IF GM's Volt, or Chrysler's EV concept ever make production, they have a solution to the problem. The vehicle travels 40 or 60 miles on its battery charge, and if you keep driving an onboard generator turns on to supply electricity and recharge the battery.
In effect your car is a complete EV for a reasonable commute, and a fossil fuel vehicle (although not a traditional one) for long trips.
But the rumors are floating around that when the GM Volt hits showrooms, the price tag on the tiny four seat sedan will be above $30,000.
In 2008 dollars, the nuclear plant cost about 16 times as much to construct, $266 million versus $4 billion. But the nuclear plant supplies 104 times as much energy per year as the 134 million kilowatt-hours from the solar plant. So the nuclear power generates more than six times as much energy per dollar spent on construction, which still dramatically offsets the ongoing maintenance and fuel cost advantages of solar.
Nuclear power supplies 20% of the electricity used in the country, and it's done that for decades. Coal kills people with black lung disease, mining collapses, and of course lung cancer, asthma, and emphysema from coal burning emissions.
If we multiply the number of Americans killed by nuclear power each year by 5 to extrapolate if nuclear power supplied all of our energy, we still end up with a number far below the deaths from coal.
Oops, my math is off, the nuclear plant costs about 16 times as much. That's still roughly 6 times as much electricity per dollar spent, an enormous cost edge for nuclear.
Solar thermal power is nice because it can supply power 24 hours a day unless you have a long stretch of highly overcast days. The problem is that solar thermal power plants are ridiculously expensive compared to coal or nuclear.
The Nevada Solar One (http://en.wikipedia.org/wiki/Nevada_Solar_One) solar thermal power plant cost $266 million and supplies 134 million kwh per year.
The Calvert Cliffs power plant in Maryland (http://en.wikipedia.org/wiki/Calvert_Cliffs_Nuclear_Generating_Station) cost $766 million to construct in the 1970s. Depending upon how you calculate inflation, that would be maybe $4 billion today. The plant supplies more than 14,000 Gwh per year, on average, or 14 billion kwh.
So the nuclear power plant costs 8 times as much to construct, but supplies 104 times as much power as the Solar Thermal plant in a year. That's 12 times as much electricity per dollar spent. I'm certain the higher maintenance and fuel costs on a nuclear plant do not nearly offset that difference. Nuclear is still the way to go.
Most of the added cost of ground source is getting the wells drilled and the piping from the wells into your home. In 20 years, you may need to replace your compressor, but barring some very unusual problems you well should be fine for many more decades.
Plus, the ground source system compressor is indoors, which improves the system longevity versus an outdoor traditional air-source heat pump.
Funny enough, we're in the middle of having a ground source heat pump installed right now. We opted for ground source over air because our house is old, some of the walls are stone, and it's in an L shape. So we have high surface area for relatively modest interior space (2100 square feet for a family of five), and some of the house is effectively impossible to insulate well. My wife and I figured that no matter what we do, we're going to lose a lot of heat. A hypothetical 25% cost savings difference between an air source heat pump and a ground source heat pump will pay for itself with our home far more quickly than most better shaped, better insulated properties.
We live between Philadelphia and Reading. Funny that you're near Tannersville - my best friend grew up outside Snydersville, PA. His parents' house is probably in your back yard.
Wikipedia says the UK summer average temperature in the warmest regions is 15.5 to 17.7 C (60-64 Fahrenheit).
Many areas in the US are significantly warmer. Telling someone to 'just open a window' doesn't work when it's 38 C (100 Fahrenheit) for 8 hours of the day or more, like most summer days in parts of Arizona.
While I agree, the counterpoint is that you get a lot of diminishing returns for your PC investment. Depending upon what you get, maybe $800 well spent (not counting the monitor) will get you 80 fps.
Bump your PC spending to $1800, and the extra $1000 gives you better graphics, maybe 120 fps, maybe 200, whatever. But unless you have money to burn, that extra grand wasn't well spent. Just put it into the bank, and buy another $800 machine in 3 years.
