I know, but I happened to be a great example of what I wanted to illustrate: it's not obvious to you, but some people use cars without imposing huge environmental burdens, and the cost to us of losing that ability to drive might be large by comparison.
I do believe that there are much more carbon efficient ways to live
Right, so do I. One of those ways is to eliminate urban building restrictions which make housing too expensive, and force people to live far from where they work. By allowing more development, we allow people to live close and drive less. The main reason why this is not allowed is because it obstructs the view of a few hilltop mansion dwellers, who are usually much more influential in local politics than the vast sea of commuters.
Another way is to allow the building of nuclear power plants. No carbon emissions, and plenty of cheap energy for all. The pollution comes out in a nice, concentrated form which we can recycle once or twice and store somewhere in an uninhabited area (I hear rumors that, per kw-h, a coal power plant generates more radioactive waste, but I don't have a source).
"Here's what we need to do": Yes, I'm afraid, that is science.
Actually, that's engineering that you describe. I'm going to stop splitting hairs here, because it's really irrelevant to the argument.
people starve whilst we have a surplus of food
People starve because there's no cheap way to distribute food to them (and sometimes the government gets in the way anyway). Getting food to a remote village in Africa is not as easy as giving them a Safeway gift account number. There may or may not be a viable road, the road may be patrolled by gangs, etc.
First of all, I'd like to thank you for making more of a point this post. Your last post didn't convince anyone.
I've read a few of the papers, and whilst climate change isn't my field (quantum mechanics) the case is compelling that the cost of not acting will be quite large.
That may be true.
also in terms of food production
Usually, problems with malnutrition/starvation are due to either an oppressive government or problems with food distribution. There's still a lot of room for more food production.
Politicians have their own agendas, so in your comparison are you really claiming the climatologists are reporting their results for personal profit, telling us of global climate change to further some selfish cause
I wasn't comparing climatologists to those politicians, I was comparing you to them. Climatologists acquire facts, develop theories, and test those theories against new facts. That helps us develop plans, analyze the costs and benefits, and hold the plan accountable, which is exactly what I want. Some climatologists also put on a political hat and start drumming up support for their research (after all, they want grants, etc.), but not all.
here's what we need to do". In other words, science.
There's nothing scientific about "here's what we need to do". That's a policy decision, and scientists observe, develop theories, and test them. Some scientists also advocate policy, but that has nothing to do with being a scientist.
Here, the science IS known.
Here's where I disagree. I think your bridge analogy is interesting, but we know a lot more about structural engineering than global climate. Knowing and acknowledging what we don't know is an important step in making good decisions.
You'll probably have to stop driving
I pay extra rent so I live 2 miles from work, and 1/8 mile from the light rail, and if I drive to lunch, usually it's 4 to a car. How much CO2 are you going to prevent by not allowing me to drive? A tank every other month? Here's the problem: you arbitrarily want to prevent me from driving, with no thought to what that will actually benefit anyone nor what it will cost me. I won't be able to drive to the doctor or the pharmacy any more, or visit my relatives. Not having a car could easily add 10 hours per week to my schedule, and save some tiny volume of gasoline.
Accountability - well, that's a hard one.
Accountability isn't always in absolutes. Who will pay the costs of reducing CO2 emissions, and how can we be sure they are reduced? What stops me from inflating the amount of CO2 that I'd like to produce so that I can say that I'm "saving" more (e.g. "my home office is saving 100 metric tons of CO2 per year")? What is an acceptable amount of CO2 production, and how is that monitored? How about in other countries? What if a country decides not to participate, do we invade? Do we fine companies for producing CO2? Do we tax them? How do we fine/tax individuals (like people with cars)? Where does this money go? If it goes to more CO2 reduction plans, how are they held accountable?
Yes, those are details. But they are important details and will mean the difference between a successful program and a failed bureaucracy flush with cash, and a new breed of lobbyists running the whole thing.
Well, that's the way everyone tries to frame their idea, because not having to explain costs is very convenient.
"Support my idea without question, or die a horrible death"... heard it before; didn't buy it then either*. The ideas I do buy have the following characteristics:
(1) Defined benefits, supported by facts and analysis (2) Expected costs, supported by economic analysis of the plan in question (3) Accountability for both the costs and benefits
The environmentalists that present their ideas and plans as outlined above (a.k.a. "grown-ups") experience infinitely more success than those that merely introduce scary scenarios or try to evoke empathy for some critter or other.
