Learning to Fix: Knowledge, Collaboration and Mobile

Phone Repair in Dhaka, Bangladesh

Syed Ishtiaque Ahmed
Information Science
Cornell University, Ithaca, NY
[email protected]

Steven J. Jackson
Information Science
Cornell University, Ithaca, NY
[email protected]

ABSTRACT

Practices of technology repair in developing country

contexts play crucial and often overlooked roles in
supporting ICTD goals of access and sustainability. They
also constitute complex and neglected sites of technical
skill, knowledge, and learning. Building on original
ethnographic fieldwork, this paper explores the explicit,
tacit, and social knowledges that shape practice and
expertise in the mobile phone repair markets of urban
Bangladesh. We document forms of learning and
collaboration central to the production and innovation of
repair skills and knowledge, and show how these processes
operate at the intersection of global knowledge flows and
local efforts to access, contextualize and situate that
knowledge. We conclude by arguing for repair as an
underappreciated site of third-world technical practice and
expertise, and reflecting on how ICTD research might
better take such practices into account.
Author Keywords

Repair; Learning; Craft; Skill; Tacit Knowledge;

Ethnography; Bangladesh
ACM Classification Keywords

H.1.2. Human Factors

INTRODUCTION

Knowledge, learning and collaboration have long been

central to the forms of technology use and adoption at the
heart ICTD research. Participatory design and action
research traditions of ICTD work have sought to build
collaborative partnerships between users and designers, in
the interests of more usable and locally appropriate forms
of design. Work on technology adoption has sought to
understand how technical skills and knowledge are shared
and transmitted among user populations, enabling or
disabling the adoption of new tools in developing country
Permission to make digital or hard copies of all or part of this work for
personal or classroom use is granted without fee provided that copies are
not made or distributed for profit or commercial advantage and that copies
bear this notice and the full citation on the first page. Copyrights for
components of this work owned by others than ACM must be honored.
Abstracting with credit is permitted. To copy otherwise, or republish, to
post on servers or to redistribute to lists, requires prior specific permission
and/or a fee. Request permissions from [email protected].
ICTD '15, May 15 - 18, 2015, Singapore, Singapore
Copyright is held by the owner/author(s). Publication rights licensed to
ACM.
ACM 978-1-4503-3163-0/15/05$15.00
http://dx.doi.org/10.1145/2737856.2738018

Md. Rashidujjaman Rifat

CSE, BUET
Dhaka, Bangladesh
[email protected]

contexts. Access advocates have explored complex and

distinct forms of sharing and collaboration that have
disrupted western assumptions around the relationship
between users and devices (and in particular, models of
individualized ownership that may be inappropriate or
limiting in many third-world contexts). Taken collectively,
these findings have driven important research and design
insights in ICTD scholarship, and helped to surface features
of post-colonial computing environments that set them
apart from their western counterparts.
But learning and collaboration are no less central or
complex in other technological moments and practices,
including those centered on technology maintenance and
repair. The fields traditional design orientation has tended
to obscure this work, leaving a long list of important and
interesting questions unasked. How do actors in third-world
environments build the skills and experience needed to deal
effectively with broken or malfunctioning devices? How do
repairers acquire the knowledge and skills needed to restore
function and value to broken tools and infrastructures? How
do people work together around repair, pooling and sharing
knowledge, skill, and experience? And how are collective
bodies of repair knowledge built, shared, updated and
sustained over time, at both the local and trans-local levels?
This paper seeks to address these questions through original
ethnographic fieldwork with mobile phone repair workers
in Dhaka, Bangladesh. We start by reviewing HCI and
social science literatures on repair, including older and
newer bodies of work that have called attention to
maintenance and repair as complex and constitutive sites of
collaborative technological practice. We then turn to
theories of learning and collaboration that have shaped our
understanding of working knowledge in contexts of skillbased material engagement. After briefly describing our
methods, we turn to findings from our fieldwork, offering
brief biographical sketches of three representative figures in
the mobile phone repair worlds of Dhaka. We describe
three distinct forms of knowledge around repair explicit,
tacit, and social and the forms of learning that
characterize the transmission of each. We conclude by tying
our findings back to larger ICTD concerns around
technology access and sustainability, and offering
recommendations to ICTD researchers for how they might
better account for practices of repair in their work.

BACKGROUND AND RELATED WORK

While its subsequent history of attention has been spotty,

work on technology repair in the broader information
science field in fact goes back several decades. Leveraging
long-standing ethnomethodological interests in social and
conversational repair (see, inter alia, Garfinkel [1],
Goffman [2], Sacks [21]), Lucy Suchmans seminal
contributions around the relationship between formalized
plans and situated actions were built on the collaborative
experience and challenge of photocopier repair work at
Xerox [23]. Julian Orr [14] extended this work in a learning
direction, studying the central role of war stories and other
informal means of knowledge flow among repair workers
as a central conduit of learning outside of the formalized
channels of official technique and procedure.
More recent work has built on these foundations to explore
motivations, values, and practices of repair across a wide
range of contexts. Research by Rosner and colleagues
[18,19,20] has explored connections between repair,
materiality and the longevity of artifacts; between repair
and gendered conceptions of technology use and identity;
and the material and ideological flows attached to repair
work in global development projects. Jackson and Kang [8]
have argued for links between repair, creativity and
collaboration in artistic production, and their connection to
questions of care and responsibility in human-object
relations more generally. Separate field programs in
Namibia [9,10], Uganda [4], and Bangladesh [7] have
explored patterns of infrastructure, practice, collaboration,
and local art and craft constituting local repair worlds
along with their connections to global flows of knowledge
and infrastructure around technology. Rangaswamy et al.s
[17] study of mobile phone repairers in Mumbai has argued
for the role of repair technicians as agile and innovative
producers (rather than passive and unskilled consumers) of
technology. Writing more broadly, Jackson [6] and
Graham and Thrift [3] have argued for moments of
breakdown, maintenance and repair as key elements in
human relationships with technology, with deep and
formative connections to questions of innovation, justice,
and sustainability.
These recent and older bodies of work suggest deep and
generative connections between repair and the forms of
technology design, use and adoption that have more
typically occupied ICTD scholarship. They point to the
relationship between repair and other values innovation,
access, and sustainability long central to ICTD concern.
They explore have pointed out tensions between the forces
and motivations that push people toward fixing, and the
complexities that confront the work once engaged. And
they have suggested the deeply collaborative nature of
repair, showing how people address complex challenges of
repair through the sharing of knowledge, skills, and
material resources at scales ranging from the workbench
and lunch table to large transnational networks. If this is
true generally, its even more pronounced under the

