Learning from the Chinese government

I was recently in China - when going through immigration you get to rate the immigration officer. There are four buttons ranging from smiley face to sad face where you can provide feedback as to the level of service you received. I don't know what they do with the data, but I clearly the immigration officer is aware of the feedback mechanism and probably does a better / nicer job as a result.

After we sold Core Mobility, I forgot to update the DMV of my address and the registration on my Tesla expired last December. A nice officer in Palo Alto reminded me of that with a ticket, and then two days later, I'm pulled over in Los Gatos for the same thing! Nuts! Four months of daily driving with no issue, and then two tickets in a week! Sure would have been nice to get those tickets a few months ago since the late fees alone on the registration renewal were over $500!

I told the second officer that I already had a ticket, but he said he is unable to verify that. Strike one! Let's see, in today's connected world they can't look up whether I received a ticket? I sure hope the systems for finding real criminals are better than that!

Anyway, I pay the $1500 for the registration and send all the info to the traffic court hoping it ends the matter. Unfortunately, they don't get an automatic notification after the registration is up to date, even though I have to both pay for the registration and have another officer sign off on the ticket.

The ticket was issued April 2nd and I mailed the proof of registration to the traffic court on April 7th, just before leaving for China.

When I got back, I found the following in my snail mail (address redacted to protect the guilty!):

More than a few things wrong with this picture!
1) The date on the letter is February 1, 2011. That's off by over 2 months!
2) The letter refers to a citation from April 12, 2011. From the point of view of when they got this, that's in the future! The date on the citation is April 2, 2011.
3) The letter head is a template that is not filled out. For example, Branch Name shows as FACILTY_NAME
4) The letter says that I may inquire about the status 45 days after the date on the letter. The date on the letter is incorrect, and 45 days after the time I received this back is after I am supposed to appear in court!

Our tax dollars at work. As Bill Gates said in his TED talk: State Budgets are huge expenditures that undergo very little scrutiny. Unfortunately, the same is true for other areas of state operations.

I'd really like to have a feedback mechanism like the Chinese immigration department has!

And, of course, don't forget to use speech to text transcription services from Speechpad!

The New Office

My new company, Speechpad, just moved into a cool new office. We have a security guard (pictured) straight from Thailand who keeps us safe 24x7. Mike Scanlin, CEO of Born To Sell, is right down the hall and has been advising us on SEO techniques to help drive organic traffic to our site via search engine results. Born To Sell focuses on strategies for selling covered calls - it's a super cool site if you are interested in selling covered calls.

If you haven't seen Speechpad, please check it out. We use crowd-sourced techniques to deliver high-quality, low-cost transcriptions quickly. We've been working on the company for just over two years and have been successfully transcribing financial calls and recorded statements for insurance for most of that time.

Our current focus is on meeting transcriptions, a new market enabled by easy audio acquisition through phones. Please check us out - if you have any questions, don't hesitate to contact me!

Your help needed: who is this?

How much information is in a picture? What if you had 1000 people interpret the picture, creating that information? How about just finding out a few key pieces of information?

I once posted a "how-to" kiting video shot in Maui with an unknown rider, and heard back from the rider within a few days! That was an easy one, though - there just aren't that many kiters on Maui, and the community is closely linked. 

While on a boat in the Maldives, were were discussing how quickly we could track down someone in a picture that has a much broader search domain, and a bet evolved.

Setup: The pictures in this post were in the background of another picture (that included a kiter, of course!). They have been blown up and enhanced to better make out the relevant details. They were shot with a high resolution camera that included GPS tagging, so we know the time the picture was taken (5:07:05 PM local time - shortly before sunset), the location where it was taken (3 52.8661'0"N, 73 27.3889'0"E at sea level) and, of course, that they were taken in high resolution (4256 x 2832) so we could pull out the background.

We were discussing the power of crowd sourcing and the power and diversity of the crowd. Eventually a bet evolved: how long will it take to identify a dive shop given only some pictures? This boat is literally on the other side of the world from the US, in the South Male Atoll in the Maldives.