* Just a couple politicians who also tried this tactic:
- Bush 43: give me absolute power or the terrorists will get you
- FDR: give me absolute power or you will die of poverty (see packing the Supreme Court and confiscating gold).
we're better off slowing CO2 output and being wrong about global warming than we are heating up the planet with CO2 and being wrong about not having a human global climate impact.
How is this insightful? This kind of thinking shows everything that's wrong with the environmental movement: a complete disregard for cost/benefit analysis. You're saying that it's somehow "better" to impose arbitrary restrictions on the economy of a completely unknown cost, in the hope that whatever you did creates some kind of unspecified benefit.
This is no better than a medieval doctor removing a few pints of blood because it's better to remove your blood and be wrong than to leave you with a cold and do nothing.
If you want to help the environment, present your solutions in the way every other idea needs to be presented: here are the expected benefits, here are the expected costs, and here is how we're going to be accountable for these benefits and these costs.
I think PHP has got beyond the stupid-design-flaws-causing-security-issues stage.
A major problem for PHP is still it's configureware mentality. No other programming language has a configure file. PHP started with it because it's also a web framework; which is somewhat understandable. However, they then proceeded to abuse the configuration file for all sorts of semantic behaviors, and the end result is that it's very HARD to program securely and portably at the same time. Make a configuration change, and that could introduce a subtle security flaw.
magic quotes is probably the worst idea that PHP ever had.
what we really need is a basic arithmetic requirement for journalists.
And economics. There aren't "100,000 jobs" that "need to be filled". The job market is controlled by supply and demand, pure and simple. If you make a better offer (money, environment, hours) you will attract the people you need. If you can't make a better offer, well, gee, looks like you didn't "need" that employee after all!
I am not in any way saying that it's a good or a bad policy to encourage foreign labor and/or immigration, that's a much more complex question. I'm just saying that anytime people use terms like "shortage" or "surplus" in a market economy they are trying to get the government to give them a better deal by intervening in the marketplace. Shortages and surpluses are what happen when you do not have a free market economy.
I personally have a shortage of jet airplanes. I've got $50, and I've have been looking all over for a jet airplane, and I can't find one. We have a national crisis!
Nobody likes having to pick up the slack for a guy who is not pulling his weight, especially not one who is making $65k / year.
This brings up the question: why does IBM need any reason at all? As soon as an employee becomes a net loss, he should be out of there. A job isn't property, it's a relationship. As soon as the relationship isn't working out, it's done. No explanation required. If he was really such a great employee, it will be IBM's loss and the employee can find work elsewhere.
Imagine a persistance layer with no SQL, no extra user management, no extra connection layer, no filesystem under it and native object suport for any PL you wish to compile in.
They've had databases like that for a long time. And then they invented Relational databases to solve the problems with the simple types of databases you describe.
The advantages of an RDBMS are:
* abstraction of the physical representation of data from the way the data is logically presented
* a model for constraints that are meaningful yet can perform well: normalization. Of course, constraints can only detract from performance, but normalization performs well in comparison with alternative implementations of constraints.
* a declarative query language that tells the database what you want not how to get it. This allows the database engine to optimize and choose a new strategy based on, e.g. the size of a table. No need to rewrite the query in the application or re-code the application to ask for the data in a different way.
I encourage you to read and think about why these things are valuable. The database can be a part of the application development process, and can make radical application changes much simpler. A simple example is that you can almost always avoid painful data format changes when your application needs to meet a new requirement.
Re:Opposite way of thinking?
on
PHP 5 in Practice
·
· Score: 2, Interesting
If you're an experienced developer, you'll be able to apply the right tools for the problem in a way that interfaces with complex business processes. You'll complete jobs quickly and have a lot of happy customers. However, sometimes customers might become unhappy if your solution doesn't continue to work for their changing needs, or becomes expensive to maintain or alter. You might occasionally surprise clients with additional cost because a tool you used doesn't work well for a new demand.
If you're a computer scientist, you'll be able to solve complex problems in a robust, flexible way. However, you might use the wrong tools, not enough tools, or take longer to accomplish the task initially. Your customers will stay content because your solution will continue to work (although you should learn to equate silence with "thank you"). You are less likely to encounter unexpected extra costs (but that's because you probably wrote too much of the code yourself).
