4 ms·

Einstein's annus mirabilis was 1905 and it took decades of determined, international promotion of the Einstein brand, World War 2, a letter from Einstein to Roo

Einstein's annus mirabilis was 1905 and it took decades of determined, international promotion of the Einstein brand, World War 2, a letter from Einstein to Roosevelt, plus another 6 years and 33 billion 1945 dollars (473B USD today), and a staggering collective of the smartest minds to ever walk the Earth in an existential race against a mortal enemy to develop nuclear weapons. NASA's shuttle project started in 1969, cost $200B, was shut down in 2011 and never got close to it's planned 50 launches per year. This stuff is hard.

tachyonbeam 7y ago
I have a book from 1982 that predicts fully self-driving cars by 1995.
dx034 7y ago
But both of these were mostly science problems. Self driving cars are about making machines interact with humans. That has never been solved before. Self driving in itself wouldn't be an issue, it's the other humans on the road that make it so hard to solve. One of the main problems for manufacturers is that we have never before solved a similar problem, so there's nothing to compare it to.
- asdff 7y ago
  >it's the other humans on the road that make it so hard to solve. Humans, cats and dogs, deer, tumbling rocks, low flying seagulls, tumbling trashbags or a toppled trashbin bin, palm fronds, drunk people jaywalking, drunk guy in a jeter jersey directing traffic after a yankees game, edgy teenagers pulling the invisible rope trick (or just yanking a sheet on a string), puddles reflecting humans on the sidewalk, blind turns and hidden driveways, competitors self driving tech, the metawars that will result by trying to raise a competitors accident rates (and not doing that would be leaving money on the table).
beefield 7y ago
> This stuff is hard. Luckily we attract our best minds with highest salaries to solve the pressing problem how to get people click more ads. (Yes, sarcasm.)

tachyonbeam 7y ago

I have a book from 1982 that predicts fully self-driving cars by 1995.

dx034 7y ago

But both of these were mostly science problems. Self driving cars are about making machines interact with humans. That has never been solved before. Self driving in itself wouldn't be an issue, it's the other humans on the road that make it so hard to solve. One of the main problems for manufacturers is that we have never before solved a similar problem, so there's nothing to compare it to.

asdff 7y ago
>it's the other humans on the road that make it so hard to solve. Humans, cats and dogs, deer, tumbling rocks, low flying seagulls, tumbling trashbags or a toppled trashbin bin, palm fronds, drunk people jaywalking, drunk guy in a jeter jersey directing traffic after a yankees game, edgy teenagers pulling the invisible rope trick (or just yanking a sheet on a string), puddles reflecting humans on the sidewalk, blind turns and hidden driveways, competitors self driving tech, the metawars that will result by trying to raise a competitors accident rates (and not doing that would be leaving money on the table).

beefield 7y ago

> This stuff is hard. Luckily we attract our best minds with highest salaries to solve the pressing problem how to get people click more ads. (Yes, sarcasm.)