conditions of many third-world computing environments,

where devices, systems, and infrastructures may be
particularly prone to breakdown and failure, resources may
be in short supply, and repair-vs.-replace decisions may
follow a very different economic calculus.
If this work explains the importance and complexity of
repair along with its distinctive role in the third-world
computing environments, it also leaves important questions
unanswered. For example, while we know of the presence
and importance of global flows of technical knowledge and
material around repair, the precise mechanisms by which
such global resources interface with repair operations at the
local level remains unclear. Nor is it clear how local repair
communities are formed, organized, and reproduced over
time. Finally, the forms of learning and collaboration by
which repair skills and knowledge are circulated and
sustained within local repair networks require further
elaboration.
To answer these questions, we turn to classic work on
learning and collaboration in the social sciences. An
important part of the skills mastered through repair fall
under the category of tacit and craft knowledges, referenced
by the fact that, as Polanyi [15,16] argues, we can know
more than we can tell. He illustrates this with the example
of learning to ride a bike. On one hand this may be
informed by abstract conceptual knowledges relating to
gravity, momentum, centripetal and centrifugal forces, the
mechanics of chain and pedal, etc. But beyond this stand
hard to describe elements of look and feel that constitute the
actual knowledge of bike riding for those who have
mastered the skill. Scholars of craft like Sennet [22] have
explored the nature of craft work, forms of skillful material
engagement that embed relations of care, value, and pride
in work that link the subject and object or medium of work.
Resistant to formal specification, tacit and craft knowledge
is instead embedded in the practical accomplishment of
work, and often flows from teacher to learner, master to
apprentice, through processes of observation and
experience wrapped in trust, care, and respect.
Repair practice may also depend on forms of social
knowledge situated in professional or craft-based
communities of work. This argument is most powerfully
made by Laves [11] and Lave and Wengers [12] analyses
of the learning processes attached to communities of
practice that may form around craft and professional work
of all kinds. Lave and Wenger show how communities of
practice are often characterized and sustained through
apprentice relations responsible not only for flows of
knowledge from master to apprentice, but also the
production of social order, norms, and hierarchies and the
reproduction of community skill and knowledge as a whole.
Under this conception, learning is not a simple and one-way
absorption of factual information, but rather a process of
building a newcomer (though legitimate peripheral
participation) into a fully functioning member of a

community through involvement. Apprentices and

newcomers learn the norms and values of the community
by helping and observing the existing members, situate
themselves in it, and earn their membership by showing
their competence to exist in and contribute to the
community.
Taken together, prior ICTD and information science work
and the contributions of Polanyi, Sennet, McCullough, and
Lave and Wenger reviewed above provide deeper
understanding of the nature of the learning processes
involved in the production and transmission of repair
knowledge. At the same time, they connect the nuances of
the learning process to broader issues associated with the
profession, culture, and social organization of repair work.
In the empirical descriptions that follow we use these
theories to see how repairers establish relationships with
their tools and techniques, how they tune their hands, eyes,
and hearts with the technical forms of digital artifacts, and
how the learning processes associated with repair work
operate to convert apprentices to skilled repairers and fullyfledged members of the repair community. We argue that
this reproductive repairing process is essential in
sustainable development, and deserves proper attention.
METHODS AND FIELDSITES

Our study was divided into three broad phases. In the first
phase we conducted 58 semi-structured interviews with
mobile phone repair workers at 10 different sites in Dhaka,
Bangladesh, outlining basic patterns of repair activities,
identifying main repair sites and networks, establishing
long-term relationships with informants, and refining the
research questions. The next phase consisted of 4 month
ethnographic field study conducted in Summer 2013,
visiting major repair sites, observing ongoing work, and
conducting an extended series of interviews with a select
subset of participants (from which the biographies below
are partly constructed). Our informants included repair
workers, novice apprentices, repair workshop owners,
engineers graduated from universities of polytechnic
institutes, repair customers, and e-waste collectors and
vendors. This phase also included participating in a monthlong training program at a local mobile phone repairtraining center, and a 3-week apprenticeship at another
repair workshop under a senior repairer. Our third and final
phase of work included 70 semi-structured interviews with
repairers and clients at 10 different repair shops in Dhaka,
focusing on questions of practice, value, and collaboration.
All of our interviews and field interactions were conducted
in Bangla by two Bangladeshi researchers, and later
partially translated and transcribed into English. The
observations produced textual notes of several hundred
pages, and more than 1000 photographs in which repairers
demonstrated and documented key techniques. Most of our
investigations focused on the craft and expertise involved in
repairing, and the process of learning those.

