typodupeerror

## Comment Re:No. (Score 1)1049

Why don't you point out exactly which of my numbers you think is wrong. Why don't you make a quantitative counter-argument. Otherwise I will assume that you are only pretending to have a counter-argument by trying to sound confident.

## Comment Re:No. (Score 3, Interesting)1049

I see a lot of people confidently asserting opinions here without actually giving arguments refuting much of anything in the source article. So let's do some basic cost calculations. Let's say that your electric car has a capacity of 85kWh. That capacity with the very heavy Tesla Model S will give you an approximate EPA range of 426km (265 miles). If your electricity cost was \$0.15/kWh, that means the cost to charge your car fully from empty would be \$0.15/kWh x 85kWh = \$12.75. Since you would seldom fully empty your car battery fully, you would typically charge less than this, and it is likely the EPA range does not bring the battery to full empty. Even so, I will assume the price of driving the range of 426km would still be \$12.75 (charged from the charger in your garage...fully charged when you get up). This gives an electric cost of \$12.75/426km = \$0.0299/km.

Now let us consider a gasoline car. I'll assume an optimistic 10L/100km. That would mean that driving 426km would use 426/100 x 10 = 42.6L of gasoline. Gasoline costs \$1.32/L where I live, but let's give it a cheaper price of \$1.11/L. This would give a cost for driving 426km of 42.6L x \$1.11/L = \$47.29. The cost per km would be \$47.29/426km = \$0.111/km. In other words gasoline costs \$0.111/\$0.0299 = 3.7 x more or 370% more than electric per km! Electric cars are simpler. The battery technology is constantly improving. There are Tesla electric cars that have driven 200000 miles with no battery replacement (the car linked to here did have its battery replaced at 200000 miles, but it actually had most of its range, and it is likely Tesla wanted to examine the battery). Recent improvements in battery technology promise batteries that will last the life of the car. The announcement referred to here was in reference to an increased voltage battery chemistry that showed 92% capacity remaining after 1200 charge cycles. If your car has a range of 230 miles per charge cycle, than that would allow the car 230 miles x 1200 = 276000 miles and still have 92% battery capacity! For most of us, that would be longer than the lifetime of a fossil fuel car.

The cost of the cells is already dropping precipitously. The trend shown over the last few years is going to continue. There is no such trend in gasoline cars. Costs are for fossil fuel cars are going up. Electric cars will appear at lower and lower points in the market, first in the used market, and later in the new car market. In the end, electric cars will be the only economical choice. It is simple physics and economics. You can deny it all you want, but in the end, physics will win. Steam won over horse transportation because it was cheaper and better. Gasoline won over steam power because it was cheaper and better. Electric will win over gasoline because it is cheaper and better.

## Comment Re:I find your lack of faith disturbing... (Score 1)389

With automobile pilots we tolerate faulty humans whose decision-making processes we absolutely don't understand such that car crashes don't even make the news, but every car AI pilot fender bender will "raise deep questions about the suitability of robots to drive cars."

If AI is better than humans, then fewer people will die in cars. Period.

## Comment Re:I find your lack of faith disturbing... (Score 1)389

It is all about past experience. If some humans drive well then we predict they will continue to drive well and give them an insurance discount. If they drive poorly, we charge them a lot for insurance, under the prediction that they will continue to have more crashes. If a particular AI has a better driving record than humans, then it would be logical to give it a lower insurance rate, based on past experience. We don't have to know the details of how the human brain works to predict these things, and we shouldn't need the exact details about how the AI works to predict its behaviour. Better is better.

## Comment Re:Netflix outspends HBO more than 2:1? (Score 1)312

Netflix Marvel: First season of Daredevil...great! Second season...still pretty good. Both were soooooo much better than the original movie. Luke Cage...unique style, excellent. Jessica Jones...not bad. Ironfist...didn't really like.

ABC Marvel: Agents of Shield...couldn't get into it, despite Joss Whedon. I don't know, there was something about the writing, the casting that did not work for me. The production values just felt cheap, like any other boring network TV series. Netflix Marvel series feel more like movies. The cinematography, the fight scenes on Netflix were often very well done. Their series have a sense of atmosphere that is lacking in the big network's shows.

I cut my cable TV a while ago, and I will never go back. Commercials seem like a slap in the face now. I hate them hate them hate them! As far as I'm concerned, the big three networks can just die.

## Comment Re:Elon Musk, Tesla, and Robotics (Score 1)297

Why not direct some of your anti-government animus towards companies like Lockheed Martin (a military contractor) that receives almost all of its revenue from the government? Most of the sugar you eat is subsidized by the government through corn subsidies (why do you think it's so cheap?). General Motors would have gone bankrupt if it weren't for government money given after the 2008 crash. And of course the biggest one would be gasoline; fossil fuel companies receive massive direct and indirect government subsidies.

