Content uploaded by Vaishnav Kameswaran
Author content
All content in this area was uploaded by Vaishnav Kameswaran on Dec 10, 2018
Content may be subject to copyright.
85
‘We can go anywhere’: Understanding Independence
through a Case Study of Ride-hailing Use by People with
Visual Impairments in metropolitan India
VAISHNAV KAMESWARAN, University of Michigan, USA
JATIN GUPTA, University of Michigan, USA
JOYOJEET PAL, University of Michigan, USA
SILE O’MODHRAIN, University of Michigan, USA
TIFFANY C. VEINOT, University of Michigan, USA
ROBIN N. BREWER, University of Michigan, USA
AAKANKSHA PARAMESHWAR, University of Michigan, USA
VIDHYA Y, Microsoft Research India, India
JACKI O’NEILL, Microsoft Research India, India
Ride-hailing services have received attention as part of the growing work around the sharing economy,
but the focus of these studies has largely been on drivers. In this paper, we examine how ride-hailing is
transforming the transportation practices of one group of passengers - people with visual impairments in
metropolitan India. Through a qualitative study consisting of interviews and observations, we examined the
use and impact of these services on our target population, who otherwise contend with chaotic, unreliable, and
largely inaccessible modes of transportation. We found that ride-hailing services positively aects participants’
notions of independence, and we tease out how independence for our participants is not just about ‘doing
things alone, without help’ but is also situated, social and relative. Furthermore, we show how accessibility, in
the case of ride-hailing in India, is a socio-technical and collaborative achievement, involving interactions
between the passenger, the driver, and the technology.
CCS Concepts: •Human-centered computing →Empirical studies in accessibility;
Keywords: Accessibility, social accessibility, collaborative accessibility, independence, stigma, social interac-
tions, ridesharing, Uber, Ola, blind users
ACM Reference Format:
Vaishnav Kameswaran, Jatin Gupta, Joyojeet Pal, Sile O’Modhrain, Tiany C. Veinot, Robin N. Brewer,
Aakanksha Parameshwar, Vidhya Y, and Jacki O’Neill. 2018. ‘We can go anywhere’: Understanding Indepen-
dence through a Case Study of Ride-hailing Use by People with Visual Impairments in metropolitan India. In
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, CSCW, Article 85 (January 2018). ACM, New
York, NY. 24 pages. https://doi.org/10.1145/3274354
Authors’ addresses: Vaishnav Kameswaran, Jatin Gupta, Joyojeet Pal, Sile O’Modhrain, Tiany C. Veinot, Robin N. Brewer,
Aakanksha Parameshwar, School of Information, University of Michigan, 105 S. State Street, Ann Arbor, MI 48105, US;
Vidhya Y, Jacki O’Neill, Microsoft Research India, 9, Vigyan, Lavelle Road, Ashok Nagar, Bengaluru, Karnataka 560001,
India;
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 prot or commercial advantage and that copies bear this notice and the
full citation on the rst page. Copyrights for components of this work owned by others than the author(s) must be honored.
Abstracting with credit is permitted. To copy otherwise, or republish, to post on servers or to redistribute to lists, requires
prior specic permission and/or a fee. Request permissions from permissions@acm.org.
©2018 Copyright held by the owner/author(s). Publication rights licensed to ACM.
2573-0142/2018/1-ART85 $15.00
https://doi.org/10.1145/3274354
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:2 V. Kameswaran et al.
1 INTRODUCTION
"We have to take sighted assistance; most of the times it was normal travelers who helped
us take a taxi and stu like that. But when this service is available, it is really very, very
much freedom [sic] for us because we can do what we want, we can avail anything, we
can go anywhere, we don’t need to take sighted assistance [...] It unveils a new door, a
new world for us - the world of independence, I must say." - Participant 10 talking about
impact of ride-hailing on his sense of independence.
According to the World Health Organization, there are 285 million people with visual impairments
in the world including 39 million blind people, over 90% of them living in the Global South [
51
].
India, with 63 million people with visual impairments, including 8 million blind people, has one
of the largest populations of people with visual impairments in the world [
37
]. Researchers have
suggested that people with visual impairments in the Global South struggle with social and economic
participation and have cited many reasons for this, including the lack of accessible educational
material [48], the lack of access to assistive technology, signicant barriers in the workplace [34],
and limited access to accessible transportation [
42
], which fosters independence by enabling access
to health, educational, and employment resources [
2
,
27
,
36
]. Further, for people with disabilities,
transportation has been recognized as an important conduit to increased community participation
and, subsequently, inclusion [19,28,49].
Although metropolitan India boasts a rich and varied transportation system, including buses,
trains, metros, auto rickshaws, and taxis, in general these systems are not accessible to people with
visual impairments. Not only are they crowded but there is little in the way of public announcement
systems and accessible resources such as maps and schedules. However, recently ride-hailing
services like Uber, and its Indian counterpart, Ola, are disrupting the transportation landscape by
providing private access to car eets. Uber has been active in India since 2013 and now operates in
33 cities
1
. Ola cabs, Uber’s main competitor in India, has been operational since 2010 and operates
in more than 100 cities
2
. Dillahunt et al. examined how ride-hailing services could positively impact
low-income individuals in the United States, who beneted from the reliability of the services and
social interactions with drivers [
13
]. Given the importance of transportation for economic and social
participation, we investigated how ride-hailing services were being used by people with visual
impairments within the wider context of the Indian metropolitan public transportation system.
To this end we conducted a two-part qualitative study, consisting of interviews and observations,
initially looking at peoples experiences with public transportation overall, then focusing in on ride-
hailing services (Ola and Uber), as we uncovered their impact on people’s sense of independence.
This paper makes several contributions to CSCW. To our knowledge this is the rst study of how
ride-hailing services are used by, and by-and-large promote the independence of people with visual
impairments. Independence is a core theme underlying accessibility research. As Wobbrock et al.
suggested, "all accessible computing approaches share a common goal of improving independence,
access, and quality of life for people with disabilities" [
52
]. As such, the idea of ‘independence’ is
woven into much of the ‘HCI and accessibility’ research, often as a metric to evaluate the value
or eectiveness of an assistive technology or technology intervention designed for people with
disabilities. However, in this research independence has been primarily understood as self-reliance,
i.e. ‘doing things alone, without help’. An artifact is considered to be eective when it allows its
intended users to perform specic tasks without, or with greatly reduced-assistance. Independence
emerged as a strong theme in our study, being articulated by our participants as a strong driver of
their use of ride-hailing services. However, we reveal a more nuanced conception of independence
1https://www.uber.com/en-IN/cities/
2https://www.olacabs.com/about.html
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:3
when it comes to ride-hailing, which extends beyond self-reliance and includes, among other
facets, its social nature. Secondly, we reveal how the accessibility of ride-hailing goes beyond
the simple functional accessibility of the service’s app and requires the driver and passenger to
collaborate. Indeed, we found the accessibility of ride-hailing services in India to be a collaborative
and socio-technical achievement, involving the driver, passenger, and technology. The degree of
success in collaborations between passengers and drivers is the cornerstone to the accessibility
of these services, and the source of any feelings of increased independence. This paper therefore
contributes to the emerging ‘turn to the social’ [20,43] in accessibility research.
2 RELATED WORK
2.1 Ride-hailing services
Most studies on ride-hailing services in HCI and CSCW focus on drivers, extending traditional
workplace studies to understand how these new technology-mediated workplaces impact drivers
[
3
,
17
,
25
,
40
]. In comparison, rider experiences are less well understood. In a study in Detroit,
Dillahunt et al. found that Uber oers certain benets to low-income populations, including
opportunities for rich social interactions with drivers, but that there are still many barriers to entry
resulting from larger social challenges, such as low digital literacy, and from technical limitations
of the platform, such as the lack of inclusive payment infrastructures [
13
]. Gloss et al. found
that Uber users were "dismissive of the traditional cab experience" with their preference for Uber
resulting from the increased in-ride security and reduced uncertainty during the hailing process
[
17
]. Kameswaran et al. examined how Uber can be a source of social and cultural capital for low-
income populations [
22
]. Meurer et al. studied the transportation habits of older adults including
their use of informal ride-shares (taking rides from friends and family) and examined how they
navigate the complexities of the driver-rider relationship to reduce their sense of dependence [
29
].
We extended these studies examining rider perspectives to focus on the use of ride-hailing by
people with disabilities, specically people with visual impairments in India.
2.2 Notions of independence
Notions of independence have been widely discussed in the HCI, CSCW, and accessibility literature,
primarily as an outcome or evaluative metric for technology interventions designed for people
with disabilities. The most common interpretation of "independence" in these studies is that it
relates to self-reliance, or "the ability to do things without ones help or assistance" [
41
]. Here,
technology as a facilitator of independence functions in a substitutive capacity, lling in for people
from whom help is sought, thus fostering increased self-reliance. Take the case of navigation, which
has long been a challenge for people with visual impairments, who often rely on human assistance
to get around. Consequently, there has been an interest in designing assistive technology to foster
independent navigation, i.e. enable people with visual impairments to get around with reduced
or no help. Examples include technologies that enable people with visual impairments to acquire
information about their immediate environment on their own, whether this is indoor landmarks
like elevators [6] or bus stop locations and arrival times [4].
Researchers have also attempted to quantify independence, so that the impact of some assistive
technology can be measured. For example, technology users might rate their degree of ‘indepen-
dence’ on a Likert scale - a representation of the extent to which the artifact allows them to perform
a task on their own [
35
,
38
]. However, this research has examined independence almost entirely as
self-reliance and has been contested because it gives undue weight to physical capacities while
ignoring aspects like autonomy - the ability to independently make decisions.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:4 V. Kameswaran et al.
2.2.1 Independence beyond ‘self-reliance’. In the few studies that looked beyond self-reliance,
independence was expressed in terms of autonomy, control, and reciprocity. Autonomy is the
freedom to exercise one’s choice and in a study of older adults, Hillcoat-Nalletamby found that
that the "self-determination in choosing what to do, how and when to do it" is closely associated
with independence [
21
]. In a study of the transportation habits of older adults, Meurer et al. found
that the use of informal ride-shares is associated with loss of decisional autonomy, which is the
freedom to plan for trips. This is because passengers become contingent on the availability of others,
resulting in a loss of independence [
29
]. Control, dened as "the capacity of people to perform
behaviors to inuence outcomes in their environment", is also associated with independence [
21
,
50
].
