Features

XXI.6 November-December 2014
Page: 42
Digital Citation

The fog phone


Authors:
Sarah Sterling, Leslie Dodson, Hawra Al-Rabaan

Over the past five years, Interactions has brought HCI for development (HCID) into focus with contributions from authors such as Kentaro Toyama, Neil Patel, Edwin Blake, and Shikoh Gitau. This work continues to espouse participatory, user-centered design paradigms to effectively remove barriers to technology access and use. In particular, HCID research focuses on meeting the distinctive needs of users and creating opportunities for empowerment in infrastructure-challenged regions. Here, we focus on rural Berber communities in Morocco. The outcomes of our research suggest it could be generalized to other low-literacy communities, as well as to communities where gender roles prohibit unrelated men and women from communicating.

To provide some background, we invite you to the Southwestern Aït Baamrane region in Morocco. A spigot is broken and leaking in the 105-degree heat, water quickly soaking into the dusty soil before it can be reclaimed. Water is a critically scarce resource in this region. Thankfully, the water system is being tested and the problem is quickly addressed. A similar leak in a neighboring village went on for two weeks—no one wanted to report the leak in case he or she was held responsible for the damage. Furthermore, no woman could have called the man on Mt. Boutmezguida who manages the water source, a series of fog curtains. It is culturally taboo for unrelated men and women to engage with one another, and cellphones are even more suspect. While the village enjoys cellphone service, husbands often do not allow their wives to own or use phones, as they feel it may encourage infidelity; women who do have phones are subject to having their call histories monitored. Not only would such a call place a woman under suspicion, but the water manager also does not want women calling him and ruining his reputation as a “good” man.

Insights

ins01.gif

Two competing forces exacerbate this situation. The fog curtain project is an internationally funded, direly needed source of water in a region where climate stress threatens to empty villages and decimate the only argan trees in the world—and the supply of valuable argan oil. Hundreds of kilometers of curtains and pipes will bring water to people’s homes and support sustainable argan agriculture. Village men, the “mobile” members of the community, will not have to go into Moroccan cities or look for seasonal work in Southern Europe, as they will be needed along the mountain. However, water gathering is traditionally “women’s work.” And though some may argue that women and girls will now be able to go to school instead of fetching water, many of these communities are too conservative for this to be a viable option. Not being in charge of water means that women have fewer visible roles in the community, which works against their empowerment. The local NGO sponsoring this project understands both pressures and envisions the men working along the fog curtains and pipes, and the women managing water distribution and maintenance at the village level. But somehow the women need to be able to contact the water manager with problems without anyone fearing repercussions.

For the past two years, doctoral student Leslie Dodson and a team of master’s students in the Information and Communication Technology for Development (ICTD) program at the Alliance for Technology, Learning and Society (ATLAS) Institute at the University of Colorado have worked with an array of stakeholders to develop a communications system that would engender women’s participation in water provisioning, as well as enable men and women to communicate about water issues in a culturally appropriate way. Drawing on the work of Medhi et al. and others, the team created a text-free code for reporting water-system status [1]. Published codebooks removed any suspicion about communication content, and the water manager did not object to receiving specific codified messages pertaining to the water system. Gender was intentionally obscured in these transactions; women were now able to communicate with men, as well as remain active participants in water management. In many development scenarios, women are excluded when the process transitions from manual to mechanical. In this case, however, women held onto one domain, water, and extended their reach into another, information and communication technology. This is critical, as the impacts of climate change are often most acute at the local level in developing communities, and the negative effects of climate change are most often borne by rural women. The United Nations ranks women’s access and control over resources such as water as crucial to the creation of a more equitable society, and scholars and practitioners have established that water projects in developing communities are strengthened when women’s effort, knowledge, and experience are included [2]. Because women in developing communities are particularly vulnerable to rapidly depleting and compromised natural resources, they must participate in resource use, resource monitoring, and resource stewardship in order to have their needs addressed.

