Not every car is going to be retrofitted at the same time, nor would non-auto cars be mandatory, you'd think.
So I think there will be two levels of automation: single-vehicle AI, which would know about other cars, traffic, etc., but would not affect other cars, and "network level" automation, where some central (or distributed?) system is able to coordinate between different vehicles. Maybe we'll get the first before the second.
It would definitely be useful for my car to do the long, boring commutes on autopilot. I don't think I'd want to be forced to use it though.
This is a good point. When I've visited the US (minus NYC) it's always seemed like the cities were built with the assumption of car ownership. There's even houses you can't walk to because there's no sidewalk. "Going to buy some groceries" seems to mandate getting in a car and driving a good few minutes, which often means buying a huge amount of stuff each time is clever. When I was living in London, you could just walk out the door and be at a shop within a minute, buy a couple of things, and be back.
Brings into the light the numbers on just how dangerous automobiles are. Few activities have these huge numbers of deaths, accidents, and property loss and damages.
If anything, this just brings to light how USEFUL cars are. Otherwise it wouldn't be worth the cost.
Because if terrorists had a reliable key distribution network, they'd already be an army, not a loosely organized criminal band with minimal transportation infrastructure? One time pads are only as good as your distribution system. And the moment you run out of key bits and reuse them, your system is broken.
This is the 21st century and some people still haven't wrapped their heads around proper safeguarding of data.
It sucks to be them, but it's their own responsibility to make sure their data is replicated on as many devices in as many places as is convenient and affordable to do so.
This is the 21st century and some people still haven't wrapped their heads around proper safeguarding of data.
It sucks to be them, but it's their own responsibility to make sure their data is replicated on as many devices in as many places as is convenient and affordable to do so.
Agreed. At the end of the day, you're responsible for your own property. Particularly when it's something irreplaceable like your own work, you can't rely on someone else, even if that's the whole point of their business. People screw up things all the time, and even if you can sue them, you're still screwed out of your work.
I wonder how much was really lost. Surely they don't just make the video, upload it, and then delete all the stuff they used produce it?
First, why do you have a health insurance policy in a country with socialized medicine?
Second, the insurers can't stay in business if everyone ends up using several multiples of what they pay. It's impossible to "make it up on volume" that way, and who wants to start a business that is guaranteed to lose money.
There's a big gap in quality between the socialized system and the private one. NHS: people wait around for hours with broken bones, not even let out of ambulances. In my case, after a very short examination, doc decides my very obvious ACL snap is nothing, sends me home. Friends have had to wait months for surgery. Private doc sends me to specialist, MRI, surgery, own room for a week, all done within a couple of weeks.
Sure, they can't stay in business if everyone takes out more than they pay in. But the whole idea is that they even out the spikes (me) with lots of people who pay but don't claim. Rumour has it that actuarial science is pretty advanced in how it predicts these kinds of probabilistic events.
Re caps, I suppose they're free to offer only policies with caps, I'm just saying there's not much point in buying a policy that has one. Why does an individual (one dice roll) guy want insurance? In case the one dice roll goes wrong. Why does the underwriter (many dice rolls) write? Because he's guaranteed some crap rolls, but also a lot of good ones, and on average he can stay afloat. If the price is right.
This isn't really so surprising. Just like Twitters, most of the world's men have only shagged a few women, while a few guys have done it with hundreds. A huge number of people live on a dollar a day, but some guys at the top can make over a billion a year. Most entertainers are unknown wedding singers, but a few are known by everyone on the planet.
Not saying it's right or wrong, just these kinds of distribution occur.
Is this the US you're talking about? What the heck is the point of insurance if there's a cap? The whole point of insuring yourself is to not mess your whole life up if something bad (but unlikely) happens.
I'm pretty sure my insurance in the UK wasn't capped. In any case, I ended up using several multiples of what I'd paid in premium.
You've heard of adverse selection right? People who are paying too much (vs what is right for their risk) for insurance will decide not to take the deal. Then you're left with people who are either overly cautious or downright unhealthy.