Yang et al. described how users of their technology - an indoor navigation tool for people with
visual impairments felt independent, the result of a sense of control they derived from knowing
what to do when they were lost [
53
]. Finally, reciprocity is also linked to independence because it
reduces the cost and feelings of dependency [
5
]. Brady et al. examined the question-asking behavior
of people with visual impairments. The authors found that a preference for getting answers from
crowdworkers ahead of friends and acquaintances on their social networks was related to a desire
for independence. In this case independence was expressed in relationship to reciprocity, with
people with visual impairments seeking to avoid unpayable social debts where possible, e.g., asking
questions without being able to answer any [
7
]. Evidently, there is no singular, all-encompassing
denition of independence, but given the importance of this concept to accessibility research it is
worth exploring further.
2.3 Accessible technology research
Assistive technology has traditionally been developed to play a functional role for people with
disabilities. This approach has its roots in the medical model of disability [
10
], a model that
emphasizes one’s physical impairments. In the medical model, assistive technology and accessibility
are a means to oset physical incapacities and HCI research often adopts a needs-based approach
to design and evaluate tools to maintain and expand these capacities. Recently, there has also been
a focus on understanding the accessibility limitations of mainstream applications and services
that impede their use, with design bridging the gap between inaccessibility and accessibility. This
includes work examining use and limitations of social media platforms [
26
,
31
,
55
] and public
transit services by people with visual impairments [11,39].
2.3.1 Social Accessibility. Accessibility research might be said to be having its own "turn to the
social" [
20
,
43
] in that there is a growing body of research examining either the situated use of
assistive technology [
12
,
45
,
56
] or the social concerns of its users [
7
,
32
]. Shinohara and Wobbrock
found that the form and function of assistive technology inuences social interactions by impacting
user self-condence and self-ecacy [
46
] and implored designers to look beyond the functional
value of the technology and design for "social acceptability" [
45
]. Through a qualitative study of a
wearable technology, Zolyomi et al. demonstrated the social dimensions of technology-assisted
sight [
56
]. Morrison et al. [
32
] conducted workshops with people with visual impairments in India
and the United Kingdom around "imagining articial intelligence". The authors found that most of
their participants’ concerns were social, from identifying people and their non-verbal behavior, to
managing social interactions. Even the more mechanistic applications, such as identifying money,
were envisioned in a social context, for example to avoid the embarrassment of giving the wrong
oering at a temple. Research examining the collaborative nature of accessibility in dierent
contexts [
8
,
9
,
54
] can also be considered to be part of this ‘turn to the social’. For instance, Branham
et al. found that environments like homes and workplaces pose accessibility challenges for people
with visual impairments who then adopt dierent strategies to circumvent them, including seeking
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:5
sighted assistance [
8
,
9
]. The authors concluded that the nature of personal relationships between
people with visual impairments and sighted partners/co-workers inuences how, when, and how
often people seek help.
This work makes it clear that it is important to look beyond simple functionality of accessibility
technology to how its use is socially situated and how that use plays out within the everyday
actions and interactions of its users. In this paper, we build on this growing body of research,
highlighting the social as opposed to wholly practical or functional concerns.
3 METHODS
We conducted this study over two phases: May-June 2016 and June-August 2017. In phase one,
we conducted in-person semi-structured interviews with six participants in Bangalore, India, to
understand their use of, and experience with, dierent transportation services including public
transportation, auto-rickshaws, and ride-hailing services. These participants (P1 - P6) were recruited
through personal contacts (n=2) and snowball sampling (n=4). Interviews lasted 60-75 minutes
and were audio-recorded and transcribed verbatim. We also observed three of the interviewees
(P1, P2, P4) on four journeys, shadowing one participant on his commute to work by bus (P4)
and two others using Uber (P1, P2). For the rst ride-hailing trip, we accompanied a participant
on a routine journey from home to work (P1); on the other ride-hailing trip, specically for the
research project, we accompanied a participant to a destination chosen by the participant - a
coee shop - and on the subsequent return journey (P2). We used jottings [
14
] to capture accounts
of participants’ trips. For the bus journey, our observation began at the participant’s home and
concluded upon the participant reaching their oce. The ride-hailing journeys started at the time
when participants used the mobile app to book a cab and concluded when they reached their
destination. The research team compiled nal notes by elaborating and commenting upon the
jottings. We conducted additional unstructured interviews after the observations to understand
aspects of the journey that were unclear. These interviews lasted between 15 and 20 minutes. The
research team paid trip costs.
In phase two, we conducted semi-structured interviews with 30 participants (P2, P4, P5, P6 and
P7 - P32) from eight metropolitan cities in India (Bengaluru, New Delhi, Kolkata, Chennai, Mumbai,
Pune, Lucknow, and Guwahati), specically focusing on ride-hailing experiences. We recruited par-
ticipants via Access India (n=15), an online listserv discussing accessibility issues and concerns for
people with disabilities, personal contacts (n=6) and subsequent snowball sampling (n=9), by which
point we had reached data saturation. Four participants (P2, P4, P5, P6) from phase one participated
in phase two of the study. Interviews, lasting 60-75 minutes, were a combination of in-person and
Skype/phone calls and were conducted in English, which all participants were comfortable and
uent with. Interviews were structured to elicit narrative accounts [
24
] of participants’ experiences
with ride-hailing services and challenges with using the services; we used conceptual questions
[
24
] to understand participants’ notions of independence in relation to the use of Uber and Ola. We
complemented the interviews with four observations of participants (P5, P13, P17, P22) using Uber
and Ola in Bengaluru. The participants were dierent from those observed during phase one.
In total over the two phases, we interviewed 32 individuals, seven of whom we also observed. All
participants identied themselves as totally blind. We obtained verbal consent from all participants,
with separate consent for photographs and video-recording for the observations. Participants
were compensated with a Rs. 250 (approximately $4 U.S.) gift voucher for their time. Interviews
gave us in-depth insights into the many aspects of ride-hailing experienced by our participants,
including how they used services to get around while navigating the many challenges that surfaced.
The data from the observations complemented the interviews, helping us better understand the
moment-by-moment experience of ride-hailing.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:6 V. Kameswaran et al.
Figure 1. Local bus service in Chennai
(Source: wikicommons - Bernard Gagnon)
Figure 2. Metro train-station in New Delhi
(Source: wikicommons - Chandradhar yadav)
We analyzed the interview transcripts and eld notes using a combination of inductive and
deductive techniques. We used open coding [
44
] to understand the journey experiences of our
participants; we developed provisional codes [
44
] using literature on independence to understand
how ride-hailing services aect perceptions of dependence and independence. Specically, we
coded for aspects of autonomy, control, self-reliance, and reciprocity, all of which are associated
with notions of independence.
4 DEMOGRAPHICS AND CONTEXT
Participants were between 24 and 53 years old. In phase one, three participants were male and
three female and all resided in Bengaluru, India. Of the 30 participants in phase two, 24 were male
and 6 female. We tried to even the gender balance through snowball sampling and by visiting two
hostels for blind women in Bengaluru, but very few reported using ride-hailing to get around,
which limited their participation. Participants in phase two lived in eight cities: Bengaluru (11), New
Delhi (10), Kolkata (2), Chennai (2), Mumbai (2), and one each from Pune, Lucknow, and Guwahati.
Most participants had used both Uber and Ola, although six had only used one service (four with
Uber and two with Ola). Most participants (n=25) used Android phones (with Talkback, the inbuilt
screen reader) and seven used an iPhone (with VoiceOver). The higher number of Android users is
in stark contrast to past research in the Global North suggesting people with visual impairments
prefer the iPhone [30]. Android phones are relatively more aordable [35], being available over a
wide cost spectrum. This dierence is important to note because it highlights a core theme running
through this research: many of the people with visual impairments in our study were concerned
about money. In the Global South, disability is often linked to lower income [
1
] - there are barriers
to nding and keeping a job because of social factors such as prejudice and also infrastructural
factors such as lack of accessible transportation and workplaces.
India has a rich transportation landscape, with major cities having multiple modes of transporta-
tion, including urban mass transit (buses, rail) and auto-rickshaws, three wheelers prevalent in many
parts of the country. People with disabilities often receive concessions while using these modes of
transportation although this varies by state. Urban mass transit services are often overcrowded and
unreliable (Figures 1 and 2)[
47
]. In terms of accessibility barriers, few bus stops and train stations
have public announcement systems, and of course it is rst necessary to arrive at the bus stop or
train station. All this makes their use challenging for people with disabilities. On demand services
are also available, such as the three-wheeled auto-rickshaws, which are often distinctly colored and
must to be hailed from the side of the street. Rickshaws are equipped with digital meters. However,
in some cities (e.g., Chennai) many drivers prefer to operate on a price negotiated at the start of
the journey. Once hailed, often the passenger and driver often negotiate the destination and price
and only on conclusion of a successful negotiation will the journey proceed. It is common to have
to stop multiple rickshaws before commencing a journey. Local cabs, operating in the same fashion
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:7
as auto-rickshaws, are also available in certain parts of the country. Call-in services are available in
some cities. On-demand services are more expensive than xed-route transportation. People also
hire drivers to drive cars they own or enter into contractual agreements with a rickshaw driver. In
both cases, drivers are paid at the end of a xed duration of time (for example, a week or a month).
5 UBER AND OLA APPLICATIONS
The Uber and Ola mobile applications are functionally very similar. Both applications allow a user
to choose pickup and destination locations, cab type, payment method and rate a driver. While the
default pickup location in both cases is set by GPS, they can be edited by the user. The destination
address is an auto-complete drop-down list, accessible through the keyboard. At the time of the
study, Uber let a user pick between three cab services (Pool, Go, X) while Ola had seven services
(Mini, Micro, Prime, Share, SUV, Auto, Outstation). Options to change payment mode and choose
the number of seats in a shared cab were also available during booking. Once booked, the passenger
is notied of driver and vehicle details, and the vehicle’s estimated time of arrival. In addition, Ola
sends a four-digit code to the rider, who communicates this code as part of a verbal handshake
with the driver, who must enter it to initiate the journey. Uber and Ola are integrated with Google
Maps, a visual resource that allows a sighted user to track the location of the cab from the time of
conrmation to when the destination is reached. However, this visual resource is inaccessible to
both screen readers (TalkBack and VoiceOver). Uber and Ola allowed cash-based or digital wallet
payments. Uber was integrated with PayTM, a third-party digital wallet service, whereas Ola uses
its own custom wallet, Ola Money. With Uber, passengers can also link their debit and credit cards
to their accounts. Card payments in India are two-factor authenticated, requiring one to enter the
CVV number and a one-time password (OTP) sent to one’s mobile phone to conrm the payment.