I would say that for short-term projects, lean in the direction of "experienced developer". There's no substitute for a developer that just gets the job done, quickly and under-budget, in a way that works seamlessly with the business needs. However, for long-term projects, lean more in the direction of computer scientist. The computer scientist will be better able to respond to new demands from the system quickly (without costly rewrites or data migration nightmares). Ideally you want someone who is both, and understands when to use which principles.
Both issues [gun rights, abortion] have little relavance to those who scream about it either way.
The right to bear arms isn't just about armed rebellion, it's also a civil right. It's the right to defend yourself. If you depend on the government to defend you, your other rights are meaningless: the government can simply let it be known that they will not protect you. Then, the local criminals will have their way with you.
This has a long history as a government tactic. It happened to blacks in the American South, for instance (although that's far from the only example).
The right to defend yourself is a very important civil right. It's protected by the 2nd Amendment for a reason. And if we let the elected officials deprive us of one right, what's to stop the rest of the Bill of Rights from following? The best thing Americans can do is stand up for the Constitution, even if you disagree with parts of it. If you really think things need changing, you can try to amend the Constitution.
that'll give them a decent pool of cash to play with
So that's an excuse to make a bad investment? The argument is the same. You have to make choices between alternatives that exist. If you deploy broadband to everyone, even people living in a remote swamp by themselves, you are using resources that might be better spent elsewhere.
Let people who want broadband pay for it. Let people who don't want it invest their money in something more urgent to them. The fact that you can locate expenses that you think aren't worthwhile doesn't mean that we should turn around with "cash to play with" and waste it on whatever you might think is a cool idea. It's not your allowance, it's the tax dollars of everyone in the country.
Free markets only work in ideal conditions, with no transaction costs, no externialities, full information, and localnonsatifiability.
I think what you mean is "free markets only work ideally in ideal conditions". Of course free markets aren't perfect.
However, supply and demand still applies, and it still provides the valuable feedback cycle and information transfer that makes our economy so strong.
You say that people can be satisfied, which can be true, but a large fraction of the people will never be satisfied. There will always be demand for: (1) better doctors (2) better medicines (3) more immediate care (4) more convenient care (5) more medical resources (6) 2nd, 3rd, and 4th opinions
So, who gets those resources? The answer, in socialized medicine, is to use queuing, rationing, and unnavigable bureaucracy to address the shortage.
Some doctors are better than other doctors. Who gets the good doctor? Some medical equipment is expensive. Who gets it? If I want to see a doctor, I want to see a superdoctor, that has graduated from harvard medical school at the top of his class and has a ridiculously high IQ. However, those are in limited supply.
This idea of people being "satisfied" is a myth. Economics deal with scarce resources. If people could be "satisfied" with health care then those people wouldn't even be players in the health economy. But clearly, they are.
Sure, free-market medicine looks bad compared with an ideal. But it looks great compared to socialized medicine.
Disclaimer: I am not suggesting that American medicine is a free market. There are all kinds of non-free elements, like the government not taxing corporations for health benefits. That means that companies pay people with health benefits, which is a really bad idea. We've got a quasi-socialist medical system here already. People are insuring themselves against stuff they know will happen, which is a bad idea (think how stupid it would be to buy gas insurance -- if your tank goes empty you file a claim). But hey, at least if you really need a good doctor you can still get one here in the US.
You have to either add a bunch of anti gells to it or hydrogenate it.
Doesn't hydrogenation increase the freezing temperature? This is a real question, I'm not saying you're wrong. But, as I understand it, hydrogenation is done to make vegetable oils freeze at a higher temperature due to the extra hydrogen bonding.
Consider it an investment. By increasing opportunities in rural or economically disadvantaged areas, by introducing freer information and allowing more people the chance to inform and educate themselves, you only enrich society and create a more vital economy. That $25,000 could result in massive returns.
Lots of things are investments. The question is, is it a good investment. We have to compare it to other things, like the capital and labor we're giving up to deploy the broadband. Maybe it would be a better investment to just let the people keep their money and invest it in capital and labor that's more urgently needed. That capital and labor could instead be used to build housing, protect people with police, or educate people. All of those things are investments, too.