The following sections detail our findings around

collaboration and learning as practiced in the mobile phone
repair worlds of Dhaka, Bangladesh. We begin with an
overview of the organization of work in this space,
describing the sites, resources, and networks through which
repair work is accomplished. We then draw on interviews
and field notes to construct biographical sketches of three
of our informants who, taken collectively, speak to the
range of backgrounds, experiences and trajectories to be
found among repair workers in our study. While later parts
of the paper will detail specific forms of skill, knowledge
and learning central to repair work, we argue that
biographical accounts of the sort offered here (and practiced
elsewhere by Linde [13], and Weiss [24]) are important to
understanding both immediate practices of learning and
skills transfer, and the broader processes of identity,
meaning and social organization that are central to repair as
practice and as a way of life.
Background: Basic Repair Activities in Dhaka

The mobile phone repair worlds of Dhaka include sites

ranging from high-end repair shops at general-purpose
shopping malls to clusters of independent repair operations
occupying a whole market. The high-end service centers
attached to multinational brand companies such as Nokia,
Siemens, and Maximus usually stand alone at large
shopping malls or business complexes, and perform
warranty fixes of phones still covered under service
contracts. While they usually recruit engineers graduating
from low-ranked universities or polytechnic institutes, they
tend to perform a simple and limited range of repairs,
usually limited to the detection of non-functioning or
problematic handsets and replacing them with new units
supplied by the company.
A much larger range and volume of repair takes place in the
independent and informal sectors. Some of these repairers
have their own mall or roadside shops, while others operate
alongside other repair operations in large informal markets
like the Gulistan Underground Market. While a subset of
these workers, like Akbar., draw on repair skills accrued
while abroad, most rely on knowledge and connections
earned through apprenticeship. Few if any have formal
academic training in technology, and their knowledge of
mobile phone operation in its theoretical or engineering
dimensions is often limited. In comparison to the brand
repairers, workers in the informal sector deal with a
remarkable range of problems, from broken screens and
non-responsive keypads to overheating and failed or spotty
network connections. They are also much more prone to
perform actual repair (as opposed to replacement) work,
going to often extraordinary and innovative lengths to
diagnose, correct, and work around the range of
breakdowns and problems encountered.
At the immediately material level, sites in the brand and
informal repair sectors are linked in turn through a third
community no less integral to the collaborative practice of

repair. Referred to locally as bhangari (from the Bangla

word bhanga, meaning broken) they circulate through the
city, visiting repair shops and buying or collecting broken
parts and devices (cases, motherboards, batteries, keypads,
etc.) at low or bulk prices. These materials are then sorted
and resold: some on to other repairers looking for parts or
raw materials, and others on to groups of Chinese recyclers,
who, it is rumored, export the materials for processing and
disposal in China.
These local repair networks are connected in turn to global
networks of consumption and repair, through both materials
and knowledge. The electronic devices used in Bangladesh
are almost entirely designed and manufactured in other
countries; as a result, forces and trends emanating from
abroad may drive local uses and repair of technology. The
tools and spare parts used by repairers (hot air guns,
soldering irons, various ICs, etc.) are mostly imported from
other countries, especially China. As we have seen, the
material by-products of breakdown and repair (including
forms of waste) may also be tied in to regional and global
recycling markets. Repair knowledge is similarly
networked at the global scale. Some local repairers receive
their training abroad, and then redistribute this knowledge
through teaching or training of apprentices. Some repairers
also draw on resources from the Internet, including the
repair boards of the GSM forum, to borrow and sometimes
contribute to global discussions around new techniques,
designs, and known flaws or issues (though barriers of
access, language, and literacy can sometimes limit this
connection). In the section that follows we explore the
stories of three representative figures, who occupy and
constitute the contemporary repair worlds of Dhaka.
Biographies
1) Akbar

Akbar is a 49-year-old senior repairer and trainer of mobile

repairing who works at Nahar Plaza, Dhaka. He was born in
Kishorganj, a town 5 hours drive away from Dhaka. He
developed a keen interest in electronic devices with the help
of his father who also loved electronics, and taught him the
basics of current, voltage, and power at a very young age.
While growing up, Akbar met an appliance technician who
taught him about motherboards and Integrated Circuits
(ICs). As a student in high school, Akbar reports always
producing small electrical devices for his science projects.
But he ended up going into accounting as this was
considered at the time a more promising job market. While
studying for his Masters degree in accounting, he took a
part-time job at company selling closed circuit cameras and
televisions. After graduating, he moved on to full-time
work in the telecom industry: first at a company providing
PABX phone connections to home and offices, and two
years later at a government office providing mobile network
support for military units. All these jobs helped Akbar learn
the basics of wired and wireless networks.

