I'm closer to 50 than 40 and have had no trouble at all finding jobs recently. In fact, after moving from programing into management over the last few years, I decided a few months ago that I wasn't enjoying life much anymore so I found a job as a programmer again. It took all of a week to get a couple offers. Maybe it's where I live.
So your argument boils down to this: any actual measurements that show overall improvements are irrelevant since they are always trumped by anecdotal evidence of particular abuses.
I agree that there's more to life than infant mortality and starvation. However, I still haven't seen any evidence that things are overall getting worse. All I see are examples of people who are suffering. That doesn't tell me much except that there's lots of room for improvement, because there were plenty of examples of people suffering 30 years ago, too
It's like, I say "the weather is cooling off" and you say "It can't be, it's still hot!". I say,"but it was hotter yesterday." You say "but, jeez, its almost 90 degrees". I say, but yesterday it was 105". You say, "But look, that fellow is sweating---it can't be cooling off."
"it is wrong to adopt policies that provide additional luxury for those who already have more than they can use, while others go without enough food -- assuming alternative policies are available."
Why? That's not a rhetorical question. Why? You seem to believe that there is some inherent good in me flushing money down the toilet to reduce income disparity, in spite of the fact that no one would benefit.
"I believe that in the long run increasing the disparity between rich and poor guarantees that the rich are going to have the power to reduce the poor to subsistence slavery."
Why would they want to do that? It is certainly not in the best interests of the rich to reduce the standard of living of the poor. The rich get that way by selling stuff. You can't sell much to poor people.
He's likely quoting verbatim from some mass-media right-wing ideologue, who similarly said it with no actual knowledge to back it up.
Oh, now there's a good argument.:-)
This is comically untrue. Check the "made in" on almost every common consumer good you own -- clothes, cars, toys, baby strollers. Indonesia, Pakistan, Bangladesh, Malaysia, Thailand, Mexico, and of course, the People's Republic of China.
And in which of those places has the standard of living dropped since, say, 1970? In which is there more starvation than there was thirty years ago? Note: I didn't ask "in which of those places are some people suffering."
1. Where is it proven? 2. It's not a troll. It's pointing out a fact: being worse off relative to the rich is not the same as being worse off in absolute terms. We could reduce global disparity by nuking Canada and Finland. Would that help the poor? Of course not. 3. Check your facts. The argument seems to be "things are getting worse, and there's globalism, so globalism makes things worse." That argument is pretty weak to begin with, but it is totally silly if things are actually getting better. And all objective data show that to be the case.
The last paragraph betrays a huge misunderstanding of reality:
As a result of globalization, the divisions between the world's rich and the poor continues to widen. According to the United Nations Development Program, the richest one percent of the world's population receives as much income as the poorest 57 percent. More than a billion people live on less than a dollar a day; nearly a billion lack any access to clean water; 826 million suffer from malnutrition; 10 million die annually due to lack of basic health care. Some of these conditions pre-dated globalization, but the new economy has hardly improved matters.
This is so far from reality that it's hard to know where to start debunking. First of all "As a result of globalization" barely qualifies a hypothesis; it certainly isn't a proven fact. "As a result of disparities in legal respect for property rights" is a better hypothesis.
Second, growing disparity between rich and poor is not necessarily bad. If you could wave a wand and improve the standard of living of the poor by 8x, but in the process make the rich 10x as rich, would you do so? If not, why not? Just because disparity would grow?
Third, by almost all objective standards, the amount and severity of poverty in the world has dropped significantly during the era of globalization. There is less starvation; infant mortality is lower; life expectancy is longer; there is less malnutrition.
Finally, the places where things haven't improved correspond not to hotbeds of globalization, but to regimes so repressive or corrupt that global investment doesn't happen. Globalization has barely touched most of Africa or North Korea because no one will invest. In those places the standard of living is wretched.
Re:Perl isn't unreadable - some Perl programs are
on
Exegesis 4 Out
·
· Score: 2
1. As of Perl5, I disagree. The prefix is not very useful because it applies to the expression, not the variable. E.g., you write "my @foo = (1,2,3); $foo[3];". I don't see how that's an improvement over "my foo = (1,2,3); foo[3]." In both cases, the type of foo is obvious from the context; the prefix characters don't add anything. In Perl6, the symbols will actually be useful.
