Archive for November, 2007

India’s 450 million

November 30, 2007

Summary: Most of India lives in villages, and 450 million people live on less than $1.50 per day. Rural Indians need to tap into new markets to raise their economic output into the trillions. My hunch is that finding ways to help rural Indians to diversify beyond agriculture may be the only way to improve their standard of living to $10/day, because the total agricultural market is top limited to perhaps $100 or $200 billion and that will not be sufficient for everyone to afford integrated package of education, health care, water, food, public works etc. Lifting the constraints of distribution (geography, information flow) using the Internet to create online markets for small scale goods/services tailored to the rural economy can tap into “long tail” of demand. Read below to find out how I arrive at this conclusion.

I got to attend a Stanford conference sponsored by the Freeman Spogli Institute for International Studies a few weeks ago discussing issues that are of relevance to all of us. See Conference explores nexus of development, global warming, nuclear proliferation. Audio transcripts are available here.

Warren Christopher, former Secretary of State, has begun a commission with Baker to examine the presidential perrogative to go to war, how to conduct a war, and how to terminate it. This is inspired by current events involving the tensions between Congress and the President in the Iraq War.

Former defense secretary Perry noted that every month 1000 US military personnell are killed, maimed, or wounded. Security is moving in the wrong direction. The top threats he sees are:

  1. Terrorist nuclear weapon
  2. New Cold War (China/Russia)
  3. Environmental Disaster
  4. Radical Fundamentalists Gaining in Influence

In the Day After conference he organized, they considered the possibility that terroroists could threaten multiple terrorist nuclear attacks (say 1 every month) until they get what they want. We have no way to defend or deter this, and confidence in the US government would evaporate if DC were to be attacked. In terms of solutins to carbon emissions there are hybrids, wind, solar, and nuclear power. There is a conflict between the need for power and the ability to prevent terrorists to get their hands on fissile material. There is also a lack of political will to get ride of nuclear weapons. It took one third of Perry’s time as defense secretary to reduce nuclear weapons, and getting rid of them completeley per the bi-partisan proposal will require even greater time commitment from future administrations.

Scott Sagan discussed Pakistani nuclear weapons expressing many of the views for which he is quoted in this WaPo article “Calculating the Risks in Pakistan.”. Prior to 2001, Stanford scholars visited Pakistan and found very little security. So Stanford scientists started working with them to institute various best practices including psychological screening. See recent article the the Wall Street Journal on Pakistan’s Personell Reliability Programs and other security measures for their nuclear weapons program. The US government officially got involved in helping Pakistan secure their nuclear weapons after 9/11, which are currently secured through physical separation, not devices known as PALs.

Rosamond L. Naylor described the Millenium Development Goals to reduce the number of people living in chronic hunger by 1/2 by 2015, using a 1990-1992 baseline. People in hunger will either find a way to live, end up with long term disability, or die. The human security report points out the number of people that die from various causes:

  1. Terrorism: 3000
  2. Battle: 20,000
  3. Conflict (Genocide): 50,000-100,000
  4. Collateral damage: 500,000-1,000,000
  5. Hunger/malnutrition: 6-8.4 million

Kofi Annan’s list of risks to human security are hunger, infectious disease, resource depletion, civil conflict, and climate change. 15% of people live on $1/day and 50% live on less than $2/day. In sub-saharan Africa most people live on less than $0.60 per day.

Shashi Tharoor talked about his new book called The elephant, the tiger, and the cell phone and he pointed out there are three ways to get what you want, sticks, carrots, and soft power. He discussed at length how India’s soft power is rising with culture, political values (when it lives up to them), foreign policies, movies, art, cuisine, fashion, etc. India’s economic growth is also enabling this. There are 8.5 million cell phones sold per month in India, and the average bill is $4 / month. Quoting the US ambassador to India from an earlier session, he points out that while 450 million people in India live on $1.50 per day, 260 million of them live on less than $0.30 per day (Indian poverty line). 540 million of the Indian population is under 25 years old, and 165 out of 602 districts have Maoist insurgencies.

My question to Shashi during the Q&A was what was India doing in comparison to China for the children who grow up in these $1.50 per day environments, because quite literally they really the future of India? His answer had to do with education, which in turn requires health care, and food to feed hungry children. He pointed out that the government is making room for NGOs (non-profits) to help here. This reminded me of the list of principles for universal health care by Paul Farmer which includes food security, water, education and so on. Human development all comes back to the same “package” of human security needs that need to be provided for in a functioning economy.

Putting all this together, my thoughts are that handouts provided by the government or NGOs may succeed in patches but will fail in scale. The only solution that makes sense for the bottom 450 million people is economic development to increase their wages so they can afford food, water, health care, education, housing, and security from crime and the natural disasters.

