A
small but highly evocative corpus of work offers ethnographic ac-
counts of repair communities and repair practices. Early work by
Suchman (1987) and Orr (1996) were central to bringing attention
to repair practices as work that holds its own technical knowledge
and social-material context. Scholars have investigated the prac-
tices of Fixer collectives in the USA, phone repair communities in
Uganda and Bangladesh, repair markets in Bangladesh, artists, and
hobbyists (Jackson 2014; Jackson and Kang 2014; Rosner and Ames
2014; Rosner and Turner 2015; Houston et al. 2016) to name a few. In
these works, repair was found to be intertwined in the socio-material
contexts we live in and to be embedded in local and global networks.
Repair also represents signicant forms of craft-based knowledge
and is an ongoing process rather than a planned script. Importantly,
Houston et al. (2016) highlight how repair is a central site where
values are performed and achieved and where people’s relations to
objects might be elevated, leading to an empowered link with arti-
facts (Rosner and Ames 2014).
Repair, reuse, and appropriation are at the center of discussions
around planned obsolescence and sustainability—bringing into focus
responsibility and opportunities to celebrate longer artifact lifecycles.
Along those lines, Tsaknaki and Fernaeus (2016) propose ways to use
Wabi-Sabi as a resource for design, building on the three evocative
principles of: ‘‘nothing lasts’, ‘nothing is nished’, and ‘nothing is perfect’’
(Tsaknaki and Fernaeus 2016).
Repair was also looked at through the lens of creativity. Maestri and
Wakkary (2011) report on acts of everyday repair, where lay peo-
ple are resourceful and creative in the ways they repair, reuse, and
repurpose everyday objects. Jackson and Kang (2014) argue that
breakdown and repair are a productive lens to extend HCI’s under-
standings of creativity. Ikemiya and Rosner (2014) created and de-
ployed ‘Broken Probes’ as a way to elicit “insights into how broken objects
and acts of breakage may be given new life” (Ikemiya and Rosner 2014).
Our work builds on this rich history in design and HCI by proposing a
probe that brings attention to the wear and tear of an everyday object,
the bicycle. We note that our work, similarly to Ikemiya and Rosner
(2014), focuses on the moments and acts of breakage as much as the
acts of repair.
Figure 1. Rep(AIR) on a mountain
pass in Mongolia.
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 6
Our main objective in this paper is to investigate
opportunities to bring to the foreground repair
and maintenance, in the case of bicycle care. To
do so, we turn to peripheral interaction as a strat-
egy for design. Peripheral interaction, or “what we
are attuned to without attending to explicitly” (Weiser
and Seely Brown 1997) was initially framed as a
means of seamlessly integrating technology into
our everyday lives under the term calm technology.
Important to interaction design, peripheral inter-
action can “be performed in the periphery of attention
when another activity is being performed simultane-
ously in the center of attention” (Bakker, van den
Hoven and Eggen 2015). As a result, Bakker and
Niemantsverdriet (2016) urge designers to utilize
the full range of what they call the interaction—at-
tention continuum to afford exibility to attend
to interactions on a level each individual deems
appropriate for the context. Bolton, Jalaliniya and
Pederson (2015) also position their work on grace-
ful interruptions to allow a choice over how and
when to engage with their technology by relying on
the “existing interruption management infrastructure in
our brains’”(Bolton, Jalaliniya and Pederson 2015).
Peripheral interaction shows promise in our case
because it allowed Key to act upon the olfactory
feedback as she saw t: focusing attention direct-
ly when it was safe and appropriate, allowing the
information to linger in the periphery when riding,
and drifting between the two as attention resources
ebbed and owed. Our work builds on the theoreti-
cal foundations of peripheral interaction and offers
a clear and novel case that exhibits how olfactory
feedback in the context of an everyday practice can
provide meaningful peripheral awareness.
Scent as interface
Olfactory perception stands apart from other sen-
sory systems in the way it is routed in the brain.
Rather than pass through the thalamus en route
to the cortex, olfactory stimuli is “relayed directly
to the limbic system, a brain region typically associated
with memory and emotional processes. This provides
olfaction with a unique and potent power to inuence
mood, acquisition of new information, and use of in-
formation in many different contexts” (Sullivan et al.
2015). Furthermore, the emotional implications of
acquiring knowledge through olfactory encoding in
the conscious brain have also been observed in the
unconscious brain (Sullivan et al. 2015), suggesting
that meaning might be layered, peripheral, and
multiplicative in an olfactory feedback system.