2. Interpolating in strings is not a good reason for forcing every variable to start with a funky character. You could instead simply use "$" in strings to mean "replace the following expression with its value". Python does something like this, if I remember right.
One problem: improving efficiency often increases total consumption. For example, refrigerators used to be very inefficient and hence expensive to operate. People got along with much less refrigeration. Then much more efficient refrigerators became common; people found all sorts of new uses for refrigeration, and the total amount of energy spent on refrigeration is now far higher.
If you really want to reduce petroleum consumption, mandate that passenger cars get at most 0.1 miles per gallon. Gasoline usage will drop to nearly zero overnight. Or mandate that oil-burning furnaces be no more than 1/10 as efficient as today: people will start wearing sweaters really quick.
Suppose some agency actually did build a machine that could crack 1024-bit RSA. How would they use it? The answer is, they would keep it very secret and use it only on very important stuff---nuclear threats, etc. They would certainly not risk revealing it's existence to crack small cases.
Think about this for a minute. Five years ago, I was running Quake on a machine about 1/40 the speed of the machine I'm using now, and Quake ran pretty well. Now assume that Java detractors are right and Java is only 1/2 as fast as C/C++. Why should it have trouble with Quake? Java runs many times faster on my current machine than carefully written C/C++ did on my machine five years ago.
That sort of depends on what ideal you pursue, doesn't it? I'm sure we can all think of historical figures who worked ceaslessly to promote horrible ideas. (Pol Pot?) Not to compare RMS with serial killers and depots, just pointing out that ceasless pursuit of an ideal is admirable exactly when the ideal is admirable.
Your second point is the nastiest. However, it seems to me that this weakness could be mitigated to some extent. The key authority could strip off the real message authentication and substitute its own. I.e., it would say, effectively, "I certify that Joe Shmoe indeed sent this message, but you will have to take my word for it." Furthermore, it could then shred it's own authentication after the expiry date. Although someone might have a screen capture of the email, both Joe Shmoe and the key authority could plausibly deny that it was real.
Well, let's see... what advantages does it have over a missle? Mainly, it's a tad faster---the difference between the speed of sound and the speed of light. There is no time for evasive maneuvers and no time for countermeasures. Either you already have defenses in place when the button is pushed or you get blown up. And it will be at least a few decades before most nations have any effective defense.
Re:Classes and APIs more important than language
on
What is .NET?
·
· Score: 2
Unfortunately, Guile got off to a very bad start. One of the very first things they did was to break compatibility with Scheme in some fairly stupid ways. The result was that the existing Scheme community, which could have been their biggest asset, mostly turned their collective backs on the project. The whole episode was really amazingly stupid: a wonderful pportunity was wasted because of the arrogance of a few people. Sad...
Re:.NET good, not evil
on
What is .NET?
·
· Score: 4, Interesting
But it's not "any language". It's a collection of languages that are nearly isomorphic with C#. In fact, there are currently many more languages, and more diverse languages, targetted to the JVM than there are to the.NET VM. See, for example, this list, which contains about 160.
It isn't all about good names, but that is certainly part of the problem with overloading. There are such a limited number of operators to overload that the meaning is almost invariably corrupted. Yes, you can write bad method names, but in my experience it's much less common, simply because the obvious name is usually best anyway.
Whether a formal system satisfies the axioms of a field is undecidable. Haskell doesn't even try. Haskell allows you to promise that a collection of operations does follow the axioms, but it has no way of enforcing that.
Sigh. I knew I'd get this response:-) Maybe this has been discussed before?:-)
First, I'll concede the obvious. Yes, if there were some way to overload operators only for things that obeyed field axioms, then it would be a Good Thing. Unfortunately, I don't know of any way to enforce that.
What happens in real languages where operator overloading is supported? Left shift is overloaded to mean print. Plus acts as sum, or bit-string concatenation, or dot product, or set union, or... In short, small snippets of code are completely indecipherable unless you are deeply familiar with all the little idioms and short cuts that the programmer used to save a few keystrokes. Give me nice, clean method calls any day.
As someone with a math degree, I disagree. Here is why.
First, what does this do?
a = b * c;
No one knows. Could be element-by-element multiply; could be matrix multiply; could be something else entirely. Any since the actual method call (if that's what it is) is implicit, I don't know where to go look to find out. I'd much rather see
a = Matrix.mult(b,c)
because I have a better hunch what it does, and I know where to look verify that hunch.