On the basis of this practical knowledge, Akbar applied to a

job at a Saudi Arabian telecoms company, and was the only
non-engineer selected from over 108 candidates. The
company sent Akbar to Thailand for a six month training
period, where he learned about the function and repair of
mobile phones from experienced Thai electrical engineers.
He then moved to Saudi Arabia for employment, working
with a multinational team that included engineers from
Bangladesh, India, Pakistan and China overseen by a team
of Korean supervisors. Akbar describes his learning during
this period as follows:
You should watch them (superiors) carefully, although
most of the time you will not understand anything. You just
see and remember. A good learner is he, who remembers
everything and tries to understand things later. Also you
need to have a good relationship with the superiors.
When they will see that you are trying to understand things,
they will help you. But if you pretend like you understand
everything, they will be pissed off. Again, if you show no
interest in learning, they will be disheartened.
After two years in Saudi Arabia, Akbar received news that
his father was sick and decided to return to Bangladesh.
Using funds and skills acquired during his time abroad, he
decided to start a mobile phone repair training and servicing
center at Dhaka. This was in 2001, when mobile phone
repair in Dhaka (and mobile phones themselves) were still
relatively scarce. Akbar built his training center around the
model he had seen in Thailand, structured around a balance
between theoretical and practical knowledge. He started
with two shops, but soon expanded his business by renting
another three. By the time of our fieldwork in 2013, Akbar
reports having graduated more than 7000 students, most of
whom went on to start repair business of their own in
Bangladesh (though others followed Akbars own earlier
path and went to Canada, the US, or Middle East to earn
more money). Beyond his training activities, Akbar also
maintains good relationships with parts and equipment
importers and e-waste collectors, and runs a small retail
shop for repair tools and accessories that is well-known
within the local repair community. He also runs his own
Bangla repair blog for local repairers who do not
understand English. While his training business has
declined somewhat since around 2008 an outcome he
blames on growing and sometimes dishonest competition in
the market and a growing flood of cheap and disposable
Chinese handsets that undercut the price and market for
repair Akbar continues to dream big. He tells us he hopes
to start his own mobile phone assembly factory in Dhaka in
the very near future.
2) Rupam

Rupam is a 35-year-old mobile phone repairer working at

Eastern Plus Market, Dhaka. He was born and brought up in
a lower middle-class family in Jinjira, a suburb of Dhaka.
He was sent to school, but never considered himself a good
fit there and left after grade 9 to start work in an ornaments

shop. The job was easy, but low paying, and faced with a
family financial crisis, he resolved to find a better paying
job. Around the same time he tells us, Rupam developed a
crush on a girl. To impress her, and to take advantage of the
rapid growth in mobile phone use then underway in the
country, Rupam decided to become a repair technician.
This proved difficult to do without prior skills, experiences
or connections. So, he started working with one of his
friends who worked as an appliance repairman. Rupams
task was to help his friend by handing him tools and taking
care of the shop. As he worked, his friend would explain to
him what he was doing, and thus he got his first lessons on
repairing. When their relationship fell apart, Rupam started
spending his time at the repair stalls of the Stadium Market,
where he would stand and silently observe the repairmen at
work. Eventually, Rupam explained,
If you stand at the same place for two days, people will
look at you. At some point somebody will come to you and
ask, Who are you? What do you want? That also
happened to me.
Rupam became friendly with the owner of one of the repair
shops. When the shop-owner learned that Rupam was fluent
in the dialect of old Dhaka, which many customers spoke
but the shopkeeper struggled in, he was given a job greeting
customers, receiving orders, and explaining the reported
problems to technicians at the shop. While working in this
way, Rupam would watch how the repairers worked. One
day Rupam brought a broken mobile phone of one of his
friends to the shop, but found no repairers available to fix it.
He decided to attempt a technique called servicing which
he had seen the repairers perform many times before. The
technique worked, and soon after the other repairers started
to take his help in their work. Rupam worked there for 2
years. During that time, he heard about a senior repairer
who had returned from abroad and was offering a training
course on ICs. He took the course and became familiar with
the names and functions of the ICs.
Soon after, Rupam became friends with an expert repairer
in Stadium Market. They decided to open a shop together at
a nearby location. Rupam worked at the new workshop
while his friend continued on at the Stadium market. Since
Rupams grasp on repair techniques was still limited, he
would often go to his friend for help, who would fix the
device but not show Rupam how to do it. Rupam thought
this had been the way his friend would try to maintain his
power over the business. However, Rupam started to learn
newer things on his own by trial-and-error. He also started
drawing on help from Internet. He recalled the first iPhone
he unlocked in this way:
A client came to me to get his iPhone unlocked. I did not
know how to do that. I went to my friend. He told me to
leave the phone to him and he would fix that later. But the
client was waiting at my shop and I had no way to leave the
phone to my friend. So, I searched on Internet and got a
video on doing this. I followed that and it worked. The

client was happy. I also learnt how to unlock the phone. It

was not a difficult task. But my friend did not want me to
learn that.
A few days later, Rupam had a quarrel with his friend and
he left the shop. Then he started his own business at Eastern
Plus market at Shantinagar. He has now been working at
that shop for more than 5 years, and has continued to
develop his skills, techniques, and customer base. He has
also continued to build connections with other repairers, ewaste collectors, IC importers, and others in the repair
world. He reports being happy with his work, and enjoys
the connections and respect it offers him in his local circles.
3) Jahangir

Jahangir is a 38 years old bhangari business owner at

Elephant Road, Dhaka. He was born and raised in a middleclass farmer family in Sylhet, a 4-hour drive from the
capital. He left school after the 4th grade due to lack of
interest and his familys financial constraints. When he was
12 years old, one of his maternal uncles, who ran a bhangari
business in Dhaka, visited Sylhet and invited Jahangir to
join him in the capital. Jahangir decided to take the offer,
both for the financial reward it offered and for the
possibility of migrating abroad from Dhaka it might offer.
From his uncle, Jahangir learned the basics of the bhangari
business, traveling the streets of Dhaka and buying
discarded objects from households, which he would then
resell at a higher price. Mr. learned how to approach
unknown people, inquire about discarded objects, and
bargain effectively on the price. He said,
People often cannot remember that they have discarded
objects at their house. So, you have to ask like, Do you
have old newspapers, broken mobile phones, or torn bags?
Do your kids have broken toys? Then they will be able to
remember. You have to be intelligent enough to guess what
sort of discarded objects a person can have at his place.
He also learned how to categorize the various objects
collected by material type and market values. He realized
that electronic devices were more profitable than other
businesses, and that offices were better places to find
electronic wastes than households. So, he stopped following
his uncle and started going to the offices. Jahangirs uncle
would sell the collected objects to a wholesale bhangari.
Mr. H noticed how his uncle would maintain good
relationship with that wholesale bhangari, and he also got
himself introduced to that person. Later that guy became the
main client of Jahangir.
It was not easy for a young boy to go to big offices and buy
discarded objects. He spent a lot of time observing and
learning how people dressed and talked around the offices
he visited. He started dressing and talking in the same way,
and found that it helped him source materials more
effectively. He also learned how to compete for tenders
when Government offices would sell their old electronic
devices at steep discount. With his new profit he was soon

able to rent his own shop and recruit younger bhangaris to

collect discarded objects on his behalf, sending them to
repair shops, government offices, universities, and
households.

