You know, if you could say, "it feels cool to use," I'd be OK with that. It's the downsides.
I'm not even talking about the hacking concerns. The reason companies are so hot to sell these things is that they view them as consumer behavior tracking and modification devices.
You know, I always try to adopt a neutral, nonjudgmental attitude toward people adopting new technology, either positive or negative. Buying some piece of tech doesn't make someone cool; but on the flip side I always remind myself that it's not my place to judge how they spend their money.
Yet somehow, when it comes to these things, I just can't manage it. If someone tells me he's bought one of these things, and the next words out of his mouth aren't "So I can take it apart and hack it," I immediately think this person must be an idiot.
Right. In a world where everyone was perfectly logical and unemotional we'd say, "Well, the AI classifier identified distinct groups of racist customers." But in the world we live in reputation matters, and the most valuable customers don't want to be associated, even tenuously, with these morons.
Well, you get what you pay for. I also note that Vancouver has some of the "worst" traffic in Canada -- the average commuter spends 30 hours annually in traffic. This however compares to a US-wide average of 42 hours/year -- and well over 80 in the Bay Area. In my hometown forget the commute -- which his horrible -- the average driver spends 53 hours per year looking for parking.
As you get older, you value your time more than money, because you realize its running out.
Right; I was just clarifying: nothing about the computerized methodology fixes the problems people have always had with statistical results, which is determining whether you are looking at a comparable sample to the one you obtained those results with.
A working knowledge of probability and statistics is probably more important than a working knowledge of calculus.
Many, many years ago when I was a student I took the famously difficult stochastic processes course at MIT. Recently I went through MIT Open Courseware's 18.05 lectures, and was amazed by how much the teaching of probability and statistics has changed... and for the better.
Well, waiting times in ER are not necessarily representative of health care quality overall. For example ER waits have traditionally been a particularly weak area for the US, because you have to be triaged along with people who use the ER as their primary contact with the health care system.
It is hard to find any statistical area other than elective surgery where the US outperforms Canada. Canadians have a longer lifespan, can expect to spend 3.3 more years of that lifespan healthy (i.e. not in a nursing facility), and have lower infant mortality rates than Americans. Now Canadians may simply lead healthier lifestyles than Americans, but the evidence is mixed. Canadians have lower rates of obesity but higher smoking rates.
Still, even giving American health care every benefit of the doubt, it's hard to statistically paint Canadian health care a inferior. It is, of course, always possible to find individual situations where US health care might have done better.
The usual methodology for training is you start with a big sample of data and you randomly divide the data into records into two subsets; the first you use to train the model and the second you use to test the results of the training.
If there is no statistical difference between your training and testing groups, a better-than-random performance on the test data indicates that your algorithm actually learned something about the original universe of data. At that point you have the same problem you always have in statistics when you try to use your results: is some set of data you encounter in the wild so to speak really comparable to the data you build the model on?
One of the advantages of regression learning is that a classification your model produces is rebuttable. This is very important in a world where some courts are using proprietary software in sentencing to classify people by how likely they are to re-offend.
The human brain sees pattern everywhere it looks too.
I'm retired now but I've been doing a lot of reading and experimentation with decision tree based classification methods. I like these because the produce models you can examine critically, as opposed to so-called "deep learning" algorithms which produce results that you pretty much have to judge by their giving you the result you expect. It's not that that isn't useful in some cases, but I don't find it as interesting.
An hour and half driving's not bad at all, especially if traffic isn't bad. And as much as I like eating fish, I like catching them more so I release everything except hatchery trout, which are no loss to sport.
If I had a job lined up. Toronto is like a smaller, cleaner, better-organized version of New York. The cost of living is 27% less too.
The only thing is I also do like a place with a little more topography than Toronto, and access to wilderness-y areas within reasonable driving distance. There probably aren't any good places for fly-fishing around Toronto, that's almost a deal-breaker. Maybe Vancouver, then. Weather's better there, too: a bit rainy in the winter but with dry, cool summers with, long, long days. Good summer trout fishing in BC, too from what I hear, and salmon runs starting as early as August in some rivers.
