Ask YC: Saving Inspiration
11 comments
The auto industry is fine, it's doing quite good. It's the American auto industry that has a problem - and this is a business problem, not really a technology problem.
If there were some dude making some way advanced beepers right now, what he has is not a technology problem, it's a business problem.
If there were some dude making some way advanced beepers right now, what he has is not a technology problem, it's a business problem.
I think prize economics is really interesting... though practically speaking, there's probably a limited class of problems that could be better solved by using prizes. (I think "the auto industry" will do just fine without any kind of prize.)
Check out these articles:
"The Bogus $1 Million Meat Prize" http://www.slate.com/id/2189693/
"Should the government start handing out prizes for science breakthroughs?"
http://www.slate.com/id/2182663/pagenum/2
Check out these articles:
"The Bogus $1 Million Meat Prize" http://www.slate.com/id/2189693/
"Should the government start handing out prizes for science breakthroughs?"
http://www.slate.com/id/2182663/pagenum/2
I think someone (Milton maybe?) proposed a solution to do this for pharmaceutical drugs. The companies are then computing for the drug discovery and as soon it's discovered it becomes open to anyone to manufacture.
This avoids pharmaceutical companies preventing the poorer people/countries from getting needed drugs.
This avoids pharmaceutical companies preventing the poorer people/countries from getting needed drugs.
Another example would be Google's Lunar X prize, and, of course, the X-Prize.
I have to admit that I never fully understood how the economics work out for the teams competing for a prize. Money for the R&D efforts of each team still has to come from somewhere, and only one team wins the prize to cover the costs. If the technology is so valuable, it makes more sense to me to attempt to take it to market rather than compete for a prize: there can be more than one winner in the market, and the potential payoff can be higher than the prize money.
Obviously, this doesn't work if there is no accessible market for the tech, or if the team is motivated by recognition rather than money.
I have to admit that I never fully understood how the economics work out for the teams competing for a prize. Money for the R&D efforts of each team still has to come from somewhere, and only one team wins the prize to cover the costs. If the technology is so valuable, it makes more sense to me to attempt to take it to market rather than compete for a prize: there can be more than one winner in the market, and the potential payoff can be higher than the prize money.
Obviously, this doesn't work if there is no accessible market for the tech, or if the team is motivated by recognition rather than money.
Nothing is stopping a losing team from taking their product to the market, Think of it as an added incentive to do it now rather than later.
Speaking in terms of time-value... We have a cost now, and a benefit in the future. If C> B/((1+r)^n) then a team will not try to design and create this product /* C = Cost, B = Expected Benefit, r is discount rate, n is number of years*/ Now, if we add a prize = P, it is possible that C<= (B+P)/((1+r)^n). Over the spread of many teams, this means that we are subsidizing additional competition to enter into the market by making available this prize.
Speaking less pedantically, if we make the prize high enough in addition to whatever rewards a team will get in the open market, then we give the team an incentive to take the gamble. Who knows, maybe you'll just win.
Speaking in terms of time-value... We have a cost now, and a benefit in the future. If C> B/((1+r)^n) then a team will not try to design and create this product /* C = Cost, B = Expected Benefit, r is discount rate, n is number of years*/ Now, if we add a prize = P, it is possible that C<= (B+P)/((1+r)^n). Over the spread of many teams, this means that we are subsidizing additional competition to enter into the market by making available this prize.
Speaking less pedantically, if we make the prize high enough in addition to whatever rewards a team will get in the open market, then we give the team an incentive to take the gamble. Who knows, maybe you'll just win.
I hate to break it to you but the current record for hydrogen powered cars is over 5000km/litre and for combustion engines it's over 1200km/litre.
The auto industry has some of the best engineers in the world and are more than capable of offering high efficiency vehicles but it's not in their best interests to do so. Frequent vehicle replacements and high energy consumption are in the industies favour.
http://www.canadiandriver.com/thenews/2008/04/13/fuel-econom...
http://www.uni-protokolle.de/nachrichten/id/102185/
The auto industry has some of the best engineers in the world and are more than capable of offering high efficiency vehicles but it's not in their best interests to do so. Frequent vehicle replacements and high energy consumption are in the industies favour.
http://www.canadiandriver.com/thenews/2008/04/13/fuel-econom...
http://www.uni-protokolle.de/nachrichten/id/102185/
A friend of mine runs innocentive.com, which is a smaller-scale instance of this sort of thing.
Justin - fascinating idea mate. I'm a huge fan of original resource, seeing pretty raw stuff before it gets "popularized" and... well, you know how it goes.
If you'd be so kind, could you point me to a couple places to start learning about prize economics? I'll check here, and my email is contactsebastian -at- gmail.com. Anyone else from HN is welcome to shoot me a line too, I feel like I'm with people I like in my living room when I'm on this site.
If you'd be so kind, could you point me to a couple places to start learning about prize economics? I'll check here, and my email is contactsebastian -at- gmail.com. Anyone else from HN is welcome to shoot me a line too, I feel like I'm with people I like in my living room when I'm on this site.
A while back I did some research on prize economics, and in general there is a 100x advantage to using a prize economic model on BIG PROBLEMS.
Right now the auto industry is facing a crisis because cars are expensive to manufacture, there are delays in manufacture and credit approval, and gas really put the pinch on us this summer.
My recommendation is that the car manufactures put out MASSIVE prizes for improvements that we value as a country. Heck we could even vote for what we as a country want to give prizes away for.
One example might be a prize of $1 billion dollars to the 1st car company that can travel cross country in 40 hours on 40 gallons of gas. The prize money would be dolled out over time, and at each stage require the car company to pass certain mile stones.
When Netflix started the Netflix prize they were unsure of who would participate, the goal was to get 10% better recommendations than Netflix did in house and the prize was 1 million dollars or 8 full time software engineers for 1 year. The result is that there are thousands of people competing for the prize and glory of wining. WIthin the 1st month they were 5.5% better than Netflix, but today people are only 8.8% better than Netflix's algorithm 3 years ago. With thousands of people working on it across the globe, we have all learned a few interesting things.
Darpa did the same for self driving robotic cars. They put up $1 million in the 1st year and had about 20 teams compete each spending about $100k-$1.5 million to win, no one one, but in the second year, Darpa vowed to add $1 million to the prize each year until someone won. In year two 12 companies completed the course and 1 group from Stanford took home the prize and a government contract. When the head of Darpa was questioned, he thought it was a bargain, because a government contract would have cost over $200 million dollars to accomplish the same thing.
Today we are faced with big problems that need to be solved in the auto industry.
Accident and death rates are so high, that for people in their 20's and 30's car accidents put people at the greatest change for injury.
US cars have low MPG ratings when the technology exists to have cars in the 50+ MPG range and getting to 100+ MPG should be viable in 1 or two years.
We have other industries that could use a boost from prizes, like health care with cancer, aids, heart disease, and diabetes.
These problems are so big that capitol markets have had problems solving them because engineering teams need to 1st think about budgets and business models before they think about solutions. Prize economics makes it noble and profitable to give the technology away to win the prize.
Instead of giving away charity, lets inspire people with a prize worth awarding, and lets get a gift worth sharing.
Justin
please pass this on.