Because it is an emotionally evocative and dimen-
sional mechanism for relaying information, its
usage has grown in design and HCI as a medium
for interfaces. While olfactory output in multime-
dia has been around since the 1960’s (Heilig 1963),
Kaye (2001) developed the rst theoretical frame-
work for scent in HCI through extensive research
and explorations such as ‘Smell Reminder’ (Kaye
2001), an olfactory notication system, and ‘Honey,
I’m Home’ (Kaye 2004), a personal messaging sys-
tem. Similarly, Strong and Gaver (1996) challenged
the dominance of sight, sound and touch in HCI
with their exploration of an olfactory interface for
initiating communication between loved ones who
are apart. Since these early explorations, olfactory
interfaces have remained oriented towards inter-
personal communications and immersive media
experiences, but have become smaller and mobile
(e.g. a pair of glasses (Choi et al., 2011) or a neck-
lace (Dobbelstein, Herrdum and Rukzio, 2017).
Further works have used olfactory data output in
the eld of mental health: Amores and Maes (2017)
developed ‘Essence’, a bio-sensing necklace and mo-
bile app to inuence mood and cognitive perfor-
mance, and Tillotson’s (2002) evocative prototype
‘SmartSecondSkin’ is a scent delivering garment for
Rep(AIR) adds to this growing eld by offering a
detailed account of a functional stand-alone scent
interface. Furthermore, Rep(AIR) is uniquely con-
cerned with direct communication of the need for
the repair, maintenance, or care of another object
rather than ambient communication between
humans, as with most examples above. Rep(AIR)’s
olfactory feedback is aimed at eliciting explicit
action rather than stimulating a biological, or emo-
Interaction design and bicycles
Furthermore, we note some previous works at the
intersection of interaction design and bicycles. For
instance, Pielot et al. (2012) and Steltenpohl and
Bouwer (2013) focused their research on exploring
haptic waynding interfaces for vacationers on an
island and in a city (respectively). These interfaces
pivot away from visual interfaces which are dis-
tracting, dangerous, and require at least one hand
to zoom or re-orient. Similarly motivated, Rowland
et al. (2009) developed two immersive audio-based
interventions for cyclists where safety concerns
remained a key insight of their analysis. They
argue interfaces should encourage cyclists to stop
when direct manipulation is required, asserting
that ‘content delivered at the wrong time will at best be
ignored’ (Rowland et al. 2009). This is an interesting
point of departure from the peripherally lingering
olfactory output of Rep(AIR) where, when the scent
is not immediately attended to it has the capacity
to resonate in the subconscious where it might still
have an effect—even while being ‘ignored’.
We use an autobiographical design approach to in-
vestigate our research question. Autobiographical
design is the design and genuine use of a system
for and by the design researcher (Neustaedter
and Sengers 2012). In autobiographical design,
insights are drawn from the rst-person experi-
ence of building and using an artifact or system.
This method has proven benecial in cases where
access to participants is difcult, because it needs
to be ongoing over a long period of time, or because
the context of the inquiry is personal, intimate,
and difcult to enter. The benets of using autobi-
ographical design include the possibility to design
a fully tailored design product for the participant
and their situation, to rapidly iterate on the built
prototype since the designer, the maker, and the
user are the same person, and to gather rich and
thorough data by being in direct contact with the
situation (Neustaedter and Sengers 2012). While
there are challenges in using autobiographical
design (Desjardins and Ball 2018), investigating
the subtle and intimate maintenance relationship
between a person and their bicycle was a research
endeavor well aligned with the qualities of autobi-
Rep(AIR) was designed as an artifact to be expe-
rienced as-is as a means to investigate a research
question. The level of nish on its form, materials,
and behavior, and the robustness and reparability
of the prototype allowed Key to fully live with and
experience the artifact during her cycle touring
trips. Rep(AIR) was designed with the intention
of asking a question—what relationship qualities
emerge when we pay more attention to acts of repair
and maintenance?—yet it was also designed to sat-
isfy the genuine curiosity and imagination of Key.
Rep(AIR) traveled with Key on an international
bikepacking (self-supported mountain bike ride/
camping) trip to Mongolia and Slovenia. Key and
her partner rode 300 miles over 6 days in Mongolia,
then she rode 270 miles over 6 days in Slovenia
alone. During each trip, Key took notes in the probe
notebook for everyday of riding. During the 6 days
in Mongolia, 4 acts of repair, maintenance or care
were made to Key’s bicycle, and one on her part-
ner’s bicycle; during the 6 days in Slovenia, 5 acts
of repair, maintenance or care were made. While
different from an everyday commute, interna-
tional bicycle travel—where resources are limited,
conditions are challenging, and repairs are time
sensitive—makes for a rich space to investigate a
relationship to maintenance with a personal object. Figure 2a, b, c, d. Deploying Rep(AIR): a) Dousing
the spools of wool with essential oil; b) Connect-
ing the sensors in Mongolia c) Rep(AIR) in everyday
use d) Key reecting in the tent.
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 8
In addition to using Rep(AIR) on her trips, Key also
developed a daily probe notebook to ensure con-
tinuous and thorough data collection about the
experience of riding with Rep(AIR). The daily probe
notebook was designed as a tool to reect on the
research question by providing a physical struc-
ture and specic provocations for recording and
collecting data. It consisted of three elements: a
daily Inspection Report page (g. 3c), a daily Notes
page (g.3a), and a Deeper Reections page every
third spread (g.3b).