Too many people seem to think that the money comes from nowhere important, as though the money would just be spent on cocaine otherwise. Everything sounds good when you compare it against nothing. That's the problem with huge income taxes on "the rich," often it doesn't reduce their consumption, but merely drains their investments. That is, the capital they once owned (directly or through a series of investment houses) is sold to China and the workers they employed (directly or indirectly) are laid off.
Says here you get 24x7 web and phone support plus 30 minute emergency response time.
Sun Microsystems offers 24x7 PostgreSQL support.
Eat that, pgsql-bugs
I've always found the mailing lists to be great. I'm sorry you didn't have that experience. By the way, pgsql-bugs is not a typical support channel, you'd be better off in pgsql-general or #postgresql if you have support needs. Unless you have an actual bug, of course.
I think that MySQL could show a little more restraint in the configuration file and other options. It's too easy to have two apps -- both of which support MySQL -- which can't work together on the same instance of MySQL. I think it's better to make changes more slowly and send signals to the users about what the better practices are, even if they change over time.
Also, the fact that a table can behave differently depending on the storage engine is quite worrisome. If a storage engine doesn't support a constraint efficiently, support it inefficiently. Then people can evaluate for themselves how important that constraint is given the alternatives that exist.
In this respect MySQL is like another popular open source project, PHP. Both technologies -- if you'll forgive the overstatement -- have configuration files that look like a FAQ: "How do I get MySQL to do xyz?" - "It's a configurable option". PHP is starting to change the trend and be more forceful to its users about using "good" practices. Not everyone agrees on every detail about what's "good" and what's not, but steering towards accepted practices and away from configureware makes PHP apps more compatible, and have less surprising results and interactions. I think MySQL will need to do the same.
It's easier to provide a configurable option (or a new table type that silently ignores whatever is inconvenient) than to educate users about how to use a pre-exising, consistent system to accomplish their task. I think there are some instances where both PHP and MySQL are guilty of that.
This is a major (if not the major) difference in philosophy between MySQL and PostgreSQL.
The person to whom I replied was showing how latency internal to the drive can cause a throughput problem. However he didn't realize it was a throughput problem, he thought it was still latency.
If he had stopped after he said "This may be true in a very technical, pedantic sense," I would have stopped. However, he continued on with confusion about latency versus throughput.
The way I remember the definition of a "bottleneck", it's basically the lowest performance part that brakes everything else.
Bottleneck is an analogy.
If you have a bottle of beer, you can drink the beer much faster than the neck of the bottle allows (at least in the short term, with normal gravitational forces propelling the beer at normal beer-flowing speeds and accounting for the need to exchange air back into the bottle). If you widen the thinnest part of the neck, the second-thinnest part will then become the point of constriction, and will be the new "bottleneck". This is a throughput problem.
However, something like shampoo (when there's only a little left in the bottle) is a latency problem.
If I continously tell the disk to seek to one extreme and read a cacheful, then seek to the other extreme and read a cacheful, it will neither be waiting for requests nor using its full bandwidth.
The disk is at maximum throughput for the given problem set (the requests you give it). A bandwidth rating on the box is for an ideal problem set, but if you change the problem set the bandwidth decreases.
Latency and throughput are unrelated only if there can be infinitely many requests produced and satisfied in parallel.
I never said they were unrelated, I said unequal.
Latency limits throughput.
In some cases.
A bottleneck is not latency. It's just not. A bottleneck can cause latency, and latency can cause a bottleneck, but they can happen independently as well.
Latency causes idle time on *both* devices. If we play tick-tack-toe by mail, that's a latency problem, and it's latent for both of us. If I write you a letter asking you to send me all the gold you find, that's a bandwidth problem (I can put it in my safe much faster than you can find it and send it to me).
What you say does not make sense in the context of the entire thread. Here is how I understand the discussion (posts are obviously paraphrased):
Post #1: I/O is going to be an even greater bottleneck Post #2: At large datacenters people often don't demand the full bandwidth of their I/O system Post #3: I think #1 meant latency, not bandwidth Post #4 (me): bottleneck refers to a bandwidth limitation, not latency.
Now you're talking about erratic request patterns that can decrease the performance of an I/O system. However, to measure both throughput and latency, you need to hold the requests constant. Latency is the time between when the requests arrive at the disk and the time the disk returns the first useful piece of data. Throughput is the rate at which the data arrives at the destination.