livelihood (though they are all these things as well). Rather

they should be understood as historical processes of sharing
and
enculturation
through
which
collaboration,
sustainability, and innovation are accomplished.

He explained the core knowledge of his business to us as

follows: The challenging part is to get the broken things
that hold value. Selling is never a problem. There are
always enough parties to sell things. Determining the
worth and condition of a discarded technical object is a
relatively challenging task. Between two discarded
motherboards, the one with more metal and less damaged
ICs often holds more value on local markets. Over time,
Jahangir had learned to estimate the value of a discarded
motherboard just by looking at it and touching the parts. He
developed this skill through his relationships with repairers
and recycling wholesalers, and through years of experience
in the collection and resale of broken electronics.

TYPES OF KNOWLEDGE AND LEARNING PROCESSES:

Summary

As these stories suggest, entry into the repair professions

follows a wide range of paths and motivations, from the
familial, affective and personal to the economic and
instrumental. But in all cases learning is central. Knowledge
and skill play a central role in shaping the prospects for
employment, stability, income, status, and future work
trajectories for all members of the broader repair worlds in
Dhaka. We have seen how people train themselves in
different ways and how local and international networks,
corresponding community, contemporary society, and
communication technology play roles there. Learning
remains an essential companion throughout the career of
these people to keep them updated with changing
technologies, skill sets, and the social and economic orders
around repair. However, several practical challenges
coming from issues including professional jealousy, lack of
jobs, and illiteracy often limit their access to the
community, and hamper the process of learning and
developing skills.
Another important takeaway from these biographical
sketches is the understanding of the reproduction process of
repairer communities. Through the practices described
above, repairers not only learn to support their existence in
the community, but also, at some point, start sharing their
knowledge with others through social practices that
contribute to sustaining and reproducing the broader repair
community. This contribution is collaborative both in the
form of its immediate practice (apprenticeship), and in its
broader contribution to sustaining and reproducing the
community of repair that is essential to computational
practice and infrastructure in Bangladesh. Beyond the
transfer of necessary knowledge and skill, learning to fix is
also about learning to be a fixer, acquitting oneself well and
responsibly within both an immediate professional and
wider social community. Learning processes in the repair
community should therefore not be understood just as
access tool, survival mechanism, or road to a sustainable

As the biographies in the previous section make clear,

knowledge and learning are central to practices of
technology repair, reuse, and recycling in Dhaka today. But
the knowledge and learning required come in several
different forms or varieties, which competent repairers must
master and blend. This includes explicit or conceptual
knowledge of phones in their technical or schematic sense
(knowledge of circuitry, function, etc.). But it also includes
forms of tacit, craft, and social knowledge no less central to
entry and long-term success in the field. The following
subsections explore each of these knowledge forms in turn,
along with the core processes and mechanisms of learning
by which they are shared and transmitted within the wider
repair community.
Explicit Knowledge

By explicit knowledge we refer to the largely conceptual

knowledge that lies behind the effective repair of broken
devices. This includes information on the different parts of
the mobile phone, how theyre related to each other, the
functional connections (including discrete paths in the
circuitry) that link them into an operational hole, and the
larger schematic similarities and differences that unite or
separate different phone models. Repairers in Dhaka gain
this knowledge in different ways, mostly from their mentors
during their days of apprenticeship, and then from sources
like Internet, manuals, and friends.
Akbars training center offers short course on repairing
mobile hardware, which includes both the theory classes
and practical sessions (the terms, theory class and
practical sessions, are adopted from Akbars training
center). One of our team members took this one-month long
course and learnt about the structure of this explicit
knowledge from the theory classes. Most of the instructions
would follow a particular format. Akbar would introduce
the students to a problem that could occur in a mobile
phone, for example: the battery is not charging, or no
network connection. Then he would tell which IC on the
motherboard would be responsible for that. Then he would
tell what action was required to fix or replace that IC. So,
essentially the structure of the information had two basic
mappings in it: i) mapping a problem with a IC, and ii)
mapping a IC with the solution. The following field note
taken at Akbars class should make this structure clear.
Akbar is explaining the repair works involved with
Polaroid capacitor today. He says,
Many times you will get complaints from the customers
that their phone is not holding charge. If you put the phone
on charge that shows the phone being charged. You can
even use the phone as long as you want while the phone is

charging. But as soon as you remove the phone from the

charger, all the charge goes away. This is a very common
problem, and an IC called polaroid capacitor is
responsible for this. The responsibility of this capacitor is
to hold the charge. When this capacitor is broken, the
phone fails to hold charges. The only solution is you have to
replace this capacitor with another one collected from
another motherboard. But, first of all, you have to identify
which one is the Polaroid capacitor on the motherboard.
Polaroid capacitor is the yellow IC on the motherboard,
with a brown patch. You can easily identify it, because it is
slightly larger than the other ones. The side that has the
brown patch is the positive side.
Akbar would try to make the information as general as
possible, so that it could work for mobile phones of any
manufacturer. If there were exceptions for any particular
manufacturer, he would mention these too. While
introducing the function of the ICs, Akbar often used
metaphors. For example, he used the metaphors of a river
and tube-well to explain the difference between the
functions of a diode and a transistor.
Besides these sorts of lessons from senior repairers, repair
workers in Dhaka often take help from online resources.
The Internet has spread out the country recently, and
repairers often browse Internet through their mobile phones
or in some cases through some very old refurbished
computer that they collected only for this purpose. Manuals
are another source of explicit knowledge for Bangladeshi
repair workers, valued in particular for their diagrams of
motherboards and crucial information on the location and
function of ICs.
Tacit Knowledge

