Content uploaded by Allison Druin
Author content
All content in this area was uploaded by Allison Druin
Content may be subject to copyright.
From New Zealand to Mongolia:
Co-Designing and Deploying a Digital Library for the World’s Children**
Allison Druin, Benjamin B. Bederson, Anne Rose, Ann Weeks
University of Maryland, Human-Computer Interaction Lab
Contact: allisond@umiacs.umd.edu, +1 301 405 7406
Abstract
The Internet has led to an explosion of users throughout the world. Low-cost computing options
are now emerging for developing countries that are changing the world’s educational landscape.
Given these conditions, there is a critical need to understand the obstacles and opportunities in
designing and deploying technologies for children worldwide. This paper discusses seven years
of strategies and methods learned in co-designing and deploying the International Children’s
Digital Library (www.childrenslibrary.org) with children in multiple countries. Our experience
with iterative international co-design, and developing world deployment shows that acquiring
site-specific knowledge is critical to adapting methods needed to be successful. In the case of
co-design, a combination of face-to-face and email collaboration is important to building on-
going partnership relationships. With deployment activities, it is important to be prepared for the
unexpected – managing complex technologies in rural settings is very difficult. Therefore, the
more site-specific knowledge that can be acquired the more likely there will be a successful
outcome.
Keywords:
Children, Co-Design, Design Methods, Digital Libraries, Educational Technology
**This article in currently “In Press” and will be published in a special issue of:
Children, Youth and Environments (
http://www.colorado.edu/journals/cye/):
Children in Technological Environments: Interaction, Development, and Design, Editors:
N.G. Freier & P. H. Kahn
The Need
“Oh, I like books from Australia, the UK, the states, heaps of places. My mum says I’m just a
world traveler with my books. I guess it’s easier than getting in a plane…”
(New Zealand girl, age 8, 2003)
These are the words of just one of our team’s partners in designing the International Children’s
Digital Library (www.childrenslibrary.org). Since 2001, an interdisciplinary, international, and
intergenerational team of adults and children has been designing a digital library for the world’s
children. Led by researchers at the University of Maryland (USA), co-designing has taken place
with partners in New Zealand, Honduras, Germany, Mongolia and the U.S. These collaborative
research experiences have led to unique online tools for searching (Figures 1a, 1b) and reading
fully digitized children’s books from around the world (Figures 2a, 2b). There are thousands of
books in the library (all completely free), in 48 languages and include picture books, chapter
books, fiction and non-fiction titles, historic and contemporary materials. These books were all
previously published in print and have since been digitized either by the contributor (e.g.,
National Libraries, publishers, authors) or by our own team (Druin et al. 2003; Druin 2005;
Druin, Weeks et al. 2007; Hutchinson et al. 2007; Massey et al. 2007).
Figure1a:AnexampleofhowtosearchforabookintheInternationalChildren’sDigitalLibrary
(ICDL):usingalocation‐basedsearch
Figure1b:AnexampleofhowtosearchforabookintheInternationalChildren’sDigitalLibrary
(ICDL):usinggraphicalsearchiconssurroundingtheresultsinthecenter
Figure2a:AnexampleofhowtoreadabookintheICDL:usingagridofthumbnailstoaccess
eachpageinabook
Figure2b:AnexampleofhowtoreadabookintheICDL:usingaspirallayoutofthebook’s
pages
With users from over 150 countries, the goal of this digital library work is to excite and inspire
the world's children, by making the best in children's literature available online, to help young
people (ages 3-13) understand the value of tolerance and respect for diverse cultures, languages
and ideas. Since the inception of the ICDL, a guiding principle has been to design the library’s
tools and collection shaped by an understanding of children as technology users, book readers,
and library visitors (Druin, Weeks et al. 2007). This is motivated not only by our own past
experience in partnering with children in the design of technology intended for children’s use –
but also by our broader concern for recognizing the right of children to participate in society.
Much has been said about giving children “meaningful participation” in the societies that they
are a part of (Skivenes and Strandbu 2006; Lambert 2007), and the use of technology for
education certainly falls into the realm of societal activities where children have a right to
participate.
While much has been written by our team and others about the use of digital libraries with
children, far less has been published on the methods needed to accomplish co-design or
participatory design with children and adults in multiple countries. Although the co-design
movement began with trade unions over 25 years ago in the Scandinavian countries (Bjerknes et
al. 1987; Greenbaum & Kyng 1991; Schuler & Namioka 1994), and the International Conference
on Interaction Design and Children regularly has contributors from Europe, Asia, and North
America, few papers if any, describe the methods needed to work with partners from multiple
countries. Yet, the need for these methods is particularly critical as the Internet has led to an
explosion of international users. In addition, as organizations such as One Laptop Per Child
(www.laptop.org) and Intel’s World Ahead Program
(
www.intel.com/intel/worldahead/classmatepc) have begun to bring low-cost computing to
children in developing countries, there is also a critical need to understand how to deploy these
technologies when electricity, a stable network, and computing expertise is challenging to find.
In addition, as is widely recognized, working with children to understand their needs is not easy
– their developing language and communication abilities often requires creative means to
understanding them (Skivenes and Strandbu 2006).