The probe notebook included one month’s worth
of full color pages, was covered in canary yellow
cardboard, and bound with yellow thread.
Key collected data during the ideation, fabrication,
prototype building, and beta deployment phases
through scans, photographs, and notes. In addi-
tion, Key collected 9 reported acts of repair in the
probe notebook, which included written thoughts,
diagram annotations, and photographs. We con-
ducted a thematic analysis to identify common
patterns across the data. We coded the data along
initial emergent themes such as exposing the life
of the object, time displacement, and trust. After
reviewing the data along those themes, we rened
our nal set of ndings under three general areas.
We present the ndings of our analysis below, after
rst detailing the Rep(AIR) system.
Figure 3a,b,c. Probe notebook: a) Notes page where space was provided to use the HP Sprocket Photo Printer
for photo documentation; b) Deeper Reections page; c) Inspection Report page where inspiration came from
rental car inspection procedures, including a simplied diagram of the bicycle in pink with blue instruc-
Figure 4a,b. Rep(AIR) detail: a) Three 12V squir-
rel cage fans attach to a wood base, three plastic
caps lled with a spool of essential oil soaked
wool screws into a plastic base allowing proper air
intake over the fan blades; b) Exploded view of the
main parts demonstrating how they t together.
Rep(AIR) is an impact-triggered scent diffuser
mounted to the handlebars of a mountain bike. Its
polystyrene housing is a 5-inch-wide by 8-inch-
tall cylinder with a concave lid with a star shaped
opening for scent to escape. The housing faces
outward, angled up towards the rider’s face and
is attached to the handlebars of a bicycle with a
Velcro strap. Inside the housing are three battery
powered fans which, when activated, each blow air
over spools of wool doused in a different essential
oil and controlled by an Arduino microprocessor.
To capture impact to the bicycle, piezoelectric
discs were placed at various areas along the frame
which frequently see damage (g. 6). Each sensor
was calibrated individually to balance the mate-
rial properties of its placement on the bike (more
or less dense steel tubing) and its tolerance (how
critical is this area of the bicycle). Balancing these
two aspects of the system to make the olfactory
feedback contextually relevant required manual
calibration of each sensor.
The scents chosen were meaningful to Key:
Cedarwood evokes a sense of accumulated use
as it smells like old books and is diffused when a
mild impact is sensed. Cinnamon evokes a sense of
urgency as it opens up the sinuses and is diffused
when a major impact is sensed. Eucalyptus evokes
a sense of a longer timescale as it is a reminder of
growing up on the California coast and is diffused
every 10th diffusion and runs for 10 milliseconds
longer each time, so as the bike ages the eucalyptus
scent lingers for longer.
Rep(AIR)’s nish balances an attention to aesthetic
detail, color, shape, as well as t: it was aestheti-
cally tailored and physically positioned to its rider
alone. All code was tested on the intended bicycle,
with the intended rider, and adapted for such. This
is not to say that the system might not work in
another situation but meant to emphasize the be-
spoke nature of the process of building this object
to reect its purpose.
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 10
Figure 5. Housing: Sheets of white polystyrene were vacuum molded,
hand molded, and textured creating a clean bright outward appearance.
The hand cut lid was meant to abstractly resemble a ower.
Figure 6. Sensor diagram: piezo discs (represented by blue
dots) were placed along the frame where damage frequently
occurs and were individually calibrated.
Analysis: Elevating Acts Of
Maintenance And Care
In this section we begin by discussing how
Rep(AIR) provoked focused and peripheral tactics
to elevate acts of maintenance and care. We then
present two qualities of the relationship between
Key and her bicycle which were revealed and nur-
tured through this process. These qualities point
towards a relationship with objects where care,
preservation, and agency are complimentary coun-
terparts to function.
A catalyst for attention—focused and peripheral
During the 12 day deployment of Rep(AIR), 9 acts
of repair, maintenance, or care were made. Some
were minor xes like adjusting a dérailleur cable,
others required more time and ingenuity such as
using a piece of discarded car tire to plug a leaky
valve. Some were attended to as a direct result of
olfactory feedback, but some came about through
other means. However, the presence of Rep(AIR), in
all cases, directed attention to these acts by either
focused or peripheral means and reectively ori-
ented perception within those moments.
Action through olfactory feedback
Although scent, in this case, is an ambiguous rep-
resentation of damage, it effectively communicated
the need for attention. For example, during the de-
ployment in Mongolia, Key noted: “the bike fell over
today, and I got cinnamon...it further damaged the rack
and pushed the strut into the tire...it took a lot of inves-
tigation to nd the issue.” On another occasion, while
riding over a section of trail where many large roots
traversed the path, Rep(AIR) triggered a cedarwood
diffusion. “The scent made me pause the audiobook
and check-in on the bike. I noticed a bit more rubbing
on the tire from the previous day. Another adjustment
xed it. I might not have noticed as soon if it weren’t for
the Rep(AIR) bringing the idea of maintenance back into
my mind.” In this case, what initially triggered the
cedarwood was not the tire rubbing (but the large
roots) and yet the olfactory feedback led directly,
though unintentionally, to an act of repair (g. 7).