Second, the real work that goes into mathematical code is done on the blackboard, not the keyboard. The fact that the typing notation is suboptimal just isn't that important.
Third, infix is not particularly more intuitive. Mathematicians use all kinds of notations, including infix, prefix, postfix, and lots of two-dimensional stuff that doesn't fit any of those.
Finally, given a choice, real mathies tend to use languages like Mathematica instead of C/C++, Java, Lisp, etc. But that has to do more with real semantic differerences, not syntactic sugar.
No. I've been inside one of those companies, and the GPL is in fact the biggest obstacle, not the biggest advantage. They don't like the viral properties of the GPL. They are much more worried about being forced to release existing trade secrets than they are about competitors adopting their code.
No, the real reason is simply buzz. Mind share. There is nothing particularly wrong with BSD, but it isn't exciting and it doesn't get the press. It's viewed as old and a little stodgy. (To Sun, especially, it seems old and stodgy---the original SunOS was based on BSD and they spent a great deal to move to SysV, so going back isn't viewed as a sensible solution. )
This is just wrong. There is no reason to send a "ship". There is no reason to slow down. Find a metallic asteroid a few hundred feet in diameter.
Accellerate it to 0.9c using any of a number of fairly crude methods of propulsion. Mount a small guidance system capable of hitting a planet the size of earth. Wait.
I don't think there's much point in fear, because there is absolutely nothing we could do about it.
However, I completely disagree with your reasoning. Xenophobia may well be the rational course. This is particularly the case if interstellar travel and trade is impractical. Destroying a nearby planet is extremely cheap. The downside is... what? If I accept your argument, then there will never be any commerce or contact anyway. The upside is that the probability that they will destroy my planet is now zero instead of some positive value. Unless there is something to be gained by letting another civilization live (as there certainly is between civilizations on Earth), it's clearly rational to destroy them.
I think "L" is the problematic one. The term "Lifetime" is deceptive, because it isn't the lifetime of the civilization, it's the span of time that the civilization wastes huge amounts of power broadcasting into space.
It may well be that most civilizations go through a brief broadcast period and then learn to use point-to-point methods of communication that aren't easily detectable. One reason is simple economics: dumping energy into space is wasteful. But it may also be that successful civilizations actively avoid broadcasting their presence to avoid hostile encounters.
Here's a depressing thought: they may also consider it prudent to quickly destroy nearby infant civilizations quickly, perhaps by accellerating small chunks of rock to near lightspeed and aiming them at noisy planets. Such an attack could obliterate life on earth with virtually no warning at all.
Slashdot Mirror
I'm closer to 50 than 40 and have had no trouble at all finding jobs recently. In fact, after moving from programing into management over the last few years, I decided a few months ago that I wasn't enjoying life much anymore so I found a job as a programmer again. It took all of a week to get a couple offers. Maybe it's where I live.
Please, someone with moderation points: mod this
funny. Very funny. Very, very funny.
So your argument boils down to this: any actual measurements that show overall improvements are irrelevant since they are always trumped by anecdotal evidence of particular abuses.
I agree that there's more to life than infant mortality and starvation. However, I still haven't seen any evidence that things are overall getting worse. All I see are examples of people who are suffering. That doesn't tell me much except that there's lots of room for improvement, because there were plenty of examples of people suffering 30 years ago, too
It's like, I say "the weather is cooling off" and you say "It can't be, it's still hot!". I say,"but it was hotter yesterday." You say "but, jeez, its almost 90 degrees". I say, but yesterday it was 105". You say, "But look, that fellow is sweating---it can't be cooling off."
Why? That's not a rhetorical question. Why? You seem to believe that there is some inherent good in me flushing money down the toilet to reduce income disparity, in spite of the fact that no one would benefit.
Why would they want to do that? It is certainly not in the best interests of the rich to reduce the standard of living of the poor. The rich get that way by selling stuff. You can't sell much to poor people.
Oh, now there's a good argument.
And in which of those places has the standard of living dropped since, say, 1970? In which is there more starvation than there was thirty years ago? Note: I didn't ask "in which of those places are some people suffering."