You have an I/O bottleneck (aka throughput problem) when the CPU idle time waiting for the full response to the request is much longer than the CPU idle time spent waiting for the first useful piece of data to arrive. You have a latency problem when the CPU idle time spent waiting for the first useful piece of data is much longer than the CPU idle time waiting for the full response.
I stand by my comment: bottleneck != latency problem. The seeking on a disk affects the throughput as well as the latency. The fact that you're not getting the theoretical maximum throughput for your series of random requests does not mean that you're experiencing a latency problem. A bottleneck can cause latency, and latency can cause a bottleneck (such as seeking within a disk, as you pointed out), but they are not equal.
...if there was a similar competition in the OS market. You wouldn't need these mammoth processors in the first place
What makes you say that? What makes you think that there are the same kind of performance gains possible from an OS? What makes you think there is not competition?
Slashdot Mirror
I'm not speaking about you specifically.
I know, but I happened to be a great example of what I wanted to illustrate: it's not obvious to you, but some people use cars without imposing huge environmental burdens, and the cost to us of losing that ability to drive might be large by comparison.
I do believe that there are much more carbon efficient ways to live
Right, so do I. One of those ways is to eliminate urban building restrictions which make housing too expensive, and force people to live far from where they work. By allowing more development, we allow people to live close and drive less. The main reason why this is not allowed is because it obstructs the view of a few hilltop mansion dwellers, who are usually much more influential in local politics than the vast sea of commuters.
Another way is to allow the building of nuclear power plants. No carbon emissions, and plenty of cheap energy for all. The pollution comes out in a nice, concentrated form which we can recycle once or twice and store somewhere in an uninhabited area (I hear rumors that, per kw-h, a coal power plant generates more radioactive waste, but I don't have a source).
"Here's what we need to do": Yes, I'm afraid, that is science.
Actually, that's engineering that you describe. I'm going to stop splitting hairs here, because it's really irrelevant to the argument.
people starve whilst we have a surplus of food
People starve because there's no cheap way to distribute food to them (and sometimes the government gets in the way anyway). Getting food to a remote village in Africa is not as easy as giving them a Safeway gift account number. There may or may not be a viable road, the road may be patrolled by gangs, etc.
Yes, I was boiling it down
First of all, I'd like to thank you for making more of a point this post. Your last post didn't convince anyone.
I've read a few of the papers, and whilst climate change isn't my field (quantum mechanics) the case is compelling that the cost of not acting will be quite large.
That may be true.
also in terms of food production
Usually, problems with malnutrition/starvation are due to either an oppressive government or problems with food distribution. There's still a lot of room for more food production.
Politicians have their own agendas, so in your comparison are you really claiming the climatologists are reporting their results for personal profit, telling us of global climate change to further some selfish cause
I wasn't comparing climatologists to those politicians, I was comparing you to them. Climatologists acquire facts, develop theories, and test those theories against new facts. That helps us develop plans, analyze the costs and benefits, and hold the plan accountable, which is exactly what I want. Some climatologists also put on a political hat and start drumming up support for their research (after all, they want grants, etc.), but not all.
here's what we need to do". In other words, science.
There's nothing scientific about "here's what we need to do". That's a policy decision, and scientists observe, develop theories, and test them. Some scientists also advocate policy, but that has nothing to do with being a scientist.
Here, the science IS known.
Here's where I disagree. I think your bridge analogy is interesting, but we know a lot more about structural engineering than global climate. Knowing and acknowledging what we don't know is an important step in making good decisions.
You'll probably have to stop driving
I pay extra rent so I live 2 miles from work, and 1/8 mile from the light rail, and if I drive to lunch, usually it's 4 to a car. How much CO2 are you going to prevent by not allowing me to drive? A tank every other month? Here's the problem: you arbitrarily want to prevent me from driving, with no thought to what that will actually benefit anyone nor what it will cost me. I won't be able to drive to the doctor or the pharmacy any more, or visit my relatives. Not having a car could easily add 10 hours per week to my schedule, and save some tiny volume of gasoline.
Accountability - well, that's a hard one.