Both services request passengers to rate a driver after the ride, and this is compulsory in Ola.
6 FINDINGS
6.1 Independence
Interviewees from phase two (n=30) were largely positive about the impact that ride-hailing services
had had on their lives. Fifteen interviewees described major changes in their ability to get out and
about since they started using ride-hailing. For these participants, a core advantage of ride-hailing
services is that they no longer need to rely on sighted people to get around, and this has a big
impact on their sense of independence.
"One of the pleasures of commutation [sic] is that you don’t have to depend on anybody for
anything, see that is what - I don’t know how familiar you are with people with disabilities
- one of the things that checks the right box when it comes to becoming a part of the
mainstream is that you would not like to disturb anybody for anything." P5
Like the classic denition [
41
], independence here is bound up with self-reliance - doing things
for yourself, on your own, without help. In comparison, nine participants were already regularly
traveling alone by auto-rickshaw. While these participants were largely happy with ride-hailing, it
did not change their lives in such a drastic way. Additionally, two participants had private drivers
driving cars they owned, and four participants were non-committal.
To understand the impact of ride-hailing services, it is important to understand what transporta-
tion was like before for people with visual impairments in India. Indeed, participants themselves
compared ride-hailing with other transportation options, observing that buses, trains, and auto-
rickshaws all required them to seek help at multiple junctures, whereas Uber and Ola required
help-seeking less often. During an observation of P4 using the local bus service for his regular
commute in Bengaluru, we saw people step in on multiple occasions to oer help. First, on the
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:8 V. Kameswaran et al.
walk from his house to the bus stop, P4 received two oers of help, which he refused. Afterwards,
he reported that although he did not seek help, it was frequently oered and usually he would take
it so as to not oend the person oering. Once at the bus stop, the only way to nd out whether
his bus had arrived was to ask someone. He asked some people nearby and they agreed to let him
know, but they were chatting, and their view of the road was blocked, so they did not alert him
when his bus arrived. Luckily an acquaintance (who regularly caught the same bus), spotted him
and together they made a dash through a maze of people and multiple buses to catch the bus just
before it left. Once on the bus, P4 asked people next to him about his current location to determine
whether he was close to his destination. Upon reaching his bus stop, he was helped o the bus
and across the road to his oce by a passerby. It was clear from our observations that it would
be almost impossible for people with visual impairments to get on a bus without asking for help
because of the chaotic nature of major bus stops, with buses coming from all angles, stopping in
any place, without any public announcements. However, asking for help does not necessarily bring
the desired results, leaving the passenger at the mercy of strangers. Fortunately, when traveling
regularly on the same route, passers-by come to recognize the passenger and often oer help,
whether to get to the bus stop or onto the bus. This is one of the reasons why regular journeys,
such as commutes, are easier and the interviewees reported being more comfortable making such
journeys. Making public transportation accessible without relying on the help of other people
would require considerable change in infrastructure and practice.
Similarly, hailing an auto-rickshaw (auto) is often an arduous multi-step task, requiring venturing
to main roads, physically signaling for the auto (e.g., waving) and bargaining with rickshaw drivers.
This process is often repeated multiple times before agreement is reached and a ride commences.
In comparison, participants could use their mobile to conveniently book an Ola or Uber cab and if
necessary use the text resources, i.e. the driver’s name and contact number on the app, to call the
driver and conrm the pickup. P8 contrasted her experience with the auto with that of Uber/Ola.
"Walking out, hiring an auto and going away on my own is not something that I am
comfortable doing, so that sort of limits me to being dependent on someone else to do it. I
would say [I prefer] Uber a little more for these reasons. For being independent and being
able to take a cab and go on my own to wherever I want to." P8
There are practical reasons for desiring independence such as being able to go out when and
where one wants, without having to wait for other people. However, there are also strong social
elements. Asking for help has social consequences and can cause discomfort, and the interviewees
were very sensitive to the irritation they might cause to strangers.
"If you frequently demand for help from somebody, that person might get somewhat
annoyed. You know, what is this guy always asking for some help. You know, always I
have to serve him [...] why he has to bother me again and again?" P6
The interviewees reported that not having to rely on other people helps them feel more equal and
empowered. However, it is interesting to note the role of the driver, who is expected to provide
help while tting comfortably within participants’ ideas of "without sighted help". For example, P12
discussed the requirement to ask for help when using other transportation and how that compares
to ride-hailing.
"Even if I [am] walking to my home also (sic) - I have to ask somebody to cross me, cross
the road - after crossing the road I can reach my home, I can use my senses and reach the
home - if there is a new location I have to ask them to guide [me] to my location. Those
things are not required at present - I have overcome those issues due to, because they will
drop me exactly wherever I want and most of the drivers are ready to even guide me, they
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:9
won’t simply run away after dropping me. They used to walk with me up to the door. [...]
Due to these cars my independence has increased a lot." P12
Many participants echoed similar comments, and while it might seem strange that for the cab
ride to take place the driver and passenger must successfully locate one another, that is part of (or
a small extension of) the driver’s job. Thus, this is seen as less burdensome than seeking help from
others without this obligation. Clearly, all help is not equal, and P11 even suggested that the role of
driver might be even further expanded,
"So if somebody has to go for shopping or has to go for a picnic or wants to visit someplace
or wants to go and watch some movie [...] so the driver would not only drive but guide as
well - like he would park the car [...] take the person along. He may charge extra but in
that way it would be benecial for the visually impaired people. They would be totally
independent." P11
These quotes begin to complicate the idea of independence as simple self-reliance; in practice, it
is a more nuanced concept. Another aspect of independence is its
social nature
. It is not just that
ride-hailing gives people with visual impairments the condence to travel - it also gives others the
condence that people with visual impairments can travel safely, and this plays a key element in
their ability to engage in independent travel.
"I think I am much more independent in terms of traveling. I think people around me
are very comfortable me going by Uber or Ola. It could be my parents or my friends, the
fact that they themselves are using it, and I guess they are condent that I will be ne or
something like that or I would reach [my destination] safe." P2
Independence goes beyond what one can and cannot do, it is also about how one appears to
‘society’, one’s perceived competence or face [
18
]. Technology has a role in leveling the playing eld
and enabling people with visual impairments to exhibit their competence, in part by demonstrating
their independence. Thus, it is not enough that one can do things for oneself, but also that others
must recognize this. Being a competent member of society means exhibiting that competence in
and through our daily actions and interactions [
15
]. The interviewees appreciated that technology
could help them demonstrate this in the face of ignorance and stigma, as P5 clearly articulated:
"You can say that you don’t worry about what other people think of you, but when it
comes to a person with disability the impression that you create on other people, especially
about your disability, largely depends on how sophisticated you are. If I am able to operate
my phone on my own, that not only communicates the fact that he is not a dumb fellow
but also the fact that he is able to, he is not as bad as I thought, or he is as good as I am or
even better. [...] So that impression came only after I started using a smartphone [...] that
gave a quantum jump in the condence level." P5
Several features of ride-hailing services together aord this extra independence, including
booking the cab, the door-to-door service, the cab journey, the exibility of the services, and the
formalization of certain elements of transportation. We address each of these in turn.
6.1.1 Booking the cab. People with visual impairments can book an Ola or Uber using their
smartphone app with a screen reader. Booking is a multi-stage process, including choosing the
destination, picking the type of cab (e.g., UberPool vs UberGo), choosing the number of seats in
the cab ride if a shared ride and conrming the booking. Notications on the mobile application
make it easy for them to determine when their vehicle has arrived. For the most part, both apps are
accessible enough that people with visual impairments can book the cab without help. Certainly,
Ola and Uber oer improved accessibility compared to the alternative of taking a bus, auto, train
or metro, which often require help to get to the right location and help to board the vehicle.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:10 V. Kameswaran et al.
Furthermore, this ability for people with visual impairments to book the cab themselves has an
added advantage: not only does it enable them to travel themselves, but they can also use these
services to book cabs for others.
"So it’s a real life changing game - wherever I had to go, I have to go previously I ask my
mom [...] just book a taxi for me, take me to the taxi etc. But now if my mom requires I
book a taxi for her [...] blind persons are, you know, still considered to be downtrodden so
they are not asked to do any service [...] for the betterment of the family [...] but I feel a
little bit condent that I am somewhat, a little bit of helping my family in that matter
where it was not possible previously." P10
Being able to book the cab for others thus goes beyond doing things for oneself and enables
people with visual impairments to contribute to their families. This is important for increasing
self-esteem and status in the family and helps in ghting the stigma of being visually impaired.
6.1.2 Door to door. That the cabs go
door-to-door
is also a major advantage because many of
the problems people with visual impairments have in getting around stem from the diculty of
getting to a location where they can nd transportation, which means circumventing potholes,
a lack of curbs and pavements, stray animals, and crowds, as well as the diculties of actually
hailing an auto-rickshaw or locating the bus.
"Well you know [to] take any other transportation whether it is auto-rickshaw, cycle-
rickshaw, certainly we have to depend on somebody. [...] So, certainly of course indepen-
dence is quite interesting - because we don’t have to leave our house, go to the main road,
talk to the people." P15
Independence is clearly
relative
; ride-hailing gives people with visual impairments more in-
dependence compared to using the bus and auto. This is important to remember later when we
examine the problematics of ride-hailing. Ride-hailing does not just reduce the physical diculties
of getting around, but also the emotional stress. Just as with asking for help, there is an element of
emotional work that comes into play, particularly with auto-rickshaws, where one may need to
bargain with several drivers before coming to a mutual agreement to travel. While this is the same
for sighted passengers, the trouble is magnied for people with visual impairments, who cannot so
easily ag down the next auto and who may face stigma because of their visual impairment.
6.1.3 The Cab Journey. Once in the cab, participants relied on the voice navigation system on
the driver’s phone to determine whether they were on the right path. If the driver’s voice navigation
system was turned o or inaudible, they either requested that the driver turn on/increase the volume
or they used Google Maps on their personal phones for navigation. In fact, as we discuss later,
ride-hailing apps are just one part of an ecosystem of technology that improves the independence of
people with visual impairments, and the use of Google Maps, and indeed smartphones themselves,
was often reported alongside Ola/Uber as being a critical component of this process. As P5 elucidated,
technology (specically computers and screen readers) transformed the workplace experience, but
there was initially a gap between the experience at work and the experience outside.