In our fieldwork with repair communities, we observed the

necessity of tacit knowledge for repairing. In most of the
cases, they learn this through observation and imitation
during their apprenticeship. They learn things first by
observing their superiors, and then by imitating them. They
also learn through their experience while making an attempt
to perform the task. The following field note taken at
Akbars practical session can give a better insight of this
learning process.
Akbar took a motherboard in his hand and started to
demonstrate us how to remove an IC from the motherboard
and how to attach an IC on a desired place on the board.
Akbar was holding the hot air gun in one hand and forceps
on the other. The forceps were gripping the IC on the board
that he was trying to remove, and the hot air gun was
blowing hot air onto that. In a few seconds the paste started
melting and the IC became lose. Akbar pulled the IC off the
board using the forceps. Then he showed us how to place
the IC back to its original position. Next, he asked us to
remove and then place back a small resistor on a
motherboard. This was the first time I was holding a hot air
gun. I was just trying to follow what he had just done.
Placing that small resistor at its original position needed

very sound attention. I tried to get eyes closer to the IC so

that I could see it better, but it was not possible because of
the hot air gun blowing hot air. My hand was trembling,
and I was struggling to grip the chip with my forceps. But
after a number of failed attempts, finally I succeeded

Figure 1: An expert repairer is practicing at his desk in

Gulistan Underground Market
The above example demonstrates how observation and
imitation contribute to the acquisition of hard-to-describe
skills. We bring in another example here to clarify the
nature of tacit knowledge and the role of the master in its
learning process. The following example is taken from a
field note written during the apprenticeship of one of our
team members at Rupams workshop.
Then I started brushing the board. Rupam was closely
looking at my work. He said, You are being too soft while
brushing. This will not remove the dirt. You have to be a
little harder, but not too much. If you are too hard, you will
brush away the ICs. Just be as hard so that the black spots
go away. You got it?
I realized that he was actually right. I started to be soft
while brushing on the ICs, but harder while brushing on the
other areas on the motherboard. Then I started looking for
black spots and trying to remove those. The whole point of
servicing now became clear to me.

Figure 2: A repairer at Eastern Plus Market fixing a

motherboard.
This example demonstrates that the appropriate level of
hardness is hard to convey through explicit instructions;
rather an apprentice has to learn that through his

experience. This sort of tacit knowledge is essential in

repair work. The repairers get the inspirations for gaining
these fine expertises from their passion for improving the
quality of their work. Getting the task done quickly and
making the repaired device visually attractive often bear the
sign of skills of the repairers. Rupam once said,

like this to an old friend. You have to value people in order

to value your business.

The way I fix the mobile phone, it is hard to tell for others
that the phone is repaired. But sometimes I get many
phones where the repairers made the board look really bad,
and you wont want to work on that. Some people just do not
care to learn good work.
Social Knowledge

Repair knowledge is not limited to technicalities. Rather, a

repairer is educated on a lot of rules inside and outside the
workshop: when to open and close the shop, when to take
lunch, how to talk with the clients, when to turn the lights
on, and when to clean the floor these all are the examples
of the social knowledge involved in the learning process of
a repairer. Essentially, these teach an apprentice the values
the repairer community holds.
A big part of knowledge in repairing is comprised of the
lessons on how to deal with the customers, and fix the
service charge. The following texts taken from the field
note written during the apprenticeship at Rupams
workshop demonstrate how social knowledge becomes a
part of learning in repair works.
An old guy just came to our shop. He was looking for a
battery that we do not sell. Rupam gently suggested that
guy to get the appropriate shop to get the battery. But that
old guy asked Rupam to get the battery for him. To my
surprise, Mr. left the shop and got the battery for him. As
the guy left, I asked Rupam why he had done so. Rupam
explained, He is my old customer. I have to do this for
him. This is how you keep your relationship with others in
the community.
The practice of situated learning is also prevalent among
bhangaris. Bhangaris need to know the social skills to keep
good relationships both with the expert repairers and their
customers. Jahangir, while explaining the reason behind his
success in bhangari profession, said to us:
I collect the e-wastes from the secretariat and cabinets.
Those are the places where you will find a lot of things. But
not everybody can go and get things there. You have to
dress properly and talk smartly. I have a good relationship
with all the people who work there. So, I make a good
business at that place.
These social skills are learnt by bhangaris while working
with their seniors. For example, Jahangirs younger cousin
recently started working with him. He was dealing with a
customer who had been an old friend of Jahangir. At one
point, his cousin said something inappropriate to that
person. Jahangir immediately reacted and apologized to that
person. He turned to his cousin and said, You do not talk

Figure 3: Repairers at Gulistan Underground Market.

