SMS Servers Replacing PCs in India

phonetractor.jpg
Sean Blagsvedt and Rajesh Veeraraghavan of Microsoft Research India (blog) presented at ETel last week. They told us about some amazing work they have been doing with mobile phones and SMS servers.

The SMS servers are being used to power an MSR project designed to test replacing PCs with SMS servers in the village of Warana (map). Mobile phones are used by farmers to access their data. In their system mobile phones become the client and SMS is used to communicate with the server.

The project is named Warana Unwired. Since the project has gone live it has processed 6000 SMS from 1238 different farmers (80% were data requests about sugarcane output). The time for farmers to get their data has dropped from 15 days originally, 2 days with a PC to immediate.

At ETel they also announced the launch of their SMS toolkit.

The SDK allows any Windows Mobile 5 phone to be used in conjunction with a PC to act as an SMS Server, allowing developers to easily write SMS applications, using just their phone and a PC. Additionally, we also provide a collection of samples that among other things, allow non-programmers to use Excel to send bulk SMSs and to build simple information lookup applications.

I asked Sean Blagsvedt and Rajesh Veeraraghavan to do a write-up of their project for Radar. Click-through to read it.

Warana Unwired

Basic Thesis:

We have run an experiment replacing a PC based system for helping a rural sugarcane cooperative with a mobile phone based system. The new mobile system replicates almost all of the PC based functionality. It is cheaper, adds additional functionality and is more popular.

We believe that this is the first project of its kind in developing regions where an entire PC setup has been replaced with mobile phones.

Warana Wired Village project is now called the “Warana Unwired.”


Context:

Warana is a village located in Rural India, in the state of Maharashtra. The sugarcane cooperative is serving about 70,000 farmers across 75 villages.

The government of India in 1998, started a pilot experiment to bridge the digital divide by setting up this project referred to as the Warana Wired Village project. It is touted as Asia’s first computing intervention in rural areas of this scale. The project was introduced in 1998 jointly with government of India funding 50%, government of Maharashtra funding 40% and 10% from the Warana Cooperative. The total amount that was spent on this pilot project was $500,000. Under this project 54 kiosks are established to connect 40,000 farmers across the different villages.

Since it was a pilot, the original goals of the project were understandably very exploratory. The original goals of the project were to give internet access to farmers, to allow farmers to check market prices so that they can sell the produce to the market that was offering them the best price, it was to setup a remote agricultural advisory system. For various reasons, these didn’t work out.

The cooperative then turned around and started using these kiosks for remote bookkeeping.

The farmers used the kiosks to check their sugarcane output each farmers produce, track their fertilizer outputs, issuing harvesting permits and to get their pay stubs. There is a kiosk operator serves as the intermediary to give access to these farmers.

Problems with the existing System:

It is indeed amazing that these kiosks are still running after 8 years after their original installation, it is rare you see rural computing projects running for this long. That said, the PC’s were running into many issues due to the rugged rural conditions and the maintenance cost were shooting up steadily. Power is a huge issue in these rural places, and they had UPS backups that would help with it. It costs money to replace them and that also added to the maintenance costs.

Research question:

Can we preserve the functionality of the existing PC based system while making the entire system cheaper and more effective?

Our solution:
smsservers.jpg

We replaced the client PCs with SMS enabled phones. On the server, we attached a smart phone through USB to their PC server. So, we effectively have an SMS gateway that receives incoming SMS messages and converts into database calls and the response was also converted to an SMS message and the result sent back to the phone that sent it. The authentication was through the SIM card (essentially the phone number).

Results:

We found that all the application scenarios they had could be converted using the SMS enabled phones. The system is now available 24 hours, and we have farmers using the data on a few occasions at odd times like 3.30 in the morning. The solution is truly mobile and the farmers are using it in places like the tea shops, front of the farmer fields and in the kiosks. In most cases they use the kiosk operator (who is now the phone operator) as the intermediary to send the SMS messages. There is a potential saving of over a million rupees($22,000, which is a big deal in these contexts) if the cooperative completely switched from the PCs to the mobile phones, this is primarily due to the savings from the maintenance costs of the PCs.

We are working with the Warana cooperative to see whether they want to scale this to all of the 54 villages they are operating. We hope this will lead to many more SMS applications throughout the world.

Sean Blagsvedt and Rajesh Veeraraghavan

tags: ,