"So, there was this workplace condence and outside you are not. So that problem is solved
with smartphones - especially with Google Maps and all that. I used to like play around
with it, like take an auto and turn on Google Maps and watch where they are going. And
those guys were like - they won’t have a clue what hit them because I will tell them, you
can take a right, why are you going straight. They will be like, how do you know that? [...]
So, the thing is that condence comes from the control you have about your situation." P5
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:11
We can see here that the judicious use of technology in general can help to level the playing
eld for people with visual impairments. Google Maps enables them, whether in cabs or autos, to
assert their competence during the journey. Importantly, it gives passengers an increased sense
of control, whether that is just through giving them extra information to enable them to know
where they were both in the journey (ETA) and in space, or whether they actually used this to
correct or direct the driver, which is one way of exhibiting competence. In addition to Google
Maps, participants used applications like BlindSquare and DotWalker to track their location, citing
their increased accuracy and availability of additional information like landmarks as reasons for
their preference. These resources are used to supplement (not replace) cues like smells, sounds,
yovers, and bridges, which come into play on known routes. Although during the interviews most
participants reported using some form of technology to track their location, during the observations
we noted that participants also relied on their drivers to communicate details about the journey,
including information about their route, their location, and journey time.
6.1.4 Flexibility. Ride-hailing oers considerable exibility compared to other transportation
services because it is on-demand. This enables people to enjoy little freedoms, such as spending a
few extra minutes at home in the mornings to catch up on news.
"I am quite condent though I get late. If [...] I should listen [to the cricket] highlights one
more time. And I have that passion, I do not want to miss and go. By the time I nish that
rst half it will be 8:45 so if at all I have that condence, I have Ola, no other worries." P4
Even where participants had regular arrangements with drivers, all remembered times when they
had been stood up without notice and were able to fall back on ride-hailing services. Such services
allow people the exibility to get around when other forms of transportation are unavailable or
dicult to access, for example late at night or, as with P4 above, when they simply wanted the
luxury of dawdling over cricket news.
Another useful resource coming from the combination of ride-hailing and Google Maps is the
Estimated Time of Arrival (ETA), which enables considerably more accurate planning than public
transport. While also experienced by sighted passengers, the magnitude of the problem is increased
for people with visual impairments, who are more likely to miss their bus and take longer to nd
an auto. As P19 said,
"If I am going for a date - I don’t want to disappoint my date by being late for an hour, or
if I am going for an interview I don’t want to disappoint my future employees by being
late by an hour. So I am condent, you know, I can reach [my destination]. That’s it -
that’s a huge boost." P19
Uber and Ola also made traveling to, and within, unfamiliar locations possible. Many participants
reported how they were now able to access transportation in a uniform and consistent manner
across geographic contexts, removing the stress of traveling to new locations, whether locally
or further aeld. Using buses and trains requires a certain amount of local knowledge while the
consistent way in which Uber and Ola services operate made them a convenient choice.
6.1.5 Formalisation. Ride-hailing services formalize some aspects of the transportation expe-
rience, and participants mostly found this benecial. Specically, interviewees discussed safety,
payment, and recourse. An increased sense of
safety
came from the ability to track the journey,
share their trips with other people (in the case of Uber) and use the panic/SOS buttons in case of
emergencies. However, one participant reported feeling safer on public transport such as the metro,
because it was busy and public, and help was routinely provided by the metro sta.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:12 V. Kameswaran et al.
Two aspects of
payments
were called out by our interviewees as contributing to feelings of
independence (1) the automated fare calculation, and (2) digital payments. Firstly, the automatic
fare calculation both liberated passengers from bargaining and reduced concerns about cheating.
"The saving of the time when we go out and bargain with the driver and [...] these types
of things we don’t have to do and to save your energy, saves your time, saves your you
know, [getting] upset due to the fact that you are not getting any transportation. So [...] it
has become quite easy." P15
This quote clearly elucidates the emotional work involved in taking auto-rickshaws and public
transportation. This is recognizable to any transport user in India, and the main motivation for
sighted passengers to use OlaAuto was because it removed the fare negotiation [
3
], but again this
challenge is magnied for people with visual impairments.
Secondly, the integration of digital payment systems allowed participants to have additional
control over payment. This is because cash transactions are time-consuming and leave people
open to cheating, given that dierent denominations of notes and coins are not easy to tell apart.
Furthermore, having cash carries a safety risk. Several participants were worried about using credit
cards because they often had to seek help to use them, whereas the digital wallet is relatively
more accessible. Nonetheless, many participants continued to use cash (Figure 3), citing a lack of
trust in digital payment platforms and driver preference for cash. The dierent ways in which the
participants distinguished between currency notes included using the length of the notes to tell
between denominations (only partially reliable because certain denominations are similar in size)
and using dierent portions of their wallet to organize notes, with some taking sighted help to
help with this process. Participants described the note identication process as being dicult and
time-consuming, with participants having to rely on drivers to verify denominations and hand
back the right change.
During the observations we also noted that participants relied on drivers to communicate the
cost of the trip to them, even though cost details were available on the app. Because this information
is on the app, and the passengers have recourse with the ride-hailing companies, passengers have a
reasonable condence in a fair transaction. Furthermore, there is also a fair amount of trust in the
passenger-driver interaction (as illustrated by P9), even if it is tempered by an awareness that not
all drivers are trustworthy.
Formalization also gave participants options for
recourse
in the event of a poor experience.
Ola had a customer support line and Uber took complaints through the app. Although only a few
participants had registered complaints with Ola and Uber, others were comforted by the fact that
this option was available. P2 described how he was overcharged on a trip, but after complaining,
got his money back. He contrasted this with his feeling of often being cheated by auto-rickshaw
drivers and being able to do little about it.
There is a tension in the relationship between the passenger and driver, which is magnied
for people with visual impairments. While they are comfortable with the driver providing them
assistance to nd the cab and arrive at the location, there is also an often-hard-won awareness that
not all drivers are trustworthy, and it is much easier to deceive people with visual impairments.
Passengers do not have an innate trust in the drivers; they are aware there are both trustworthy
and untrustworthy drivers and that it is not always easy to tell them apart. However, some recourse
is found in technology - both in the features of the apps themselves (location tracking, payment,
etc.), and the ways in which they formalize certain aspects of the service, and through resources
such as the competence and condence instilled by Google Maps. These resources give people with
visual impairments more control over their traveling environment.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:13
Figure 3. Cash transaction in Ola cab Figure 4. Participant (in red) taking sighted help to nd cab
6.2 Challenges to independence
We have described the advantages of ride-hailing services and how these aord a level of indepen-
dence for people with visual impairments in India. However, our observations and participants’
descriptions showed a number of challenges to using these services, including: (1) app inaccessibility,
(2) nding the cab, (3) choosing and negotiating drop locations, and (4) driver troubles.
6.2.1 App inaccessibility. The apps themselves, while basically usable, are not fully accessible.
One inaccessible feature in both apps was the map view, which shows where the car is in relation to
the passenger and in which direction it is going. As a result, most participants relied on notications
or sighted help to determine where the car was on the map.
More features were accessible in the Uber app than the Ola app, perhaps because Uber, as a
multi-national corporation, might face more pressure to conform to accessibility standards that are
prevalent in other countries where it operates. In the Ola app, the buttons were not always well
labeled. Interviewees found it dicult to choose among the dierent types of vehicles, forcing them
to rely on sighted help by either handing over their phones to sighted acquaintances (although this
means having to turn the screen reader o), or asking acquaintances to book cabs for them on their
own phones (enabled by cash payments). The relative accessibility of the Uber app was one of the
main reasons for participants preferring it over Ola. In addition, software updates were a concern
in both applications because they could herald reduced accessibility without notice. Interviewees
either desisted from updating altogether (at the cost of missing out on new features) or sought out
information about interface changes from online and oine sources before updating.
6.2.2 Finding the cab. Almost all participants mentioned that the biggest challenge with ride-
hailing was identifying the precise location of the vehicle once it had arrived. Ola and Uber display
the cab’s registration number; however, the only practical way to nd the cab is to match this
provided number to the cab itself by visually searching for the car with the right number. Although
ride-hailing is ostensibly a door-to-door service, in practice many factors intervene, such that the
cab could easily be 50-100 meters away (or farther). The cab could also be parked on the other side
of the road from the rider, and its orientation is unknown to the participants. Several factors play a
part in this happening: (1) GPS accuracy is not always perfect; (2) it is not always easy to establish
the correspondence between booking location (when inside an oce building, apartment block or
college campus) and the best access point for cabs; and, (3) the cab needs to nd a safe place to wait
for the passenger. For example, during one observation, P5 read out the cab number from the Uber
app to his mother who was accompanying him to nd the cab. She located the cab less than 20
meters away on an adjacent perpendicular road and asked the cab driver to back up. P5, apparently
alerted by the sound of the cab, then moved toward it and only avoided colliding with it because of
his mother’s intervention. In this case, nding the cab was a joint achievement between P5, his
mother, and the driver. Our interviewees frequently
took sighted help
to locate the cab (Figure
4); however, it was not always eciently achieved. In another observation, P2 sought help from the
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:14 V. Kameswaran et al.
security guard in his apartment building to nd the cab, after phoning the driver and being told the
cab had already arrived. Although the cab was the only car on the road, the guard, did not identify
it and spent a few minutes looking. P2 then used the app to call out the cab number, and the guard
located the cab and escorted P2 to it.
The same people who helped locate the cab, typically escorted participants to it. They also helped
locate the unoccupied seat in the case of shared rides and opened the car door to show passengers
in. An alternate strategy to taking sighted help was to
involve the driver
in locating the rider.
Typically, participants called the driver from the app, described their location, and gave cues to the
drivers that they could use to identify both their location, such as landmarks, and them, e.g., shirt
color or white cane. P12 described his experience with both strategies:
"Initially, I had no idea how to direct my driver to my oce - so I used to talk to my friends,
this is the vehicle number, tell me if it comes because driver may not be able to identify
me. They will expect me to wave my hands - so I took few people’s help initially, then later
[...] I saved my location. If I select that then I can pinpoint my driver nowadays. I will
stand with my walking stick [and tell him] you can identify me and pick me." P12
However, it was common for drivers to try to circumvent road challenges like U-turns by asking
passengers to cross the road or walk from the narrow side streets to main roads to nd the cab.
