Bear in mind statistics was a barely developed science at the time such studies were developed, and actually doing a study which clearly shows the difference between smokers and non-smokers is hard to do well (and takes a fair amount of time).
As an aside, Ronald Fisher, probably the biggest single figure in modern statistics, was absolutely convinced of the safety of tobacco.
What rubbish. Science is about testing the predictions made by a hypothesis. Your hypothesis is that a man in the sky made the universe. Great. Do tell me how I can predict something with that so I can decide it is a better hypothesis than, say, the creation of the universe by a being made of egg and fine flour?
Chavs existed long before Maggie; they just had no visibility. Crime in the UK has declined since the implementation of the 1980s reforms. And I don't think anyone is talking about repealing the HSE; but the fact is that when a country becomes rich enough, people refuse to work in poor conditions. It happened in the UK, in the US, and it will happen in China.
*really?* I remember what industry used to look like - in the first world - back in the heyday of the trade union. Health and safety didn't exist. Good luck if you were black, or Jewish, or gay, or female. If you fell out with the union rep, you were on your own. And in the meantime, you were being tapped for union dues that went to fund a party espousing some of the most fucked-up economic policies the UK ever saw.
What creates better conditions for workers is economic progress, so that workers don't have to go and work in a paper mill where they lose fingers as a matter of routine, or a shipyard where workers fell from gantries every week.
I don't know. I feel rather sentimental about this whole sorry affair, since I installed Linux for the first time around two weeks before it all kicked off. Happy times.
I seem to recall (it may have been here, actually) a study where people's driving skills were independently assessed alongside how they drove. Driving skill was very strongly correlated to their tendency to follow the rules of the road.
Point one, there are queries that can only be written using a sub-query (unusual, but they exist); point two, computers are supposed to help, not hinder. What's the point of having a declaritive language if I can't declare a query in the form I think of it in?
The problem is that mortality is strongly heteroskedastic and it's hard to make useful inferences at higher ages. The largest studies actually indicate that mortality hits a constant at high ages, but the effect is difficult to pick out.
That's a pretty common sentiment. However, you won't just wake up one morning as a bent, hobble old man with nothing to live for; it will be a slow, multi-year descent and you won't be aware of it.
Sorry if that's a bit bleak, but it's what happens. Nobody wants to be senile and bed-ridden, but we're not smart enough to dodge it either.
Yes and no. As far as resources go, we can use our resources much, much more efficiently; the barriers to this are political. If we let it, economics is very good at regulating resource usage.
The real question is personal: if you've never experienced old age close up, it's not pleasant. You're talking about potentially spending forty years of your life or more reliant on other people to dress, wash, and feed you. Apart from the expense, which people will not acknowledge until it's too late, it is a terrible way to live.
The cohort effect is fascinating; there's a generation in the UK born around 1935 who have extraordinary mortality, and every year it gets better and better. Yet there is no real causal explanation for it (lots of theories, such as WWII rationing, but no verifiable causal links).
On the present topic, there is a well known theory in mortality studies that death rates stabilise at high ages, around 110 in humans. Research done on the US Social Security Master Death File has suggested that this rate is (IIRC) about 46%. Given that of a population of 300 million, only a few thousand (rough estimate from ELT15, the only life table I have handy that runs that high) reach 110, it seems pretty reasonable you'll get very few lives surviving beyond 114 given the compounding of death rates.
Losing weight via exercise requires quite a lot of work. It is doable, and worth doing, but that paper suggests 250 mins/week of fairly intense exercise. That's quite a lot - I get it through running 25 mins/day during the week and a fifteen minute cycle commute, but not everyone has the luxury of being able to find that much time in a week.
Here's an easier (and more effective version): sprint for 20 seconds. Jog for ten seconds. Repeat eight times (you won't manage, but keep trying). If you're badly out of shape, do some jogging for a few weeks before you try this.
The problem is not relly one of efficient allocation of plants to those who need them most, but the efficient allocation of inputs to those plants - in particular, water. If farmers globally were actually charged for the water they used, the cost of producing meat would jump quite sharply, and supply and demand would take care of the rest.
You're not alone - musicians are easy prey for this fallacy. Martin made thousands and thousands of factory produced guitars in the twenties and thirties, and of course they now have a legendary reputation simply because only the good ones are still around.
I don't think air wars are a counter example. Political conflicts are never solved from 30,000 feet. Yes, most wars are about resources on some level, but that is down to the politics of war. To get the crops, water, or copper out, you still need to be controlling the ground.
The main issue with cyber war, it seems to me, is one of controllable intent and resources. Stuxnet blew a massive amount of resource on a single shot. Some of that resource can be solved by throwing money and time at it but you cannot just trot down to BAE and buy another truckload of zero-day exploits. I'm not saying that internet based hostilities can have no effect - indeed, the vulnerability of infrastructure to what the industry terms "advanced persistent threats" is probably understated. But as an effective warfighting tool, I just don't see it. No country is ever going to roll over because someone hacked into their defence ministry.
Rubbish. Warfare is ultimately the art of taking and controlling territory. Cyber attacks may play some part in this but when you win a "cyberwar" you've got no more territory and no more control. Further reading.