1. Where is it proven?
2. It's not a troll. It's pointing out a fact: being worse off relative to the rich is not the same as being worse off in absolute terms. We could reduce global disparity by nuking Canada and Finland. Would that help the poor? Of course not.
3. Check your facts. The argument seems to be "things are getting worse, and there's globalism, so globalism makes things worse." That argument is pretty weak to begin with, but it is totally silly if things are actually getting better. And all objective data show that to be the case.
This is so far from reality that it's hard to know where to start debunking. First of all "As a result of globalization" barely qualifies a hypothesis; it certainly isn't a proven fact. "As a result of disparities in legal respect for property rights" is a better hypothesis.
Second, growing disparity between rich and poor is not necessarily bad. If you could wave a wand and improve the standard of living of the poor by 8x, but in the process make the rich 10x as rich, would you do so? If not, why not? Just because disparity would grow?
Third, by almost all objective standards, the amount and severity of poverty in the world has dropped significantly during the era of globalization. There is less starvation; infant mortality is lower; life expectancy is longer; there is less malnutrition.
Finally, the places where things haven't improved correspond not to hotbeds of globalization, but to regimes so repressive or corrupt that global investment doesn't happen. Globalization has barely touched most of Africa or North Korea because no one will invest. In those places the standard of living is wretched.
1. As of Perl5, I disagree. The prefix is not very useful because it applies to the expression, not the variable. E.g., you
write "my @foo = (1,2,3); $foo[3];". I don't see how that's an improvement over "my foo = (1,2,3); foo[3]." In both cases, the type of foo is obvious from the context; the prefix characters don't add anything. In Perl6, the symbols will actually be useful.
2. Interpolating in strings is not a good reason for forcing every variable to start with a funky character. You could instead simply use "$" in strings to mean "replace the following expression with its value". Python does something like this, if I remember right.
Life is not as simple as you assume.
One problem: improving efficiency often increases total consumption. For example, refrigerators used to be very inefficient and hence expensive to operate. People got along with much less refrigeration. Then much more efficient refrigerators became common; people found all sorts of new uses for refrigeration, and the total amount of energy spent on refrigeration is now far higher.
If you really want to reduce petroleum consumption, mandate that passenger cars get at most 0.1 miles per gallon. Gasoline usage will drop to nearly zero overnight. Or mandate that oil-burning furnaces be no more than 1/10 as efficient as today: people will start wearing sweaters really quick.
Suppose some agency actually did build a machine that could crack 1024-bit RSA. How would they use it? The answer is, they would keep it very secret and use it only on very important stuff---nuclear threats, etc. They would certainly not risk revealing it's existence to crack small cases.
Think about this for a minute. Five years ago, I was running Quake on a machine about 1/40 the speed of the machine I'm using now, and Quake ran pretty well. Now assume that Java detractors are right and Java is only 1/2 as fast as C/C++. Why should it have trouble with Quake? Java runs many times faster on my current machine than carefully written C/C++ did on my machine five years ago.
That sort of depends on what ideal you pursue, doesn't it? I'm sure we can all think of historical figures who worked ceaslessly to promote horrible ideas. (Pol Pot?) Not to compare RMS with serial killers and depots, just pointing out that ceasless pursuit of an ideal is admirable exactly when the ideal is admirable.
Your second point is the nastiest. However, it seems to me that this weakness could be mitigated to some extent. The key authority could strip off the real message authentication and substitute its own. I.e., it would say, effectively, "I certify that Joe Shmoe indeed sent this message, but you will have to take my word for it." Furthermore, it could then shred it's own authentication after the expiry date. Although someone might have a screen capture of the email, both Joe Shmoe and the key authority could plausibly deny that it was real.
Well, let's see... what advantages does it have over a missle? Mainly, it's a tad faster---the difference between the speed of sound and the speed of light. There is no time for evasive maneuvers and no time for countermeasures. Either you already have defenses in place when the button is pushed or you get blown up. And it will be at least a few decades before most nations have any effective defense.
Unfortunately, Guile got off to a very bad start. One of the very first things they did was to break compatibility with Scheme in some fairly stupid ways. The result was that the existing Scheme community, which could have been their biggest asset, mostly turned their collective backs on the project. The whole episode was really amazingly stupid: a wonderful pportunity was wasted because of the arrogance of a few people. Sad...