DISCUSSION

The above discussion reveals the nature of explicit, tacit,

and social knowledges at play in the repair worlds of
Dhaka, Bangladesh. We have described how explicit
knowledge arrives at, flows through, and is sustained
among repairers through people, manuals, and Internet
resources, keeping repairers up to date with global changes
in technology. We have pointed to how repairers build tacit
knowledge through observation, imitation, and repetition.
We have explored the craft-based nature of repair
knowledges, and how these skills were acquired through
observation, repetition, and passionate engagement. We
have also shown how apprentices in repair communities are
converted to skilled repairers through their assimilation and
absorption of core social practices and values. Beyond the
simple transfer or acquisition of skills and knowledges, it is
this essential component that accounts for the production
and reproduction of vocational identities and wider
communities of practice in the repair worlds of
contemporary Dhaka.
The act of repairing itself is not very easy either; rather it is
painstaking and laborious. Many repairers in the Gulistan
Underground Market reported that mobile phone repair
work involves very fine eye-work, and a person cannot
continue in hardware repair for more than 20 years without
significant damage to sight and body. Working conditions
in the market are hot, crowded, and noisy, with
questionable air quality tied to the frequent use of glues,
solder, and solvents coupled with generally poor
ventilation. Economic livelihoods in the profession may
also be precarious, including in recent years as new and
cheaper handsets and growing competition in the sector
drive down the prices that can be charged for repair. At the
same time, repair workers face the challenge of
continuously updating themselves to keep track of newer
technologies and changing environments. For this reason,
learning remains a central and ongoing challenge for repair
workers, who must continuously build their skills and
networks throughout their career. In a very real sense,
learning is a survival skill.

These findings carry important implications for ICTD

scholarship. At the most immediate level, our study draws
attention to the struggle that repairers confront due to the
design and documentation strategies of global IT producers.
New handsets are often designed not to be taken apart and
fixed, guarded by a combination of glues, non-standardized
fasteners (proprietary screws, etc.), and design decisions
that locate core elements in hard-to-access portions of the
device. Documentation of new models is often spotty or
hidden (outside of the proprietary channels of the brand
shops) raising the time and difficulty involved in engaging
new models. Improved accessibility, repair documentation,
and other design-for-repair strategies could substantially
change this situation (and would align naturally with the
sustainability and emerging market commitments of global
IT manufacturers).
Second, in many developing countries repair (and the forms
of tacit, and social knowledge that support it) is often
neglected and devalued in formal engineering education,
routinely ignored in higher end universities and only partly
addressed by the underfunded and low status polytechnic
schools. National education policies may need to consider
paying more attention to these areas to build stronger
support infrastructures for technology use and consumption.
Third, transnational collaboration and information sharing
through online sources is often less accessible for lowliterate
repairers.
Automatic
translation,
video
demonstration, and Bangla-lanaguage forums for local
repairers might enhance learning, collaboration and wider
access to knowledge within this segment of the local repair
community.
Beyond this immediate level, the forms of knowledge,
learning, and collaboration described in the paper reveal
two parallel yet opposite realities with important
implications for how we think about, design for, and
practice collaboration and learning in transnational and
third-world computing environments. On one hand, there is
a structured network of knowledge present in Dhaka that
connects local repair knowledges and practices with global
hubs of learning and activity. People trained abroad bring
their knowledge to the country and disseminate through
teaching. The contents and methods of their teaching often
reflect the foreign mechanisms of repair education. Also,
the repairers often gather knowledge from Internet, and
books written abroad. Knowledge that flows through this
network from global knowledge hubs to the local repair
community, attempts to align the local repair knowledge
and practices with their global counterparts. This process
points towards the ongoing role of global sources in
shaping and supporting (or failing to) practices of technical
learning and collaboration in developing countries, and the
transnational politics of access, support, and dependency
that may be central and often ignored features of ICTD [6].
On the other hand, local trainers and senior repairers make
a significant effort to situate and adapt foreign knowledge

to the exigencies of the Bangladeshi context. They invent

metaphors and find alternative techniques to address the
high cost of tools, unavailability of materials, associated
monetary risks, and other social, economic, and
infrastructural factors related to repair work. Through long
processes of apprenticeship, repairers learn from existing
practitioners how to translate and adapt foreign knowledge
to local needs, values, and contexts. This part of the story
reveals the collaborative effort to protect and enhance local
cultures of technical skill and knowledge. In this way, topdown forces of global knowledge and bottom-up forces of
localization combine to continuously reshape and update
the terrain of repair learning and collaboration in Dhaka,
and in many other developing sites. Now, the interplay
between these two ongoing forces can be explained through
postcolonial computing terms [5], and can be seen as a local
resistance against foreign intrusion of technology,
knowledge, and culture. This tension essentially calls for
local IT and foreign policies that should consider the
balance between these two sides based on national interest.
This tension connects ICTD interests with the international
politics through repairing works, and demonstrates how
national foreign policies and transnational collaboration are
instrumental in sustainable development through
technology an issue yet to be developed in ICTD
research.
The biographies and descriptions of collaboration and
learning above make it clear that tacit knowledge and
master-apprentice relations sit at the center of learning to
fix. These in turn depend on wider social relations of trust,
standing, and respect. To make the learning process
successful, an expert repairer is required to maintain a
social standing and image that commands the respect of his
apprentices. Similarly, apprentices must uphold a social
standard and image to maintain the support of their masters.
As described by both Polanyi [15,16] and Lave and Wenger
[12], the transmission of tacit and social knowledges
through observation, imitation and enculturation depends
on trust, care, and respect. Hence, persons social skills are
essential in developing the technical skills for repairing.
Furthermore, our study also demonstrated internal
practices and valuations of quality of work that went well
beyond what was required to satisfy or indeed would even
be visible or recognizable to customers. Repairers took
great pride in the quality of their work and recognized (or
criticized) such qualities in others, features that align well
with Sennets [22] analysis of craftsmanship. Important
parts of the apprenticeship and learning process were in fact
dedicated to inculcating and reproducing this value in
others. The social value of repairers is often threatened by
the competition with engineers, and many repairers left
their job because of the lack of proper recognition. By
strengthening the standing and reputational economy of
repair work, recognition of the craft-based quality and
values of repair work could therefore improve sustainability
and reinforce dedications to quality and reliability in repair