The whole idea of a healthcare insurance is to spread the risk between people... therefore it's pretty much necessary that healthy and unhealthy people pay the same.
That's not how car insurance works. People with a higher risk of accidents pay more. That's why your premiums go up after a speeding ticket, and go down when you get married / have kids. Your premiums change with your statistical level of risk.
I don't know how other forms of insurance work, so maybe someone can enlighten me, but I assume they also charge based on risk.
So, why would health insurance be different?
The difference is the extreme skewness of the risk to the insurer. The worst you can do with a car is to write it off. Even if it's a $2M Veyron, that's the most you can end up claiming on the car insurance. Also, the premium is a good few percent of that per year, so the insurer has a chance of making money if he has a pool of car owners, and the law of large numbers is helping him. Basically, there's no reason to think the accident rate will go up, and in addition, the value of the car is is sliding down.
If a 21 year old guy gets diabetes, that's a lifetime of insulin treatments. People seem to be getting more and more unhealthy, if you believe the periodic news story about fat people. Once they're signed on, they will be a customer for years. And the chance they get ill increases as they get old. There's also a chance that the treatment cost will increase as pharma firms invent new ways to treat people. So the numbers on which you're basing your business are changing (ie more uncertainty), with costs likely to go up, and you're locked in.
So you're right, insurance is based on risk, but the numbers are going different ways for your two examples.
As for what to do about it, I'm not convinced of either side being right, ie all private or all socialized. The times I've been in touch with the NHS in the UK have been disappointing, while the times I've used insurance have been really good. Yet it seems clear not everyone will be able to pay for their own treatment, and that a fully private model has certain undesired consequences.
It is a general fact about any kind of insurance that the interests of the insurer are misaligned with the interests of the insuree. They're predatory industries who rely upon promising more than they will deliver and tricking their customers wherever possible.
Citation needed...
Only in the circumstance that the insurer is required to insure everyone does the profit motive go in the direction of the patient's interests (in the form of preventative care). Preventative care is a long term investment that wall street doesn't see.
And why does the system magically work if they have to insure everyone? Surely they'll just offer a deal to everyone, but a crap deal for everyone they don't want?
If he really wants to shake things up, create 'micro-startups' inside Google. Put it in a separate building, isolated area, whatever. Shoot any managers or bean-counters that approach the area
Worked for Apple
You still can't get around the thing that makes it one company, rather than a bunch of companies: if your internal startup is crap, it won't die. You have to kill it, or wait for the whole firm to go down. This means management will, despite your best efforts at separation, meddle with which ones it likes and which ones it doesn't. You will get good ideas fighting with bad ones for resources. You have the same problem of "how do I pick a winner" and the same incentives for management ("if I let project x live, and it loses money, I'll look bad. If I don't have some interesting sounding projects, I'll look bad. Where are my dice?").
Meh, security is a bit of a cross-cutting concern. People who are thinking about how read/write rows of data quickly might not have given it much thought that their product can be abused in this way.
I will give you that injection attack is a rather basic hack they should have thought about.
But why would devs not have the required personality traits? There's more than one personality profile that suits management, and there's more than one personality type that is suitable for a dev. You'd surely be able to find an overlap in there?
Unfortunately developers rarely rise to any BA role.
What's unfortunate about it? It's a vastly different set of skills, and should be done by people trained accordingly. Let everyone do the job they're good at.
Meh, it's a very modern conceit that one can learn how to manage by getting a degree. Fact is, you can't. The sad thing is thousands of MBAs graduate each year thinking they know what business is. No, what you really need to manage is experience, not a load of acronyms and buzzwords. You want a good dev manager? Get a dev, start giving him management responsibilities. Same goes with other jobs, eg. very few soccer managers were not players on some level (they didn't have to be good players, but they do need to know what it's like).
Unfortunately, the whole world seems to be going the MBA way, mainly because business is becoming more and more BIG business. This favours hierarchies and internal politics, and MBAs are the most hungry to climb this pyramid.