The emotional associations afforded by the use of
scent also helped direct attention immediately; in
her notes, Key comments ‘the cinnamon overpowers...
it’s alarming at rst until you realize what’s happening.’
These examples show how Rep(AIR) functioned as
intended: by using scent as a way to focus Key’s at-
tention to care and maintenance issues while riding.
Reection through delayed investigation
In the examples above we see how attention was di-
rected towards potential damage immediately, but
because stopping to investigate what might have
caused the olfactory feedback was not always prac-
tical, thoughts of maintenance and repair tended
to linger in Key’s mind. For example, in Mongolia
when descending a mountain pass on a steep
trail badly washed out by a recent rain storm, Key
recalled smelling Rep(AIR) ‘“wice during very rough
descents on single track today, both cinnamon, but I was
in no position to check it out.” In conditions like these,
dedicating focused attention to a scent is at least
impractical and at most dangerous. However, this
does not erase the knowledge that some breakage
has occurred, it simply redirects it to the periphery
of the mind. Earlier on that trip, when reecting
on the method, Key remarked on the un-feasibility
of stopping to write in the probe notebook every
time she had thoughts of maintenance, saying
“Maintenance is on my mind all day, but I can’t stop to
order them or pay much attention.” The constraints of
the situation meant that reections were conned
to the quiet moments inside the tent at night or
early morning (g. 2d) and that daily ruminations
were forced to steep in the conscious and subcon-
scious meanderings of the mind. On the last page
of the probe notebook, when reecting on the delay
between scent and documentation, Key succinctly
concluded that “the delayed scent is cool - it’s curious,
investigative, inviting more questions.” Because of the
delays in both making and documenting repairs,
thoughts of maintenance and its effect on the rela-
tionship between bicycle and rider extended over
a longer time period, inviting different questions,
reections, and heightening the rider’s curiosity.
Figure 7. Site of repair: A cracked and bent front
rack caused rubbing on the tire, Rep(AIR) diffused
cinnamon to communicate the impact.
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 12
Awareness through its actuality
Rep(AIR)’s physical presence on the handlebars
of the bicycle brought attention to, and over time
became a symbol of, its intended function: elevat-
ing acts of maintenance. Even while not actively
diffusing scent, Rep(AIR) served as a reminder to
consider wear and tear as it was happening. Key
notes: “Although Rep(AIR) was off today (due to heavy
rain) I was more attentive to any new sounds or feelings.
The project itself has me more repair minded.” During
the rst deployment in Mongolia, Key distressed
over the difculty in drawing a clear line between
cause and effect with the project, writing “it’s hard
to separate the design/repair/diffusion effects. It’s creat-
ing more awareness but what’s doing that?” Although
the lineages of causation remained difcult to trace
throughout the rest of that trip and the next, the
subliminal usefulness of the artifact solidied as
Key came to see and use it as an icon as well as an
interface: “came to use other kinds of interactions with it
to drive thought about repair. It became much more sym-
bolic. It’s like the rubber band on your wrist that reminds
you not to smoke or whatever. It’s aiding in maintenance
indirectly.” In this way Rep(AIR) as an icon was kept
alive through each activation as well as through
acts of maintenance on the device itself.
These three tactics for awareness became import-
ant modes with which to engage in the research
question and goal of the inquiry. Some were
planned and explicit, others emerged over time as
natural responses to constraints in the environ-
ment and were ambient. By adapting the usage of
Rep(AIR) to suit her needs and situation, Key was
more exibly equipped to reect on emergent
qualities of the relationship with her bicycle, as
described in the sections below.
Shared stories—the bike experiences too
The stories of living on and traveling with a bicycle
constitute a rich accumulation of experiences for
both the rider and her bicycle. In this section we
discuss how acts of maintenance and care re-
veal those non-human parts of the story typically
unseen and how this added dimension extends
the appreciation of a thing to fundamentally draw
the human and the thing closer. We reveal two
emergent behaviors which surfaced signs of those
experiences outside of human perception through
reunions after time apart and through a ritual of
bathing and inspection.
Over the length of the trip the bicycle was packed,
shipped, and unpacked half a dozen times. At each
leg of the journey the bicycle would have to be tak-
en apart for transport then rebuilt once it landed in
the new destination. Each of these events became
an opportunity to carefully inspect the bicycle.
These investigations were focused on what dam-
age had occurred in transit (while Key was com-
fortably in her seat eating peanuts) or for damage
Key missed while riding. This was in some ways a
pragmatic practice—uncovering potentially critical
damage is self-serving, but in reality, the motive
came more from curiosity than a fear: “This is the
frame damage I noticed when unpacking the bike before.
