Software Development for Tsunami Victims

Sanjiva Weerawarana writes about his efforts to develop software to aid in the coordination of Sri Lanka's relief efforts for the victims of the tsunami catastrophe. It's one thing to write software knowing your livelihood and everyone else in your department or company's livelihood is at stake. That's the pressure that's upon you when you join a small startup company. However I simply cannot even imagine the pressure felt in building something when lives are at stake. Tens of thousands of lives are at stake and the clock is ticking. Every minor delay could cost hundreds of lives. It's an unbelievable endeavor.

I've met Sanjiva only once, that when we were both working for IBM research in Hawthorne, NY. That was when his research group released the first java implementation of SOAP. I don't know if he still works for IBM, however I was surprised to learn that he's now based in Sri Lanka. It takes a lot of idealism to leave a comfortable research position in IBM and trade it to do work to improve the lives and opportunities for people in a third world country. However, when your work has true meaning and value for so many, then it makes it all worthwhile.

Now obviously with such stringent time constraints nobody in his right mind would build an application from scratch. The development team has chosen to build their application on top of Mambo a LAMP solution. It'll be interesting to know how they came to the decision to use Mambo over another LAMP application like CivicSpace.

CodeHaus an open source project repository has gratiously offered its services. This is testament to how agile open source development has become to support rapid distributed development. Try to set up the development environment using conventional approaches and it can take weeks if not months.

There are many open question of course. Does leveraging an open source code base over a commercial solution makes sense? It it worth the risk or is it the only viable choice? Can you develop functionality faster and more robustly on LAMP versus Java, Python or .NET? Unfortunately, there's very little time for debate. There's no time for theoretical discussion, functionality is needed and it's needed yesterday. I wish the development team the best of luck and pray that they've made the correct technical decisions.

So if you are a PHP developer and know Mambo please contact Sanjiva.

One final thought, does Indonesia have a similar development effort?