Such requests were frustrating because navigating the roads is arduous and often requires sighted
intervention. At this point interviewees typically disclosed their disability to the drivers if they had
not already done so.
"I just ask the driver "where have you parked exactly?" I am a blind person [...] I am
wearing such and such color of shirt, so you know, can you see me? I ask the driver. When
the driver says "Yes, I can spot you" and then he would be able to guide me. Most of the
times the drivers have cooperated with me, but there have been some cases where the
drivers have been sort of little rude, because rst of all if you are traveling in a shared
taxi - the driver also has, you know, time issues." P6
Most noted that disclosing their disability resulted in drivers traveling to them, although in some
cases participants reported that drivers were still not accommodating. For instance, P22 explained
how drivers were usually more helpful once he revealed his disability but that on one occasion
his driver still refused to take a U-turn and he had to take a friend’s help to cross the road. For
some participants, often those working in colleges or other large grounds where cab drivers were
unwilling to venture, this was a major barrier to using ride-hailing. They therefore preferred other
forms of transportation. As P28 described:
"And mostly the drivers is really not like a sensitive so, if you are like a visually challenged.
So when I just ask the pickup [...] just come to my hostel so like they ask me [...] please
come to the main gate. [...] So I have to like walk 1 kilometer and then I have to catch this
cab thing. So it’s horrible." P28
P11 described how he had to frequently fall back on his students to take him to the college gate
to nd his cab, and so he returned to using his own car.
"For the whole month I am using Ola and Uber, but this serious problem was so much
there because from my college I had to depend on my students at least on 6-7 occasions.
So, that was very problematic - so from March I again [...] started using my own car." P11
Although P11 did say that having access to ride-sharing boosted his self-esteem and condence,
the practical problems associated with nding the cab, which are particularly large because he works
at a college, meant his private car (with a driver) was more convenient. Not only is ride-hailing not
always "door-to-door" enough, it is not always "on-demand" enough.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:15
"So, I would love it if I could ditch my car. [...] It is a time away because I’ll tell you what
happened: I go with Uber to the supermarket okay, I buy my stu. I can’t predict how fast
the checkout lines will move, right, and I call the driver and he’s waiting there and then
the challenge of nding him with all the shopping, so that’s a bit of a nuisance." P14
People with visual impairments are often prepared to pay extra for independence and will stretch
their budget to take the transportation that aords the greatest independence. For P11 and P14 it
was their private car (with drivers), but as the most expensive option, it was not available to all. For
others it was ride-hailing, but if too much help is required to nd the cab, the value of ride-hailing
is impacted. P11 explicated the role independence plays in transportation choice.
"Sometimes if some colleague is going in that direction he will denitely give lift but that
basically defeats the whole purpose of taking the cab, because it is the independence it
aords. [...] [the] point is that he doesn’t have to ask for help from anybody, then it’s worth
paying money [...] But you have to pay money and then you have to depend on others
also, then it’s no point." P11
Finding the cab was the main barrier to independent travel, leading interviewees to speculate
about whether a technical solution would help. A few proposed that they would like to indicate
disability in the app. Others proposed that the app might support them in locating the car.
"For example my Ola app is saying, the Ola which you have booked is 10 feet right from
you [...] the cab driver can track me or I can track the car and they can give me proper
directions." P10
While this could be a useful extension of the app, it could be technically dicult to get enough
precision and would not solve all the problems because the cab could still be across a busy road, for
example. Any solution would likely need to be a collaborative one helping driver and passenger
locate each other, and therefore involve both technology and driver sensitization to the needs of
passengers who might require extra help. We will return to this in the discussion.
6.2.3 Choosing and negotiating destinations. In addition to nding the cab, another diculty
faced by people with visual impairments is getting to the right drop location. Destination addresses
had to be typed in, and while participants saved frequent destinations, many addresses are either
unlisted or dicult to identify. Roads in Indian cities frequently share the same names (e.g., 4th
Main Road or 6th Cross are common street names) and the list provided by the app does not provide
enough detail to distinguish among them.
"And, sometimes there are even too many areas with same name. There is Chandra Layout
in Vijaynagar and there is another Chandra Layout near Marathalli - so the last time
I booked cab to Chandra Layout Vijaynagar, I ended up booking to Chandra Layout
Marathalli. And the driver told me this is where you have booked and I don’t think you
are planning to go there. So I had to get down and book another cab." P16
P16’s experience brings to light another challenge of using ride-hailing: the inability to change a
destination address once it has been conrmed, a exibility one has with auto-rickshaws, where
destinations can be changed on-the-y through negotiation with the driver. In comparison, drivers
cannot work around the Uber and Ola workow.
Lacking access to the visual resource that allows a sighted person to drag a pin on a map to mark
a destination address, many participants picked the address of the closest landmark listed in the
apps, such as restaurants, ATMs and oce buildings. Interviewees identied landmarks through an
active information-seeking process in advance of the trip; this involved asking friends and relatives
or searching the web. However, the problem with choosing a landmark, is that passengers must
then navigate from there to their actual destination. They might ask the driver to take them to the
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:16 V. Kameswaran et al.
actual destination, but drivers were not always willing to do so. Some interviewees talked of the
importance of being condent and asserting one’s right to get to the real destination.
"If I am stuck I go in condently saying "Boss, I have told you this destination get me to
this place" it’s as simple as that. [...] I will ensure that he drops me at the right point. I will
not get down only from the cab - what will he do? For those situations I book UberGo, so
that is not a trouble for the driver because if he spends additional minutes with me, that
fare is compensated." P18
This is not something that could easily be done with shared rides, which although cheaper, are
more rigid because the concerns of multiple passengers need to be balanced. Another strategy
involved participants using the landmark as an exchange or hand-o point, coordinating with
people to pick them up, although this reduces the independence aorded by ride-hailing services.
Even where not using a landmark, drivers often parked a few meters away. Passengers sought
information from the driver to orient themselves, but this was not always sucient. For example, in
one observation with P5, as the driver passed the destination, he asked "Where exactly do you want
to go?" P5 said: "XYZ oce building". Driver: "Exactly the XYZ building?". P5: "Yes". The driver
brought the vehicle to a screeching halt and remarked "XYZ oce building is about 50 meters
behind". In reality, it was about 200 meters behind. As he exited the vehicle P5 inquired "Where
exactly have you dropped me o?" to which the driver responded, "In front of the complex, the
XYZ oce is about 100 meters behind". Unfortunately, the road was full of commercial complexes
and P5 then asked the rst author to escort him to his oce building. Other times people in the
vicinity noticed the participants needed help and gave them information to orient themselves,
warned them about impending dangers, and even escorted them to their destinations. Take the
case of P22, who needed a timely intervention to prevent a fall:
"I asked [the driver] the landmark and he said everything was ne, so I got down there
but it was very far. [...] I thought that there is my home gate, I kept walking there and
some people told me that there was a gutter. [...]. Someone stopped me, otherwise I would
have fell down in that gutter." P22
Along with these core themes around independence, a couple of other themes came up in our
data around drivers’ reactions to disability and accessibility and technology.
6.2.4 Drivers Troubles. Interviewees often rely on the driver to help them achieve a successful
ride. However, while some drivers were helpful, overall they were something of a mixed bag.
Participants encountered drivers with little education and understanding of disability.
"Most of them don’t even have basic IQ levels, fundamentally understand that the passenger
can’t see so he will need additional help." P14
As discussed, some drivers have even cheated people with visual impairments. Although some
interviewees wanted to be able to identify as visually impaired on the app if this would get them
additional help, they were aware that because disability is not widely understood in India, this in
itself might cause problems. Interviewees suggested that drivers needed to be sensitized through
training. One suggested that further automation might be employed to reduce trust concerns, e.g.,
with notications from Uber if the driver changes or takes a longer route. Of course, trust can never
be fully automated, although putting checks in place could help enhance the perception of it from
the passenger side. However, it should not be forgotten that over-formalization can remove human
negotiation and judgment and be bad for both passenger and driver [3].
6.2.5 Accessibility and Technology. When discussing independence, Ola and Uber comprise just
one type of option in an ecosystem of technology that has positively impacted the lives of people
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:17
with visual impairments in India. As mentioned, Google Maps is commonly seen as a boon, not
just for directing the cab and instilling condence in one’s location, but also for checking ETAs,
giving directions and walking. Wallets and digital payments, in general, as well as online shopping
were also cited as having had a big impact in the lives of people with visual impairments. And the
smartphone itself has, especially for older participants, noticeably changed their lives.
"The kind of enormous leapfrogging they are having from 2000’s towards independent use
of technology has been tremendous. [...] Since the advent of the smartphone, you have a
technological solution for everything." P5
All of these are mainstream technologies enabled by the screen reader. Indeed, it was notable in
the interviews that aside from the screen reader, there was little mention of accessibility technologies.
As mentioned, as a mainstream technology, ride-hailing enables participants to call cabs for family,
as well as giving the family more condence in participants’ use of it because they used it too. Our
participants were well aware that they were now dependent on these services for their independence,
and although this is always a fragile situation somewhat outside their control (for example, the
widespread strikes by drivers impacted participants), they believed that as a mainstream service it
was sustainable.
7 DISCUSSION
In the previous section, we examined how ride-hailing services are transforming people’s lives by
enabling them to get out and about more, at their convenience, to unfamiliar places and even to
new cities. In this section, we examine what independence means for our participants regarding
transportation and what makes a service such as ride-hailing accessible. We nish with a discussion
on ride-hailing in India and discuss possible improvements.
7.1 Independence
Independence in HCI, CSCW, and accessibility research has primarily been understood in its
relationship to self-reliance (‘doing things alone, without help’ [
41
]) or less frequently, in its
relationship to autonomy (‘the freedom to exercise decisions’ [
21
]). Both self-reliance and autonomy
featured in our interviewees descriptions of how ride-hailing made them more independent. The
convenience of booking a cab through a mobile app allowed them to depend less on other people
and gave them more freedom to decide when to venture out and where to go. Ola and Uber aorded
a new, often greater, level of independence by formalizing the booking and payment of the cab
through a smartphone app that is largely accessible. Most participants talked in-the-large about
being able to take the cab by themselves, and although the ideal and reality do not quite match,
ride-hailing oers a much improved experience when compared to other transportation services.