> Then the insurer could rely on positive selection (as opposed to adverse selection of people who didn't consent) as well as monitoring to give you a better rate.
Nope. If you allow positive selection for those who volunteer, that implies negative selection for anyone who refuses to volunteer, and it would be a short hop from there to assume anyone refusing to share has something to hide.
It's not really different, you know, adverse/positive selection. Perhaps differential selection would be a better term.
What if they share that info with insurance companies, and you end up paying more for life or car insurance because they flag you for buying alcohol in an amount they consider excessive? Or condoms, or pregnancy tests.
If that information was passed without consent, yes, it would be sinister. But what if you willingly allowed the information to be passed to your insurer? Then the insurer could rely on positive selection (as opposed to adverse selection of people who didn't consent) as well as monitoring to give you a better rate.
Our Grocery stores track what we purchase, and everyone said "oh well, cheaper prices" (BS But okay).
Our ISPs track our information, even hijack DNS error pages now. Everyone said "Oh well, they are a business"
Now this, and I guarantee it will be "Oh well, they are a business that needs to make money"
Consumers let this happen.
Well, yeah, they let it happen because they can see the use of collecting that info and therefore consent to it. The real question is whether this information is sent to other organisations, such as the government? I wonder how long it will take before someone's movements are tracked and used for police investigations. Perhaps it's already occurred.
There's just no way any MMO is going to "beat" World of Warcraft
As far as I'm concerned, any MMO is better than WoW. I never got why people liked it so much. It's the most terrible one I've ever played. It's pure grinding, has little to no character customization, uninteresting classes that are all the same, very poor graphics... It's like an old school MMORPG but without the roleplaying nor the old-school feel.
I played WoW for quite a while, and I can understand some of what you say. To begin with, I was between jobs, with a lot of time to kill. I thought I'd play WoW to see some interesting landscapes, which of course meant you had to level. It was a bit of a grind, but at least it changed somewhat every couple of levels, and there were new abilities, new instances, etc. I could play the game as I wanted, in my own time, without relying on other players, while still playing it as a multiplayer game.
But when I got to the max level, all that changed! When you're raiding, the game plays you! You have schedules, people are expecting you to have done your homework (hotkeys, spec, reading about encounters). On top of that, it's quite frustrating to wait for the recruitment of replacements, as people inevitable will decide not to turn up. And also, rerunning a boss encounter over and over is tiring.
That doesn't mean you need government on the scale it is now. Most of what government does these days is NOT the competition regulation that you're talking about.
Being from another Scandinavian country myself, I have to disagree. The nanny state is huge, and people are no longer able to be responsible for themselves. Everyone thinks about their rights, not their obligations to society. Also, there's a great deal of Jantelov thinking, which is basically an institutional form of jealousy. The scandies need to consider that they're no longer in an isolated, homogeneous part of the world, where everyone agrees about what the public pot should be spent on.
But anyway, that's not the main point I wanted to make. It's not that giving power to the wealthy is the main problem. (Sure, it can be a problem, no doubt.) The problem is giving power to large institutions. Microsoft, AT&T, Shell, etc... government is yet another example. If you ever work with or for one of these behemoths, it's understandable why you're frustrated. Large organisations lack common sense in their decision making, and they lack common empathy in their dealings with ordinary individuals. Due to their size (and influence) they're also able to live beyond their useful age, holding up resources (people, mainly) from more productive uses. Unfortunately, the west has institutionalized a system where the big firms work with the big governments to make sure neither of them is ever renewed. If organisations were smaller, we'd have a much more healthy society, where useful firms live, and old, unproductive ones die.
Well, if we can't, and this problem has come up before, the population will naturally stop growing. It won't be pretty if it's a hard landing, but various population explosions in the past are connected to innovations in farming, and why would it be any different in the future?
Slashdot Mirror
Not every car is going to be retrofitted at the same time, nor would non-auto cars be mandatory, you'd think.