But it's not "any language". It's a collection of languages that are nearly isomorphic with C#. In fact, there are currently many more languages, and more diverse languages, targetted to the JVM than there are to the .NET VM. See, for example, this list, which contains about 160.
It isn't all about good names, but that is certainly part of the problem with overloading. There are such a limited number of operators to overload that the meaning is almost invariably corrupted. Yes, you can write bad method names, but in my experience it's much less common, simply because the obvious name is usually best anyway.
Whether a formal system satisfies the axioms of a field is undecidable. Haskell doesn't even try. Haskell allows you to promise that a collection of operations does follow the axioms, but it has no way of enforcing that.
Sigh. I knew I'd get this response :-) Maybe this has been discussed before? :-)
... In short, small snippets of code are completely indecipherable unless you are deeply familiar with all the little idioms and short cuts that the programmer used to save a few keystrokes. Give me nice, clean method calls any day.
First, I'll concede the obvious. Yes, if there were some way to overload operators only for things that obeyed field axioms, then it would be a Good Thing. Unfortunately, I don't know of any way to enforce that.
What happens in real languages where operator overloading is supported? Left shift is overloaded to mean print. Plus acts as sum, or bit-string concatenation, or dot product, or set union, or
As someone with a math degree, I disagree. Here is why.
First, what does this do?
a = b * c;
No one knows. Could be element-by-element multiply; could be matrix multiply; could be something else entirely. Any since the actual method call (if that's what it is) is implicit, I don't know where to go look to find out. I'd much rather see
a = Matrix.mult(b,c)
because I have a better hunch what it does, and I know where to look verify that hunch.
Second, the real work that goes into mathematical code is done on the blackboard, not the keyboard. The fact that the typing notation is suboptimal just isn't that important.
Third, infix is not particularly more intuitive. Mathematicians use all kinds of notations, including infix, prefix, postfix, and lots of two-dimensional stuff that doesn't fit any of those.
Finally, given a choice, real mathies tend to use languages like Mathematica instead of C/C++, Java, Lisp, etc. But that has to do more with real semantic differerences, not syntactic sugar.
No. I've been inside one of those companies, and the GPL is in fact the biggest obstacle, not the biggest advantage. They don't like the viral properties of the GPL. They are much more worried about being forced to release existing trade secrets than they are about competitors adopting their code.
No, the real reason is simply buzz. Mind share. There is nothing particularly wrong with BSD, but it isn't exciting and it doesn't get the press. It's viewed as old and a little stodgy. (To Sun, especially, it seems old and stodgy---the original SunOS was based on BSD and they spent a great deal to move to SysV, so going back isn't viewed as a sensible solution. )
This is just wrong. There is no reason to send a "ship". There is no reason to slow down. Find a metallic asteroid a few hundred feet in diameter.
Accellerate it to 0.9c using any of a number of fairly crude methods of propulsion. Mount a small guidance system capable of hitting a planet the size of earth. Wait.
I don't think there's much point in fear, because there is absolutely nothing we could do about it.
... what? If I accept your argument, then there will never be any commerce or contact anyway. The upside is that the probability that they will destroy my planet is now zero instead of some positive value. Unless there is something to be gained by letting another civilization live (as there certainly is between civilizations on Earth), it's clearly rational to destroy them.
However, I completely disagree with your reasoning. Xenophobia may well be the rational course. This is particularly the case if interstellar travel and trade is impractical. Destroying a nearby planet is extremely cheap. The downside is
I think "L" is the problematic one. The term "Lifetime" is deceptive, because it isn't the lifetime of the civilization, it's the span of time that the civilization wastes huge amounts of power broadcasting into space.
It may well be that most civilizations go through a brief broadcast period and then learn to use point-to-point methods of communication that aren't easily detectable. One reason is simple economics: dumping energy into space is wasteful. But it may also be that successful civilizations actively avoid broadcasting their presence to avoid hostile encounters.
Here's a depressing thought: they may also consider it prudent to quickly destroy nearby infant civilizations quickly, perhaps by accellerating small chunks of rock to near lightspeed and aiming them at noisy planets. Such an attack could obliterate life on earth with virtually no warning at all.
Which is why it's important to get the local Office Superstore to sell machines loaded with Linux and $150 dollars cheaper.
Voltage is not energy. There are lots of ways to
increase the voltage across a pair of terminals without increasing the energy in the system.