I didn’t have time to investigate a bunch that day, but I
thought a lot about it today. I really would like to gure
this out.” In this instance, Rep(AIR) was not actively
tracking, and Key was not there to witness these
breakages, yet the investigation itself was an inti-
mate exposing of the bicycle’s experience in par-
allel to the rider’s. Although this investigation was
not prompted directly from Rep(AIR), Key’s more
mindful and inquisitive attitude towards wear and
tear on the bicycle most certainly was.
On several occasions throughout the journey the
opportunity arose to bathe the bicycle, and in some
cases thoroughly clean its various components.
This served to expose additional wear and tear, but
also to reciprocate care for the object which was
the rider’s literal, and at times emotional, support.
When remarking on a particularly symbolic bath
conducted in the “deep puddle” of a parking lot, Key
wrote “This does not improve function - it’s not even
profoundly different, but it just feels good to do” (g. 8).
The desire to give back and acknowledge the ex-
istence of an object extended beyond moments of
convenience, as when Key recalled, “I had a dream
last night that I got back to my hotel, in between trips,
and put Cal [the bicycle’s nickname] in the shower, haha.
I wanted to do something nice and make her shiny. Very
Figure 8. Bathing ritual: Key bathes her bike with
care in a puddle.
appreciative of her today.”Bathing became a ritual of
investigation but more so a celebration of shared
accomplishments by recognizing the individual
role the bike played in the journey. This was appar-
ent in the last bath of the trip when Key expressed
her gratitude by noting: “Removed all parts for deep
cleaning. Felt good to give her a proper cleaning though.
She earned it.”
Through seeing the bicycle as an entity whose ex-
periences are interwoven, overlapping, but also in-
dependent of the rider’s, the stories created about
each trip are enriched with a second, non-human
perspective. By using maintenance and care to
distinguish or celebrate the bicycle’s individu-
al existence, wear and tear become meaningful
physical traces like battle scars which symbolize
this parallel life. Bringing emphasis to those marks
can then be a powerful tool to see that objects exist
apart from us as well as with us, that they shape us
as well as are shaped by us.
A collaborative existence—flattening the
In the previous section we discussed how viewing
the bicycle as an entity with its own experiences
can affect its relationship to the rider. Here we
bring to light the entity which is the bicycle and
rider together. In these examples we show how
maintenance and care foster and strengthen a
non-hierarchical exchange which manifests itself
alongside its constituent parts.
One of the most salient evolutions seen in the
analysis of the probe notebook was how the bicycle
went from a tool to a team member. This change
was gradual; not a conscious desire, but an other-
ly force born of many factors of which repair and
care were foundational. On the rst day in Slovenia
Key describes how this bond began to emerge,
“This was my rst day alone with Cal and Rep(AIR) and
it seemed different. When I noticed a strange sound and
xed and diagnosed it right away I was so proud, and I
just felt more connected to Cal. As if I relied on her more.
I am just generally more in tune with her today. Maybe
it’s because I am relying solely on her and my ability to
x her... either way it is being expressed through main-
tenance and repair.” Key took pride in being able to
repair the bicycle but even more so in her ability to
quickly diagnose the issue. This intimacy of knowl-
edge is directly related to the combined prior acts
of maintenance, investigation, and awareness. This
particular act of repair illuminated for her how that
intimacy of knowledge was strengthening their
Later during that same trip, after a particularly
harrowing incident, this nascent understanding of
how the bicycle and rider can operate as one was
solidied. Key details the event in the probe note-
book: “Today was crazy, just nuts. It was a very weird
powerful bonding day for me…on the biggest dirt descent
I’ve ever seen a massive storm hits. Hail, rain, lightning,
etc. I was legitimately scared. For hours I was repeat-
ing the mantra ‘we will get down this mountain’. At the
bottom, back on pavement I let myself feel for a minute
- I was so thankful for Cal, so proud that we made it. So
thankful that it worked, it survived, it saved me! I really
felt like we were going through this together.” During
this descent Rep(AIR)’s olfactory communications
were dampened by its rainproof plastic cover (g.
9), and yet it periodically entered the mind of Key
as she worried that it too might be damaged in
the storm. Once safely down, drying in the tent at
night and reecting on this event, she realized how
much she relied on and trusted her bicycle to do its
part in getting them both down the mountain. This
trust was largely built through making repairs and
adjustments, such as ling down and cleaning the
brakes the day before (making them more suited
for a steep descent), having that tness tested, and
succeeding. The account of this descent illustrates
how trust in the rider-bicycle team superseded
that of the individuals as they operated as a unit to
survive the event.
Figure 9. Rep(AIR) covered: The waterproof bag
dampened but did not eliminate communication.
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 14
As Key rode with, repaired, maintained, and cared
for her bicycle, their collective skill developed.
They collaborated to strengthen their mutual abil-
ity: the bicycle affording muscle development for
the rider, and the rider adjusting and maintaining
the bicycle’s tness for the current terrain. From
that co-created capability grew an empowerment
to accomplish ever more challenging undertakings.