Indians need to find ways of getting out of agriculture into new diversified economic activities –giving loans/handouts to farmers (see microcredit) or using the internet optimize the agricultural supply chain (E-Choupal) will only work for the same old agrarian market which is top-limited ($100-200B annually?). they will need access to new larger markets (local or international) if things are going to change for them. three generations ago, my family was completely in farming and today my cousins are all in IT/medical/etc – many in the US. that model of migration or outsourcing will never scale to most of india.

The growth of ITC’s E-Choupal initiative is impressive, focused on agricultural trading and supply chain automation. on the other hand it its scope is limited to the top economic activity, so no avenue to buy/sell other goods/services like craig’s list. There are also barriers to usage – it relies on training/access by one farmer in each village, does not allow open/self registration like most web 2.0 sites, and appears to be in english not local languages. in seven years they have reached 31,000 villages through 5200 kiosks – in total 3.5 million farmers. across india there are half a million villages and 450 million people, so large scale initiatives need to reach 10-100x of E-Choupal.

Launched in June 2000, ‘e-Choupal’, has already become the largest initiative among all Internet-based interventions in rural India. ‘e-Choupal’ services today reach out to more than 3.5 million farmers growing a range of crops – soyabean, coffee, wheat, rice, pulses, shrimp – in over 31,000 villages through 5200 kiosks across six states (Madhya Pradesh, Karnataka, Andhra Pradesh, Uttar Pradesh, Maharashtra and Rajasthan) The problems encountered while setting up and managing these ‘e-Choupals’ are primarily of infrastructural inadequacies, including power supply, telecom connectivity and bandwidth, apart from the challenge of imparting skills to the first time internet users in remote and inaccessible areas of rural India.

the agrarian economy can still continue to grow year on year, but how far can that go towards changing the majority of Indians’ daily wages from $1-3 per day? see “Field of greens: India is undergoing a second agricultural revolution – building the infrastructure that connects farm to supermarket.” if we are to hope to multiply daily wages by 10x, we need a local solution (maybe like Craig’s list) to connect buyers to sellers in new ways, not just agriculture.

Amartya Sen’s work on famines suggests that the bottleneck to economic growth at these levels is distribution – which in turns is limited by geography and information flow. From wikipedia…

In 1981, Sen published Poverty and Famines: An Essay on Entitlement and Deprivation, [read it here] a book in which he demonstrated that famine occurs not only from a lack of food, but from inequalities built into mechanisms for distributing food. Sen’s interest in famine stemmed from personal experience. As a nine-year-old boy, he witnessed the Bengal famine of 1943, in which three million people perished. This staggering loss of life was unnecessary, Sen later concluded. He believed that there was an adequate food supply in India at the time, but that its distribution was hindered because particular groups of people—in this case rural labourers—lost their jobs and therefore their ability to purchase the food. In his book Poverty and Famines: An Essay on Entitlement and Deprivation (1981), Sen revealed that in many cases of famine, food supplies were not significantly reduced. In Bengal, for example, food production, while down on the previous year, was higher than in previous non-famine years. Thus, Sen points to a number of social and economic factors, such as declining wages, unemployment, rising food prices, and poor food-distribution systems.

We are talking about vast numbers of people. There has to be a long tail stretching out in this demand curve. Until their net worth improves they are not going to have credit cards anytime soon, so transaction-oriented sites like ebay won’t reach them in ten years at least. one possibility is mobile banking for these 450 million people to participate in online transactions (instead of credit cards). See

Craig’s list in India. here are the cities in India… only the very biggest ones are covered (ahmedabad, Bangalore, Chennai, delhi, goa, Hyderabad, kerala, Kolkata, Mumbai, pune). they are all in English with the same US-centric categories (personals, jobs, housing, etc). not aware of any similar site in india. is a fee-based job board site kind of like, except they use social networking among high-income people and paid referrals to find jobs for low wage workers who make up the majority of the indian population. here is a podcast by their founder/CEO Blagsvedt (formerly at Microsoft Research) who explain’s the company model/strategy… babalife is a social network to attract wealthy folks who automatically get signed up for babajobs – the site for jobseekers.

there is a nice article in the NYT about them and how they stumbled on their ideas. “In India, Poverty Inspires Technology Workers to Altruism”. babajobs also has an interesting slide deck/pitch contrasting US social networking sites to low income Indians’ needs, including internet vs cell phone subs – unlike US people making < $250 per month are the vast majority of the population. in india,