Here's what we know:

The pictures show the boat registration number and we have a blown up picture of presumably the dive master.

The satellite map is tagged with where the boat was seen in the South Male Atoll based on the GPS coordinates in the picture. The boat was heading north at the time and it looks like it's returning from a dive trip.



There is an MTurk job that points to this post and the associated geo-stamped pictures - assuming that information was kept in the upload - (we'll know soon enough - but the satellite picture and the coordinates in this post provide everything we need).

Cash for Clunkers: A Good Deal?

The cash for clunkers program cost tax payers $3 billion. In exchange, 700,000 Americans were able to get a new car at below market rates. Was this a good deal? Are there better ways to allocate this capital?

When making a cash outlay, one of the first questions should be to understand whether the money is being invested or spent. Expenses are things such as movies or dinners which have no residual economic value. Investments are things such as houses or stocks which generally have substantial residual, and even increasing, economic value. Cars are an interesting hybrid as they have residual value, but unless it becomes a collectors item, depreciates quickly and if not fully realized as an expense now, will become an expense as it’s fully depreciated over future years.

Providing discounted cars to a subset of Americans is something that benefits the few and has small, if negligible, ongoing economic value to the larger populace. The program was justified through the benefit of less pollution and better energy efficiency which benefits everyone, as well as spurring more auto sales to help struggling US auto makers.

If these are our goals: improve energy efficiency, reduce pollution, and help US auto makers, is there a better way to spend $3 billion? Rather than a one-time stimulus that benefits the few (one in 500), is there a program that can have on-going benefits for all Americans?

Tesla Motors, a US manufacturer of electric vehicles, recently announced installing five charging stations between San Francisco and LA so that their flagship car, the Tesla Roadster which has a 250 mile range, can make the trip. Total cost for these 5 charging stations: about $10000 each on average.

The energy dispensed by the charging station is free, justified by local merchants who foot the bill by driving traffic to their establishments. Roadster owners, who stop to charge their cars for three hours, will likely engage in commerce near the charging station. The cost of the electricity to charge the Roadster is around $2, so the merchants are paying $2 to have a potential customer in the vicinity for several hours looking for something to do.

Suppose we used the $3B for the clunkers program and built out electric charging stations instead? Could that achieve our goals of improving energy efficiency, reducing pollution, and helping US auto makers?

There are around 110,000 gas stations in the US. At the current price of $10,000 per charging station, we could install about 30,000 charging stations around the US. Assuming economies of scale, it’s not hard to imagine that 100,000 or more charging stations could be installed for the price of the clunkers program.

If we assume 50% or more of the cost of installing a charging station is labor, such a program would create a number of jobs, directly helping unemployment.

What we are left with is 50,000 to 100,000 free charging stations, subsidized by local merchants, to drive traffic to their stores. But what good does this do? Very few people can afford a Tesla!

This is actually where the large benefits take place. Over the last year, I have seen numerous companies that are building electric vehicles: bikes, motorcycles, and cars. In fact, there are over 100 companies that have been started to just build electric cars. With free charging stations covering the US, consumers will be very interested buying a new electric vehicle that takes advantage of this handout. And a number of entrepreneurs will be focused on building vehicles that can exploit the new grid, from a consumption as well as generation and storage standpoints. In short, we have created a new innovation platform.

But how do automakers benefit? Here we can look at other industries that have undergone a large technology leap, and examine what has happened. Recently, mobile phones went from being able to simply make basic calls to having cameras on them. This caused a rapid upgrade cycle, moving the average time a consumer keeps his phone from 24 months down to 18. The widespread availability of wireless data networks and high quality phone software is now causing another upgrade cycle to devices such as the iPhone and Android.

Whether these upgrades are captured by the existing players or new players depends on how well the companies are run, but in any case it’s a huge boon for the players who can get it right. Just look at Apple’s stock price or the buzz Google is getting for its Android operating system.