In reecting on the emotional aftermath of the
stormy descent mentioned above, Key recalled
“Today seemed sweeter because of the storm. As shitty as
it was it was empowering. I feel like me and this bike can
handle anything.” Not only did Key feel like she and
her bicycle were acting as a team, she felt empow-
ered by that relationship, that as a team they could
accomplish more than as an individual using a
tool—even a powerful and well adapted one. As
acts of repair and maintenance physically attuned
the bicycle to Key and created an intimate knowl-
edge of how they operated as a team, the bicycle, at
times, became an extension of the rider rather than
a separate entity. An account in the probe note-
book illustrates this well: “The rst half of the day was
just brilliant riding. Me and Cal were fused - responsive,
efcient, climbing huge hills, everything fabulous.” Every
act of repair maintenance, and care moved the bi-
cycle physically and emotionally closer to Key. The
net effect was a different mental model with which
to see her bicycle—as no longer a bicycle and a self,
but an empowered us. On the last day of riding,
Key describes a moment of empowerment where
she expressly acknowledged the emergence of this
third entity: “Last entry – weird, this trip was quite
different than the last in terms of relationship build-
ing. When I got in to town and had time to spare before
meeting with Airbnb I rode around. Came across a set of
cobbled steps and, fully loaded, thought ‘we can do that’.
There is an invincibility that’s emerged. An identity too.”
Over the course of the two deployments and all the
packing and unpacking in between, Key’s relation-
ship with the bicycle evolved into something more
cohesive, and identiable.
Through these examples we see how Rep(AIR)
catalyzed renewed and reective attention to acts
of maintenance and care, and how over time those
became ways to acknowledge and reinforce the
mutuality of the relationship’s evolution into an
amalgamated existence. As this relationship devel-
oped, maintenance shifted to be not only self-serv-
ing, or even as a means to express appreciation for
what the bicycle brings to the experience, but to
maintaining the relationship itself. As Key ex-
pressed in a probe notebook entry, “You want to x
it to get it working again - but something changes, and
you want to x it for other reasons.” Likewise, as Key’s
relationship to Rep(AIR) developed, olfactory feed-
back shifted to be not only about surfacing break-
age and wear but to continuously and peripherally
supporting that relationship development.
Our ndings show how Rep(AIR) brought to the
foreground moments of breakage, ongoing wear,
maintenance, and repair acts. Below we discuss
how Rep(AIR) and our autobiographical design
inquiry opens up new understandings regarding
practices of repair with a focus on ongoing pro-
cesses of breakage. In addition, we discuss the
more tted relationship that has emerged overtime
between the rider and the bicycle, in part support-
ed by ongoing acts of maintenance.
An ongoing process
Rep(AIR) was a catalyst to think about repair while
Key was riding. While our original intention was to
document discrete moments of repair, we found
that Rep(AIR) instead focused attention to the
ongoing processes of wear and tear, breakage and
eventually repair. Our work adds nuance to current
discussions about repair in design and HCI which
often portray repair as an act that begins once an
artifact is broken. Our ndings certainly point
to clearly dened moments of active repair, but
importantly show how they are inherently part of a
longer history and at times a peripheral awareness
of breakages and use. By better understanding
the slow and accumulative wear and tear on the
bicycle, moments of maintenance and care likely
prevented some more unexpected breakdowns.
Simultaneously, tracking the decay of some parts
of the bicycle meant that breakages didn’t come as
a surprise, but were expected and could explicitly
be mitigated, as when, after the audiobook inci-
dent mentioned above, Key decided to replace the
damaged rack for fear it would degrade the tire and
cause a at. Our aim is not to display an ‘idealized’
process of repair; our ndings also highlight the
challenging parts of this process. For instance,
nding the cause of a particular worn area took
time and investigation—suggesting that repair is
not only the physical act of replacing a part or rear-
ranging material, but also includes an inquisitive
practice. The ways in which maintenance was on
Key’s mind all day, how smells lingered and hinted
at ongoing wear, or how deeper repair investiga-
tions were delayed, start to paint a different narra-
tive around how breakage and repair can happen
when there is an ongoing path of communication
between object and human.
A more tted relationship
Part of our study was to investigate the qualities
of the relationship that would emerge if we paid
more attention to care and repair. Our ndings,
particularly the last two qualities, point to a strong
bond and a attened hierarchy between bicycle
and rider. We found that repair not only restored
or enhanced some functionality of the bicycle, it
also made it more attuned to its rider, in a piece-
meal process. Functionality, in this sense, was not
abstracted to the bicycle alone, but in fact depend-
ed on a good t with the rider. This echoes and
expands Maestri and Wakkary’s idea of ‘adaptation’
(2011) as well as Odom et al.’s notion of ‘augmen-
tation’ (2009), where both concepts articulate how
repair might change the original function of an
object. Our case, however, shows that although
the basic working function of the object was not
dramatically transformed, the very denition of
functionality was. As parts were subtly adjusted to-
wards a better t between the human-object team
its meaning grew to encompass that relationship.