This work was not trivial, given the limited exposure the women had to technology and the types of mobile phones they had access to. Old, broken, and counterfeit phones constitute the technology at hand—phones with keys whose letters and numbers have worn off, “Mobiola” phones (instead of Motorola), and phones with many different settings and menus. Additionally, most of the women identified themselves as illiterate and were unable to read or write any of the numerous languages spoken and written around them: Arabic, Darija, Tifinagh (Berber), or French. (There is scant use of the Tifinagh script in ICTD. Few mobile phones support the Tifinagh language, and older mobile phones do not support 16-bit Unicode, which is necessary to display Tifinagh script and symbols.) Also, direction and letter shape were confusing to many women, as Arabic is written and read right to left and French is written and read left to right. Some women were able to read a limited amount of Arabic but owned phones that did not support Arabic script. On phones that could support Arabic script, the device had to be set to display that language. In order to display Arabic-script SMS messages, phone parameters must be set to “full character support.” (Factory settings are normally set to “partial character support,” the minimum required to display Roman languages.) Without full character support, an incoming Arabic text message will either not display at all or display garbled symbols. On phones that did not support Arabic script, other difficulties were evident. Capital Roman letters look significantly different from their lowercase counterparts, leading some women in the workshops to mistake the letters for two different alphabets. Likewise, due to the bidirectional complexities of multiple languages, suggestions such as “8 comes after 7” were not useful. For some, it was difficult to discriminate between the similarly shaped numbers 6 and 9 and similarly shaped lowercase letters such as b, d, p, and q.

The mobile form factor lends itself to shorthand and symbols, which stymied many women. This convention, which keeps text messages shorter and less costly, makes it difficult for semi-literate and emerging literate mobile users to “correctly” learn new words and phrases. SMS messages often use numbers to represent Arabic letters and phonemes. Because there is no direct transliteration of Arabic words into Latin letters, and neither Darija nor Berber has a written form, text messages are often written in Arabish—a creative, nonstandard, informal writing style that blends alphanumeric code and mixed-language words. Thus, for low-literate users in oral-language communities, texting syntax becomes another language in an already complex linguistic environment. Despite these barriers, study participants appeared to have a well-developed ability to recognize patterns (e.g., number sequences, emoticons, design elements) on their phone screen and in tiny phonebooks that family members had created for them. Dodson conducted several training sessions to help women more easily navigate their mobile phones. Participants also learned to identify letters on phone keypads and on a chalkboard and how to write their names, using their stored name on the phone for reference. To support the use of SMS, participants devised a list of short, simple, relevant, easy-to-understand SMS messages that included Berber-language phrases for “call me” or “come home now.” These SMS messages had to be expressed in letters borrowed from either the Arabic or Latin alphabet, depending on the phone. Berber words had to be transliterated into Arabic script or Latin letters, adding an additional layer of complexity to the training.

Workshop outcomes echo previous studies of how illiteracy affects users’ ability to effectively use their phone: Jan Chipchase has identified common coping strategies including rote memorization, visual cues, and the use of intermediaries to assist in phone use [3]. He suggests that the adaptive and creative strategies employed by low-literate mobile owners may be harnessed to design ICT or interfaces for the developing world, and that illiterate participants may be “lead users for the rest of us.”

To this end, following the workshops, Dodson and ICTD master’s student Hawra Al-Rabaan, a native Arabic speaker, worked with women to first understand their information needs vis-à-vis the water system and then create a shorthand SMS syntax in response to the requirements. Dodson, Al-Rabaan, women participants, and NGO employees created various iterations of posters featuring four common water issues that had occurred during tests of the water system: no water, leaking tap, wet interior wall where the pipe connected to the tap, and wet exterior wall near the water meter. Each water issue was assigned a letter corresponding to a number on a mobile phone keypad. For example, the letter A/a indicated that the household had no water service. The poster carried an image of a mobile keypad showing that the letter A/a could be found on the number 2 key on a mobile phone. Each participating household was assigned a water meter number so service requests could be identified by number rather than a name or a phone number, further protecting gendered societal norms while providing crucial information.