With this approach, for the same $3B, we would have:
• a national charging grid that enables electric vehicles to be charged at as many locations as the corner gas station
• an opportunity for merchants to drive traffic to their stores by subsidizing the energy using to charge the vehicles
• a massive upgrade cycle as entrepreneurs and consumers find innovative ways to use the electric grid
• a spike in employment as factories, stores, and other infrastructure are upgraded and built to fulfill the demand for electric vehicles
• much cleaner environment, as electric power is substantially cleaner than the internal combustion engine
• most likely, safer roads, as electric vehicles tend to be lighter and built out of modern materials
• drive innovation in battery and grid storage technologies, reducing costs and further increasing efficiency
• a tremendous opportunity for existing as well as new vehicle manufacturers to fill the demand to drive profits and employment

Rather than allocate another $3B, this program could be paid for via a national gas tax of two cents/gallon. This would further incentivize consumers to move toward electric vehicles since it would further pressure operating cost differences between gas and electricity, and bring prices closer in line with true costs, when deficits, national defense and pollution are taken into account.

Overall, upgrading the grid is likely a much better way to invest tax payer dollars and reap long term rewards compared to subsidizing car purchases for Americans who own “clunkers”.

NOTES:
From: http://wiki.answers.com/Q/How_many_gallons_of_gas_does_america_use_a_year
US uses 9.3M barrels a day
42 gallons of gas per barrel

From: http://www.atr.org/presidential-math-cash-clunkers-spent-billion-a3816#
Removing clunkers will save 5M barrels a year

From: http://www.wired.com/autopia/2009/09/charging-corridor/
Chargers cost 7-12k each
Subsidized by local merchants

John Mauldin 12/4/2009 newsletter:
All we did with this "Cash for Clunkers" thing was move cars forward that would have been bought later. You're not increasing sales down the road. Yeah, you're taking cars off the road and spare parts and stuff, but I think it's kind of a silly investment in dollars. But, what's $3 billion when we're wasting a trillion here and a trillion there? Still, it's disappointing.

Bruce Leak 12/9/2009 email:
Politics isn't about efficient use of capital. It is about special interest groups and playing to populace emotions. Subsidizing "green" car purchases pleases car manufacturers, suppliers, finance institutions, labor unions, dealers, environmentalists, local and state tax authorities, scrap recyclers and consumers. What's not to like?

When we're not driving autorickshaws

It's been a while since I've paid any attention to this blog and haven't really figured out a good reason to justify the time it takes to blog until I talked to Eric Ries who writes the blog at http://www.startuplessonslearned.com/. He said it's a great way to get feedback on your ideas, and immediately learn what was communicated clearly and what is up for debate. He mentioned he'd learned a lot from writing the blog and getting feedback.

Anyway, I'll try it and see what happens, or if anyone cares. Maybe I'll care in 10 years when I can go back and see what I was thinking and where I went wrong.

After riding around in an Autorickshaw for nearly 1000 miles, I good transition would be to talk about the Tesla, a car that's on the opposite end of the spectrum from a performance and pollution standpoint. When I first learned about the battery pack, I had trouble relating the battery consumption to things I was familiar with, so I did a few calculations that made it more clear to me, at least.

Like all cars, Tesla’s mileage varies by how hard you drive; “normal driving” (ie, the way the EPA measures gas mileage for cars), the Tesla requires 334 watt hours to go a mile. That means that one kilowatt hour will propel the Tesla about 3 miles. Hair dryers are generally 1000 – 1500 watts, so you have a choice: you can dry your hair for an hour, are drive the Tesla for 3-5 miles.

My Dell laptop battery says it stores 43 watt hours. That’s enough energy to move the Tesla about 700 feet, or, if I’m going 60mph, power the engine for about 8.5 seconds.

The Tesla battery pack holds 57 kilowatt hours, so it’s equivalent to about 1300 Dell laptop batteries. Cross checking, that says that we can power the car for 1300*8.5 seconds or 11,200 seconds, or just over 3 hours, for a battery pack life of about 200 miles.