While on one hand the bicycle is being adapted to
t the rider, on the other hand care and mainte-
nance also bring the rider closer to the bicycle’s
own experience (as seen in ‘Shared Stories – the bike
experiences too’). Hence, repairs and maintenance
are factors which might close the gap between hu-
man and object, starting to erase the hierarchical
relationship between the two and opening a space
for a collaborative relationship to develop in paral-
lel to individual experience. While previous works
around repair in HCI have surfaced narratives of
empowerment by engaging with repair commu-
nities, learning about technology, or using repair
tools, our work reveals empowerment through the
relation to the artifact itself and through the newly
formed human-object team.
In this paper, we presented Rep(AIR), an olfactory
interface that brings attention to wear and break-
age on a bicycle. Key’s autobiographical design
inquiry, culminating in 12 days of bicycle touring
with Rep(AIR), has offered rich and nuanced in-
sights into how paying more attention to breakage
over time and ongoing repairs can lead to a re-
newed understanding of repair, and to the descrip-
tion of a more tted relationship with an artifact
(through ritual care, teamwork, and investigative
reunions, for example).
We wish to conclude with a reection on the voice
that was given to the bicycle through Rep(AIR).
Akin to artifact-oriented ndings in research
approaches like Thing Ethnography and Interview
with Things (Giaccardi et al. 2016; Chang et al.
2017), our approach to sensing and translating the
bicycle’s data into scents supported a new expres-
sion of its wear during daily use. This example
points to the possibility of engaging in ‘conversa-
tions’ with more artifacts in our everyday lives. As
a result, we might ask: how would we change the
way we use and care for things if we knew more
about their experiences of wear and breakages?
This new channel of communication might also be
an appropriate approach to actively recognize the
ongoing changes that artifacts experience—chang-
es that contribute to their wabi-sabi. By constantly
‘listening’ to artifacts, we, as humans, might be-
come more aware that they won’t last, they are not
nished, and that they are not perfect. This might
also prompt and encourage ongoing, caring, and
inquisitive practices around the maintenance and
repair of everyday artifacts.
Thank you to Domanic Muren for support during
many phases of this project. We also wish to thank
Aubree Ball, Heidi Briggs, as well as the reviewers.
Amores, J. and Maes, P. (2017) ‘Essence: Olfactory Interfaces
for Unconscious Influence of Mood and Cognitive Performance’.
In: Proceedings of the 2017 CHI Conference on Human Factors in
Computing Systems. New York, NY: ACM Press, pp. 28–34.
Bakker, S., van den Hoven, E. and Eggen, B. (2015) ‘Peripheral
interaction: characteristics and considerations’, Personal and
Ubiquitous Computing, 19(1), pp. 239–254.
Bakker, S. and Niemantsverdriet, K., 2016. The interaction-atten-
tion continuum: considering various levels of human attention in
interaction design. International Journal of Design, 10(2), pp.1-14.
Bell, G., Blythe, M. and Sengers, P. (2005) ‘Making by making
strange: Defamiliarization and the design of domestic technolo-
gies’, ACM Trans. Comput.-Hum. Interact., 12(2), pp. 149–173.
Blevis, E. (2007) ‘Sustainable interaction design: invention &
disposal, renewal & reuse’. In: Proceedings of the SIGCHI Conference
on Human Factors in Computing Systems. New York, NY: ACM Press,
(CHI ’07), pp. 503–512.
Bolton, F., Jalaliniya, S. and Pederson, T. (2015) ‘A Wrist-Worn
Thermohaptic Device for Graceful Interruption’, IxD&A 26, pp.
Chang, W.-W. et al. (2017) ‘“Interview with Things”: A First-thing
Perspective to Understand the Scooter’s Everyday Socio-material
Network in Taiwan’. In: Proceedings of the 2017 Conference on
Designing Interactive Systems. New York, NY: ACM Press, pp.
Choi, Y. et al. (2011) ‘Sound perfume: designing a wearable sound
and fragrance media for face-to-face interpersonal interaction’.
In: Proceedings of the 8th International Conference on Advances in
Computer Entertainment Technology. New York, NY: ACM Press, p. 1.
Desjardins, A. and Ball, A. (2018) ‘Revealing Tensions in Autobi-
ographical Design in HCI’. In: Proceedings of the 2018 Designing
Interactive Systems Conference. New York, NY: ACM Press, pp.
Frictions and Shifts in RTD
#rtd2019 #researchthroughdesign #delft #rotterdam 16
Dobbelstein, D., Herrdum, S. and Rukzio, E. (2017) ‘inScent: a
wearable olfactory display as an amplication for mobile notica-
tions’. In: Proceedings of the 2017 ACM International Symposium on
Wearable Computers. New York, NY: ACM Press, pp. 130–137.
