SMS and interactive voice response systems are emerging as a significant methodology for gathering and spreading information in the developing countries. This is important to note because in 2010, more than 4 billion people paid for mobile phone service. That’€™s 6 of every 10 people on the planet compared to 3 out of every 10 people on the planet who has internet access 1.

One such initiative that utilizes SMS and IVR is NextDrop which gathers water supply information from the valvemen and spreads the information to residents of those areas.

Why does this matter? Because outside of major cities, water gets delivered once in 5 days or once in 15 days, and on those “water days,” someone (usually the women) has to stay back and ensure to collect enough water for the next 4 or 14 days. They usually don’t have enough water storage capacity for that many days and end up having to collect as much water as possible in vessels and pots. So, as you may expect, it is critical for someone to be at home on “water days” and even keep back children from going to school on those days. This is such a loss.

A valveman releasing water to an area by opening the valve

This is where NextDrop comes in and tries to solve this information asymmetry and hopefully solve corruption problems as well. NextDrop is running its pilot in Hubli, Karnataka in collaboration with the Hubli Water Board.

You can get a good overview of NextDrop in their winning presentation at the Global Social Venture Competition :

NextDrop was started by a group of Berkeley students including a Ph.D researcher in environmental science who noticed this problem when researching water quality in Hubli, Karnataka and other Berkeley School of Information students. One of the team members, Thejovardhana Kote did his masters thesis on NextDrop and made the initial prototype. Eventually he had to make the hard decision of moving on to something else and he asked me to step in. Since I had just then made the decision of going freelancing, it was perfect for me to earn money as well as make a social contribution at the same time, so I readily agreed.

It took me some time to figure out a different domain (SMS, IVR, non-computer-literate users) but eventually I got the hang of it.

The challenge for me was to get the system to a production level and subsequently, I embarked on a rewrite of the system because it was clear one had to be thrown away which I will write about in more detail later.

There are some really interesting stories on the Official NextDrop blog during this time frame:

I’ll end with Rohini Nilekani’s writeup on future water challenges for India :

India may have to ready itself for perennial freshwater shortages. The country is among the wettest in the world, with an average annual rainfall of 1170 milimeters and total water resources of around 4000 billion cubic meters per year. Of this total, a little more than a quarter is pegged as usable. With India’€™s high rate of population growth and intensifying water consumption, per capita availability of water, one of many indicators of an oncoming crisis, has declined steadily over the years. Thanks to indiscriminate withdrawal from rivers and underground aquifers, without adequate thought to recharge and regeneration, India could become an officially water-stressed country within this decade, dipping below the common indicator of 1700 cubic meters per person per year. Going beyond a merely human-centric position, it’™s critical to understand that water is a key element of nature in its own right.

1: Amazing ventures such as VillageTelco will further those numbers.

joshua schachter

The first person I started freelancing for is Joshua Schachter. Joshua is well known as the founder of, the bookmarking website which made “social filtering”, “tagging” (at the same time as Flickr) and “Web 2.0” common buzzwords.

I stumbled upon his tweet one day that he was looking for remote Python developers for his pet projects, I said “Hey, I’m interested” and next thing I know he is throwing ideas (and he has lots of ideas) and I’m furiously thinking about the fastest way to implement it. Over time, what interested me the most was not the idea itself but rather the evolution of the idea while iterating over prototypes and how Joshua creates this evolution. It has been fascinating to be a “fly on the wall” in this process.

The first project was ClueDB, a “tips and tricks” website, or a “lifehacker, by the people” as I like to call it. Built using MongoDB, Flask, my flask-boilerplate project, HTML5 Boilerplate, jQuery, jQueryUI, Twitter/OAuth integration, and Fabric. Standard stuff.

Joshua started with a simple answers / tricks site and slowly hacked on features. For example, he said “let there be tags” but there was a twist. The tagging would not be done by the submitter of the clue him/herself but by other people visiting the site. “It would be less spammy this way” he said. Boy, was that an eye-opener to me on how a simple social “tweak” could make a big change to the content and hence the trustworthiness of a site.

The second aha moment for me was when he said “let there be votes” :

cluedb voting scheme

This was a very interesting scheme of voting where we ask for the direct action by the user and not a simple plus/minus counter which doesn’t say anything about the usefulness of a clue (as opposed to popularity).

Things were going along, and then Joshua said “let there be user profiles” and things were going along again.. and then Joshua said “let there be a page listing all the clues that the user has voted ‘Plan to try it’” and we created such a page. The beauty of this was that now there was a single page that a user can use as a bookmark, homepage or share it with friends – voila, viral and useful at the same time.

Similar aha moments was creating a link called “random clue you haven’t voted” which would encourage the user to both explore more content as well as encourage to vote at the same time, and so on.

The latest project is – a simple catalog of parts for a web stack so that developers can weigh in their options and discover new ones for their next webapp. Built using Tornado, Python, Fabric and YAML files as the NoSQL database ;-).

