Android & Google Maps

Came across this announcement about Google Maps on Android:
http://googlemobile.blogspot.com/2009/10/announcing-google-maps-navigation-for.html

I thought this is really cool given it is bringing the power of Cloud Computing to everybody's palm. Most important of all, this is free to the end consumer. I really hope the service works, and hope Google invests enough money to make it available to everybody in the world.

I have a Garmin in my car, but I always felt Google Maps are much better:

• Information is more accurate and up-to-date (Garmin/Navteq has 1000 employees to collect map data, Google has millions - advantage of crowd-sourcing)
• User-interface (Garmin is pretty bad here - no word-completions, no good search interface)
• Computing power (You can never beat the cloud - it takes 5 minutes to zoom the picture on Garmin)
• Cost - I paid $400 for my Garmin, have to update the data every year and more if I want POI data

If I can get more accurate information for free - I don't see a reason why I should buy another GPS device. Effectively Google is killing the entire auto/mobile GPS industry (>$10billion?) with this one announcement. Garmin makes 70% of its revenues from auto segment - and I am sure it will see some rapid declines there. You should feel for Nokia - it bought Navteq for $8billion in 2007 (on a company that was making ~$500 mil in annual revenues) - I can bet Navteq will be worth lot less than $8B in the coming years.

Overall, I think this is a significant event to the mobile industry, and very good for the world -  it redefines mobility and it will have a similar effect that email had during the 1990s. It would have taken decades for this democratization of location data - if we left it to the current GPS vendors.

This service can be a very good move from Google's point of view:

• More android based devices in the market, and more users for Google Maps
• Better map accuracy (crowd sourcing) than competing vendors
• More ad-revenues for Google (they now know where you are in addition to what you think)
• More cloud-based services for the mobile users from Google (economies of scale & scope for Google)

Of course, I am discounting the fact that Google will become the super cop - and will know every bit about you: what you think (your search terms), all your favorite locations (including where your friends live), all the businesses you visit etc. But again, most telecom companies already track lot more information about you (how much time are you on phone/online/watching-TV/resting, which urls you visit, whom do you call & with what frequency, who are your friends etc.), but never give you anything for free in return. So - not much to complain there.

On a philosophical note, it makes you wonder what will happen to humans in another 100 years: everybody follows directions from one of these GPS clouds, and there is a possibility we completely lose our map reading skills - the genetic code that our ancestors so patiently built over the past 100,000 years.

I can imagine a couple of science-fiction movies on this theme: somebody hacking the GPS cloud and making drones out of humanity, all GPS systems destroyed by terrorists causing mayhem everywhere (airlines/marine/auto navigation systems all driven by GPS), far into the future - humans lose their dominant species status to rats - because of the poor map-reading skills. If these movies are already made, pardon my ignorance.

I had Garmin for the past 4 years, and  I am not proud to admit that I am addicted to it - I just follow the directions given out my Garmin, and don't care to remember if I visited the place a hundred times before.

Running, Evolution & Health

After I heard about persistent hunting, and how early humans used their ability to regulate body temperatures (through sweating) as a competitive advantage in hunting; I thought running is part of our genetic code, and that every one of us can run long distances.

I also believed that running and exercise are good for health, and I used to spend 2 hours a day in the gym burning 800 to 900 calories a day (4 miles of running, 6 to 10 miles of cycling, and 30 minutes of weight-lifting - four days a week). I used to feel good about my health during these periods of exercise (possible runner's rush). It has been a while since I ran that long, and feel guilty these days for bowing to worldly pleasures and not going to the gym.

But, I came across a couple of articles over the past few weeks that made me re-think about both of these assumptions, and gave me a reason to maintain the statusquo (running vs. sitting on the couch - it is an easy choice I guess :)) :

Here are the data points I am talking about:

• Recent deaths of marathoners: Detroit (3 deaths), San Jose(2 deaths), NY (2 deaths) - in all the these incidents, the runners who died went through rigorous training for the event. While 7 deaths is a very small percentage of the millions who run marathons (statistically insignificant to draw any conclusions), it still is something to think about (psychological impact) for a would-be marathon runner. (Imagine a gamble with the following payoffs - you will win $1 mil 99.9% of the time,  you will lose & die 0.1% of the time - will you play that game?)
• A blog in NYT (14th Oct, 2009) talks about how intense exercise reduces our immunity: The article cites studies done by multiple research groups about the relationship between exercise and immunity - and concludes that intense exercise (defined as a workout or race of an hour or more during which your heart rate and respiration soar and you feel as if you are working hard) can significantly reduce your immunity; and that you are better off doing something productive (like watching TV) than running.