So I think there will be two levels of automation: single-vehicle AI, which would know about other cars, traffic, etc., but would not affect other cars, and "network level" automation, where some central (or distributed?) system is able to coordinate between different vehicles. Maybe we'll get the first before the second.
It would definitely be useful for my car to do the long, boring commutes on autopilot. I don't think I'd want to be forced to use it though.
This is a good point. When I've visited the US (minus NYC) it's always seemed like the cities were built with the assumption of car ownership. There's even houses you can't walk to because there's no sidewalk. "Going to buy some groceries" seems to mandate getting in a car and driving a good few minutes, which often means buying a huge amount of stuff each time is clever. When I was living in London, you could just walk out the door and be at a shop within a minute, buy a couple of things, and be back.
Brings into the light the numbers on just how dangerous automobiles are. Few activities have these huge numbers of deaths, accidents, and property loss and damages.
If anything, this just brings to light how USEFUL cars are. Otherwise it wouldn't be worth the cost.
Because if terrorists had a reliable key distribution network, they'd already be an army, not a loosely organized criminal band with minimal transportation infrastructure? One time pads are only as good as your distribution system. And the moment you run out of key bits and reuse them, your system is broken.
Could a book code be used?
Gee, I wonder which book they'd use...
This is the 21st century and some people still haven't wrapped their heads around proper safeguarding of data.
It sucks to be them, but it's their own responsibility to make sure their data is replicated on as many devices in as many places as is convenient and affordable to do so.
This is the 21st century and some people still haven't wrapped their heads around proper safeguarding of data.
It sucks to be them, but it's their own responsibility to make sure their data is replicated on as many devices in as many places as is convenient and affordable to do so.
Agreed. At the end of the day, you're responsible for your own property. Particularly when it's something irreplaceable like your own work, you can't rely on someone else, even if that's the whole point of their business. People screw up things all the time, and even if you can sue them, you're still screwed out of your work.
I wonder how much was really lost. Surely they don't just make the video, upload it, and then delete all the stuff they used produce it?
First, why do you have a health insurance policy in a country with socialized medicine?
Second, the insurers can't stay in business if everyone ends up using several multiples of what they pay. It's impossible to "make it up on volume" that way, and who wants to start a business that is guaranteed to lose money.
There's a big gap in quality between the socialized system and the private one. NHS: people wait around for hours with broken bones, not even let out of ambulances. In my case, after a very short examination, doc decides my very obvious ACL snap is nothing, sends me home. Friends have had to wait months for surgery. Private doc sends me to specialist, MRI, surgery, own room for a week, all done within a couple of weeks.
Sure, they can't stay in business if everyone takes out more than they pay in. But the whole idea is that they even out the spikes (me) with lots of people who pay but don't claim. Rumour has it that actuarial science is pretty advanced in how it predicts these kinds of probabilistic events.
Re caps, I suppose they're free to offer only policies with caps, I'm just saying there's not much point in buying a policy that has one. Why does an individual (one dice roll) guy want insurance? In case the one dice roll goes wrong. Why does the underwriter (many dice rolls) write? Because he's guaranteed some crap rolls, but also a lot of good ones, and on average he can stay afloat. If the price is right.
This isn't really so surprising. Just like Twitters, most of the world's men have only shagged a few women, while a few guys have done it with hundreds. A huge number of people live on a dollar a day, but some guys at the top can make over a billion a year. Most entertainers are unknown wedding singers, but a few are known by everyone on the planet.
Not saying it's right or wrong, just these kinds of distribution occur.
Is this the US you're talking about? What the heck is the point of insurance if there's a cap? The whole point of insuring yourself is to not mess your whole life up if something bad (but unlikely) happens.
I'm pretty sure my insurance in the UK wasn't capped. In any case, I ended up using several multiples of what I'd paid in premium.
You've heard of adverse selection right? People who are paying too much (vs what is right for their risk) for insurance will decide not to take the deal. Then you're left with people who are either overly cautious or downright unhealthy.
I can't imagine the uproar that would occur if having a good wank would incur a license fee...