Therefore, this paper will discuss the real-world obstacles and opportunities for designing and
deploying online technologies for children worldwide. Strategies, methods, and lessons learned
will be presented in this case study describing our research over a 7-year period, from classrooms
in urban New Zealand, to rural Mongolia. This paper will begin by describing how our design
methods evolved and were adapted in distributed countries. We conclude with a discussion of
our deployment experiences and lessons learned for future researchers and developers.
Evolving Co-Design Methods for Children
The Lessons of Social Democracy
“The goal was to ‘give the end users a voice’ in design and development of computer support in
work places, thus enhancing the quality of the resulting system” (Bødker et al. 2000, 1). These
were the ideas and values of the Scandinavian researchers who pioneered the UTOPIA project.
This was a unique effort over 25 years ago to create better technologies for newspaper text and
image pre-press production by trying to understand what workers currently did and what they
wanted in the future (Bjerknes, et al. 1987; Bødker et al. 2000; Greenbaum & Kyng 1991;
Schuler & Namioka 1994).
In the early 1980s, the trade unions in Sweden were an ever-growing force throughout the
country. Laws and agreements were established to support strong worker representation and co-
determination in the work environment (Bjerknes, et al. 1987). Industrial “democracy” and
economic “democracy” were goals that seemed within reach. This was the political and social
context that researchers were working within. As UTOPIA researchers have explained, “(we)
did this out of a political commitment to the idea of democracy at work” (Bødker et al. 2000, 9)
While the central goal of this ambitious collaboration was to create better technologies in the
workplace, its legacy has become the establishment of what has come to be called co-operative
design, participatory design, or co-design methods that created a respect for the technology user
within the design process (Beyer & Holtzblatt 1999; Taxén 2003). The idea of moving beyond
“usability” to the notion of collaborative design required moving beyond traditional “testing”
techniques to methods that could give all participants a chance to be heard in the design process.
Such methods as “low-tech prototyping” were used to give equal footing to all partners in the
process by sketching with art supplies and materials which most people know how to use. When
these principles and processes are extended to children as co-designers, it means that regardless
of age or experience, young people can be empowered to bring their points of view to the design
of new technologies which can impact future learning and living.
The Lessons of Children
A decade after co-design empowered trade unions, our team began partnering with children in
New Mexico (Druin 1997) and later in Maryland (Druin 1999). However, the methods used for
adults with cooperative design needed to be adapted and further specified to apply to children.
Over many years, we created a process of working with children called Cooperative Inquiry
(Druin 1999; Druin 2002). It did not happen suddenly one day, but rather, these concepts and
understandings evolved slowly over the years with numerous research and development
experiences with children. At the University of Maryland’s Human-Computer Interaction Lab,
we have used this process to create new storytelling worlds (e.g., Benford et al. 2000;
Montemayor et al. 2004); travelled to new outdoor places with mobile technologies (Chipman et
al. 2006); taken new digital library journeys (Druin 2005; Hutchinson et al. 2006; Hutchinson et
al. 2005; Druin, Weeks et al. 2007); and built bridges between children from different cultures
(Komlodi et al. 2008; Komlodi et al. 2007). Others have also built on and validated this approach
(Large et al. 2004).
We continue to work today with children ages 7-11, who come to the lab after school twice a
week and for two intensive weeks, six hours-a-day over the summer. We have found that a
group of seven children at this age range is ideal for a design partnership because at these ages,
children can already be quite reflective in their design ideas, but not old enough to have so many
preconceived ideas that it is hard to consider paradigm shifts (Druin 2002; Druin 2005).
Children are selected for the team based on their interest in technology, their ability to
communicate, and their desire to collaborate with their peers and adults. Children have worked
with the team as long as five years and as short as one year, and since 1998, over 40 children
have been involved as design partners in Maryland. The children join researchers from computer
science, education, psychology, art, and robotics. Together this interdisciplinary and
intergenerational team brainstorms, sets project directions, tests new ideas, and implements
technologies.
We have found that different research methods are needed for the team depending on the stage of
the design process (Druin 2005). For example, in the early stages when ideas need to be
explored, the team of both adults and children might observe and/or conduct interviews with
other children in local public libraries, museums, or after school programs. Not only do adults
ask questions and observe behaviors but children do as well. Example issues we have examined
with this technique include how children choose books and how they search on the web (Druin
2005). Note-taking procedures and interview styles are by necessity different for children and
adults (Druin 1999). Children may draw more and write less than adults, but together ideas and
perspectives are found that might not otherwise if only adults did the research.
Another method used extensively by our team is to critique other systems by writing our
thoughts on Post-It or “sticky” notes. We write one idea per note, including what we like, what
we don’t like, or how we think the design might be changed (Figure 3). Then a frequency
analysis is done and similar ideas are grouped together. From these results, new design
directions are encouraged (Druin 2005). This method was adapted from many of the
participatory design methods described in (Schuler & Namioka 1994).
Figure3:Critiquingtechnologiesbywritingonpost‐itnotessomethingthattheteammember
likesordislikesorisadesignsuggestion.Thentheseideasaregroupedonawhiteboardfor
analysis
The oldest and most commonly used method adapted from cooperative design (e.g., Bjerknes et
al. 1987; Greenbaum & Kyng 1991) is what the team now calls “low-tech prototyping,” or “bags
of stuff.” This is where art supplies are used with groups of design partners to create models or
sketches of new technologies (Figure 4). Once these models are created, they are presented to
the full team while one person writes on the white board “the big ideas” (Figure 5a, 5b). These
ideas are noted on the board due to their originality or importance to the group. (Table 1 shows
how these co-design methods changed the ICDL).