This project was again interesting for the use of tags to organize the projects and the relationships between them, which you can see in action when you visit any project page on

There was a good discussion on HN about StackParts and was featured on O’Reilly Radar, and if Wayne Seguin likes it, there must be some value in it.

(As an aside, the site is also a testament to my total lack of design skills.)

And all the while, Joshua is busy with his own startup!

In summary, it’s been a fascinating experience where Joshua is doing the product management and I’m doing the coding.

My two favorite lessons from this experience are:

“You can’t reason about products – great products are emotional.”


joshua startupquote idea log

I finally got to check one particular item out of my bucket list – working out of Goa. This was possible purely because of freelancing which means I can be anywhere as long as I’m getting the work done.

When my wife and myself were planning this month-long road trip – driving from Bangalore to Mysore (so that I could run a half marathon to Mulki (so that the wife can learn surfing in the ocean to Goa, we charted out the course and stay, but I insisted we don’t book any hotels in Goa. “Why?” she asked, and I said I wanted to test travelling the way the foreigners do as I had written earlier. That turned out to be a great idea, because we found the perfect room for us – clean, spacious, reasonable tariff, separate entrance, undisturbed by neighbours, and most of all, we can see the beach and the ocean from the bed! And there was no way we could have ever found out about this place online when the owner was puzzled what “WiFi” means…

This was the view from our porch:

2011-09-26 14.11.55

2011-09-30 18.14.12

2011-09-27 18.21.36

2011-10-05 18.14.10

The owners even had a friendly cat:

2011-10-04 06.52.33

The best part about having a room almost on the beach was the amazing evening walks after a long day of work and hot weather.

2011-09-24 18.05.56

2011-10-03 18.40.09

2011-10-03 18.40.03

2011-10-03 18.44.07

2011-09-30 18.19.55

Of course, there had to be a “but.” It couldn’t be all perfect. I had a Reliance NetConnect+ as well as a Tata Photon+ connection. The former hardly worked throughout the trip after we left Bangalore, so I cancelled that. However, even the Tata Photon+ connection didn’t work in our room :( … after much calls and even a visit from a Vodafone personnel, it seemed that nobody gave connectivity towards the beach! How ironic for me when I said “All I need is my laptop and an Internet connection for me to work anywhere”… so we started searching for rooms in the main market area where we could get connectivity.

I literally held my laptop with the Photon+ USB stick attached and checking for signal at each place we checked out, heh! But the quality and location of rooms were pathetic. So I decided that we’ll continue to stay at our perfect room, so what about connectivity? Everyday morning, I drove 3–4 km to the main market area, parked the car, sat in the backseat, and had access to full Internet connectivity signal as long as the laptop battery lasted! Then, I would go back to the room and continue to work mostly offline – I connected my laptop to the WiFi hotspot from my Android phone and used the GPRS connection for the occasional technical searches I ended up doing. That was interesting for sure.

Two things I relearned – having a deadline (the battery is running out!) made me get more stuff done, and having no internet connectivity is good for your productivity (there, I said the P-word, hadn’t written about it in a long time! :-P)

Another aspect of the trip was the major expense – food. Initially, we ended up going to bad restaurants. Then, we resorted to the simplest trick in the travel book – ask the locals! The best restaurants we ended up going was Bed Rock, Infantaria and Jay Jays.

2011-10-08 16.58.31

2011-10-08 17.01.20

2011-10-08 18.09.15

… as well as Cape Town Cafe which had great salads and fantastic live performances (on Tuesdays and Thursdays):

2011-09-27 21.59.18

On days when my brain was tired, we would escape to nearby sights such as Fort Aguada:

2011-10-11 13.06.39

Overall, I surprisingly did get a lot of work done and yet had fun in visiting places, enjoying the night life and the general relaxed atmosphere. We had got so used to it that we had difficulty in motivating ourselves to start our journey again from Goa!

After the three weeks were up, we continued our journey visiting places like Ambaulim Falls where I dunked my head in flowing cold water…

2011-10-15 12.42.41

… and visiting a colourful Kolhapur, and so on.

2011-10-16 12.39.03

2011-10-16 14.20.12

Oh yeah, there was an unintended side-effect of writing about this trip – one fine day, my Dad called me up to tell me that I had received a snail-mail from the police that I jumped a traffic signal near Ulsoor on the very same day that we were on the beaches of Mulki, near Mangalore. I consulted a friend who advised me to email the head of traffic police which I promptly did and also included the blog links which had photos of us driving the very same car. A few days later, the Ulsoor traffic police inspector called me and said “nice pics” and informed me that the traffic violation ticket will be cancelled! (It was heartening to see that the entire process was over email and to note that the Bangalore traffic police is actually using modern ways to connect with citizens).

Update: Also see Traveling, Writing and Programming by the awesome Alex MacCaw.