Tapan Parikh of Washington University has created CAM, a smart way of using mobile phones equipped with digital cameras to capture microfinance data in developing countries. In Africa, the shortage or inadequacy of such data is a major reason why financial institutions are unable or reluctant to extend loans, savings, insurance and other essential financial services to rural and peri-urban groups. Parikh's system could help change all that if deployed in a way that exploits mobile phones’ utility, usability, and growing ubiquity to integrate paper-based tools--such as loan application forms, repayment records, ledgers, and the like--with computing and other modern information resources. The Figure below illustrates how this integration is achieved within a 3-tiered structure:
The user tier comprises a set of paper forms and artifacts that people in remote locations can use to record information, perform queries, and conduct transactions. The server tier is a standard Web application server, which can reside in a given village, a nearby town, or somewhere on the Internet. The mobile phone acts as a roving middleware, playing the role of scanner, user interface, network, cache, and preprocessor. Parikh explains why the use of smart mobile phones makes a lot of sense, compared to a PC:
Mobile phones have been cited as the most likely modern digital tool to support economic development in underdeveloped regions. As shown in the example of Grameen Phone, if a phone is shared by a group of people, it can be afforded by even the poorest communities. For rural computing applications, a mobile phone has inherent advantages over a PC in terms of cost, portability and familiarity to users.
The specific components of the CAM framework are: the CamForm augmented documents, which users interact with; the CamBrowser mobile phone application, which interprets these documents; the CamShell scripting language, which ties the two together; and the CamServer, which receives and organizes CamForm submissions via the CamBrowser and supports additional data entry and document processing. Parikh articulates very well the rationale for combining the paper and computing components:
...Having daily lives filled with physical tasks, rural users are much more comfortable with artifacts they can handle and touch. Paper affords a physical representation of information that can be communally viewed, edited, and stored. In some cases, illiterate users have even learned to interpret data from paper forms. For the time being, it’s unlikely that any affordable digital interface will be suitable for this kind of use...However, paper is notoriously bad for other uses. You can’t easily search, index, or archive it. Using it to perform calculations and create consolidated reports is difficult. In our work with community microfinance groups around the world, we found that data consolidation and reporting using manual, ledger-based systems is virtually impossible. We needed a way to digitally capture information from paper and let computing systems do the things they’re good at. So, we had to find a way to link the physical world of paper with the digital world of information.
To better appreciate the relevance of CAM to African needs, consider the following scenario: An international program has been launched to finance solar-powered pumps for rural farmers in an African country. The loans are to be offered to agricultural self-help groups (ASHGs) for redistribution amongst the members. The money is to be repaid with interest at regular intervals. This calls for proper management of internal accounts and lending. However, most ASHG members are minimally literate. Hence, they are not able to produce the needed financial statements and lending reports. They also tend to have inadequate accounting controls and systems for monitoring risk in the loan portfolio. Under these conditions the local bank responsible for managing the program finds it difficult to assess ASHGs’ stability and financial history when considering loan applications. To address the problem, local people who have some basic education are hired to provide the needed support to the ASHGs. But even for these field staff, preparing the needed financial and accounting reports for any one group is a difficult task, severely limiting the number of ASHGs that each of them can effectively support. This is one instance where CAM could come in handy: By automating a large part of the accounting and reporting process, CAM can dramatically reduce the work load on field staff while improving the efficiency and transparency of all ASHGs.
For instance, field staff equipped with mobile phones will record member-level transactions during scheduled meetings with ASHG members, using CamForms similar to that shown here.
Transactions are entered on the phone using the CamBrowser application, at the time of the meeting (or afterward) from the CamForms as illustrated below. In operation, the CamBrowser--which has been implemented for phones based on Nokia’s Series 60 platform--acts like a virtual window and one-button mouse for paper documents. To deal with the limitations of a mobile phone’s physical interface (small screen, a 12-button numeric, etc), the CamBrowser uses the phone's camera to recognize visual codes, effectively extending its interface to a nearly infinite number of widgets situated in the real world. This useful feature is based on software developed earlier by Rohs and Gfeller (2004).
Once captured by CamBrowser, the data is posted by the field staff to an on-line server via an SMS message. The resulting reports and financial statements may be printed from a secure access point at the field office where they may be picked up by a representative of the ASHG or delivered by the staff. Such reports can be used by the group to monitor their portfolio and to apply for loans and other services.
Pilot testing of the system in India last year returned promising results. Parikh reports that:
The users we tested had never used a camera or a PC. Most had never used a mobile phone, although they’d seen others using one. We found that most of these users were using the CamBrowser with ease in five to 15 minutes. This was a remarkable improvement in the learning times we’d noted in our earlier research with a PC-based interface. The accessible locus of interaction afforded by paper CamForms and the guided interaction driven by the audio-enabled CamBrowser made introducing the system almost seamless. Out of the 14 users with whom we conducted detailed usability tests, all found the system either easy or very easy to use.
We hope to develop--possibly with Parikh's help--a CAM architecture into the management information system of the eCARE project. It will be interesting to see how it works out, and i am looking forward to reporting our experiences here :)
For more information on the CAM architecture, follow the link below to Parikh's article:
Note: If the link fails, just type the full title into a search engine and it should take you there.