Figure4:Sketchingnewideasfortechnologywithartsupplies
Figure5a:Presentingtheirlow‐techprototypesfornewtechnology
Figure5b:TheBigIdeasfrompresentinglow‐techprototypes.
Table1:ExamplesofhowtheICDLevolvedbyusingco‐designmethods
DesignMethod InitialIdeas ImplementedFeatureinICDL
Childreninterviewingother
children(conducting
fieldwork)
“Ilookforbooksbyhowthey
makemefeel.Iliketoread
booksthatmakemehappyor
scared.”
(girl,age8)
SearchCategory:Feelings
Using“stickynotes”towrite
onethingyoulikeaboutthe
softwareandonethingyou
don’tlike.
“Idon’tlikethesoftwareto
findthingswith.What
happensifIwanttofindthe
redbooksIalwaysliketo
read?Ican’tjustaskforthe
redbooks.”
(girl,age7)
SearchCategory:Colors
Fullgroupdiscussionsession “Idon’tthinkanyofusare
goingtoagree.Can’twejust
dowhatwewant?”
(boy,age9)
CustomizableBookReader
Colors
Creating“low‐tech
prototypes”usingartsupplies
tosketchnewideasfor
technology.
“Thisishowyouwouldread
thebook.Almostlikebeing
onamerry‐go‐round.See,
hereiswhereyoucanflip
throughthepages.”
(boy,age8)
SpiralBookReader
One of the challenges of partnering across generations in the design process is that adults are not
in charge, but neither are children. Design partners must negotiate team decisions. Over the
years, we have found that the most important goal of any partnership between adults and children
is idea elaboration. This is when one team member (adult or child) shares an idea with the team
and it is extended by others. A new thought or direction may be inspired by another adult or
child. When these ideas build upon each other to create new ideas, ultimately it may be difficult
to remember whose ideas they were originally. What matters is that both adults and children
share in the process together. It can be said that this elaboration process is the hallmark of any
good design team with or without children. However, what makes this so important to
partnerships with children is that idea elaboration is so difficult to have with young people.
Children communicate differently from adults and it can be difficult to understand what they
mean (Druin 2005). In addition, children can change not just in height but in abilities—they may
not be the same design partners you initially signed up to work with a year ago. What is more
common is that adults conceive of ideas and either teach them to children, or ask for feedback.
The notion of elaborating on each other’s ideas is less common and more difficult. Children have
so few experiences in their lives where they can contribute their opinions and see that adults take
them seriously.
However, getting past the traditional power relationships between adults and children, the
hierarchies and traditional respect expected of adults can make it difficult to work with children
(West 2007). To support the best circumstances for idea elaboration with children, our team has
changed the way we set expectations, brainstorm, and reflect as a group. To set expectations, our
team insists upon four basic practices: no raising hands, using first names (no last names or
titles), wearing informal clothing, and sitting on the floor. We do this to minimize what we call
the “existing power structures” where adults are in charge and children are following directions,
or where children are in charge and adults only observe. In the case of classically trained
researchers, we have found that many so deeply want to know what children care about that they
are content to just observe children. However other adults who are much more used to leading
the way have almost quit our team rather than let children lead. As one adult explained, “I can’t
deal with kids calling all the shots! They change their minds. They don’t focus on what needs to
be done. It would be so much easier to get things done without them.”
A few years ago one of our child design partners tried to “help” the adults we work with be
better listeners. We were listing different ideas on the board that ranged from “microphones that
make the children’s voices louder,” to “checking the adults’ ears” to make sure they could hear.
But as we sat and looked at the list, one of the children (Alex) grew quiet. Finally he simply
said, “You just have to be patient with them, since they only know what adults know. But when
we’re patient you can learn from adults and they’ll learn too. We all need to talk together and
listen together. Sometimes people have to remember to hear first and then talk,” (Druin 1997).
This is consistent with other researcher’s assertions that children have a lot to offer and are more
sophisticated than many adults think. For example, in discussing this general issue, Andy West
said that “Adults had not expected to find that children have extensive knowledge and can
present sophisticated opinions, analyses and understanding of situations” (West 2007).
International Distributed Cooperative Inquiry
Design Partnering Moves out of the Lab
In the Fall of 2002, our team had the opportunity to work with children in multiple countries on
further developing and understanding our digital libraries work. Since our library contains books
from around the world and is for an international community of users, working with children in
multiple countries was something we very much wanted to see happen, but had to wait until
funding became available.
Back in 1999, we had the chance to work with adult and child partners in Sweden and England
(Benford et al. 2000), but since our co-design methods were still evolving and we were not
focusing on digital libraries, it is hard to say how much we fully understood from those early
years. If one lesson stands out from our work then, it was that co-design with children can be
quite challenging to support within the structures of a typical school day. If a team is working
within multiple schools within multiple countries, the challenges seem to multiply. The more
defined the school curriculum is, the more difficult it is to fit the interdisciplinary co-design
experience into traditional subject areas of language arts, math, or science. Since that time, our
co-design experiences have been developed to take place after the school day or as a special
activity for a few children pulled out of their school day.