## Comment Elon Musk, Tesla, and Robotics (Score 5, Interesting)297

I don't see any mention of Elon Musk and Tesla in this discussion. Musk is bringing a new level of automation to his car factories. The interior of the new Model 3 will be designed for full robotic assembly. For example, typical wiring harnesses that appear in other cars will be avoided as they are not suitable for robotic manipulation. Instead, wiring connections are likely to be more pluggable by robots. Their new cars feature full glass roofs. I suspect this is because it will leave the top of the car open for robots to work until close to the end of assembly. Most cars weld their roofs on during frame assembly (which is typically robotic for most car manufacturers). This limits access to the interior during final assembly.

Musk has talked about the machine that makes the machine as the most important engineering challenge to be solved in manufacturing. He says the final version of his factories will look like an "alien dreadnought". Humans will be involved only in maintaining the robots, and not in the actual assembly process, since they slow the entire process down to "human speed". I'm not sure how many people are aware of the level of innovation that is occurring right now in America at Tesla's factories. There is no company in the world that is doing what Tesla is doing in automobile manufacturing.

## Comment Re: Dont Buy It (Score 1)641

The reason the engine revs constantly is because the Prius has a clever mechanism using two electric motors and a differential that keeps the gasoline engine reving at exactly its optimal RPM.

That said, yes, the Prius is slow as shit.

My 2010 Prius

does 0 to 60 in 9.7 seconds. Now that isn't massively fast, but I'm not sure I'd call it "slow as shit". I think I'd call the acceleration "average". It is certainly enough to safely merge on the freeway, and when I accelerate with passengers, they are usually surprised how fast it goes.

## Can We Pollinate Flowers With Tiny Flying Drones? (economist.com) 130

An anonymous reader writes: An engineer in Japan has built a 1.6-inch "pollinator-bot" and successfully tested it in his lab. The drone's creator "has armed it with paintbrush hairs that are covered in a special gel sticky enough to pick pollen up, but not so sticky that it holds on to that pollen when it brushes up against something else," reports The Economist. They write that his experiments with the tiny drone "show that the drone can indeed carry pollen from flower to flower in the way an insect would -- though he has yet to confirm that seeds result from this pollination." While flown by a human pilot, next he hopes to equip the drones with their own flower-recognizing technology.

The Christian Science Monitor followed up with four experts, asking "Could a fleet of robo-pollinators replace, or at least supplement, the bees?" One said "There is no substitute for bees." Another pointed out that even if robo-bees are developed, some flowers will prove harder to pollinate than others. A third expert thought the technology could scale, though it would need to be mass-produced, and the engineers would need to develop a reusable pollen-collecting gel. But a fourth expert remained worried that it just couldn't scale without becoming too expensive. "I'm not sure that's going to be cheap enough to not make blueberries hundreds of dollars a pint."

Three of those experts also agreed that the best solution is just wild bees, because domesticated or not, "All they have to do is make sure to set aside enough land conducive to the bees' habitat."

## Comment Re:This is not a serious issue. This is very minor (Score 2)139

Yep. With non-recoverable rockets that end up on the ocean floor, we never know if the rocket engines were consistently on the edge of catastrophic failure. Without looking at the used engines, the only thing we really know about expendable rockets is that they generated nominal telemetry during operation.

## Comment Re:Short-term numbers versus long-term (Score 1)167

I'm not up on state of the art on computer image/object recognition but the experience I have from about 10 years ago leads me to believe that there are still challenges to be solved, especially when it comes to recognizing movements and intentions.

Neural networks have come a LONG way in ten years, due in large part to the exponential growth in processing power in GPU's. Neural nets can perform the same or better as humans in a variety of image recognition tasks. For example, neural nets have been trained to give the prognosis for cancer patients based on images of tumors. The networks were trained on thousands of known images of previous cancer patients along with medical histories. When new images were passed through the network, the prognosis, including likelihood of survival was given, based upon the images of previous patients.

If you really think about it, when humans drive, we are largely doing simple image recognition. White line, yellow line, double line, car in front, car in back, pedestrian about to cross, bicyclist riding on side of road, etc. There is some context, yes, but really in most cases the decisions we make in driving are quite automatic and shallow. They are in large part rule based, learned from long habit. There are still likely to be cases at times when simple image recognition and habitual rule based behaviour will not suffice; in such cases, yes self driving cars might have trouble. However I would assert that such corner cases are likely to be rare. The advantages of having a computer driving will be that they won't get drowsy or distracted, and will have far more information input about the surroundings of the car, gleaned from eight cameras, several ultrasound sensors, and a radar system that can detect cars in front of the car in front of you. To a certain extent, driving is simple. If there is an object in front of you, or beside you, don't hit it; stay in the correct lane; don't go too fast into corners. I think that on the whole, computers are likely to be better at this than humans.