You should never make any health decisions based on individual studies, and most importantly never, ever go looking for studies which support a decision you are going to make, because you'll find them.
I like to say that science isn't about truth, it's about evidence. The difference is there is logically no such thing as conflicting truths, although two different truths may make you feel conflicted. Evidence, on the other hand, contradicts other evidence all the time. It's the normal state of things.
What you want to do is base decisions on systematic review papers published in high impact journals. The purpose of a systematic review is to review the state of evidence at the current point in time. You may disagree with the author's conclusions, but the evidence should be laid out right there.
Well I'll tell you one thing that's more important than CPU speed: battery. If your battery doesn't last the day then you're carrying a piece of expensive glass in your pocket.
Well, it can be both. People do, after all, speculate in ordinary currency. An hour a day spent in currency speculation made John Maynard Keynes a rich man.
The whole point of Bitcoin is that governments don't control it (although this event demonstrates that this doesn't mean governments can't take actions that alter its value). This sounds like utopia to some people, but it's got its drawbacks. One of them is that there's no central bank taking actions to stabilize the value of the currency. If you can live with that, a decentralized cryptocurrency like bitcoin has its uses.
Well, not every aggregation of capital is a corporation; corporations are a specific kind of aggregation of capital which are created in law.
So it's possible to draw distinctions between the two.
Another way of looking at this is what is the market function of the thing you're talking about? Companies (whether corporations or something else) exist primarily to reduce transaction costs. Rather than negotiate with you on every task I want you to perform, I *hire* you with a broad job description. Corporate companies exist to promote capital formation.
Labor unions are essentially a labor cartel. I say this as a genuinely pro-labor liberal, so obviously I don't think a cartel is something which is necessarily always bad. I think unions are sometimes good because they offset the economic and political power of capital; however if unions ever got too powerful (far, far from what we have today) then I'd be against them. I particularly approve of unions of particularly powerless people, like those who perform low-status jobs in wealthy enclaves.
I should also point out that employers *do* attempt to form informal (because it's illegal) monopsonist cartels, as they have historically done with respect to engineering labor in Silicon Valley.
You know, this is one of the reasons you have market research and focus groups: so you don't do things and use messaging that reveals the world what an ignorant asshole you are.
It's not a job for amateurs. You need a professional asshole.
And easily modifiable to the user's needs too. That was possibly the coolest thing about the Model T, which is why the basic design lasted for twenty years, well after it had become hopelessly archaic. It was easy to hack. For the modern equivalent of three grand you got a basic, functional vehicle that could be turned into a cargo van, farm truck, or ski slope rope tow by anyone who could wield a hacksaw.
Here in the US most of our roadways were designed around cars. Even in older cities, the pre-automobile streets tend to be wider than their counterparts in older, say European cities. That's what drives our mania for large cars that simply would be impractical in many countries.
Japan has a lot of ancient roadways that are extremely narrow and sometimes windy. This has prompted the evolution of motor vehicles that seem amusingly tiny to American eyes, but are in fact the only practical solution in many cases for getting a motorized vehicle where it has to go.
If you want a bus that will reach rural elderly people on the other side of a bridge that was built just wide enough to accomodate a tractor, it's got to be a very small bus -- more the size that Americans would consider normal for a car, and not maybe not even a large car at that. But it is a bus in that it plays the role of a bus.
Japanese engineers seem to be more willing to think outside the box when it comes to system concepts; or perhaps Japanese consumers are more open-minded than American consumers. It's not just that they're good at miniaturization; they're also good at making ridiculously large things. Size is something that culturally speaking they're more at home playing with.
But people who can do better often don't, especially when their emotions run high. So the fact that you can be smart and have exhibited a pattern of smart behavior -- even critical thinking -- doesn't mean you're immune.
Slashdot Mirror
You know, if you could say, "it feels cool to use," I'd be OK with that. It's the downsides.
I'm not even talking about the hacking concerns. The reason companies are so hot to sell these things is that they view them as consumer behavior tracking and modification devices.
p>Sure, you don't need any of this, but for the price ($50), it's already paid for itself...