Gaver, W. W., Beaver, J. and Benford, S. (2003) ‘Ambiguity as a
Resource for Design’. In: Proceedings of the SIGCHI Conference on
Human Factors in Computing Systems (CHI’03). New York, NY: ACM
Press, pp. 233–240.
Giaccardi, E. et al. (2016) ‘Thing Ethnography: Doing Design
Research with Non-Humans’. In: Proceedings of the 2016 ACM
Conference on Designing Interactive Systems. New York, NY: ACM
Press, pp. 377–387.
Heilig, M. (1963) ‘Sensorama’. Available at: http://www.mortonheil-
Houston, L. et al. (2016) ‘Values in Repair’. In: Proceedings of the
2016 CHI Conference on Human Factors in Computing Systems. New
York, NY: ACM Press, pp. 1403–1414.
Ikemiya, M. and Rosner, D. K. (2014) ‘Broken probes: toward the
design of worn media’, Personal and Ubiquitous Computing, 18(3),
Jackson, S. J. (2014) ‘Rethinking Repair’, in Gillespie, T., Boczkow-
ski, P. J., and Foot, K. A. (eds) Media Technologies. The MIT Press,
Jackson, S. J. and Kang, L. (2014) ‘Breakdown, obsolescence and
reuse: HCI and the art of repair’. In: Proceedings of the SIGCHI
Conference on Human Factors in Computing Systems. New York, NY:
ACM Press, pp. 449–458.
Kaye, J. ‘Josh’ (2004) ‘Making Scents: aromatic output for HCI’,
interactions, 11(1), pp. 48–61.
Kaye, J. N. (2001) Symbolic Olfactory Display. Master of Science at
the Massachusetts Institute of Technology.
Maestri, L. and Wakkary, R. (2011) ‘Understanding repair as a
creative process of everyday design’. In: Proceedings of the 8th ACM
conference on Creativity and cognition. New York, NY: ACM Press pp.
Neustaedter, C. and Sengers, P. (2012) ‘Autobiographical Design in
HCI Research: Designing and Learning Through Use-it-yourself’. In:
Proceedings of the Designing Interactive Systems Conference. New
York, NY: ACM pp. 514–523.
Odom, W. et al. (2009) ‘Understanding Why We Preserve Some
Things and Discard Others in the Context of Interaction Design’.
In: Proceedings of the SIGCHI Conference on Human Factors in
Computing Systems. New York, NY: ACM pp. 1053–1062.
Orr, J. E. (1996) Talking about Machines: An Ethnography of a
Modern Job. Ithaca, NY: Cornell University Press (Collection on
Technology and Work).
Pielot, M. et al. (2012) ‘Tacticycle: supporting exploratory bicycle
trips’. In: Proceedings of the 14th international conference on
Human-computer interaction with mobile devices and services. New
York, NY: ACM Press, p. 369.
Rosner, D.K. and Ames, M., 2014, February. Designing for repair?:
infrastructures and materialities of breakdown. In: Proceedings of
the 17th ACM conference on Computer supported cooperative work &
social computing New York, NY: ACM Press, pp. 319-331.
Rosner, D. K. and Turner, F. (2015) ‘Theaters of Alternative
Industry: Hobbyist Repair Collectives and the Legacy of the 1960s
American Counterculture’, in Plattner, H., Meinel, C., and Leifer, L.
(eds) Design Thinking Research. Springer International Publishing
(Understanding Innovation), pp. 59–69.
Rowland, D. et al. (2009) ‘Ubikequitous computing: designing
interactive experiences for cyclists’. In: Proceedings of the 11th
International Conference on Human-Computer Interaction with
Mobile Devices and Services. New York, NY: ACM Press, p. 21.
Shklovsky, V., Art as Technique Russian Formalist Criticism: Four
Essays ed. Lee T. Lemon and Marion J. Reis (Lincoln: University of
Nebraska Press, 1965).
Steltenpohl, H. and Bouwer, A. (2013) ‘Vibrobelt: tactile navigation
support for cyclists’. In: Proceedings of the 2013 international
conference on Intelligent user interfaces. New York, NY: ACM Press,
Strong, R. and Gaver, B., 1996, November. Feather, scent and
shaker: supporting simple intimacy. In: Proceedings of CSCW (Vol.
96, No. 96, pp. 29-30).
Suchman, L. A. (1987) Plans and situated actions: the problem of
human-machine communication. Cambridge; New York: Cambridge
Sullivan, R. M. et al. (2015) ‘Olfactory memory networks: from
emotional learning to social behaviors’, Frontiers in Behavioral
Neuroscience, 9, p.36.
Tillotson, J. (2002) Smart Second Skin. Available at: https://www.
Tsaknaki, V. and Fernaeus, Y. (2016) ‘Expanding on Wabi-Sabi as a
Design Resource in HCI’. In: Proceedings of the 2016 CHI Conference
on Human Factors in Computing Systems. New York, NY: ACM Press,
Weiser, M. and Brown, J. S. (1997) ‘The coming age of calm
technology’. In: Beyond Calculation. Springer, pp. 75–85.