I guess - after thousands of years of relative-sedentary life style of the agrarian societies, running is no more an evolutionary in-thing; and it is possible mate-selection is not happening based on good running abilities.

Homeless People in Bay Area

Came across an interesting fact this afternoon, and I thought I should blog about it. A co-worker tells me about an article in the Time magazine about HomeLess in America. He says there are more than thousand homeless in San Jose alone, and similar number in San Franciso.  I didn't believe him, and started searching for the right numbers.

I came across this article in SF Chronicle that counts homeless in SFO alone at: 6500.

http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2009/04/04/BAG416SGRA.DTL

Another site that had more info:

http://www.endhomelessness.org/section/news/press/2009countsmap

Came across this article on wikipedia that counts the total US homeless population at 1% of the total population (~3mil), and of these 23% are war veterans:  http://en.wikipedia.org/wiki/Homelessness_in_the_United_States

Just for comparison, this number (3 million) is about 

• 30,000 times more than the number of people who died from H1N1 virus (swine flu) in US but it still receives very little media attention
  
• 60 times more than the number of people who die from common flu every year in US (~50,000), yet there is no flu-vaccine like drive to prevent this 
• 60 times more than the number of road fatalities in US (~50,000), but still doesn't have a trillion dollar insurance like industry to safe-guard

It is sad to see these numbers - the richest country in the world not being able to provide for its citizens - especially for veterans who fought for the country & mentally disabled who can't fend themselves. But again, we are living in a market-driven economy, and we believe in Darwin's theory of survival of the fittest - or should we?

Food for the world

Long back when I was in Business School, I put forward a simple hypothesis to some of my classmates about how world food prices are going to fall in the next 50 to 100 years.

The reasons I cited were simple: 

• with the advances we have been making over the past few decades (computers,  global commerce & what not), we will find better & efficient ways to produce, store & distribute food over the next few decades  (if we can double PC speeds every year, how much time does it take to double world food production ;))
• there are still untapped food production sources that we currently don't know of (GM crops, arrid lands in Siberia/Africa and what not)
• in the near-term: India & China (~38% of the world population) - will be investing heavily in food production as their populations become richer

There are two data points that helped my argument at the time:

• the gradual increase in world cereal production - rice, wheat & maize - over the past few decades
• the way food prices have gone down in US because of improved food production, storage & distribution (restaurants like McDonalds' & BurgerKings were able to offer buck a meal, and still make a buck because of scale economies).
  

I hypothesized that - in the distant future - world will evolve to a place where food will not be a main concern - and that food production will outpace population growth (the exact opposite of what Malthus predicted).

A couple of my classmates laughed at my theory - and asked me to watch out for food prices over the next decade before I come to any conclusions. And looking at the markets now, their caution does sound right. There has been a dramatic increase in the food prices over the past few years, and it is mostly effecting the poorest of the world. 

The emergence of India & China, and the resulting hunger for food & oil - seems to have put lot more pressure  on the demand side of the equation - than the supply side for both food & oil. Some of the new technologies (GM Foods) are facing lot of resistance from the civic bodies in the emerging world. Rise of the middle-class is effecting the arable land - people are looking for bigger and better houses, and urban sprawl is becoming an issue.

I really hope this is a short-term predicament for the world, and that we will survive this imbalance without any major famines, and will take a few steps towards the Shangrila I was dreaming of in the next couple of decades.

Elections and Knowledge of the masses

Elections and the accompanying results always amaze me. 

In the evolutionary game - can't believe somebody thought about giving everybody a piece of paper, and asking for their opinion on who should be their leader. Somewhere somebody tried this,  it worked and viola - we are in a world where a lot of countries boast about being democratic by passing around pieces of paper every 5 or 6 years. I am sure this is a very recent evolutionary trend, and yet to be seen if it will survive the tides of time.  If it indeed is an optimal solution, it is unclear why we haven't seen many public corporations conducting elections of employees to decide their CEO (..voting among share-holders is not necessarily same).

