Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Helgoland]

Nope. If you know the others are going to buy the widgets, you'd be better of not buying one yourself. That's the whole point, really. If you don't believe me, go look up the frickin' prisoners' dilemma on Wikipedia. Buying the widget corresponds to keeping mum, not buying it corresponds to ratting out the other guy.

[smjjames]

The real issue here would be whether the total labor value of doing this process would be less than just planting more plants to make CO₂ into stuff. That's questionable.

We can already make methane from biosources, so if we want to do that I'm pretty sure bio-engineering some microbes might be more efficient than this.

Intensively growing microbes is still a fairly energy-hungry method I believe, and of course runs into the "GM is evil!" Bogeyman who would strongly resist the use of these.

You wouldn't need to use genetic modification here. If all you're doing is turning dead plant matter into methane, you can find bacteria which do just that. After all, what do you find in swamps? Dead plant matter and swamp gas (which has methane). Of course though, you'd have to try and cultivate just that bacteria and there might be other byproducts you don't want.

[Maximum Spin]

Nope. If you know the others are going to buy the widgets, you'd be better of not buying one yourself. That's the whole point, really. If you don't believe me, go look up the frickin' prisoners' dilemma on Wikipedia. Buying the widget corresponds to keeping mum, not buying it corresponds to ratting out the other guy.

No, you don't know the others are going to buy the widgets, you just know that everyone will make the same decision, whatever it is, in which case the logical thing for everyone to do is buy the widget. This is a real thing, I'm not making it up, it's just a higher level concept than you'll get on Wikipedia.

[McTraveller]

We were talking around the lunch table today about fantasy tech.

We decided that we really need to start working on advancing our replicator technology, so we can take any matter feedstock (say, carbon) and turn it into something else (say, lithium, or iron, or whatever).  After all, theoretically, you can split any atom into quarks with enough energy, and then just smash them back together in other configurations.

We figure it will probably take politically-unencumbered cheap fusion reactors to do it, but if (super)novae can do it, why can't we!?  (This is the kind of thing that I would love to see our AI overlords establish, actually.)

[smjjames]

That's far future Sci-Fi, I object to calling it fantasy tech.

The thing though is that it takes a hell of a lot of energy to create heavier atoms and we already do it the atomic version of slamming two rocks together in particle accelerators. It'd be a lot easier to just provide the elements in feedstock rather than trying to smash atoms apart and reassemble them with nanomachnes or something.

[Trekkin]

We figure it will probably take politically-unencumbered cheap fusion reactors to do it, but if (super)novae can do it, why can't we!? 

Because the least energetic known type of supernova still releases about 247,252,040,818,345,000,000,000 times the energy all of humanity used in 2015 (as a rough average) and exploding stars don't have any shielding requirements.

Obviously.

(A type 1a supernova emits about 1.5*10⁴⁴ joules. (http://adsabs.harvard.edu/abs/1993A&A...270..223K) We used about 6.06×10²⁰ joules in 2015. (https://www.eia.gov/outlooks/ieo/pdf/0484(2017).pdf) )

[Reelya]

Nope. If you know the others are going to buy the widgets, you'd be better of not buying one yourself. That's the whole point, really. If you don't believe me, go look up the frickin' prisoners' dilemma on Wikipedia. Buying the widget corresponds to keeping mum, not buying it corresponds to ratting out the other guy.

No, you don't know the others are going to buy the widgets, you just know that everyone will make the same decision, whatever it is, in which case the logical thing for everyone to do is buy the widget. This is a real thing, I'm not making it up, it's just a higher level concept than you'll get on Wikipedia.

There's actually a TED talk about something highly related to this that's relevant to the thread. I'd forgotten about it until your post.

There's a scenario, similar to the Trolley Problem. The problem is for self-driving cars. The car can either run over 10 people if it goes on it's current trajectory, or it can kill 1 person. Which should it do? This is just the basic Trolley Problem, then, and almost people immediately say "kill the one person", e.g. they take the utilitarian choice: kill the least amount possible, even if it means taking action that deliberately chooses someone to die. At some point, self-driving cars will be involved in accidents and will have to take hard choices. Applying the "trolley problem" is just a way of thinking about the ethics involved.

however, what if the "one" who must be sacrificed to save other lives is you, the driver, who paid for the car? Suddenly, most people say they will not buy a car that makes that trade-off, even if everyone else agrees to buy one. e.g. if everyone had a utilitarian self-driving car that would in fact kill the driver if that was the way that minimizes total casualties, total road casualties would be objectively the lowest possible. However people don't want the cars that do this. And, objectively when you do the maths, we're all worse off if we don't have the cars that would willingly self-sacrifice if that is the choice that truly minimizes casualties. e.g. every driver is in fact more likely to die, individually if we all choose cars that put "driver first" rather than optimize for complete minimization of road casualties.

if you consider the trade-offs here it's clearly the same problem as the "widgets" example I put up. "Logically" everyone should buy the robot car that's willing to kill it's own driver rather than kill two other drivers. However people seem to balk at this, even if it in fact means their total chance of death is higher: they want everyone else, except not themselves, to have the self-sacrificing "suicidal" cars. This is from surveys the guy doing the Ted talk did btw, people overwhelmingly want other people to get the death cars, because that makes them safer, but they overwhelmingly reject the idea that they themselves should get one.