Some other interesting data points: a rechargeable AA battery is rated at about 2000mAh. These rechargeable batteries are typically about 1.3 volts (slightly less than the standard 1.5 volts), for a total capacity of 2.6 watt hours. So about 18 rechargeable AA are equivalent to a Dell laptop battery, which seems about right judging by the relative sizes. And that AA battery will power the Tesla for about .46 seconds (just under ½ second) at 60 mph, moving it about 44.5 feet; it takes about 128 AA batteries to move the Tesla a mile.

Or, in dollar terms:

• Off peak charging is 5 cents/kilowatt hour which is about 3 miles, just under 2 cents per mile for the energy cost. NOTE: the battery cost is additional. The Tesla battery lasts about 100,000 miles and currently costs $25,000 for an additional 25 cents per mile in battery cost.
• AA batteries are about 50 cents, which means it would cost about $60 to go a mile (including battery costs) if you bought your energy at a convenience store.

Kuttalam to Kanyakumari

Check out today's video! Be sure to click the "watch in high quality" link at the bottom right! The final day opens with a sheet full of challenges and Aravind’s briefing “you can not complete them all. Choose wisely.” The instructions are on both sides of the sheet, and quite chaotic even by ARC standards. There is no official flag off or flag down – we are free to leave anytime. Inspired, we decide to end on a high note and resolve to complete all the tasks – just to prove that we can complete them all. We figure out our first destination and are the second team on the road. When we stop to ask directions, we are pointed on our way. Giamp did not trust the directions we were given, so asked again. And then again. Every 100 meters or so. Always the same answer. After the third time I was getting annoyed at stopping all the time but Giamp said he still didn’t trust the directions. Stops four and five confirmed our direction was correct yet again. And again. Our sixth stop was shortly later when the clutch stopped working right next to a giant trash heap and a lot of flies.

When the mechanics show up we find out we are of course, on the wrong road and the clutch repair (2^nd in as many days) would take about 45 minutes. All was not lost because this gave us 45 minutes to decipher the day’s instructions, complete with eight GPS waypoints. But were they accurate? Were the challenges in chronological order? The repair finished just as we completed our driving plan for the day: even with the 45 minute handicap, we’re going to attempt every challenge unless we can’t get to the waterfall by 3:30, in which case we’ll skip the waterfall – the one “must see” item according to Arivind. A classic “see India” vs. “win race” tradeoff, though there is also a “survival” tradeoff as anyone who has driven at night without a headlight on potholed streets can tell you.

As it turns out, the GPS waypoints are accurate but interpreting them is problematic. At one point the road goes left or right, and the next waypoint is 5km to the left. We decide to turn left and in 1km come to a parking lot for a temple. The waypoint does indeed signify a temple, and there is a path going in the correct direction. Just a short 4km hike up the hill?

I decide to run the path for a bit to see if anything makes sense. I run into another team which provides very cryptic answers to my questions about whether there is a temple ahead. Finally, I decide to turn around but a local says: “temple, up around the corner”. OK. I’ll keep running. About ½ km later I finally come to a temple but there is no Shiva. Wrong temple! Giamp is having a field day telling me how useless the GPS, and all technology is. I keep trying to explain the problem is not the technology, but the interpretation of the information, and it is in fact me that is useless. I guess reading the sheet, and asking locals can work better than a GPS.

We take the road the other direction and sure enough, exactly as described, we find the correct temple. Then it’s a question of whether we go up the hill another 18km to get to a dam, or whether we continue. We run into the Hanumaniacs Team and they decide to skip the dam. We decide to go for it. Fortunately, the 18km was exaggerated – it was about 5km more. The directions out are the same as the directions in, and we are surprised by how few teams we encounter. In the 56km round trip, we should pretty much run into every team, but we saw only four. Maybe they all skipped this detour and went straight to the waterfall?

When we have to make our waterfall go-nogo decision, it’s right at 3:30, our cutoff time. We decide to go for it, and are treated to an awesome sight complete with a monkey that tried to hijack our rickshaw when we were at the falls. Two tasks left, and they’re both on the main path, and we are done! It’s beginning to look like a triumphal march into town.