Set a guard over my mouth, oh Lord; keep watch over the door of my lips.
You know, I always try to adopt a neutral, nonjudgmental attitude toward people adopting new technology, either positive or negative. Buying some piece of tech doesn't make someone cool; but on the flip side I always remind myself that it's not my place to judge how they spend their money.
Yet somehow, when it comes to these things, I just can't manage it. If someone tells me he's bought one of these things, and the next words out of his mouth aren't "So I can take it apart and hack it," I immediately think this person must be an idiot.
Right. In a world where everyone was perfectly logical and unemotional we'd say, "Well, the AI classifier identified distinct groups of racist customers." But in the world we live in reputation matters, and the most valuable customers don't want to be associated, even tenuously, with these morons.
Well, you get what you pay for. I also note that Vancouver has some of the "worst" traffic in Canada -- the average commuter spends 30 hours annually in traffic. This however compares to a US-wide average of 42 hours/year -- and well over 80 in the Bay Area. In my hometown forget the commute -- which his horrible -- the average driver spends 53 hours per year looking for parking.
As you get older, you value your time more than money, because you realize its running out.
Right; I was just clarifying: nothing about the computerized methodology fixes the problems people have always had with statistical results, which is determining whether you are looking at a comparable sample to the one you obtained those results with.
Well, eliminating racism is kind of a straw argument. Is anyone actually arguing that racism will disappear because Google took action in this case?
I suspect the reason that Google stepped in here was profit maximization, not social engineering.
A working knowledge of probability and statistics is probably more important than a working knowledge of calculus.
Many, many years ago when I was a student I took the famously difficult stochastic processes course at MIT. Recently I went through MIT Open Courseware's 18.05 lectures, and was amazed by how much the teaching of probability and statistics has changed... and for the better.
Well, waiting times in ER are not necessarily representative of health care quality overall. For example ER waits have traditionally been a particularly weak area for the US, because you have to be triaged along with people who use the ER as their primary contact with the health care system.
It is hard to find any statistical area other than elective surgery where the US outperforms Canada. Canadians have a longer lifespan, can expect to spend 3.3 more years of that lifespan healthy (i.e. not in a nursing facility), and have lower infant mortality rates than Americans. Now Canadians may simply lead healthier lifestyles than Americans, but the evidence is mixed. Canadians have lower rates of obesity but higher smoking rates.
Still, even giving American health care every benefit of the doubt, it's hard to statistically paint Canadian health care a inferior. It is, of course, always possible to find individual situations where US health care might have done better.
The usual methodology for training is you start with a big sample of data and you randomly divide the data into records into two subsets; the first you use to train the model and the second you use to test the results of the training.
If there is no statistical difference between your training and testing groups, a better-than-random performance on the test data indicates that your algorithm actually learned something about the original universe of data. At that point you have the same problem you always have in statistics when you try to use your results: is some set of data you encounter in the wild so to speak really comparable to the data you build the model on?
One of the advantages of regression learning is that a classification your model produces is rebuttable. This is very important in a world where some courts are using proprietary software in sentencing to classify people by how likely they are to re-offend.
The human brain sees pattern everywhere it looks too.
I'm retired now but I've been doing a lot of reading and experimentation with decision tree based classification methods. I like these because the produce models you can examine critically, as opposed to so-called "deep learning" algorithms which produce results that you pretty much have to judge by their giving you the result you expect. It's not that that isn't useful in some cases, but I don't find it as interesting.
An hour and half driving's not bad at all, especially if traffic isn't bad. And as much as I like eating fish, I like catching them more so I release everything except hatchery trout, which are no loss to sport.
If I had a job lined up. Toronto is like a smaller, cleaner, better-organized version of New York. The cost of living is 27% less too.
The only thing is I also do like a place with a little more topography than Toronto, and access to wilderness-y areas within reasonable driving distance. There probably aren't any good places for fly-fishing around Toronto, that's almost a deal-breaker. Maybe Vancouver, then. Weather's better there, too: a bit rainy in the winter but with dry, cool summers with, long, long days. Good summer trout fishing in BC, too from what I hear, and salmon runs starting as early as August in some rivers.