practice. These all suggest ICTD research to focus on

building sustainable local social infrastructure to make
technology sustainable.
Another important takeaway of this study is that the social
landscape of repair is not purely collaborative and smooth,
and
this
often
inspires
technology
invention.
Apprenticeships are not easy to get and never free of cost in
Dhaka. Rupams story revealed how difficult it was to get a
job in a workshop even as an apprentice. Apprenticeship in
repair shops is often organized through familial lines, or
financial contracts. Also, apprenticeship is mostly practiced
among the men, resulting into almost no presence of
women in repairing market. Thus the flow of knowledge is
sometimes strategic, transactional, and selective. Being cut
off from such flows can push repairers towards
experimentation, innovation and discovery, or to
frustration, stasis, and decline. Even within the confines of
the apprentice relation, knowledge may be jealousy
guarded. In some cases, apprentices are held in extended
positions of dependency and exploitation, performing large
volumes of work at low levels of compensation. The
repairers need to develop different social skills to survive in
these challenging conditions. These examples demonstrate
how inventions occur in challenged conditions of repair
world in developing countries.
CONCLUSION

In this paper, we have shown the central and sometimes

contested roles of knowledge, collaboration and learning in
the worlds of mobile phone repair in Dhaka, Bangladesh.
This study demonstrates the creative, social, and
challenging nature of repairing and positions this as a
prospective source of technical innovation in future. At the
same time, this paper provides with important design and
policy implications that can improve the repairing
infrastructure of a developing country. In a broader sense,
this paper calls for more attention to strengthen the repair
and maintenance side of developing countries that have
been significantly understudied within ICTD scholarship to
date.
REFERENCES

1.
2.
3.
4.
5.

Garfinkel,
H.
Garfinkel,
H.,
Studies
in
Ethnomethodology. Englewood Cliffs NJ: Prentice
Hall, 1967.
Goffman, E. The Interaction Order. American.
Sociological Review 48, (1983), 117.
Graham, S. and Thrift, N. Out of order understanding
repair and maintenance. Theory, Culture & Society 24,
3 (2007), 125.
Houston, L. Inventive Infrastructure: An Exploration
of Mobile Phone. PhD Thesis, Lancaster University,
2014.
Irani, L., Vertesi, J., Dourish, P., Philip, K., and
Grinter, R.E. Postcolonial computing: a lens on design
and development. In Proc. CHI10, ACM, 13111320.

6.

7.

8.
9.

10.

11.
12.
13.
14.
15.
16.
17.

18.

19.
20.
21.
22.
23.
24.

Jackson, S.J. Rethinking Repair. In T. Gillespie, P.

Boczkowski and K. Foot, eds., Media Meets
Technology: Essays on Communication, Materiality
and Society. MIT Press, 2013.
Jackson, S.J., Ahmed, S.I., and Rifat, M.R. Learning,
innovation, and sustainability among mobile phone
repairers in Dhaka, Bangladesh. In Proc. DIS14,
ACM (2014), 905914.
Jackson, S.J. and Kang, L. Breakdown, Obsolescence,
and Reuse: HCI and the Art of Repair. In Proc.
CHI14, ACM (2014), 449458.
Jackson, S.J., Pompe, A., and Krieshok, G. Things fall
apart: maintenance, repair, and technology for
education initiatives in rural Namibia. In Proc.
iConference11, ACM (2012), 107116.
Jackson, S.J., Pompe, A., and Krieshok, G. Repair
worlds: maintenance, repair, and ICT for development
in rural Namibia. In Proc. CSCW12, ACM (2012),
107116.
Lave, J. Apprenticeship in critical ethnographic
practice. University of Chicago Press, 2011.
Lave, J. and Wenger, E. Situated learning: Legitimate
peripheral participation. Cambridge University Press,
1991.
Linde, C. Life stories: The creation of coherence.
Oxford University Press, 1993.
Orr, J.E. Talking about machines: An ethnography of a
modern job. Cornell University Press, 1996.
Polanyi, M. Personal knowledge: Towards a postcritical philosophy. Harper Torch Books, 1964.
Polanyi, M. The Tacit Dimension. Doubleday &
Company Inc., 1966.
Rangaswamy, N. and Nair, S. The PC in an Indian
Urban Slum: Enterprise and Entrepreneurship in
ICT4D 2.0. Information technology for Development
18, 2 (2012), 163180.
Rosner, D.K. Making Citizens, Reassembling Devices:
On Gender and the Development of Contemporary
Public Sites of Repair in Northern California. Public
Culture 26, 1 72 (2014), 5177.
Rosner, D.K. and Ames, M. Designing for repair?:
infrastructures and materialities of breakdown. In
Proc. CSCW14, ACM (2014), 319331.
Rosner, D.K. and Taylor, A.S. Antiquarian answers:
book restoration as a resource for design. In Proc.
CHI11, ACM (2011), 26652668.
Sacks, H. Lectures on Conversation. Blackwell
Publishing, 1995.
Sennet, R. The craftsman. Yale university Press, 2008.
Suchman, L. Human-machine reconfigurations: Plans
and situated actions. Cambridge University Press,
2007.
Weiss, R.S. Learning from strangers: The art and
method of qualitative interview studies. Simon and
Schuster, 1995.