The 6.7% unemployment figure does not factor people who have stopped looking for new work because they had no luck for over a year. And unfortunately, the market only makes that worse. "This applicant hasn't worked in 8 months. He must not be any good, let's just throw out his resume without even conducting a phone interview." The problem becomes self-perpetuating.
The 6.7% figure also doesn't factor people who went from software analyst, machinist, or corporate accountant to janitor, grocery store cashier, or burger flipper. For that matter, it doesn't count people who went from full time at one company to part time at another.
Depending upon the news source you trust, the real unemployment figure in the US is closer to 15%.
Your solution requires the post submitter to do all of the work to create his solution for both kernels, and then compare them.
If someone asked whether to build a reasonably complex website in Python or PHP would you recommend that they build both and then performance test them? That's a lot of extra work.
In both the original post submitter's case and the hypothetical one I suggested, it would be much easier to gather as much information you reasonably can about both solutions and then make an educated guess as to the best option. I'm not sure Slashdot is the best place for his information gathering, but I understand what he is doing.
I'll admit most people are short sighted and stupid. But you're assigning blame where none is needed.
In order for the US to transition to rail, our massive roads infrastructure to function properly during the whole transition. That means tax revenue, in addition to covering all of the colossal debt and expenses we already have from other sources, would also have to pay for maintaining all existing roads WHILE ALSO paying to build rail.
Try pitching that plan in an election. "I know most of you are having a hard time finding work, and those that have work are working longer hours for less pay. But I have a plan: we're going to tax you an extra $5,000 per year per person, and in 5 years the 30% of you that still own a home will be able to use trains to commute!"
The original writer specifically said 128 MB of RAM. That's the issue, not the processor. I had a 700 MHz Duron that ran XP plenty fast for my needs - but it had 768 MB of RAM. Drop down to 128 MB, and XP runs quickly right until you actually try to run a program.
It depends upon your circle of friends, I guess. DVDs have been on the market about 10 years, so 200 DVDs represents two purchases per month.
If you buy everything brand new anywhere you can get it, that's maybe $40-$50 per month on DVDs. If you get stuff in the 'previously rented' DVD bin at Blockbuster and wait a year after releases hit DVD so prices come down, figure an average of maybe $8 per DVD. That's $16 per month, something almost anyone can afford.
Plus, of course, some first person shooters, roleplaying games, and real time strategy games allow fan-generated content, and it's sure as hell easier to mix and match new features and new levels with a keyboard, mouse, and multi-window editor and file explorer at your disposal than a game pad.
A console is less work to set up and has less hassle for operating system maintenance, firewalls, and anti-virus. It's also cheap. And when a generation of consoles is relatively new, they also have graphics performance reasonably close to the top end for PCs. But the PC is far from dead.
PostgreSQL has the GUI management tool pgAdmin 3, which is pretty intuitive and has a lot of features.
I haven't seen the Microsoft SQL Server or its GUI tools, so I can't give you a feature-for-feature comparison. But if you're bored, I would suggest giving PostgreSQL and pgAdmin 3 a try.
Our problem is that we have thousands of man hours invested in complex existing Crystal Reports documents that are still in heavy use. I hate using Crystal with an unholy passion.
But paying them their fucking $15,000 for the report server software we need (and we are a tiny company, less than 20 employees) is dramatically cheaper than paying a consultant - even a foreign one - to port all of our stuff to Jasper or BIRT.
All of our future development is planned for Jasper, but I would be shocked if we can ditch Crystal compatibility within five years.
and most of those students will drop out of the class and/or change their major by the end of the semester. so why waste the time/tuition money of the students who have demonstrated initiative in their own academic career?
You're out of your mind.
Some high schools don't offer software development courses at all.
Some high schools offer software development courses taught by people who know nothing of the discipline and who will pass on false information or bad habits that need to be unlearned.
Being self taught is good, but maybe the first teaching websites or development books or mentors you found were likewise very poor choices.