When we had the opportunity to work in New Zealand, Honduras, Germany, and Chicago (USA)
in 2003-2006, we asked schools to choose just three 7-year old children per school who along
with their parents, teachers, librarians, and principals would agree to work with us over a 3-year
time. However, three years quickly became four, thanks to the messiness of working with
children in the real world in multiple countries (Druin, Weeks et al. 2007). Between hurricanes
in Honduras, a homeless child in Chicago, a school library that was torn apart in New Zealand,
and the replacement of all site coordinators at each location, it made doing our weekly design
sessions in the lab seem almost easy by comparison. These and other issues happened throughout
the project. But we wanted to know more than what just one group in the US could tell us, so we
adapted our co-design methods to best address the circumstances. We wanted to understand how
truly global a child’s world is now that the Internet can be a gateway to so many diverse places.
We wanted to understand if physical libraries still matter when technology competes for
children’s attention. We wanted to understand how children would categorize and rate the
ICDL’s collection of books if they were the librarians (Massey et al. 2007).
At least one member of our team physically visited each site at least once per year to meet with
the site’s children and adults to talk about the future of libraries and computers. In between
visits, we asked them to draw self-portraits, computers of the future, and libraries of the future.
We asked them questions such as what they did for fun, the kinds of things they did in their
libraries, and if they could live anywhere in the world where would it be? We also asked our
design partners to ask their own classmates to draw pictures of their libraries of the future. In
addition, we asked the children to read and review books in the ICDL. Therefore, the site
coordinators (a local teacher or school librarian) were kept busy sending our team monthly
drawings, reviews, and answers to our questions. The full details of this study including exactly
what data were recorded and how the conclusions were drawn are reported elsewhere (Druin,
Weeks et al. 2007). During our visits, we asked the children to talk to us about their drawings,
reviews, or their classmate’s work (Figures 6-8)
Figure6:NewZealandgirlpreparestotalkabouthersecondsetofdrawings
Figure7:AnexampledrawingofaschoollibrarybyboyinHonduras
Figure8:AnexampleofaschoollibraryofthefuturebyaNewZealandchild.
In addition to these kinds of drawings, the children were asked to write weekly or monthly book
reviews as well as a “like/dislike activity matrix” (Figure 9).
Figure9:ALike/DislikeActivityMatrix
Table2:ExamplesofhowtheICDLevolvedbyusingdistributedCooperativeInquiry
DesignMethod InitialIdeas ImplementedFeatureinICDL
Drawingsdonebychildren’s
peersinclass
“Thelibraryofthefuturehas
heapsofcomfortablenooks
andspacestoread.”
(boy‐NewZealand,age8)
Currentlyworkingonversonsof
theICDLthatbeusedonmobile
phonesanddevelopingworld
laptopssuchastheOLPCXO
andtheIntelClassmatePC.
Usinglike/dislikematrixto
writeonethingyoulikeone
thingyoudon’tlike.
“Ilikethesearchingusingthe
ICDL,butIstillliketoread
regularpaperbooksbetter.”
(girl,‐Germanyage11)
Revisedbookpreviewpage,
simplifiedreadingoptions,and
madethebooktextmore
readablewithnewtechnology.
BookReviews “Ilikedthisbookwellenough,
butIdidn’tlikethatIcouldn’t
readthewordsbecauseitwas
inCroatian.”
(boy‐Honduras,age9)
Creatednewtechnologieto
showbooksinmuliple
languagestranslatedby
volunteertranslatorsfrom
aroundtheworld.
DiscussionswithChildrenover
theirDrawings
“Iwouldchangethelibrar y of
thefuturetohavemore
placestoread.”
(boy‐NewZealand,age8)
EnhancedUIonreadingdigital
booksintheICDL
Differences in Co-Design: Local vs. Distributive Cooperative Inquiry
The children at each site did not usually have the opportunity to work together as a design team
unless our Maryland team members were physically present. Therefore, the design methods
needed to rely less on real-time elaboration, and more on mixing the ideas in a more loosely-
coupled fashion. For example, the “like/dislike activity matrix” replaced the use of sticky-note
frequency analysis. And low-tech prototypes were replaced by drawings which could be more
easily scanned and sent by email. This portfolio approach to reflection was challenging but
necessary due to the high cost of travel. When we tried to bring the children from each of the
sites together via video conferencing, this failed to materialize due to the difficulty of time zone
differences, and the already heavy burden placed on local site coordinators, not to mention the
technological difficulty inherent in video conferencing.
This distributive cooperative inquiry offers opportunities and challenges in many ways. The
strength of these distributive methods lies in the ability for researchers to gain insights from
children in multiple countries simultaneously. A relationship can be built over time by taking
advantage of a mix of methods, including both face-to-face partnering and email portfolios. It is
important to note that without both face-to-face and email, this would not have worked. If we
had relied only on face-to-face, it would have been too long a period of time between visits to
sustain a partnership relationship. However, if we had relied only on email, it is likely that we
would never have been able to form a trusting partnership relationship.
