Hacker News Reader

Comment by cowmix | Hacker News Reader

I remember during the 2004 presidential election I was was on some political forums arguing how EV would reach tipping point in sales and technology within about 10 years. All these people would go on and on about how batteries will always be expensive, the laws of physics state EV will never truly be viable, EVs are slow, batteries are undependable, EVs pollute more, etc, etc.

One by one all those reasons have be discredited. The EV was and is inevitable.

View on HN

The important thing to take away from that experience is that the detractors were all correct, as far as any extrapolation of then current technology into the future could take them.

I've experienced that cycle several times where I thought "oh this will be big in 10-20 years" and had people say "No, reason x, y, and z all mean it won't happen." and then some discovery or change in the situation made x, y, and z irrelevant.

As a result I've trained my self to look at it "in reverse" as it were, which is to say "currently x, y, and z makes it impossible, what would have to change to make x, y, and z irrelevant?" and then look for changes in the margins. Things like "making oil from corn" is impractical when oil is $35 a barrel but quite reasonable when oil is $150 a barrel. Or 'cracking hydrogen and oxygen' with electrolysis takes more energy than you get back, so how can you exploit 'wasted' energy like sunlight to help it along? etc.

The important thing is to avoid trapping yourself into believing nothing will change that isn't already known. People I've met seem to do this all the time.

I think you're right. But I also think this is a fundamental limit of human thinking, to some extent.

If you consider the space of all possible imaginable futures, there are parts of this space that are easily accessible to us, and there are parts that just don't seem to be accessible at all. For example, people were able to foresee the technical development of the internet decades ago, but we're not able to foresee how internet culture will develop with any accuracy at all.

It's like the philosophical idea that you can't accurately imagine what it's like to be a genius, because if you could, then you could become a genius yourself, just by imagining what a genius would do in each situation. If we could accurately imagine the future, then the future would already be here, to some extent. E.g. if I could predict what the next hit smartphone app would be, then I could just build the app myself and get rich.

I disagree that it is a 'fundamental' limit, I think you can practice holding "impossible" things in your head so that you can work out where the places are where the "possible" things are at the edge. Using your example, you can't accurately imagine what it's like to be a genius but you can "pretend" you're a genius, and imagine that if you were a genius you would know lots of stuff about a lot of different things, and from there recognize you would need to learn about lots of different things. And from there recognize that you'd need to know how to learn quickly so that you could know lots of different things. And from there find techniques for learning things quickly and start trying different ones to see if any work for you.

That is the 'walking it in reverse' idea, if you start with this impossible idea (your own or someone else has presented it to you) and you want to figure out sort of that impossible idea might come to be real.

Just to continue this tangent: The skill you mention (holding impossible things in your head) is closely related to debugging. You observe "impossible" things (returned signal power is negative, etc.), and accept them, and work backwards to how it could happen.

How many times have you had a junior person come to you with a bug, and you ask, did you check A, B, and C? And they will say, "I checked everything and it's all OK." It's hard to kick their brain out of the rut of thinking that everything is nominal. They make no progress because they're wasting all their mental energy arguing that everything is, in fact, OK.

Thats one if the best comments ive ever read here

This heavily applies to software development too.

In the inverse, people for some reason with software like to assume the natural state of code is to work. If you assume it doesn't and look for spots where it could fail then, you find the problems.

It's amazing what a state of mind can do.

Many years ago my lecturer used to hammer this point home (he wrote telecoms systems back in the 'old days' (80's/early 90's).

Good guy, bit odd but a hell of a programmer.

It is amazing to watch Wright's Law in action as volumes grow to saturation, pulling down the price, to drive saturation. It will be interesting to see if batteries are like processing with self-generating demand, or if it is like prior industrial technologies that seem to hit ceiling based on some sort of inherent demand limit. It is hard to imagine wanting 3 electric cars...

So the analysis focuses on a single variable: cost, specifically the inflation-adjusted price of one “unit.” (A “unit” is itself sometimes a fluid concept in a rapidly changing field. Consider what “one transistor” meant in 1969, and what it meant in 2005.) So, for example, recasting Moore’s Law to translate computing power into unit cost morphed the familiar “computing power doubles every 18 months” into “transistor costs drop by 50 percent every 1.4 years.”

(Wright's Law is surprisingly hard to google)

source: http://spectrum.ieee.org/tech-talk/at-work/test-and-measurem...

That article has really good examples of the problems of overfitting. Better than the standard abstract data science stuff I've seen anyway.

I think there'll be an economic incentive for at least miniaturizing battery cells, while keeping capacity. Even if there is a limit to EV range, where consumers simply won't be interested in cars that go much further, reducing the weight of the battery itself, will be a direct economic boon for the consumer. There might also be some unforeseen boons to cheaper and lighter batteries, kind of how the personal drone market sprung out of the advances made in lithium-ion batteries for Smartphones/Laptops.

> It is hard to imagine wanting 3 electric cars

It's no more difficult than wanting 3 internal combustion engine cars - one for husband, one for wife, one for driving age children.

Or the weekend sportscar/roadster, or the "classic" car (though I wonder if any modern electric will ever achieve classic/collectable status).

As Jay Leno happily points out, some of the oldest and coolest "classic" cars are turn of the last century electric cars. [1]

Whether a current electric car may ever turn heads or have gearhead fans a century from now, you never know, retro fashions are inscrutable from this distance.

[1] An example: https://transportevolved.com/2015/12/29/jay-leno-prepares-to...

EV West seems to be doing a good job of building up a catalog of conversion kits. Model S running gear in a tri-5 Chevy won't be as efficient as the original package, but it is good enough (except for that V8 rumble)

If technology gets good enough batteries become attractive for a wide variety of engines that currently require either electric cords or fossil fuels. Think drones, bicycles, lawn appliances, vacuum cleaners, home UPS, solar energy storage, automated greenhouses, etc. It also could have knock-on effects on the type of powered appliances possible, eg. perhaps you'd want your toaster & cabinet to be mobile so you could have breakfast in bed, but this is impractical with electric cords. I wouldn't worry too much about demand for batteries.

Certainly replacing helicopters with drones seems a pretty obvious step [0].

[0]: http://time.com/4171329/drone-helicopter-ehang-184-ces/

The EV is still "evitable" while it's dependent on subsidy and favourable tax treatment, unfortunately.

It's the ICE cars that are subsidy dependent, people just have real problems accepting that because they've been subsidised for so long and in such pervasive ways that even the death tolls just seems like statistics to be waved away.

Reply on news.ycombinator.com