 And think about it, if everyone is required to have perfectly utilitarian cars, the overall chance of death is minimized. However ... if you cheated by hacking your robot car with the instruction "save me and only me at all costs" your chances of survival would increase compared to everyone else. So it would be the logical thing to do as hacking your car always minimizes your chance of death compared to not hacking your car. However, everyone is in fact optimizing their own chance of survival at the expense of increased death chances across the board. So everyone hacks their cars to be selflish, because that's just "prudent" isn't it? However once all cars are hacked, everyone is in fact less safe.

And this isn't a toy problem either. When you get into a robot car in the future you're damn well going to want to know how they work and what basis for their decisions they make.

[smjjames]

I've read about that actually, not from a TED talk but elsewhere. It is indeed a variant of the Trolley Problem. The basic situation is that the car has to avoid something (a pedestrian or other killable thing), however, the only solution to avoid the person has a very high chance of death to the driver, so, which should the car choose? As in the Trolley Problem, death is unavoidable.

The only difference is that it's an unthinking computer making that choice and it runs into the problem of 'do we want the computer making the decision?' Since its in the computers hands and not the drivers, it changes the moral dilemma

Obviously the driver might choose to save themselves, but if you think about it, automated cars killing pedestrians would definetly generate backlash and a setback in the development of them.

Edit: Yeah, you summed it up in longer form while I posted.

[Maximum Spin]

Nah, there's some flaws in that. Most importantly, nobody is truly utilitarian. (and for good reason, it's a terrible nonsense philosophy)

But also, you must see that the "run over 10 people / kill the driver" situation is asymmetrical: you can't be run over if you are in a car. If cars will always prefer to run people over than to kill the driver, pedestrians are more likely to die, but drivers are less likely. In principle, you can handwave it away by making it a premise of the thought experiment that everyone is equally likely to be a driver or pedestrian at any given time, but that's the kind of thing that most humans can't easily abstract away, so they'll always answer the question under their everyday prejudices instead of adhering to the rules of the thought experiment correctly.

Still, to expand on what I originally said, the core problem is that "minimising casualties" isn't what people want; most people would genuinely prefer a higher overall risk of death by car accident than to be killed by their own cars. You can't say that one outcome is "logically" the best if you don't understand what people actually value. Also, utilitarianism is for stupids.

[Reelya]

you're pointing out things that aren't actually important points. It could just easily be said as "cause an accident killing two other drivers" as running over people. The only reason it was phrased as "run over 10 people" was because that was how the Trolley Problem from 1967 is phrased: the original dilemma is that the train / Trolley can run over 5 people or just 1 person, but you have to make a conscious decision to kill the one person and not the five. It's just an introductory narrative which introduced the structure of the orginal "trolley problem", but links it to the idea of robot cars, purely because the Trolley Problem is one of the standard thought experiments in the field of Ethics.

So no, the thing about running over pedestrians isn't the point here. It's like complaining that the Trolley Problem is b.s. because you're not a train driver. It's meant to be a starting point for discussing ethical decision-making, not an example to 100% be taken literally.

the point is, should people get robot cars programmed to kill others to save themselves or will we legislate that everyone be forced to buy cars with programming that put community safety as a whole first. Drivers don't seem keen on this part, but it's going to be a big topic in a few years. Who decides?

[smjjames]

Alternatively, we can separate pedestrian paths from the roads, but it's going to take a very major restructuring of how our cities are designed and laid out for that to work absolutely everywhere. Probably easier to do in a space colony or a newly planned city.

[Reelya]

Instead of self-driving cars let's replace all of them with the Trolley from the Trolley problem and set it up so that 5 pedestrians are down 1 street, while only 1 is down the other. Then we can have fun always optimizing the Trolley Problem when we commute.

[Maximum Spin]

you're pointing out things that aren't actually important points.

That's the point, most people genuinely cannot just view these thought experiments in a vacuum, because they can't exclude their own experience from colouring their interpretations of the premises, so changing the phrasing can totally change the responses and all of these surveys are futile.

[Helgoland]

Oh boy, you really live up to your name, don't you?

a starting point [...], not an example to 100% be taken literally

People in this thread appear to have a difficult time understanding this concept.

[Maximum Spin]

People in this thread appear to have a difficult time understanding this concept.

Literally my point
people everywhere have difficulty understanding the concept.