We have to assume a few things for democracy to work:

1. there is an easy way to conduct fair elections
2. participating voters are rational 
3. voters think as a collective, and vote for the long-terms interests of this collective (goes back to point 2 above)
4. Assumption that majority opinion is the best opinion

Unfortunately, lot of these assumptions don't hold true for most of the elections.

When some of these assumptions go wrong, elections are: 

• won by leaders who buy votes (literally distribute money and buy votes or introduce welfare programs that appeal to the majority), intimidate voters, create divisions in the society (religion, color, class, caste come into play)
  
• won by a small section of the society who know how to engineer the public opinion (The Clintons, Bushes, Gandhis of the world)
• won by people who know how to engineer the voting machinery (hijack polling booths, rig elections, manipulate people through newspapers/radio-tv stations)

And I clearly see some of these being played out in the recent election of the biggest democracy in the world. 

I also think lot of current democracies in the world are ruled by a few select political families - and the illuminati conspiracy theories may not too far from the truth.  It will be a  good topic for a social scientist to look at the facts and prove this wrong. 

Here are a few examples I have in mind: 

• Nehru-Gandhi family ruling India for the ~50 years of since its 60 years of independence. The fourth generation of Gandhi's (Rahul & Priyanka) are waiting on the sidelines now. Independent India - under this family - became one of the most corrupt nations in the world, and is still one of the poorest in the world
• Bhutto's ruled Pakistan for most of its democractic existence (even though democracy came in patches to Pakistan) - now 3rd generation claiming their power
• The others include Bushes, Clintons, Kennedies in US, Sukarnos in Indonesia, Rahmans in  Bangladesh - and we covered more than 60% of the world population.

While having these strong political families bring some semblance of stability to the democracy of these countries, it is surprising how these political families kept their grip on power over such periods  in democratic elections (>50 years, 4 to 5 generations) - even after their sub-par governance. Possible, humans are genetically wired to follow - and only few understood how to capitalize this inherent human characterstic. Possible - all that literature on leadership - is a waste of paper it was written on.

I am not suggesting that democracy is bad - but I feel elections need to  evolve to reduce some of these issues - or in the long-term, it can affect the human evolution (a bad leader for China or India can make zombies out of a huge percentage of the human-race, and can have a significant impact on the human evolution).

If I have to write the constitution for a country, at the least I would like the following codes for an  election:

• Make sure only people who know the value of their vote can participate in an election
• If you get elected, nobody in your immediate family (sons, daughters, spouses, parents) can contest for elections (for some reasonable time-limit)
• You can't contest an election if you already won the past two elections
• People who can engineer polls through power, intimidation, money,  etc. can't contest elections (way too many rich people & criminals own the democracy these days - and another trend is to buy newspapers/tv-radio stations for coming to power)

Wolfram Alpha

I tried the wolfram alpha search engine (www.wolframalpha.com) this morning, and the marketing video was really cool - it blew me away. I used to do data analysis as a marketeer, and I understand what we can glean from structured data. But extending it to every structured data out there, providing a natural language interface and giving it away for free to the citizens of the world sounded like an incredible leap for the man-kind. 

I tried a couple of searches with Google & wolframalpha for population data (mobile phone penetraion in China), and wolframalpha won it hands-down.

Using goolge, I have to

• visit multiple sites before, 
• read through some junk, 
• kill some annoying pop-up ads, 
• scroll down to get to the information, and 
• finally interpreting what the author is trying to say

I tried a couple of searches that are tough to fit in a structured format (e.g: paracetamol dosage, flu incidence in US), and Google is much better here (wolframalpha turns up blank page for these queries).

On the negative side:

Wolfram still has to add lot more information to their database, and the natural language query is not very user-friendly. Lot of my queries turned blank pages on wolframalpha, and I found it difficult to create queries for simple questions.

Overall, wolframalpha introduces a great new way of looking at search algorithms - it is not sufficient to present all the webpages that match a particular string, but what is more useful is intrepreting these pages, collecting facts, analyzing them, building a database,  and presenting them in a useable format.

I really hope Google works on something similar, and integrates it with their current search. I feel Google can do a much better job at this given its strong financial resources, and vast data that they analyze.