The route tracking, digital payments and recourse features gave the interviewees a sense of control
over the ride, which contributed to their sense of independence. However, problems remained
around reading the map, and in locating the cab and the destination. As discussed, formalization
itself also exacerbates some accessibility problems, for example by impacting people’s ability to
choose landmarks where exact addresses are hard to specify. The workarounds found in other
on-demand transportation systems to overcome these local conditions, such as loose specication
and negotiation between driver and passenger, have been largely superseded by Uber and Ola’s rigid
workow, particularly in shared rides, causing problems for passengers with visual impairments.
Our research reveals a clear social dimension to independence. Brady et al. [
7
] touched on the
socially situated nature of independence in their study of question-asking behavior of people with
visual impairments. In the rather dierent context of ride-hailing, we too found independence to be
a socially situated phenomenon. For example, take P4’s articulation of accepting help: although he
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:18 V. Kameswaran et al.
was able to walk to the bus stop on his own, he said that he normally would not refuse help for risk
of oending the person oering. It is clear that in the real world, independence is a rather more
complex phenomenon than can be captured by self-reliance, autonomy, and control. Certainly,
with ride-hailing in India independence involves not just what you can do for yourself, but what
you can do for others and their condence in your ability in doing it. Participants talked about
how technology in general and ride-hailing specically, made them more demonstrably equal to
sighted people. Independence appears to be socially situated, which is in stark contrast to notions
of independence as understood in much of the research at the intersection of HCI and accessibility
- which is independence as perceptions of the self, i.e. how independent one feels in performing
a specic task using a certain technology artifact [
4
,
6
,
32
,
53
]. Interestingly Morrison et al. [
32
]
discuss social independence, but they dened it as "the ability to be free from the constraints of
social interaction through independent abilities", which while relevant here is rather more closely
related to traditional ideas of self-reliance. In this study, independence is not just about how people
with visual impairments feel about themselves but also extends to how they feel society sees them.
This was crucial for our participants who are living in a society where there is a lack of education
around disability and ignorance about what people with dierent abilities can and cannot do.
Such nuances, can help us think more deeply about the ways in which assistive technologies
might impact the lives of people with disabilities: Do they help the user do more things, in this
case getting out an about? Of course. But further, do they help the user do things for other people?
Or help the user demonstrate to the world that they are a competent member of society [
15
]? All
of these contributed to our users’ sense of independence and self-worth and might provide a wider
focus for designing and evaluating assistive technology.
Although self-reliance was indeed an important factor in independence, even this has a social
element. In this study, interviewees talked about not needing to take sighted help, but this did not
include the driver. It is clear, just as with [
7
] that all help is not equal. Asking for help is social,
situated, and has an order to it - in this case a preference for whom you might ask for help. Roughly
it would appear to be the driver (or conductor on buses, or other sta on metros), family/friends,
and then strangers. The preference to seek help from the driver (and transportation sta) over
others is likely explained by ride-hailing being a paid service, allowing riders to seek assistance
without the feeling of reduced independence because they see it as part of the driver’s job to help.
Additionally, in the case of cabs, for a ride to take place the driver and passenger must successfully
locate each other. Transportation, whether bus, train, cab or auto-rickshaw ride, is a collaborative
achievement between the driver and possibly other sta, and the customer, whether the customer is
visually impaired or not. In these situations, it is not that independence requires doing everything
alone; rather, independence here is about who needs to be called upon, in what context, and what
are the consequences of help-seeking for the various social relationships involved. We can relate
this to the notion of "interdependence" found in disability studies. Interdependence highlights
aspects of these social relationships by recognizing the mutual connectedness between people and
their dependence on one another as a counterpoint to challenge traditional ideas of independence
(which highlights self-suciency)[
16
,
41
]. Although the interviewees did not characterize their
reliance on the driver as interdependent, it is worth considering how this concept relates to their
notions of independence. As members of society, we are all interdependent on one another, from
the people who provide services and goods, to families, friends, and sometimes strangers. It is the
way that these interdependencies play out - with greater or lesser equity - that gives us our sense of
independence despite our practical day-to-day interdependence. In our study, our participants’ sense
of independence in part stemmed from an interdependent relationship with the driver. Certainly, it
should not be thought that there were no social consequences of asking help from the driver, it is
just as important for the passenger to demonstrate competence in the driver-passenger relationship
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:19
as in any other. Hence, interviewees talked about needing condence to ensure they got to the
right destination, of being able to demonstrate mastery of route, avoid cheating, and so on. Of
course, demonstrating expertise and competence is part of establishing a collaborative relationship
[
33
]. Unfortunately, in the case of ride-hailing, drivers do not always cooperate because they are
not aware of the needs of people with visual impairments or they are not willing or able to adapt to
them. Additionally, shared rides, while being more aordable, put extra constraints on the ability
of the driver to help, because they must consider the other passengers and there is no leeway in
the system for changing the destination or increasing the fare.
Independence is also relative; ride-hailing was not without its problems, however, for many
interviewees it was still considerably easier to get around with ride-hailing than with other means
of transportation. Interviewees who already used auto-rickshaws frequently, and either had rela-
tionships with drivers or were used to hailing autos and traveling alone, said that Uber and Ola
had not drastically changed their lives because they were already rather independent. However,
ride-hailing services oer certain improvements over the auto: they are a little safer and can be
hailed a little more easily, with no bargaining required. This relative dimension also serves as a
counterpoint to more deterministic representations of independence that are evident in accessibility
work using self-reporting Likert scales to understand the extent of one’s independence [35,38].
Finally, it is important to reect on independence for whom. People with disabilities in India
often have low literacy levels and social stigma and a large percentage (approximately 64%) of
adults with a disability in the country, including people with visual impairments, are unemployed
and subsequently lack access to a steady income [
1
]. On the contrary, our sample of interviewees
belonged to the middle and upper-middle classes, were educated and employed, and could hence
aord to use ride-hailing services. Even so, the cost of ridesharing was a burden for them, hence
the preference by some for shared rides like UberPool and OlaShare, despite the disadvantages
they conferred. Although research has underlined the potential economic benets of ride-hailing
for riders [
17
], as mentioned previously most cities in India have access to robust urban mass
transit infrastructures, which in spite of being much less accessible are considerably cheaper. This
likely explains why in most cases ride-hailing did not entirely replace participants’ use of public
transportation services. Rather, they used Uber and Ola judiciously alongside buses and trains,
especially when they desired the additional exibility (traveling to and in unfamiliar locations) and,
most importantly, greater independence. That interviewees were willing to pay for ride-hailing
services even when "it pinches" (P18), shows the value they put on independence. It would be
interesting to further investigate any gender imbalance in ride-hailing because we had trouble
locating many female participants. This might be because the main researcher was male, but it
could also be because there is an underlying dierence in access.
7.2 On Accessibility
The accessibility of ride-hailing services is a collaborative and socio-technical achievement, between
the driver, passenger, and technology. The technologies themselves were largely accessible, in that
interviewees could use them with a screen reader. Various textual resources made available useful
information like ETA and notications about the arrival of the cab. However, inaccessible features
included the map, unlabeled buttons, and software updates. However, accessibility is not only about
the technology, it is about the service provided. While technology is integral to it and a major
enabler of access, if passengers can’t nd the cab easily or struggle to get to a useful destination, the
accessibility of the app becomes irrelevant. Accessibility is achieved through the interplay of the
social and technical, it is not inherent in the technology alone. This is in contrast to the "functional
accessibility" [
45
], inherent in many HCI studies, and we would argue that this is likely to be a
feature of all accessible technologies, because they are never used in isolation. People with visual
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:20 V. Kameswaran et al.
impairments, like everyone else, are operating in the world, which is full of other people, animals,
obstacles, and unexpected happenings. A cab ride is a collaborative achievement anywhere in the
world [
23
] for anyone (except perhaps with driver-less cars, but even these operate in the social
world). Its achievement depends on the interplay between driver, passenger, and in the case of
ride-hailing, technology. Despite the many advantages of ride-hailing services and their positive
impact on perceptions of independence, ride-hailing was not without its challenges for people with
visual impairments in India. To overcome these challenges, participants had to work with drivers
and others. In contrast to collaborations between people with visual impairments and sighted
partners/co-workers in homes and workplaces, where assistance was carefully sought and weighted
so as to not aect relationships [
8
,
9
], in ride-hailing, as a paid service, participants expected the
driver to assist them. However, drivers were largely unfamiliar to riders (although a few reported
encountering the same driver on occasion) and receiving assistance from the driver often meant
disclosing their disability and needs, which they were not always comfortable with. This is likely
dierent from such exchanges in homes and workplaces, where people grow familiar with one
another and their needs over time. Where help was not forthcoming from the driver, participants
often had to fall back on the kindness of strangers, helpfulness of acquaintances, and availability
of colleagues, friends, and family. While needs are often understood by familiar people, it can be
challenging to communicate needs and elicit help from strangers.
Finally ride-hailing is just one in an ecosystem of mainstream technologies transforming the
lives of (at least middle-class) people with visual impairments in India. It is important therefore
to make all technologies accessible because mainstream technologies can almost incidentally
3
provide great leaps for the independence of people with visual impairments. Of course, in the
Global North the accessibility of technology is a legal matter, but it is often neglected in parts of
the Global South. Furthermore, even for the Global South, it is important to remember that the
accessibility of the technology is only one part of the puzzle; any services such apps provide access
to must also be accessible, which can spill over into social rather than technical concerns. In this
case, the accessibility of ride-hailing was tied to the vagaries of the drivers, not all of whom were
helpful. The unhelpfulness is likely the result of the stigma associated with disability, which results
in perceptions that people with disabilities are a liability, as well as a lack of understanding of
their needs. Accessibility here is rather ad hoc, and we would suggest that Ola and Uber take on
responsibility for providing an accessible service rather than leaving it to the individual drivers.
7.3 Understanding ride-hailing in India
This is one of only a few studies of ride-hailing outside the Global North. Whereas Ahmed et al. [
3
],
like the majority of ride-hailing studies focused on drivers, in that case the auto-rickshaw drivers
of OlaAuto in Bangalore, India, in this study we examined the passengers’ experience. As such,
we bring to light two sets of ndings that we believe are of wide interest: rst, the experience of
ride-hailing for passengers in India and, second, that of passengers with visual impairments in
India. We examine each of these in turn.