Ron Paul is not in the slightest bit serious. If he were ever to be elected President, he would have no real power to carry out the reforms he claims he will. Most of what he wants to do require legislation, and there is no way Congress will come close to passing what he wants. Ever. Ron Paul knows this - there's no way he can't - and he's clearly having a whale of a time on his libertarian soapbox. I have a lot of sympathy for his politics, but for effectiveness? No.
Slashdot Mirror
As an aside, Ronald Fisher, probably the biggest single figure in modern statistics, was absolutely convinced of the safety of tobacco.
What rubbish. Science is about testing the predictions made by a hypothesis. Your hypothesis is that a man in the sky made the universe. Great. Do tell me how I can predict something with that so I can decide it is a better hypothesis than, say, the creation of the universe by a being made of egg and fine flour?
Or perhaps the people that package these things have the sense to put them in something rather safer than a cardboard box.
Chavs existed long before Maggie; they just had no visibility. Crime in the UK has declined since the implementation of the 1980s reforms. And I don't think anyone is talking about repealing the HSE; but the fact is that when a country becomes rich enough, people refuse to work in poor conditions. It happened in the UK, in the US, and it will happen in China.
Lolzypops. When you buy on the stock exchange, your money isn't going anywhere near the company. Do keep up.
What creates better conditions for workers is economic progress, so that workers don't have to go and work in a paper mill where they lose fingers as a matter of routine, or a shipyard where workers fell from gantries every week.
And in the meantime the children of the third world can go die on a rubbish dump? Fuck you.
I don't know. I feel rather sentimental about this whole sorry affair, since I installed Linux for the first time around two weeks before it all kicked off. Happy times.
I seem to recall (it may have been here, actually) a study where people's driving skills were independently assessed alongside how they drove. Driving skill was very strongly correlated to their tendency to follow the rules of the road.
Idiot. What do you think a $5 bill is worth?
Point one, there are queries that can only be written using a sub-query (unusual, but they exist); point two, computers are supposed to help, not hinder. What's the point of having a declaritive language if I can't declare a query in the form I think of it in?
The problem is that mortality is strongly heteroskedastic and it's hard to make useful inferences at higher ages. The largest studies actually indicate that mortality hits a constant at high ages, but the effect is difficult to pick out.
Sorry if that's a bit bleak, but it's what happens. Nobody wants to be senile and bed-ridden, but we're not smart enough to dodge it either.
The real question is personal: if you've never experienced old age close up, it's not pleasant. You're talking about potentially spending forty years of your life or more reliant on other people to dress, wash, and feed you. Apart from the expense, which people will not acknowledge until it's too late, it is a terrible way to live.
The cohort effect is fascinating; there's a generation in the UK born around 1935 who have extraordinary mortality, and every year it gets better and better. Yet there is no real causal explanation for it (lots of theories, such as WWII rationing, but no verifiable causal links). On the present topic, there is a well known theory in mortality studies that death rates stabilise at high ages, around 110 in humans. Research done on the US Social Security Master Death File has suggested that this rate is (IIRC) about 46%. Given that of a population of 300 million, only a few thousand (rough estimate from ELT15, the only life table I have handy that runs that high) reach 110, it seems pretty reasonable you'll get very few lives surviving beyond 114 given the compounding of death rates.
That's hilarious. In the real world, Foxconn's suicide rate is something like a third of America's. It only gets noticed because it's so big.
Losing weight via exercise requires quite a lot of work. It is doable, and worth doing, but that paper suggests 250 mins/week of fairly intense exercise. That's quite a lot - I get it through running 25 mins/day during the week and a fifteen minute cycle commute, but not everyone has the luxury of being able to find that much time in a week.
Here's an easier (and more effective version): sprint for 20 seconds. Jog for ten seconds. Repeat eight times (you won't manage, but keep trying). If you're badly out of shape, do some jogging for a few weeks before you try this.
The problem is not relly one of efficient allocation of plants to those who need them most, but the efficient allocation of inputs to those plants - in particular, water. If farmers globally were actually charged for the water they used, the cost of producing meat would jump quite sharply, and supply and demand would take care of the rest.
You're not alone - musicians are easy prey for this fallacy. Martin made thousands and thousands of factory produced guitars in the twenties and thirties, and of course they now have a legendary reputation simply because only the good ones are still around.
And of course, there is a censoring issue. All old things that we see are good quality, simply because everything that broke has been thrown away.
Personally, I find it highly offensive that they used a fixed width layout.
The main issue with cyber war, it seems to me, is one of controllable intent and resources. Stuxnet blew a massive amount of resource on a single shot. Some of that resource can be solved by throwing money and time at it but you cannot just trot down to BAE and buy another truckload of zero-day exploits. I'm not saying that internet based hostilities can have no effect - indeed, the vulnerability of infrastructure to what the industry terms "advanced persistent threats" is probably understated. But as an effective warfighting tool, I just don't see it. No country is ever going to roll over because someone hacked into their defence ministry.
Rubbish. Warfare is ultimately the art of taking and controlling territory. Cyber attacks may play some part in this but when you win a "cyberwar" you've got no more territory and no more control. Further reading.
Ron Paul is not in the slightest bit serious. If he were ever to be elected President, he would have no real power to carry out the reforms he claims he will. Most of what he wants to do require legislation, and there is no way Congress will come close to passing what he wants. Ever. Ron Paul knows this - there's no way he can't - and he's clearly having a whale of a time on his libertarian soapbox. I have a lot of sympathy for his politics, but for effectiveness? No.