  • 18% on mobile, virtually all on SMS

  • 3% on mobile internet

  • 0.25% has broadband via PC

  • 4-7% shared network connection

towards the end of this presentation they have a section on how low-income workers find jobs in india… word of mouth (multiple hops). using the babajobs site workers can pay Rs 100 of their 1st months’ salary to “connectors” to locate jobs. for < rs 300 they issue payments in mobile minutes, for more than Rs 300 they send a State Bank of India check.

what would happen to the cost of distribution in india if a free information network/service/billboard was made accessible indians, modeled after craig’s list:

1. at the village, town level (also metro, district, state, and national levels)

2. in local vernacular (hindi, telugu, etc) and in multiple interfaces including web-based and cell phones to the extent possible

3. with categories customized to both agricultural and urban indian markets? crop prices, labor needs, equipment leases, services, etc?

internet-enabled computers kiosks are not everywhere, but there are enough around to reach a sizeable fraction of the population under $1.50 per day in cities and some villages.


Anyone, Anywhere, Anytime

November 26, 2007

A most inspiring vision I came across this weekend… “I want [the GT-R] to make supercar performance in all conditions” says Mizuno. In his estimation, the current crop of supercars are designed to perform only in optimal conditions (in the dry, at the track) with an experienced driver behind the wheel. His GT-R concept is the exact opposite: “anyone, anywhere, anytime.”

First Look: 2008 Nissan GT-R

Anyone, Anywhere, Anytime: Nissan targets Porsche, but ends up building a world beater

found these while I was stuck at O’Hare airport last night… new M3 rocks, the GT-R may beat the 911 Turbo at 2/3 the price, and the Mazda 3 may be almost as good as the new M3 at half the price – the Mazda may even match older M3 performance in many respects. 

Comparison: 2007 Audi RS4 vs 2008 BMW M3 vs 2008 Mercedes-Benz C63 AMG Subaru WRX v Dodge Caliber SRT4 v Mazdaspeed 3 v Volvo C30 – Comparison Tests    

YouTube and new ideas/concepts/products

November 26, 2007

Lights! Camera! Sales!

How to use video to expand your business in a YouTube world

November 26, 2007; Page R3

With several good examples to illustrate:

  1. Be funny

  2. Tap into current events

  3. Find a partner

  4. Be useful

  5. Get your customer involved

MIT video lectures can be searched by audio content

November 11, 2007

MIT develops lecture search engine to aid students

With a number of video lectures becoming available through MIT’s Open Course Ware program, this new search engine allows people to search for any spoken word inside these lectures! Try it — just trying searching in the textbox, don’t worry if it says “real player” must be installed.

Also see the article in Tech Review describing how it is novel/different than other video search engines on the Internet

Searching Video Lectures

A tool from MIT finds keywords so that students can efficiently review lectures.

30th Anniversary of the Internet

November 11, 2007

Major Internet Milestones:A 30th Anniversary Celebration of the First Three-network Transmission

  • Bob Kahn DARPA
  • Jim Garret Collins Radio – still in the business. Wi-Fi packet radio (for military)
  • Irwin Jacobs Linkabit (now at Qualcomm)
  • Vint Cerf DARPA (now Google)
  • Donald Neilson SRI
  • Paal Spilling Norwary
  • Ginny Travers BBN

The Internet Problem: Internetworking of disparate networks

Kahn gave introductory remarks

Internet Demonstration: Nov 22, 1977 data was flowing from a van in silicon valley from SRI to USC via London

  1. Satellite network (Packet Satellite), 64/128 kbps, 8000 bit packets – Irwin was the architect
  2. Digital packet radio (Packet Radio) 100/400kbps, 2000 bit packets, omnidirectional broadcast, DSSS. – Kahn was the architect
  3. Wired ground network (Arpanet) 50kbps landlines

Kahn inspired by “Poisson Shannon and the Amateur Radio”

Routing modules could be plugged into the ARPANET, congestion and deadlocks would be an issue. AT&T did not see it as a business.

Asked for 20 MHz portion of spectrum – got the band they used for radar at SFO

Internet router added 3 boxes instead of one.

Kahn – communications with applied math and EE. Cerf – computer scientist. Background in Leiner paper. Collaborated on the development of TCP.