Upon reflection, there are many challenges in working with children in multiple countries that
need to be addressed. Perhaps the most critical challenge is how to support idea elaboration with
the children. Most often, our team back in Maryland had to elaborate on what we were seeing
from children’s email portfolios. Some elaboration could be done onsite when talking to the
children about their ideas and those of their peers. But because there was so much time between
physical visits, the majority of idea elaboration had to happen off-site. Another challenge was
the need to rely heavily on the local site coordinator (either teacher or librarian). We had to ask
already stretched and tired teachers/librarians to put more into their already busy day and meet
with children, send portfolios, and report to us on issues that might arise locally. But many were
so excited by the project, they happily helped when and how they could. However, when a site
coordinator moved on, and they all did at one point, it was difficult to get the new person filled
in on the processes and responsibilities. It was also difficult to get these new people to be just as
excited as the initial people. Therefore, with uneven support from the countries, we often had
email portfolios that were late, missing, or of questionable quality.
Another challenge in attempting to do this kind of distributed co-design is the amount of
resources required to make this happen. The cost of travel, laptops for each of the children, and
presents for schools, was considerable. In addition, the amount of time needed to accomplish
this kind of co-design was also considerable. For example, for each of our three visits to New
Zealand, we couldn’t accomplish all that was needed in less than 8 days each time. On top of the
two days needed for travel, at least three days were needed to discuss the children’s portfolios.
In addition, meetings with parents, teachers, librarians, and principals to discuss their
perspectives of the children and the ICDL more than took up the rest of our time. We found that
in order to have a sustained relationship with these children, multiple years of visits and email
were needed to build shared experiences for design elaboration. In comparison, with our lab co-
design, we saw children twice a week for months which enabled us to build a team
understanding in less than 6 months.
Table 3 summarizes the differences between our lab co-design methods with our distributed co-
design experiences in multiple countries.
Table3:ComparingLocalvs.DistributedCooperativeInquiry
ChangedforCo‐Design
Methods:
LocalLabCo‐Design DistributedCo‐DesigninMultiple
Countries
1.IDEAEXPLORATION
• Adultsandchildren
observed/interviewed
• Childdesignersaskedclassmates
tohelpwithgeneratingideas
• Critiquedothersystemsusing
stickynotesandfrequency
analysisonwhiteboard
• Usedlikes/dislikematrix
• Low‐techprototypingwith“bags
ofstuff”fullofartsupplies
• Sketchedideasindividually
2.IDEAELABORATION Ideasharingwithinteam Onsiteyearlyelaborationwithother
childrenandresearcherorhad
Marylandteamelaborateonemail.
3.SETTINGEXPECTATIONS Nohandraising
Usedfirstnames
WoreInformalClothing
Therearenorightanswers
Beganwithparent‐teacher‐child
meetingsineachcountrytosetthe
stageforworktogether.Dressand
titleuseweredependentonthe
countrystandards.Nohandraising
andnorightanswersstressedin
meetings.
4.REFLECTINGASAGROUP GroupPresenting
AdultDe‐briefing
Onsitediscussionsoverdrawings
Mixingideasoccurredthrough
Maryland
From Design to Deployment: Lessons Learned
A Launch for Learning
In 2002, we launched the ICDL with a ceremony at the United States Library of Congress (Druin
2005). It could not have been more exciting, but within weeks of the celebration we soon
learned, that we executed one critical aspect of the launch very poorly—the installation of the
software. We spent months working with children and adults on how to search for books, how to
read books, and even the website entry to the digital library, but we didn’t focus on a very this
very important point, because we did not test the initial library outside of the United States.
People around the world could not install Java Web Start (which is what the ICDL originally
used when it relied on the Java web plug-in) and if they could, the ICDL ran so slowly, that it
was almost unusable due to the very visual nature of the interface. We had done all of our co-
design and testing once the user got to the library, but not the installation of it. And we had done
all of our homework on users in the United States (Druin et al. 2003).
We learned this hard lesson first: deploying the ICDL in different country contexts was as
important to our understanding of the system use as any co-design we could do. To make up for
lost ground, within three months, our technical team quickly rewrote the software to run using
HTML and JavaScript with the Java plug-in being entirely optional so that a user with even a
800x600 screen and 56K dial-up modem could use the ICDL, although at a slow speed. In
addition, our work in New Zealand, Honduras, Germany, and the US soon helped in this respect.
Configuring laptops, understanding Internet access and capabilities in each of the countries, even
learning when the electrical power was at its least reliable in Honduras taught us much.
We also have learned about the challenges of designing the interface for an international
audience of children. Computer interfaces intended for people from different cultures can be both
“internationalized” (meaning a single design is accessible by people from many cultures) and
“localized (meaning the design is adapted for use by a specific culture). As described in studies
looking at how Arabic-speaking children are able to use the ICDL (Bilal 2006; Bilal & Bachir
2007a; Bilal & Bachir 2007b), early versions of the ICDL were moderately successful, but
clearly had shortcomings as indicated by the difficulty that younger children had understanding
how to get started (i.e., they didn’t understand the difference between “Simple” and “Advanced”
search given either the textual or iconic representations).
Given these studies and our own analysis of use, The ICDL design has been refined over time,
and is now both internationalized and localized. For example, some changes that we have made
in recent years are that the home page now has a clear entry point to access books for which no
language skills are necessary (a clear and central animated image of a book is shown on the
home page). And once you start looking for books, the entire interface can be displayed in any
of 16 languages.