Few studies have examined passengers’ experiences of ride-hailing, and those that have have
been in the United States [
13
,
17
,
22
]. There are certainly similarities between the U.S. and India, for
example similar motivations for using ride-hailing apps, such as greater convenience than public
transit or hailing a cab in the street. However, a core motivation in the U.S. was price: compared to
traditional taxis ride-hailing services were much cheaper, nonetheless money remained a concern,
although pitched more as not being too extravagant [
17
]. In India, ride-hailing is certainly not a
cheap option, compared to more traditional forms of transport, and like in Dillahunt’s study [
13
], a
3Incidentally, since neither Ola and Uber were designed with this aim in mind, nor do much to promote it
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:21
concern with price permeates the ride-hailing discourse for Indian passengers, as evidenced by the
popularity of shared rides. In fact, the aordability of urban mass transit services in addition to their
regularity are crucial in countering some of the challenges associated with their use. In the U.S., too,
there was concern with not being "ripped o" during the ride, which was mitigated by the use of
Google Maps by passengers during the ride, plus app features like the receipt, which included the
route map [
17
]. An interesting dierence between the U.S. and India is that drivers in the U.S. have
been depicted by passengers as being just "like me", this did not come up at all in our study, perhaps
because the demographics of drivers is rather dierent in the two countries. In India driving a
ride-hailing cab is a full-time job, with most drivers working 12-14 hour days, the same spread of
part-time and hobby drivers found in the U.S. is not present. Further ride-hailing passengers in
India are typically middle-class because of the relative expense of ride-hailing and the requirement
to own a smartphone, whereas drivers often come from the working classes, with many coming in
from the villages. Two other factors dierentiate ride-hailing in India from the Global North: the
formalization of the service through the app and the problem of addresses. Unlike the Global North,
many of India’s rich transportation services are informal (including many of the buses). From
the drivers’ perspective, working on Uber or Ola keeps them in the informal marketplace (while
removing some of their freedom and agency [
3
]; however, for the passengers, the app formalizes a
number of features of the service, one of the most important being the price, and this both reduces
the emotional work of bargaining and gives the passenger a greater sense of fairness [
3
], which
is particularly important given the price-consciousness of many passengers. Secondly, addresses
can be highly problematic in India, many are poorly specied or hard to distinguish from multiple
similar options. The traditional route around this is to specify landmarks (and even Internet-based
grocery delivery services such as Big Basket enable this); however, this is not enabled in Ola or
Uber. This means that choosing an address on a ride-hailing app can be rather problematic. The
formalization of the app further confounds this by removing the negotiation between driver and
passenger, making working around wrongly chosen addresses almost impossible.
For passengers with visual impairments in India, many of the same advantages and disadvantages
are magnied. For example, the convenience of ride-hailing over other forms of transportation is
enhanced because the others are so much more inaccessible and require so much more help to
take. Hailing a cab away from the main streets in San Francisco might be dicult [
17
], hailing
an auto-rickshaw as a sighted passenger might be time-consuming and eortful, but the number
of obstacles that must be negotiated by a passenger with visual impairment is many times worse
because even just getting to the main street can be fraught with hazards. So although ride-hailing
services oer an improvement over other forms of transportation, much can be done to improve
ride-hailing, including: (1) driver training and sensitization to enable drivers to better understand
and accommodate the needs of passengers with disabilities; (2) ride-hailing company policies to
help people with disabilities, for example enabling drivers to go up to 200 meters beyond the
specied destination to accommodate them, even in shared rides; (3) more sensitive and better
documented software updates; and (4) ways to make information from the map available.
8 CONCLUSION
In this paper we present a qualitative investigation into ride-hailing use by people with visual
impairments in metropolitan India. We found that ride-hailing oers people with visual impairments
several benets over modes of transportation like buses and auto-rickshaws, making daily travel
more convenient than before. We studied the notions of independence resulting from people with
visual impairments’ use of ride-hailing and found that independence is social, relative, and situated,
thus contributing to a broader understanding of the concept. Finally, we also gained understanding
of how interactions between people with visual impairments and drivers makes it possible for
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:22 V. Kameswaran et al.
people to use Uber and Ola, leading us to conclude that accessibility in the case of ride-hailing is a
socio-technical and collaborative achievement.
9 ACKNOWLEDGEMENTS
We would like to thank Megh Marathe, Priyank Chandra, and Robert Markum who provided helpful
comments on previous versions of the paper.
REFERENCES
[1] 2007. People with disabilities in India: From Commitments to Outcomes - The World Bank. (May 2007).
[2] 2017. Accessible Mass Transit - American Foundation for the Blind. (2017).
[3]
Syed Ishtiaque Ahmed, Nicola J Bidwell, Himanshu Zade, Srihari H Muralidhar, Anupama Dhareshwar, Baneen
Karachiwala, Cedrick N Tandong, and Jacki O’Neill. 2016. Peer-to-peer in the Workplace: A View from the Road. In
Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems (CHI ’16). ACM, New York, NY, USA,
5063–5075. https://doi.org/10.1145/2858036.2858393
[4]
Shiri Azenkot, Sanjana Prasain, Alan Borning, Emily Fortuna, Richard E Ladner, and Jacob O Wobbrock. 2011. Enhancing
Independence and Safety for Blind and Deaf-blind Public Transit Riders. In Proceedings of the SIGCHI Conference on
Human Factors in Computing Systems (CHI ’11). ACM, New York, NY, USA, 3247–3256. https://doi.org/10.1145/1978942.
1979424
[5]
Mary M. Ball, Molly M. Perkins, Frank J. Whittington, Carole Hollingsworth, Sharon V. King, and Bess L. Combs. 2004.
Independence in assisted living. Journal of Aging Studies 18, 4 (nov 2004), 467–483. https://doi.org/10.1016/j.jaging.
2004.06.002
[6]
Erin L. Brady, Daisuke Sato, Chengxiong Ruan, Hironobu Takagi, and Chieko Asakawa. 2015. Exploring Interface
Design for Independent Navigation by People with Visual Impairments. In Proceedings of the 17th International
ACM SIGACCESS Conference on Computers & Accessibility (ASSETS ’15). ACM, New York, NY, USA, 387–388. https:
//doi.org/10.1145/2700648.2811383
[7]
Erin L. Brady, Yu Zhong, Meredith Ringel Morris, and Jerey P. Bigham. 2013. Investigating the Appropriateness of
Social Network Question Asking As a Resource for Blind Users. In Proceedings of the 2013 Conference on Computer
Supported Cooperative Work (CSCW ’13). ACM, New York, N Y, USA, 1225–1236. https://doi.org/10.1145/2441776.2441915
[8]
Stacy M. Branham and Shaun K. Kane. 2015. Collaborative Accessibility: How Blind and Sighted Companions Co-Create
Accessible Home Spaces. In Proceedings of the 33rd Annual ACM Conference on Human Factors in Computing Systems
(CHI ’15). ACM, New York, NY, USA, 2373–2382. https://doi.org/10.1145/2702123.2702511
[9]
Stacy M. Branham and Shaun K. Kane. 2015. The Invisible Work of Accessibility: How Blind Employees Manage
Accessibility in Mixed-Ability Workplaces. In Proceedings of the 17th International ACM SIGACCESS Conference on
Computers & Accessibility (ASSETS ’15). ACM, New York, NY, USA, 163–171. https://doi.org/10.1145/2700648.2809864
[10]
Simon Brisenden. 1986. Independent living and the medical model of disability. Disability, Handicap & Society 1, 2
(1986), 173–178.
[11]
Megan Campbell, Cynthia Bennett, Caitlin Bonnar, and Alan Borning. 2014. Where’s my bus stop?: supporting
independence of blind transit riders with StopInfo. In Proceedings of the 16th international ACM SIGACCESS conference
on Computers & accessibility. ACM, 11–18.
[12]
Guy Dewsbury, Karen Clarke, Dave Randall, Mark Rounceeld, and Ian Sommerville. 2004. The anti-social model of
disability. Disability & society 19, 2 (2004), 145–158.
[13]
Tawanna R. Dillahunt, Vaishnav Kameswaran, Linfeng Li, and Tanya Rosenblat. 2017. Uncovering the Values and
Constraints of Real-time Ridesharing for Low-resource Populations. In Proceedings of the 2017 CHI Conference on Human
Factors in Computing Systems (CHI ’17). ACM, New York, NY, USA, 2757–2769. https://doi.org/10.1145/3025453.3025470
[14]
Robert M Emerson, R Fretz, and L Shaw. 1995. In the eld: Participating, observing, and jotting notes. Writing
ethnographic eldnotes (1995), 17–35.
[15] Harold Garnkel. 1967. Studies in Ethnomethodology. (1967).
[16]
Barbara E Gibson, Franco A Carnevale, and Gillian King. 2012. "This is my way": reimagining disability, independence
and interconnectedness of persons and assistive technologies. Disability and Rehabilitation 34, 22 (2012), 1894–1899.
[17]
Mareike Glöss, Moira McGregor, and Barry Brown. 2016. Designing for Labour: Uber and the On-Demand Mobile
Workforce. In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems (CHI ’16). ACM, New
York, NY, USA, 1632–1643. https://doi.org/10.1145/2858036.2858476
[18]
Erving Goman. 1959. The Presentation of Self in Everyday Life. Garden City, N.Y. : Doubleday, 1959. https://search.
library.wisc.edu/catalog/999467804702121
[19]
Reginald G Golledge. 1993. Geography and the disabled: a survey with special reference to vision impaired and blind
populations. Transactions of the Institute of British Geographers (1993), 63–85.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
‘We can go anywhere’: Understanding Independence ... 85:23
[20]
Jonathan Grudin. 1990. The Computer Reaches out: The Historical Continuity of Interface Design. In Proceedings
of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’90). ACM, New York, NY, USA, 261–268.
https://doi.org/10.1145/97243.97284
[21]
Sarah Hillcoat-Nalletamby. 2014. The Meaning of "Independence" for Older People in Dierent Residential Settings.
The Journals of Gerontology: Series B 69, 3 (2014), 419–430. https://doi.org/10.1093/geronb/gbu008
[22]
Vaishnav Kameswaran, Lindsey Cameron, and Tawanna R. Dillahunt. 2018. Support for Social and Cultural Capital
Development in Real-time Ridesharing Services. In Proceedings of the 2018 CHI Conference on Human Factors in
Computing Systems (CHI ’18). ACM, New York, NY, USA, Article 342, 12 pages. https://doi.org/10.1145/3173574.3173916
[23]
Joseph Kasera, Jacki O’Neill, and Nicola J Bidwell. 2016. Sociality, Tempo & Flow: Learning from Namibian Ride-
sharing. Proceedings of the First African Conference on Human Computer Interaction - AfriCHI’16 (2016), 36–47.
https://doi.org/10.1145/2998581.2998582
[24] Steinar Kvale. 2008. Doing interviews. Sage.