A lot of the best senior developers in the industry night now are self taught because the CS teaching as we know it did not exist when they were younger.
At that stage in the game, the budding student doesn't even understand what he doesn't understand, so he has no ground for knowing whether his previous self-taught, website, book, mentor, or school-provided education is gold or garbage.
A good professor will take people from ground zero on day one and move at a blistering pace that still lets the hard workers keep up. I started a college CS program not knowing anything of the discipline, with two good friends that basically slept through the first two semesters and got better grades. I graduated from the program, and they did not.
Yes, but if the project is big enough and the company keeps you around to maintain your own code, the lack of documentation and the clever designs actually will screw you up.
Get too smart with the way you write the code, and anybody trying to maintain it gets burned - even you. The best way to keep the jobs where you are is to make damn good software, and convince your management with your products that you are worth every penny they pay.
Actually, the doctor who first authored the study linking vaccines with mercury to autism and other adverse reactions owned a company that sold non-mercury vaccines.
So you're right to follow the money. You just followed it the wrong way. Andrew Wakefield is the lead author of the 1998 medical article that first conjectured a link between vaccines and autism. He was paid hefty sums to write his controversial article by companies working on non-mercury vaccinatons. http://en.wikipedia.org/wiki/Andrew_Wakefield
That 106F fever, while no doubt terrifying to you and very worrisome, would not have been far worse than Measles, Mumps, or Rubella.
I have three kids and one has been in the hospital repeatedly. I understand your fears. But failing to vaccinate is far more dangerous to your child's life, and the life of other people that she meets.
There is a relatively small but active community of open source functional programming languages. The big names (for this relatively small group) are Scheme, Haskell, ML and Ocaml, and Erlang. Lisp and the new open source JVM language Scala also support functional-only programming styles. So does Microsoft's language F#. (But Lisp, Scala, and F# can be used in non-functional ways.)
I am not very experienced with functional programming at all. But if I understand it correctly, one of the ways functional programming makes parallel processing easier is the fact that variables are immutable. That is, variable values can't be changed once they are assigned.
So if you want to do a String reverse in a functional style with immutable variables, it could be (Java): public String reverse(String s) { if (s == null || s.length() < 2) return s; else return (s.substring(s.length() - 2, s.length() - 1) + reverse (s.substring(0,s.length() - 2); }
That's awkward an inefficient in Java, but you hopefully see how it works. A lot of the difficulties with parallel programming come from having multiple threads accessing and changing shared mutable (changeable) variables. You have to coordinate access carefully or there can be big problems. With immutable variables, that particular source of bugs does not exist.
Very, very few people employed in the software industry are truly incapable of doing their work properly.
The real problem is people in the software industry who are unwilling. Unwilling to accept suggestions. Unwilling to read about different solutions to the problems they solve. Unwilling to discuss new ideas. Unwilling to spend even a little time reading websites or technical books to get better. It doesn't matter if their problem is arrogance, and they think they know everything they need; or insecurity, and they're afraid to reveal what they don't know; or laziness, and they just don't care to take the effort to get better.
The difference between a good or bad novice software developer is largely a question of raw talent. The difference between a good or bad experienced software developer is largely a question of emotional maturity. If you think you already know everything, you can't improve.
I write all this because I'm doing pretty well as a senior developer with plenty of production code with a pretty low bug rate. I think I'm doing okay, although I know I still have plenty more to learn (among other things, I want to be competent at functional programming just for fun - and I find it exceedingly difficult to grasp). Quite honestly, most of the code I wrote in my first few years as a software developer was atrocious nonsense, some of it well worth prime spots on thedailywtf.com. If I could learn to be better, then almost anyone can do it.
Slashdot Mirror
IF GM's Volt, or Chrysler's EV concept ever make production, they have a solution to the problem. The vehicle travels 40 or 60 miles on its battery charge, and if you keep driving an onboard generator turns on to supply electricity and recharge the battery.
In effect your car is a complete EV for a reasonable commute, and a fossil fuel vehicle (although not a traditional one) for long trips.