Connected 2 networks, but no documentation

Cerf — was DARPA PM, Director of IPTO

3-network demo was historic – 50 people made it happen

“A ride in the packet radio van [SRI] is a religious experience – it converts everyone who rides in it.” – Vint Cerf

D&R – demonstration and recovery (instead of R&D) when US starred generals would be taken for a ride in it

Irwin Jacobs involved in packet/satellite radio – challenges

  • Working with companies all around the world
  • Short packets, long packets, protocols to handle that
  • Fighting tdma issues. Cdma would solve a lot of those problem

Donald Nielson

Weren’t many codes in spread spectrum – only one (Kahn)

Ginny built the first gateways (routers) ever

  • Surprise is the Internet of today
  • In those days we were connecting networks at the rate of every six months
  • No one could ever envision
  • Installing gateways in Europe, Norway, etc. based on passport stamps. Didn’t realize how momentous it is going to be
  • Biggest challenge we had was low bandwidth and mismatched bandwidth (9.6 kbps to 56 kbps backbone, packets would fill up and drop)

Reasons for international participation

  • Resource sharing – computers, lines,
  • Command & Control – militaries will have to work with other countries militaries – hence international participation, requested approval from DARPA and got it

Cerf & Kahn on TCP/IP

  • No patents to prevent people from adopting its use
  • Not secret to make it a standard and make it as widely used as possible
  • Social processes for standards body was a challenge and new thing
  • Secure packet technology developed at NSA was classified so could not tell anyone about it – PKI was not available until much later

Given problems with security in IP, what you do if you could start over with IP v8?


  • End to end security and authentication so you know who you are talking to
  • 128 bit address space


  • Had we tried to solve security up front, we would not have invented the Internet

Universal Health Care in Rwanda

November 11, 2007

Dr. Paul Farmer spoke today at Stanford University about “scaling up” health care in Rwanda based on the health care delivery models that his organization “Partners in Health” developed in Haiti. Minus the funny anecdotes and striking visual examples he showed on his slides, a good summary of his talk can be seen in this newsletter in

Rwanda Scales Up PIH Model and also here.

The challenge is how to provide uniformly high quality health care across the entire nations. The government of Rwanda is basing their efforts on a set of ten principles listed in the link.


  1. High quality health care requires a truly comprehensive and integrated approach at all levels
  2. A comprehensive supply chain and procurement system should be in place for drugs, diagnostics and other commodities
  3. Patients should be able to access healthcare without regard to their ability to pay
  4. Healthcare workers should be highly trained and compensated fairly
  5. All health centers should have decent basic infrastructure and functional equipment to support the services they provide
  6. Community health workers are a vital component of the health system
  7. Nutrition is an essential element of any comprehensive health care service
  8. Information and communication technology should be integrated into existing health systems to improve the delivery of health care
  9. Health institutions should be held to the highest standards of care
  10. Comprehensive rural health care must go beyond the purely clinical by providing socio-economic support as well

Point 7 is simple — the best cure for hunger is food because patients can’t recover without proper nutrition. In his slides, Farmer mentioned that point 10 includes education, drinking water, jobs, and other essentials needed for living a decent, productive life. He did not mention microcredit or microlending. When I asked him about that he said they do microlending already, and it should probably be mentioned.

For point 8, he mentioned that the remote hospitals are equipped with satellite uplinks powered by solar panels and generators to communicate back to the PIH center in Boston.

Personally, I would add access to information and communication (the Internet) to this list in point 10 in addition to microcredit.

The common thread that emerged listening to him was the basic human need for “security” in the broadest sense of the word — health security, food security, water security, security from crime, national security, access to information, financial security, job security, personal safety, and so on. A functioning national economy is supposed to provide all this for most if not all of its people. Farmer talked about health security being a “right” versus a “commodity” — in some developed countries a small fraction of the population doesn’t have health security and in other developing countries most of the population doesn’t. I think the idea of listing all these things that together define “human” security needs would help us clarify what can be done to achieve them.

update on feb 13, 2009: compare to Clinics in the US, Expansion of Clinics Shapes Bush Legacy

NASHVILLE — Although the number of uninsured and the cost of coverage have ballooned under his watch, President Bush leaves office with a health care legacy in bricks and mortar: he has doubled federal financing for community health centers, enabling the creation or expansion of 1,297 clinics in medically underserved areas.

MIT hosts first ever Elevator Pitch Contest

November 1, 2007

MIT hosts first ever Elevator Pitch Contest

October 31, 2007

An elevator pitch, so called because it can be delivered in the time span of an elevator ride, is a brief overview of an idea for a product, service or project. More than 170 contestants from 12 different schools were given 60 seconds to pitch their idea, and winners were chosen representing different fields of innovation, including health care and life sciences, Internet and communications, products and services, energy/green technologies and development.

The winners included several members of the MIT and MIT Sloan community, with MIT sophomore Jessica Laviolette collecting the grand prize of $2,500 with a pitch on automotive production. “Painting an automobile is the most expensive and time-consuming portion of the car manufacturing process. Our patent-pending method takes one-tenth the time and one-twentieth the cost of current paint-drying technology used today,” she said.