The whole idea of a healthcare insurance is to spread the risk between people... therefore it's pretty much necessary that healthy and unhealthy people pay the same.
That's not how car insurance works. People with a higher risk of accidents pay more. That's why your premiums go up after a speeding ticket, and go down when you get married / have kids. Your premiums change with your statistical level of risk.
I don't know how other forms of insurance work, so maybe someone can enlighten me, but I assume they also charge based on risk.
So, why would health insurance be different?
The difference is the extreme skewness of the risk to the insurer. The worst you can do with a car is to write it off. Even if it's a $2M Veyron, that's the most you can end up claiming on the car insurance. Also, the premium is a good few percent of that per year, so the insurer has a chance of making money if he has a pool of car owners, and the law of large numbers is helping him. Basically, there's no reason to think the accident rate will go up, and in addition, the value of the car is is sliding down.
If a 21 year old guy gets diabetes, that's a lifetime of insulin treatments. People seem to be getting more and more unhealthy, if you believe the periodic news story about fat people. Once they're signed on, they will be a customer for years. And the chance they get ill increases as they get old. There's also a chance that the treatment cost will increase as pharma firms invent new ways to treat people. So the numbers on which you're basing your business are changing (ie more uncertainty), with costs likely to go up, and you're locked in.
So you're right, insurance is based on risk, but the numbers are going different ways for your two examples.
As for what to do about it, I'm not convinced of either side being right, ie all private or all socialized. The times I've been in touch with the NHS in the UK have been disappointing, while the times I've used insurance have been really good. Yet it seems clear not everyone will be able to pay for their own treatment, and that a fully private model has certain undesired consequences.
It is a general fact about any kind of insurance that the interests of the insurer are misaligned with the interests of the insuree. They're predatory industries who rely upon promising more than they will deliver and tricking their customers wherever possible.
Citation needed...
Only in the circumstance that the insurer is required to insure everyone does the profit motive go in the direction of the patient's interests (in the form of preventative care). Preventative care is a long term investment that wall street doesn't see.
And why does the system magically work if they have to insure everyone? Surely they'll just offer a deal to everyone, but a crap deal for everyone they don't want?
You can do that internally
If he really wants to shake things up, create 'micro-startups' inside Google. Put it in a separate building, isolated area, whatever. Shoot any managers or bean-counters that approach the area
Worked for Apple
You still can't get around the thing that makes it one company, rather than a bunch of companies: if your internal startup is crap, it won't die. You have to kill it, or wait for the whole firm to go down. This means management will, despite your best efforts at separation, meddle with which ones it likes and which ones it doesn't. You will get good ideas fighting with bad ones for resources. You have the same problem of "how do I pick a winner" and the same incentives for management ("if I let project x live, and it loses money, I'll look bad. If I don't have some interesting sounding projects, I'll look bad. Where are my dice?").
Meh, security is a bit of a cross-cutting concern. People who are thinking about how read/write rows of data quickly might not have given it much thought that their product can be abused in this way.
I will give you that injection attack is a rather basic hack they should have thought about.
But why would devs not have the required personality traits? There's more than one personality profile that suits management, and there's more than one personality type that is suitable for a dev. You'd surely be able to find an overlap in there?
Unfortunately developers rarely rise to any BA role.
What's unfortunate about it? It's a vastly different set of skills, and should be done by people trained accordingly. Let everyone do the job they're good at.
Meh, it's a very modern conceit that one can learn how to manage by getting a degree. Fact is, you can't. The sad thing is thousands of MBAs graduate each year thinking they know what business is. No, what you really need to manage is experience, not a load of acronyms and buzzwords. You want a good dev manager? Get a dev, start giving him management responsibilities. Same goes with other jobs, eg. very few soccer managers were not players on some level (they didn't have to be good players, but they do need to know what it's like).
Unfortunately, the whole world seems to be going the MBA way, mainly because business is becoming more and more BIG business. This favours hierarchies and internal politics, and MBAs are the most hungry to climb this pyramid.