But the rumors are floating around that when the GM Volt hits showrooms, the price tag on the tiny four seat sedan will be above $30,000.
I've posted it before, but it bears repeating:
Nevada Solar One solar plant: http://en.wikipedia.org/wiki/Nevada_Solar_One
One US nuclear power plant: http://en.wikipedia.org/wiki/Calvert_Cliffs_Nuclear_Generating_Station
In 2008 dollars, the nuclear plant cost about 16 times as much to construct, $266 million versus $4 billion. But the nuclear plant supplies 104 times as much energy per year as the 134 million kilowatt-hours from the solar plant. So the nuclear power generates more than six times as much energy per dollar spent on construction, which still dramatically offsets the ongoing maintenance and fuel cost advantages of solar.
Nuclear power supplies 20% of the electricity used in the country, and it's done that for decades. Coal kills people with black lung disease, mining collapses, and of course lung cancer, asthma, and emphysema from coal burning emissions.
If we multiply the number of Americans killed by nuclear power each year by 5 to extrapolate if nuclear power supplied all of our energy, we still end up with a number far below the deaths from coal.
Oops, my math is off, the nuclear plant costs about 16 times as much. That's still roughly 6 times as much electricity per dollar spent, an enormous cost edge for nuclear.
Solar thermal power is nice because it can supply power 24 hours a day unless you have a long stretch of highly overcast days. The problem is that solar thermal power plants are ridiculously expensive compared to coal or nuclear.
The Nevada Solar One (http://en.wikipedia.org/wiki/Nevada_Solar_One) solar thermal power plant cost $266 million and supplies 134 million kwh per year.
The Calvert Cliffs power plant in Maryland (http://en.wikipedia.org/wiki/Calvert_Cliffs_Nuclear_Generating_Station) cost $766 million to construct in the 1970s. Depending upon how you calculate inflation, that would be maybe $4 billion today. The plant supplies more than 14,000 Gwh per year, on average, or 14 billion kwh.
So the nuclear power plant costs 8 times as much to construct, but supplies 104 times as much power as the Solar Thermal plant in a year. That's 12 times as much electricity per dollar spent. I'm certain the higher maintenance and fuel costs on a nuclear plant do not nearly offset that difference. Nuclear is still the way to go.
Most of the added cost of ground source is getting the wells drilled and the piping from the wells into your home. In 20 years, you may need to replace your compressor, but barring some very unusual problems you well should be fine for many more decades.
Plus, the ground source system compressor is indoors, which improves the system longevity versus an outdoor traditional air-source heat pump.
Funny enough, we're in the middle of having a ground source heat pump installed right now. We opted for ground source over air because our house is old, some of the walls are stone, and it's in an L shape. So we have high surface area for relatively modest interior space (2100 square feet for a family of five), and some of the house is effectively impossible to insulate well. My wife and I figured that no matter what we do, we're going to lose a lot of heat. A hypothetical 25% cost savings difference between an air source heat pump and a ground source heat pump will pay for itself with our home far more quickly than most better shaped, better insulated properties.
We live between Philadelphia and Reading. Funny that you're near Tannersville - my best friend grew up outside Snydersville, PA. His parents' house is probably in your back yard.
Wikipedia says the UK summer average temperature in the warmest regions is 15.5 to 17.7 C (60-64 Fahrenheit).
Many areas in the US are significantly warmer. Telling someone to 'just open a window' doesn't work when it's 38 C (100 Fahrenheit) for 8 hours of the day or more, like most summer days in parts of Arizona.
While I agree, the counterpoint is that you get a lot of diminishing returns for your PC investment. Depending upon what you get, maybe $800 well spent (not counting the monitor) will get you 80 fps.
Bump your PC spending to $1800, and the extra $1000 gives you better graphics, maybe 120 fps, maybe 200, whatever. But unless you have money to burn, that extra grand wasn't well spent. Just put it into the bank, and buy another $800 machine in 3 years.