Of course, with 50km to go, and a straight shot to Kanyakumari, our clutch goes again. We break down in a very rural area, and one of the homeowners comes out and invites us in for food and drink. We are too flustered and decide to wait roadside for the mechanics. 45 minutes later, we’re on the road again but it’s getting dark. The mechanics van is following us, and when we pull into a petrol station, they tune up the engine a bit. After that little tune-up, our top speed when from 50km to over 56km. If we ever do this again, I’m going to learn to tune an engine!

We arrive in Kanyakumari about 6:45, just after the 6:30 time when the judges start to calculate the results and the winner. We figure we’re in fine shape since we did all the challenges, and being just 15 minutes late should be well within any reasonable grace period.

At night there is an awards ceremony, complete with radio, newspaper, and TV reporters. Each team gives a 3 minute speech and then the organizers give awards to the top 8 teams. After 7 days, over 1400km, 3 clutches, two tires, a gear box, two muffler repairs, 28 teams from 12 countries, and over 65 participants, we’re proud to say Team Namascar was awarded the hotly contested and highly prized 2007-2008 Indian Autorickshaw Challenge World Championship. WORLD championship. J

Tutticorin to Kuttalam - January 6

Lack of internet coverage means I have not checked the stock market all week. That’s a record for me…and I’m quite curious what has happened on Wall Street. Has the sub-prime battle with inflation crushed the market or is it still bubbling along. I hardly care. We have our daily rickshaw vs. the animal and mechanized kingdom battle as can be seen in today's video.

The morning starts with the “International Pongol Festival” at Aravind’s farm. We’re divided into five teams each of which, with the help of a local family, has to make a batch of Pongol from water, rice, sugar cane, cashews, spices, and a live chicken. As Pongol is a sweet, vegetarian dish, the chicken appears to be there only to lend moral support. We are to be judged on speed and quality, though our primary purpose is to look ridiculous. With the men wearing dhotis, the women wearing Sari’s and the SS Music crew chasing everyone around we make the Indian equivalent of an MTV special.

The recipe starts with making fire to get the water boiling; we’re doing great until the gods take our pot. Water spills everywhere and our fire is out. We elect to focus on quality, and make some darn good Pongol using some “borrowed” ingredients from other teams since we were a bit short due to the hi-jinx with the first batch. After a formal offering to the gods and some cows wearing robes, our Pongol is rated 2^nd by a single point. The winners received two bunches of miniature bananas – we received one bunch. And for anyone who likes cashews, as Giamp does – see yesterday’s video – our Pongol was clearly superior as the winning team didn’t use them.

As all the teams are getting ready to go we discover our rickshaw's clutch is broken. The mechanics are right there, but this is a serious operation. Half-way through we discover the gearbox is broken as well. Arivind’s dad entertains us with coconut drinks and food as the mechanics repair the car. We start about an hour later, but quickly make up time as there are no challenges and it’s now Saturday, so again there is no school visit. As we roll into the hotel in Kuttralam, one of our tires deflates. For a day with two potentially serious problems, they could not have happened at better times. Our secret prayers to Ganesha – the remover of obstacles - are paying huge dividends.

For the adventurous among us, dinner is at a local truck stop and our transportation is the luggage truck. We’re shipped there in the dark container like a bunch of illegals and their cruel coyote. After surviving the fast food restaurant on the way to Rameswaram, we were feeling confident, but here we are taken to a whole new level. As usual, the meal starts with a banana leaf with water splashed on it, but then a fat, greasy, sweating cook comes round with pots of very spicy roasted chickens and quails. He digs around the pots with his bare hands and tosses them onto your banana leaf. Need more squab? No problem, I just happen to have a handful here. There’s no ordering. No asking. No communication whatsoever, probably because you can’t talk over all the noise. Other servers come by with paratha, vegetables, rice, etc. Just food coming at your right and left. The meal ends when you fold your leaf over – then no one can put more on your “plate”. Super efficient, and the food was delicious. The mayhem was a sight to see, as you can do for yourself in today’s video. If someone ran a petrol station like this, they’d be millionaires.