Each water issue on the poster had an explanation written in both Arabic and Latin script next to the image. The poster instructed water users to send a text to a specified number, indicating the water meter number and the water report. In focus groups, women solicited their design suggestions for the posters, resulting in a final poster that used pictures of the actual water system along with stock images of plumbing fixtures from the Internet as well as slightly abstracted icons and clipart elements. These became the basis for communicating system status on what became known as the Fog Phone system, a communications network comprising the male water manager, women in the villages, and the NGO overseeing the entire project. Principles of easy operability and immediate utility were of paramount importance throughout the design of the system. This approach to designing the Fog Phone enabled stakeholders to connect an information system to their daily tasks and activities in a culturally appropriate and amenable way. The system has since been extended through Al-Rabaan’s (and classmates’) work to include a FrontlineSMS backend to help queue and prioritize water system reports, as well as track the status of ongoing issues.

Since the deployment of the posters and the Fog Phone system, women have been reporting system status as water is extended to all 17 communities. There have been ongoing issues with typos and syntax errors, which have been addressed with ongoing training and “pushed” SMS messages that remind users how to submit reports. Long-term, we will monitor the utility of the system and the complex human dynamics underlying system design to determine if women are creating a sustainable niche in water provision. We will also evaluate whether existing social structures are being challenged and redefined with the greater use of mobile phones. While this research highlights many of the “utility gaps” [4] that can occur in complex socioeconomic, linguistic, and cultural environments, it is also another case study that reiterates the importance of taking tradition, gender roles, and language/dialects into account, not just infrastructure and cost.

While this work was contained to a specific group of low-literacy Berber women in an oral-language community, it may also apply to other communities in developing countries where there is high mobile phone ownership but low usage of text-based phone features. There are thousands of languages without a written form, which arguably makes text-based technology a barrier to ICT use. Remote and marginalized communities (and community members) are often the last to access newer technology, making feature (or “dumb”) phones a reality. The Berber women who participated in this study may appear to be “unique” users, but their experiences may be generalized to other predominantly oral-language societies. Low-literate women are acutely aware that their feature phones are not simple to use. There is an opportunity for interaction designers and those working at the crossroads on HCI and development to focus on alternative input systems in order to reach millions of low-literacy and oral-based communities who may fall further behind their neighbors due to gaps in design and understanding. Employed appropriately, ICT may enable women in Morocco and other developing countries to expand their participation in community development. In turn, new skills and opportunities may create conditions that facilitate faster, more enduring, and broader economic benefits for themselves and their communities.

References

1. Medhi, I., Cutrell, E. and Toyama, K. It’s not just illiteracy. Proc. of the 2010 International Conference on Interaction Design & International Development. A. Joshi and A. Dearden, eds. British Computer Society, Swinton, UK, 1–10.

2. United Nations. 2009 World Survey on the Role of Women in Development; http://www.un.org/womenwatch/daw/ws2009/

3. Chipchase, J. How do you manage your contacts if you can’t read or write? Interactions 13, 6 (Nov. 2006), 16–17.

4. GSMA. Mobile for Development Programmes; http://www.gsma.com/mobilefordevelopment/

Authors

Sarah Revi Sterling is founder and director of the ICTD master’s program at the ATLAS Institute at the University of Colorado, Boulder. She consults for the U.N., development agencies, and companies interested in utilizing technology for societal benefit.

Leslie Dodson received a Ph.D. in technology, media, and society from the ATLAS Institute, where her ICTD research focused on the intersection of natural resources, gender, and technology. She is an award-winning environmental and financial journalist.

Hawra Al-Rabaan is a recent graduate of the MS-ICTD program at the ATLAS Institute. Originally from the east coast of Saudi Arabia, she is passionate about women’s rights and empowerment, youth, and education.

Figures

UF1Figure. Phonebook of numbers and symbols.

UF2Figure. Studying the water poster.

©2014 ACM  1072-5220/14/11  $15.00

Permission to make digital or hard copies of all or part of this work for personal or classroom use is granted without fee provided that copies are not made or distributed for profit or commercial advantage and that copies bear this notice and the full citation on the first page. To copy otherwise, to republish, to post on servers or to redistribute to lists, requires prior specific permission and/or a fee.

The Digital Library is published by the Association for Computing Machinery. Copyright © 2014 ACM, Inc.

Post Comment


No Comments Found