[25]
Min Kyung Lee, Daniel Kusbit, Evan Metsky, and Laura Dabbish. 2015. Working with Machines: The Impact of
Algorithmic and Data-Driven Management on Human Workers. In Proceedings of the 33rd Annual ACM Conference on
Human Factors in Computing Systems (CHI ’15). ACM, New York, NY, USA, 1603–1612. https://doi.org/10.1145/2702123.
2702548
[26]
Haley MacLeod, Cynthia L Bennett, Meredith Ringel Morris, and Edward Cutrell. 2017. Understanding Blind People’s
Experiences with Computer-Generated Captions of Social Media Images. In Proceedings of the 2017 CHI Conference on
Human Factors in Computing Systems (CHI ’17). ACM, New York, NY, USA, 5988–5999. https://doi.org/10.1145/3025453.
3025814
[27]
James R Marston and Reginald G Golledge. 2003. The hidden demand for participation in activities and travel by
persons who are visually impaired. Journal of Visual Impairment and Blindness 97, 8 (2003), 475–488.
[28]
D A C Maunder, C J Venter, T Rickert, and J Sentinella. 2004. Improving transport access and mobility for people
with disabilities. In CILT Regional Conference, International Logistics and Transport: The Challenges Ahead, Number
PA4061/04, Dubai, UAE. Transport Research Laboratory.
[29]
Johanna Meurer, Martin Stein, David Randall, Markus Rohde, and Volker Wulf. 2014. Social Dependency and Mobile
Autonomy - Supporting Older Adults Mobility with Ridesharing ICT. (2014).
[30]
J Morris and J Mueller. 2014. Blind and Deaf Consumer Preferences for Android and iOS Smartphones. In Inclusive
Designing, P M Langdon, J Lazar, A Heylighen, and H Dong (Eds.). Springer International Publishing, Cham, 69–79.
[31]
Meredith Ringel Morris, Annuska Zolyomi, Catherine Yao, Sina Bahram, Jerey P. Bigham, and Shaun K. Kane. 2016.
"With Most of It Being Pictures Now, I Rarely Use It": Understanding Twitter’s Evolving Accessibility to Blind Users.
In Proceedings of the 2016 CHI Conference on Human Factors in Computing Systems (CHI ’16). ACM, New York, NY, USA,
5506–5516. https://doi.org/10.1145/2858036.2858116
[32]
Cecily Morrison, Edward Cutrell, Anupama Dhareshwar, Kevin Doherty, Anja Thieme, and Alex Taylor. 2017. Imagining
Articial Intelligence Applications with People with Visual Disabilities Using Tactile Ideation. In Proceedings of the
19th International ACM SIGACCESS Conference on Computers & Accessibility (ASSETS ’17). ACM, New York, NY, USA,
81–90. https://doi.org/10.1145/3132525.3132530
[33]
Jacki O’Neill. 2010. Making and breaking troubleshooting logics: Diagnosis in oce settings. In Ethnographies of
Diagnostic Work. Springer, 35–53.
[34]
Joyojeet Pal and Meera Lakshmanan. 2012. Assistive Technology and the Employment of People with Vision Impair-
ments in India. In Proceedings of the Fifth International Conference on Information and Communication Technologies and
Development (ICTD ’12). ACM, New York, NY, USA, 307–317. https://doi.org/10.1145/2160673.2160711
[35]
Joyojeet Pal, Anandhi Viswanathan, Priyank Chandra, Anisha Nazareth, Vaishnav Kameswaran, Hariharan Subra-
monyam, Aditya Johri, Mark S. Ackerman, and Sile O’Modhrain. 2017. Agency in Assistive Technology Adoption:
Visual Impairment and Smartphone Use in Bangalore. In Proceedings of the 2017 CHI Conference on Human Factors in
Computing Systems (CHI ’17). ACM, New York, NY, USA, 5929–5940. https://doi.org/10.1145/3025453.3025895
[36]
Joyojeet Pal, Maura Youngman, Terence O’Neill, Priyank Chandra, and Cyprien Semushi. 2015. Gender and Accessibility
in Rwanda and Malawi. In Proceedings of the Seventh International Conference on Information and Communication
Technologies and Development (ICTD ’15). ACM, New York, NY, USA, 5:1–5:9. https://doi.org/10.1145/2737856.2738020
[37]
D. Pascolini and S. P. Mariotti. 2012. Global estimates of visual impairment: 2010. British Journal of Ophthalmology 96,
5 (2012), 614–618. https://doi.org/10.1136/bjophthalmol-2011- 300539 arXiv:WHO/NMH/PBD/12.01
[38]
Anne Marie Piper and James D. Hollan. 2008. Supporting Medical Conversations Between Deaf and Hearing Individuals
with Tabletop Displays. In Proceedings of the 2008 ACM Conference on Computer Supported Cooperative Work (CSCW
’08). ACM, New York, NY, USA, 147–156. https://doi.org/10.1145/1460563.1460587
[39]
Sanjana Prasain. 2011. StopFinder: improving the experience of blind public transit riders with crowdsourcing. In The
proceedings of the 13th international ACM SIGACCESS conference on Computers and accessibility. ACM, 323–324.
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
85:24 V. Kameswaran et al.
[40]
Noopur Raval and Paul Dourish. 2016. Standing Out from the Crowd: Emotional Labor, Body Labor, and Temporal
Labor in Ridesharing. Proceedings of the 19th ACM Conference on Computer-Supported Cooperative Work & Social
Computing - CSCW ’16 (2016), 97–107. https://doi.org/10.1145/2818048.2820026
[41]
Solveig Magnus Reindal. 1999. Independence, dependence, interdependence: Some reections on the subject and
personal autonomy. Disability & Society 14, 3 (1999), 353–367.
[42] Peter Roberts and Julie Babinard. 2004. Transport strategy to improve accessibility in developing countries. (2004).
[43]
Yvonne Rogers, Liam Bannon, and Graham Button. 1994. Rethinking Theoretical Frameworks for HCI: Report on an
INTERCHI ’93 Workshop, Amsterdam; 25th April, 1993. SIGCHI Bull. 26, 1 (jan 1994), 28–30. https://doi.org/10.1145/
181526.181530
[44] Johnny Saldaña. 2015. The coding manual for qualitative researchers. Sage.
[45]
Kristen Shinohara and Jacob O. Wobbrock. 2011. In the Shadow of Misperception: Assistive Technology Use and Social
Interactions. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems (CHI ’11). ACM, New
York, NY, USA, 705–714. https://doi.org/10.1145/1978942.1979044
[46]
Kristen Shinohara and Jacob O Wobbrock. 2016. Self-conscious or self-condent? A diary study conceptualizing the
social accessibility of assistive technology. ACM Transactions on Accessible Computing (TACCESS) 8, 2 (2016), 5.
[47] Sanjay K Singh. 2005. Review of urban transportation in India. Journal of public transportation 8, 1 (2005), 5.
[48]
Aditya Vashistha, Erin Brady, William Thies, and Edward Cutrell. 2014. Educational Content Creation and Sharing
by Low-Income Visually Impaired People in India. In Proceedings of the Fifth ACM Symposium on Computing for
Development (ACM DEV-5 ’14). ACM, New York, NY, USA, 63–72. https://doi.org/10.1145/2674377.2674385
[49]
Manon M L Verdonschot, L P De Witte, E Reichrath, W H E Buntinx, and L M G Curfs. 2009. Impact of environmental
factors on community participation of persons with an intellectual disability: a systematic review. Journal of Intellectual
Disability Research 53, 1 (2009), 54–64.
[50]
Michael L Wehmeyer, Kathy Kelchner, and Sandy Richards. 1996. Essential characteristics of self-determined behavior
of individuals with mental retardation. (1996).
[51]
WHO. 2013. Factsheet on Visual impairment and blindness. (2013). http://www.who.int/mediacentre/factsheets/fs282/
en/
[52]
Jacob O Wobbrock, Shaun K Kane, Krzysztof Z Gajos, Susumu Harada, and Jon Froehlich. 2011. Ability-based design:
Concept, principles and examples. ACM Transactions on Accessible Computing (TACCESS) 3, 3 (2011), 9.
[53]
Rayoung Yang, Sangmi Park, Sonali R Mishra, Zhenan Hong, Clint Newsom, Hyeon Joo, Erik Hofer, and Mark W
Newman. 2011. Supporting Spatial Awareness and Independent Waynding for Pedestrians with Visual Impairments.
In The Proceedings of the 13th International ACM SIGACCESS Conference on Computers & Accessibility (ASSETS ’11).
ACM, New York, NY, USA, 27–34. https://doi.org/10.1145/2049536.2049544
[54]
Chien Wen Yuan, Benjamin V. Hanrahan, Sooyeon Lee, Mary Beth Rosson, and John M. Carroll. 2017. I Didn’T Know
That You Knew I Knew: Collaborative Shopping Practices Between People with Visual Impairment and People with
Vision. Proc. ACM Hum.-Comput. Interact. 1, CSCW, Article 118 (Dec. 2017), 18 pages. https://doi.org/10.1145/3134753
[55]
Yuhang Zhao, Shaomei Wu, Lindsay Reynolds, and Shiri Azenkot. 2017. The Eect of Computer-Generated Descriptions
on Photo-Sharing Experiences of People with Visual Impairments. Proc. ACM Hum.-Comput. Interact. 1, CSCW (dec
2017), 121:1–121:22. https://doi.org/10.1145/3134756
[56]
Annuska Zolyomi, Anushree Shukla, and Jaime Snyder. 2017. Technology-Mediated Sight: A Case Study of Early
Adopters of a Low Vision Assistive Technology. In Proceedings of the 19th International ACM SIGACCESS Conference on
Computers & Accessibility (ASSETS ’17). ACM, New York, NY, USA, 220–229. https://doi.org/10.1145/3132525.3132552
Received April 2018; revised July 2018; accepted September 2018
Proceedings of the ACM on Human-Computer Interaction, Vol. 2, No. CSCW, Article 85. Publication date: January 2018.