You should never make any health decisions based on individual studies, and most importantly never, ever go looking for studies which support a decision you are going to make, because you'll find them.
I like to say that science isn't about truth, it's about evidence. The difference is there is logically no such thing as conflicting truths, although two different truths may make you feel conflicted. Evidence, on the other hand, contradicts other evidence all the time. It's the normal state of things.
What you want to do is base decisions on systematic review papers published in high impact journals. The purpose of a systematic review is to review the state of evidence at the current point in time. You may disagree with the author's conclusions, but the evidence should be laid out right there.
Well I'll tell you one thing that's more important than CPU speed: battery. If your battery doesn't last the day then you're carrying a piece of expensive glass in your pocket.
Well, it can be both. People do, after all, speculate in ordinary currency. An hour a day spent in currency speculation made John Maynard Keynes a rich man.
The whole point of Bitcoin is that governments don't control it (although this event demonstrates that this doesn't mean governments can't take actions that alter its value). This sounds like utopia to some people, but it's got its drawbacks. One of them is that there's no central bank taking actions to stabilize the value of the currency. If you can live with that, a decentralized cryptocurrency like bitcoin has its uses.
You are fed up, I take it.
Well, not every aggregation of capital is a corporation; corporations are a specific kind of aggregation of capital which are created in law.
So it's possible to draw distinctions between the two.
Another way of looking at this is what is the market function of the thing you're talking about? Companies (whether corporations or something else) exist primarily to reduce transaction costs. Rather than negotiate with you on every task I want you to perform, I *hire* you with a broad job description. Corporate companies exist to promote capital formation.
Labor unions are essentially a labor cartel. I say this as a genuinely pro-labor liberal, so obviously I don't think a cartel is something which is necessarily always bad. I think unions are sometimes good because they offset the economic and political power of capital; however if unions ever got too powerful (far, far from what we have today) then I'd be against them. I particularly approve of unions of particularly powerless people, like those who perform low-status jobs in wealthy enclaves.
I should also point out that employers *do* attempt to form informal (because it's illegal) monopsonist cartels, as they have historically done with respect to engineering labor in Silicon Valley.
Methinks the AC doth protest too much.
When are you idiots going to learn that logical fallacy arguments are in and of themselves fallacious
And when are you going to develop a self-protective sense of irony?
Once again the magical belief in the etymological fallacy rears its head.
You know, this is one of the reasons you have market research and focus groups: so you don't do things and use messaging that reveals the world what an ignorant asshole you are.
It's not a job for amateurs. You need a professional asshole.
And easily modifiable to the user's needs too. That was possibly the coolest thing about the Model T, which is why the basic design lasted for twenty years, well after it had become hopelessly archaic. It was easy to hack. For the modern equivalent of three grand you got a basic, functional vehicle that could be turned into a cargo van, farm truck, or ski slope rope tow by anyone who could wield a hacksaw.
Where does the dividing line come?
Here in the US most of our roadways were designed around cars. Even in older cities, the pre-automobile streets tend to be wider than their counterparts in older, say European cities. That's what drives our mania for large cars that simply would be impractical in many countries.
Japan has a lot of ancient roadways that are extremely narrow and sometimes windy. This has prompted the evolution of motor vehicles that seem amusingly tiny to American eyes, but are in fact the only practical solution in many cases for getting a motorized vehicle where it has to go.
If you want a bus that will reach rural elderly people on the other side of a bridge that was built just wide enough to accomodate a tractor, it's got to be a very small bus -- more the size that Americans would consider normal for a car, and not maybe not even a large car at that. But it is a bus in that it plays the role of a bus.
Japanese engineers seem to be more willing to think outside the box when it comes to system concepts; or perhaps Japanese consumers are more open-minded than American consumers. It's not just that they're good at miniaturization; they're also good at making ridiculously large things. Size is something that culturally speaking they're more at home playing with.
Situationally dumb, yes.
But people who can do better often don't, especially when their emotions run high. So the fact that you can be smart and have exhibited a pattern of smart behavior -- even critical thinking -- doesn't mean you're immune.