However, even with both the technical and cultural understanding that we have gained through
years of the ICDL’s use, we were challenged with new problems when working in rural
Mongolia.
Mongolia the Next Frontier
Mongolia is a land of dichotomies – warm sun and cold wind, green pastures and arid steppes,
industrialization and nomadic life. With 47% of Mongolia’s population being under age 20
(Lambert 2007), it is a place where supporting effective education is particularly important. In
June of 2006, Ben Bederson, director of technology development for the ICDL and a professor at
the University of Maryland, traveled to Mongolia to install a server with a special version of the
ICDL we built for the children of Mongolia. The Mongolian language version of the ICDL
(www.read.mn) is part of a larger collaborative project between our lab, the World Bank and the
Mongolian Ministry of Education, Culture and Science. The result of the collaboration is the
Rural Education And Development (READ) project. In an attempt to rebuild a culture of reading
and publishing that was greatly harmed by the fall of the Soviet Union, the READ project is
designed to increase the production and dissemination of children’s literature across the country.
Through this project funding, several computer labs and thousands of physical books are being
distributed to primary schools (grades 1-5) for classroom and community use.
The version of ICDL in Mongolia has been localized a bit more than the primary website. It’s
home page is in Mongolian, and it describes the website within the context of READ – which is
widely known among teachers in Mongolia. Because of resource limitations, we have not
translated the broader website (with information about our research, volunteering for us, etc.) –
but the “library” section of the website that supports searching and reading is all in Mongolian,
and there are links back to the primary (English) website.
This approach was predicated on the fact that internet connectivity within Mongolia was fast, but
getting outside of Mongolia was slow (which was true when we started this project in early
2006). It also required us to maintain a server running in Ulaan Baatar. It turns out that the first
assumption is less important now (in late 2008), as world-wide internet connectivity is
improving, and the challenge of remotely maintaining a server half-way around the world turns
out to be significantly difficult. When there are hardware problems (as we have had), it is
extraordinarily difficult to get the computer repaired, and software reinstalled. So, we are in the
middle of transitioning to a new strategy which is to keep the localized URL (www.read.mn), but
redirect it back to our primary servers in the U.S. – and provide the same localized experience
that people were offered when the server was in Mongolia. Time will tell if this strategy is better
overall.
In any case, neither of these strategies support children in rural areas without general internet
connectivity at all. So, in November 2007, Ben Bederson and Sheri Massey, a Ph.D. student in
the College of Information Studies traveled back to Mongolia for ten days in November of 2007
to install local versions of the library in rural schools and to provide teacher training that would
serve as a foundation for using the digital books to support teaching and learning in the schools.
To prepare for the trip our technical team “ingested” the 200 newly digitized Mongolian books,
and created metadata in English and Mongolian. They also had to modify the Linux-based
technology to run entirely on Windows. Bederson would be setting up the first three pilot
schools to test deployment processes in an isolated and rural setting. These schools had labs of
10 computers each that were connected on a local network, but not to the internet. The approach
was to install the ICDL web server on the teacher’s computer, and have the student computers
access the ICDL from the teacher’s computer via a standard web browser. Afterwards, the
Ministry of Education in Mongolia would take on the installation for other schools. The goal
was to consider how to install things easily and keep them running reliably. Bederson brought
multiple backups of the ICDL on many difference kinds of media: laptop, external hard drives,
DVDs and thumb drives. He brought power strips, adapters, and even an American keyboard
(Bederson 2008a; Bederson 2008b).
Despite what we thought was tremendous preparation with so many backups, and multiple tests
on different systems, it turned out that Bederson was preparing for the last trip and not this one.
On his last trip, the server that was hand-carried from the United States had a disk failure. It took
days and a complete rebuild to bring the server back to life.
But it turned out this time there were new problems that presented themselves: encryption
problems for the books, viruses which crashed servers, fluctuating power which destroyed
hardware, missing software drivers, and of course – no internet to help debug the problems.
Official ceremonies with government officials and dancing children warmly greeted our team at
each stop they made, putting additional public pressure on the installation success. In addition,
teacher workshops were dependent on getting the ICDL running at each rural school our team
was to work with. Therefore, at times, in order to make systems work, emergency multi-hour
trips were taken to local cities with internet access. Hours were spent on international phone calls
with our team back in Maryland. Whole computer systems had to be reconfigured or rebuilt.
Code had to be modified in the field to get around security-related issues. Demos had to be
given to 50 expectant people lifting a laptop in the air. The lack of site-specific knowledge
coupled with a lack of resources to fight viruses was endlessly frustrating.
But armed with determination, a Ph.D. in computer science, patient support staff back in
Maryland, and a talented graduate student who could teach even when the computers weren’t
working at times, Bederson succeeded in deploying our digital library in three rural villages in
Mongolia (Figure 10).
Figure 10: The ICDL in Mongolia
Lessons Learned
We have just received word that the next phase of the work in Mongolia will be funded and
therefore, we will be going back with the ICDL running on One Laptop Per Child’s small green
XO laptops. This brings up one of the essential gratifying results of this project – that there were
surprisingly few cultural hurdles in making things work. There were very challenging
technological and logistical issues, but aside from the expected burden of communicating across
a language barrier, our experiences went quite smoothly. At the outset, we did not know how
Mongolians, especially rural ones, would respond to a technological approach to reading and
literacy. We thought there was a risk that they would see technology as something foreign and
irrelevant to their lives. However, the teachers in the training workshops in every school we
visited were deeply engaged. They skipped lunch breaks and kept us on well past the planned
completion times. The schools had welcoming ceremonies with song and dance. Most
importantly, the children were very excited to have what seemed like an unending supply of
books and technology. They clearly saw the broad potential that computers could have in their
lives, and wanted to explore what it has to offer them.