The 6.7% unemployment figure does not factor people who have stopped looking for new work because they had no luck for over a year. And unfortunately, the market only makes that worse. "This applicant hasn't worked in 8 months. He must not be any good, let's just throw out his resume without even conducting a phone interview." The problem becomes self-perpetuating.
The 6.7% figure also doesn't factor people who went from software analyst, machinist, or corporate accountant to janitor, grocery store cashier, or burger flipper. For that matter, it doesn't count people who went from full time at one company to part time at another.
Depending upon the news source you trust, the real unemployment figure in the US is closer to 15%.
Your solution requires the post submitter to do all of the work to create his solution for both kernels, and then compare them.
If someone asked whether to build a reasonably complex website in Python or PHP would you recommend that they build both and then performance test them? That's a lot of extra work.
In both the original post submitter's case and the hypothetical one I suggested, it would be much easier to gather as much information you reasonably can about both solutions and then make an educated guess as to the best option. I'm not sure Slashdot is the best place for his information gathering, but I understand what he is doing.
I'll admit most people are short sighted and stupid. But you're assigning blame where none is needed.
In order for the US to transition to rail, our massive roads infrastructure to function properly during the whole transition. That means tax revenue, in addition to covering all of the colossal debt and expenses we already have from other sources, would also have to pay for maintaining all existing roads WHILE ALSO paying to build rail.
Try pitching that plan in an election. "I know most of you are having a hard time finding work, and those that have work are working longer hours for less pay. But I have a plan: we're going to tax you an extra $5,000 per year per person, and in 5 years the 30% of you that still own a home will be able to use trains to commute!"
The original writer specifically said 128 MB of RAM. That's the issue, not the processor. I had a 700 MHz Duron that ran XP plenty fast for my needs - but it had 768 MB of RAM. Drop down to 128 MB, and XP runs quickly right until you actually try to run a program.
It depends upon your circle of friends, I guess. DVDs have been on the market about 10 years, so 200 DVDs represents two purchases per month.
If you buy everything brand new anywhere you can get it, that's maybe $40-$50 per month on DVDs. If you get stuff in the 'previously rented' DVD bin at Blockbuster and wait a year after releases hit DVD so prices come down, figure an average of maybe $8 per DVD. That's $16 per month, something almost anyone can afford.
Absolutely.
Plus, of course, some first person shooters, roleplaying games, and real time strategy games allow fan-generated content, and it's sure as hell easier to mix and match new features and new levels with a keyboard, mouse, and multi-window editor and file explorer at your disposal than a game pad.
A console is less work to set up and has less hassle for operating system maintenance, firewalls, and anti-virus. It's also cheap. And when a generation of consoles is relatively new, they also have graphics performance reasonably close to the top end for PCs. But the PC is far from dead.
PostgreSQL has the GUI management tool pgAdmin 3, which is pretty intuitive and has a lot of features.
I haven't seen the Microsoft SQL Server or its GUI tools, so I can't give you a feature-for-feature comparison. But if you're bored, I would suggest giving PostgreSQL and pgAdmin 3 a try.
I'm curious, what features do you use in Oracle that aren't available in PostgreSQL?
Our problem is that we have thousands of man hours invested in complex existing Crystal Reports documents that are still in heavy use. I hate using Crystal with an unholy passion.
But paying them their fucking $15,000 for the report server software we need (and we are a tiny company, less than 20 employees) is dramatically cheaper than paying a consultant - even a foreign one - to port all of our stuff to Jasper or BIRT.
All of our future development is planned for Jasper, but I would be shocked if we can ditch Crystal compatibility within five years.
You're out of your mind.
At that stage in the game, the budding student doesn't even understand what he doesn't understand, so he has no ground for knowing whether his previous self-taught, website, book, mentor, or school-provided education is gold or garbage.
A good professor will take people from ground zero on day one and move at a blistering pace that still lets the hard workers keep up. I started a college CS program not knowing anything of the discipline, with two good friends that basically slept through the first two semesters and got better grades. I graduated from the program, and they did not.