> Then the insurer could rely on positive selection (as opposed to adverse selection of people who didn't consent) as well as monitoring to give you a better rate.
Nope. If you allow positive selection for those who volunteer, that implies negative selection for anyone who refuses to volunteer, and it would be a short hop from there to assume anyone refusing to share has something to hide.
It's not really different, you know, adverse/positive selection. Perhaps differential selection would be a better term.
What if they share that info with insurance companies, and you end up paying more for life or car insurance because they flag you for buying alcohol in an amount they consider excessive? Or condoms, or pregnancy tests.
If that information was passed without consent, yes, it would be sinister. But what if you willingly allowed the information to be passed to your insurer? Then the insurer could rely on positive selection (as opposed to adverse selection of people who didn't consent) as well as monitoring to give you a better rate.
Should we be surprised?
Our Grocery stores track what we purchase, and everyone said "oh well, cheaper prices" (BS But okay).
Our ISPs track our information, even hijack DNS error pages now. Everyone said "Oh well, they are a business"
Now this, and I guarantee it will be "Oh well, they are a business that needs to make money"
Consumers let this happen.
Well, yeah, they let it happen because they can see the use of collecting that info and therefore consent to it. The real question is whether this information is sent to other organisations, such as the government? I wonder how long it will take before someone's movements are tracked and used for police investigations. Perhaps it's already occurred.
As far as I'm concerned, any MMO is better than WoW. I never got why people liked it so much. It's the most terrible one I've ever played.
It's pure grinding, has little to no character customization, uninteresting classes that are all the same, very poor graphics...
It's like an old school MMORPG but without the roleplaying nor the old-school feel.
I played WoW for quite a while, and I can understand some of what you say. To begin with, I was between jobs, with a lot of time to kill. I thought I'd play WoW to see some interesting landscapes, which of course meant you had to level. It was a bit of a grind, but at least it changed somewhat every couple of levels, and there were new abilities, new instances, etc. I could play the game as I wanted, in my own time, without relying on other players, while still playing it as a multiplayer game.
But when I got to the max level, all that changed! When you're raiding, the game plays you! You have schedules, people are expecting you to have done your homework (hotkeys, spec, reading about encounters). On top of that, it's quite frustrating to wait for the recruitment of replacements, as people inevitable will decide not to turn up. And also, rerunning a boss encounter over and over is tiring.
I've never bought anything from any of these "artists", yet they foul the air "for my enjoyment" constantly.
That doesn't mean you need government on the scale it is now. Most of what government does these days is NOT the competition regulation that you're talking about.
Being from another Scandinavian country myself, I have to disagree. The nanny state is huge, and people are no longer able to be responsible for themselves. Everyone thinks about their rights, not their obligations to society. Also, there's a great deal of Jantelov thinking, which is basically an institutional form of jealousy. The scandies need to consider that they're no longer in an isolated, homogeneous part of the world, where everyone agrees about what the public pot should be spent on.
But anyway, that's not the main point I wanted to make. It's not that giving power to the wealthy is the main problem. (Sure, it can be a problem, no doubt.) The problem is giving power to large institutions. Microsoft, AT&T, Shell, etc... government is yet another example. If you ever work with or for one of these behemoths, it's understandable why you're frustrated. Large organisations lack common sense in their decision making, and they lack common empathy in their dealings with ordinary individuals. Due to their size (and influence) they're also able to live beyond their useful age, holding up resources (people, mainly) from more productive uses. Unfortunately, the west has institutionalized a system where the big firms work with the big governments to make sure neither of them is ever renewed. If organisations were smaller, we'd have a much more healthy society, where useful firms live, and old, unproductive ones die.
Well, if we can't, and this problem has come up before, the population will naturally stop growing. It won't be pretty if it's a hard landing, but various population explosions in the past are connected to innovations in farming, and why would it be any different in the future?
You could also just limit yourself to 2 children, which is just below the 2.1 replacement rate. (Read it many places)