Microcredit -- the practice of lending poor people small amounts in order to start cottage industries and small businesses -- is a model that, generally speaking, has been a wild success.
Microcredit is exstimated to have helped as many as 100 million people to escape poverty. While the model has problems -- that it doesn't work for the desperately poor because they can't afford to pay the loans back; that microcredit doesn't meet all the financial needs of the developing world's poor (they also need financial institutions which will help them save money at interest, transfer funds at low charges (remittances from workers in the developed world to their relatives in the developing world account for more money than all foreign aid programs put together), and receive short-term or emergency credit at non-usurious rates); and that microcredit does little to secure people's economic and property rights, leaving the poor continually vulnerable to oppression and corruption -- that the debate has reached a stage where we're able to focus on those problems in some ways just highlight how effective the model as a whole has been.
One of the difficulties remaining is building the software to support microfinance.
As Tomas Krag puts it on Multiplicity:
"[K]eeping tabs on thousands of tiny loans is a daunting task without a computer system. One of the problems has long been that most software packages fro Micro Finance come from the traditional banking and finance software market, and are most often prohibitively expensive, or have been hacked together by a smal ngo somewher in the world as they gradually expanded their own microfinance services, and is hence anything but stable, adaptable and well documented.
"This has lead me, and many others to the inevitable conclusion that there would be a interesting market for an Open Source, Free (as in beer and probably also as in speech) microfinance package with a well-designed, stable architecture, and good documentation. The problem, which is also the reason why good Open Source financial packages in general have been few and far between, is that most geeks like to work on software that interests them, especially when doing this work in their spare time. And few real geeks (at least ones that I know) supplement their interest in computers with an intense fascination of finance and bookkeeping."
The Micro-Finance Open Architecture project is one effort to change that:
"The Mo-Ap Project is conceived as an open source effort to develop standards, tools, libraries and solutions for the international micro-finance and micro-credit communities. We are just getting started, please click on the links below to read our documentation and subscribe to our mailing list to find out how you can get involved.
"High volume transactions, group solidarity lending, and field operations in poor rural locations all make Microfinance a difficult market niche to support with commercial portfolio management and accounting products. At the same time, the grassroots philosophy, low margins, and innovative tendencies of the emerging microfinance industry make it a good match for Open Source Systems development. Moreover, the lack of data standards makes long term financing more difficult and prevents necessary collaboration within the sector for credit checks, securitization, and poverty impact assessment."