Your elitism is misplaced.
Yes, but if the project is big enough and the company keeps you around to maintain your own code, the lack of documentation and the clever designs actually will screw you up.
Get too smart with the way you write the code, and anybody trying to maintain it gets burned - even you. The best way to keep the jobs where you are is to make damn good software, and convince your management with your products that you are worth every penny they pay.
Actually, the doctor who first authored the study linking vaccines with mercury to autism and other adverse reactions owned a company that sold non-mercury vaccines.
So you're right to follow the money. You just followed it the wrong way. Andrew Wakefield is the lead author of the 1998 medical article that first conjectured a link between vaccines and autism. He was paid hefty sums to write his controversial article by companies working on non-mercury vaccinatons. http://en.wikipedia.org/wiki/Andrew_Wakefield
I think a good analogy would be cars.
A good software developer is like a skilled mechanic or manufacturer. He knows which pre-existing parts to use when fixing or creating a car.
A great software developer builds new kinds of cars. If existing parts were sufficient for that task, he isn't building something new.
And ultimately, even though both jobs can be interesting and challenging, creating something new is more interesting and rewarding.
Funny, my siblings and I were breast-fed and one is a severe asthmatic, and my children were breast-fed and one is an asthmatic.
Do you have any statistical evidence to back up your assertion that it reduces the risk of asthma?
That 106F fever, while no doubt terrifying to you and very worrisome, would not have been far worse than Measles, Mumps, or Rubella.
I have three kids and one has been in the hospital repeatedly. I understand your fears. But failing to vaccinate is far more dangerous to your child's life, and the life of other people that she meets.
There is a relatively small but active community of open source functional programming languages. The big names (for this relatively small group) are Scheme, Haskell, ML and Ocaml, and Erlang. Lisp and the new open source JVM language Scala also support functional-only programming styles. So does Microsoft's language F#. (But Lisp, Scala, and F# can be used in non-functional ways.)
There are some great discussions on the advantages of functional programming. Here's one on the Haskell website: http://www.haskell.org/haskellwiki/Introduction Another one is this paper on the subject: http://www.md.chalmers.se/~rjmh/Papers/whyfp.pdf
I am not very experienced with functional programming at all. But if I understand it correctly, one of the ways functional programming makes parallel processing easier is the fact that variables are immutable. That is, variable values can't be changed once they are assigned.
So if you want to do a String reverse in a functional style with immutable variables, it could be (Java):
public String reverse(String s) {
if (s == null || s.length() < 2) return s;
else return (s.substring(s.length() - 2, s.length() - 1) + reverse (s.substring(0,s.length() - 2);
}
That's awkward an inefficient in Java, but you hopefully see how it works. A lot of the difficulties with parallel programming come from having multiple threads accessing and changing shared mutable (changeable) variables. You have to coordinate access carefully or there can be big problems. With immutable variables, that particular source of bugs does not exist.
incapable of doing it properly
Very, very few people employed in the software industry are truly incapable of doing their work properly.
The real problem is people in the software industry who are unwilling. Unwilling to accept suggestions. Unwilling to read about different solutions to the problems they solve. Unwilling to discuss new ideas. Unwilling to spend even a little time reading websites or technical books to get better. It doesn't matter if their problem is arrogance, and they think they know everything they need; or insecurity, and they're afraid to reveal what they don't know; or laziness, and they just don't care to take the effort to get better.
The difference between a good or bad novice software developer is largely a question of raw talent. The difference between a good or bad experienced software developer is largely a question of emotional maturity. If you think you already know everything, you can't improve.
I write all this because I'm doing pretty well as a senior developer with plenty of production code with a pretty low bug rate. I think I'm doing okay, although I know I still have plenty more to learn (among other things, I want to be competent at functional programming just for fun - and I find it exceedingly difficult to grasp). Quite honestly, most of the code I wrote in my first few years as a software developer was atrocious nonsense, some of it well worth prime spots on thedailywtf.com. If I could learn to be better, then almost anyone can do it.