As Bogdan & Bilken suggest, “a qualitative researcher does not put together a puzzle whose
picture she already knows; rather she is constructing a picture that takes shape as she collects and
examines pieces of data” (Bogdan & Bilken 1998, 47). This has been the path in this research.
Whether we are deploying systems in Mongolia or co-designing with children in New Zealand,
our main lesson has been that site-specific knowledge is critical to success. Understanding where
the roadblocks might be (e.g., lack of electricity, virus protection, mobile phone
communications), anticipating the cultural expectations for welcomes or revisits (e.g., bringing
gifts, planning in time for ceremonies, eating the local food) and knowing what local resources
might be available if and when needed (e.g., interpreters, drivers, tech support), are all important
to co-design and deployment.
Just as important is that the technical and educational team members sent to support these
activities must be respectful of cultural differences, and be flexible, creative, problem-solvers
who understand that there are no right answers, only good paths to move forward.
Other lessons learned whether co-designing or deploying:
- Co-design within the school day is terribly challenging to arrange.
- A combination of face-to-face and email methods are needed to sustain a partnership
relationship.
- Deploying complex technologies in settings without reliable power and Internet is very
difficult, and requires technical and problem-solving expertise.
For the Future
We heard throughout our research that children love to read in comfortable spaces.
Unfortunately if this means being tethered to a desktop computer while sitting at a desk, children
are less interested. They want to read sitting in beds, on floors, under tables, etc. Therefore, our
current and future work is focused on adapting the ICDL for any platform that enables children
to be mobile. This includes small laptops such as OLPC’s XO and Intel’s ClassmatePC as well
as mobile phones. With our work on mobile phones, we are also focusing on supporting
intergenerational sharing of stories with children and older adults. To do this, we have begun to
develop new co-design methods for users at the extremes of life, children working together with
grandparents or grand friends (Druin, Xie, et al. 2007).
References
Bederson, B. B. (2008a) “Experience the International Children’s Digital Library”, Interactions
Magazine, ACM Press: 50-54.
Bederson, B. B. (2008b). “No Ride Drive: The International Children’s Digital Library Goes to
Mongolia.” Maryland International Magazine, University of Maryland, College Park Press.
Benford, S., Bederson, B. B., Akesson, K., Bayon, V., Druin, A., Hansson, P., Hourcade, J.
P., Ingram, R., Neale, H., O'Malley, C., Simsarian, K., Stanton, D., Sundblad, Y., and
Taxén , G. (2000). “Designing Storytelling Technologies to Encourage Collaboration Between
Young Children.” CHI Letters 2(1): 556-563.
Beyer, H., & Holtzblatt, K. (1999). “Contextual Design”, in ACM Interactions, 6 (1): 32-42.
Bilal, D. (2006). Measuring the usability of an international user interface: Culture and design
representations. Papers of the ASIS&T Annual Meeting, Human-Computer Interaction
Symposium.
Bilal, D., & Bachir, I. (2007a). Children’s interaction with international and multilingual digital
libraries. I. Understanding interface system design representations. Information Processing &
Management, 43(1): 47-64.
Bilal, D., & Bachir, I. (2007b). Children’s interaction with international and multilingual digital
libraries. II. Information seeking, success, and affective experience. Information Processing &
Management, 43(1): 65-80.
Bjerknes, G., Ehn, P. and Kyng, M. (1987). Computers and Democracy: A Scandinavian
Challenge, Alebury, Aldershot, UK.
Bogdan, R., C. and Bilken,S., K. (1998). Qualitative Research for Education: An Introduction
to Theory and Methods. MA: Allyn & Bacon.
Bødker, S., Ehn, P., Sjögren, D. and Sundblad, Y. (2000). “Co-operative Design —
Perspectives on 20 Years with ‘the Scandinavian IT Design Model.” Proceedings of NordiCHI
2000, Stockholm, Sweden.
Chipman, G., Druin, A., Fails, J., Guha, M. L., Simms, S., Beers, D. (2006). A Case Study of
Tangible Flags: A Collaborative Technology to Enhance field Trips. In Proceedings of
Interaction Design and Children (IDC’2006). Finland.
Druin, A., Stewart, J., Proft, D., Bederson, B. B., and Hollan, J. D. (1997). “KidPad: A
Design Collaboration between Children, Technologists, and Educators.” In Proceedings of ACM
CHI 97.
Atlanta, GA: 463-470.
Druin, A. (1997). “Computers and Kids: Why do We Make Technologies for Kids?” SIGCHI Bulletin,
29(2), Accessed: http://bulletin2.sigchi.org/archive/1997.2/kids.html, June 05, 2008.
Druin, A. (1999). “Cooperative Inquiry: Developing new technologies for children with
children.” In Proceedings of ACM CHI99. Pittsburgh, PA: 592-599.
Druin, A. (2002). “The Role of Children in the Design of New Technology. Behaviour and
Information Technology 21(1): 1-25.
Druin, A. (2005). “What Children Can Teach Us: Developing Digital Libraries for Children.”
Library Quarterly 75(1): 20-41.
Druin, A., Bederson, B. B., Weeks, A., Farber, A., Grosjean, J., Guha, M. L., Hourcade, J.
P., Lee, J., Liao, S., Reuter, K., Rose, A., Takayama, Y., and Zhang, L. (2003). “The
International Children’s Digital Library: Description and Analysis of First Use.” First Monday:
8(5) Accessed June 5, 2008:
http://firstmonday.org/issues/issue8_5/.
Druin, A., Weeks, A., Massey, S., and Bederson, B. B. (2007). “Children’s Interests and
Concerns When Using the International Children’s Digital Library: A Four Country Case
Study.” In Proceedings of Joint Conference on Digital Libraries (JCDL’2007) Vancouver,
British Columbia, Canada: 167-176.
Druin, A., Weeks, A., Massey, S., and Bederson, B.B. (2007). “Final Report on IMLS Funded
Grant #LG02020026-02 APPENDICES.” University of Maryland, HCIL Tech Report #HCIL-
2007-01. www.cs.umd.edu/hcil.
Druin, A., Xie, B., Fails, J., Golub, E., & Massey, S. (2007). “Connecting Generations:
Developing Co-design Methods for Older Adults and Children.” University of Maryland, HCIL
Tech Report #HCIL-2007-53.
Ehn, P. (1988). Work-Oriented Design of Computer Artifacts. Stockholm, Sweden:
Arbetslivcentrum.
Greenbaum, J. and Kyng, M. (1991). Design at Work: Cooperative Design of Computer
Systems. NJ: Lawrence Erlbaum Associates.
Hutchinson, H., Bederson, B. B., and Druin, A. (2006). “The Evolution of the International
Children’s Digital Library Searching and Browsing Interface.” In Proceedings of Interaction
Design and Children (IDC’2006). Finland.
Hutchinson, H., Rose, A., Bederson, B. B., Weeks, A., & Druin, A. (2005). “The International Children’s
Digital Library: A Case study in Designing for a Multilingual, Multicultural, Multigenerational Audience.”
Information Technology and Libraries, 1: 4-13.
Komlodi, A., Hou, W. Druin, A., Preece, J., and Golub, E. (2008). “Cross-cultural
Communication through Pictures in an International Children’s Online Community.” 6th
International Conference on Cultural Attitudes towards Technology and Communication,
Universite de Nîmes, France.
Komlodi, A., Hou, W., Preece, J., Druin, A., Golub, E., Alburo, J., Liao, S., Elkiss, A. , and
Resnik, P. (2007). “Evaluating a Cross-cultural Children's Online Book Community: Lessons
Learned for Sociability, Usability, and Cultural Exchange.” Interacting with Computers 19 (4):
494-511.
Lambert, L. (2007). “Releasing the True Potential: Adolescent Participation in Mongolia.”
Children, Youth and Environments 17(1): 62-87. Retrieved November 2008 from
http://www.colorado.edu/journals/cye.
Large, A., Beheshti, J., Nesset, V., & Bowler, L. (2004). “Designing Web Portals in
Intergenerational Teams: Two Prototype Portals for Elementary School Students.” Journal of the
American Society for Information Science & Technology, November 2004, 55 (13): 1140-1154.
Massey, S., Druin, A., and Weeks, A. (2007). “Emotion, Response, and Recommendation: The
Role of Children's Book Reviews in a Digital Library.” In D. Nahl & D. Bilal (Ed.) Information
and emotion: The emergent affective paradigm in information behavior research and theory.
Medford, NJ: Information Today: 135-160.
Schuler, D. & Namioka, A., eds. (1994). Participatory Design -Principles and Practices. NJ:
Lawrence Erlbaum Associates Publishers.
Skivenes, M. and Astrid S. (2006). “A Child Perspective and Children’s Participation.”
Children, Youth and Environments 16(2): 10-27. Retrieved November 2008 from
http://www.colorado.edu/journals/cye.
Taxén, G, Druin, A., Fast, C, and Kjellin, M. (2001). “KidStory: A Technology Design
Partnership with Children,” Behaviour and Information Technology 20(2): 119–125.
Taxén, T. (2003). Towards Living Exhibitions (Dissertation). Royal Institute of Technology,
Sweden (KTH), Centre for User-Oriented IT Design (CID).
West, A. (2007). “Power Relationships and Adult Resistance to Children’s Participation.”
Children, Youth and Environments 17(1): 123-135. Retrieved November 2008 from
http://www.colorado.edu/journals/cye.
Acknowledgements
We are in debt to all of the children, parents, teachers, librarians, and principals who gave so
generously of their time to partner with us over the years. However, this research could not have
taken place without the extraordinary efforts of the ICDL development team (that do not include
the PIs): Alex Quinn, Hilary Hutchinson, Juan Pablo Hourcade, Sabrina Liao, and Tim Browne.
And finally this work could not have been done without the generous support of the National
Science Foundation, Institute of Museum and Library Services, the Ministry of Education of
Mongolia (through the World Bank), Discovery Communications, and the Microsoft
Corporation.