Accountability isn't always in absolutes. Who will pay the costs of reducing CO2 emissions, and how can we be sure they are reduced? What stops me from inflating the amount of CO2 that I'd like to produce so that I can say that I'm "saving" more (e.g. "my home office is saving 100 metric tons of CO2 per year")? What is an acceptable amount of CO2 production, and how is that monitored? How about in other countries? What if a country decides not to participate, do we invade? Do we fine companies for producing CO2? Do we tax them? How do we fine/tax individuals (like people with cars)? Where does this money go? If it goes to more CO2 reduction plans, how are they held accountable?
Yes, those are details. But they are important details and will mean the difference between a successful program and a failed bureaucracy flush with cash, and a new breed of lobbyists running the whole thing.
Benefit: We continue to live.
Well, that's the way everyone tries to frame their idea, because not having to explain costs is very convenient.
"Support my idea without question, or die a horrible death"... heard it before; didn't buy it then either*. The ideas I do buy have the following characteristics:
(1) Defined benefits, supported by facts and analysis
(2) Expected costs, supported by economic analysis of the plan in question
(3) Accountability for both the costs and benefits
The environmentalists that present their ideas and plans as outlined above (a.k.a. "grown-ups") experience infinitely more success than those that merely introduce scary scenarios or try to evoke empathy for some critter or other.
* Just a couple politicians who also tried this tactic:
- Bush 43: give me absolute power or the terrorists will get you
- FDR: give me absolute power or you will die of poverty (see packing the Supreme Court and confiscating gold).
Should we not take steps to normalize the temperature anyways to stabilize the environment we prefer?
We should take the steps, provided that we know what those steps are, and that the benefits are worth the costs.
we're better off slowing CO2 output and being wrong about global warming than we are heating up the planet with CO2 and being wrong about not having a human global climate impact.
How is this insightful? This kind of thinking shows everything that's wrong with the environmental movement: a complete disregard for cost/benefit analysis. You're saying that it's somehow "better" to impose arbitrary restrictions on the economy of a completely unknown cost, in the hope that whatever you did creates some kind of unspecified benefit.
This is no better than a medieval doctor removing a few pints of blood because it's better to remove your blood and be wrong than to leave you with a cold and do nothing.
If you want to help the environment, present your solutions in the way every other idea needs to be presented: here are the expected benefits, here are the expected costs, and here is how we're going to be accountable for these benefits and these costs.
I think PHP has got beyond the stupid-design-flaws-causing-security-issues stage.
A major problem for PHP is still it's configureware mentality. No other programming language has a configure file. PHP started with it because it's also a web framework; which is somewhat understandable. However, they then proceeded to abuse the configuration file for all sorts of semantic behaviors, and the end result is that it's very HARD to program securely and portably at the same time. Make a configuration change, and that could introduce a subtle security flaw.
magic quotes is probably the worst idea that PHP ever had.
SQL
OCaml
what we really need is a basic arithmetic requirement for journalists.
And economics. There aren't "100,000 jobs" that "need to be filled". The job market is controlled by supply and demand, pure and simple. If you make a better offer (money, environment, hours) you will attract the people you need. If you can't make a better offer, well, gee, looks like you didn't "need" that employee after all!
I am not in any way saying that it's a good or a bad policy to encourage foreign labor and/or immigration, that's a much more complex question. I'm just saying that anytime people use terms like "shortage" or "surplus" in a market economy they are trying to get the government to give them a better deal by intervening in the marketplace. Shortages and surpluses are what happen when you do not have a free market economy.
I personally have a shortage of jet airplanes. I've got $50, and I've have been looking all over for a jet airplane, and I can't find one. We have a national crisis!
Nobody likes having to pick up the slack for a guy who is not pulling his weight, especially not one who is making $65k / year.
This brings up the question: why does IBM need any reason at all? As soon as an employee becomes a net loss, he should be out of there. A job isn't property, it's a relationship. As soon as the relationship isn't working out, it's done. No explanation required. If he was really such a great employee, it will be IBM's loss and the employee can find work elsewhere.
Imagine a persistance layer with no SQL, no extra user management, no extra connection layer, no filesystem under it and native object suport for any PL you wish to compile in.
They've had databases like that for a long time. And then they invented Relational databases to solve the problems with the simple types of databases you describe.
The advantages of an RDBMS are:
* abstraction of the physical representation of data from the way the data is logically presented
* a model for constraints that are meaningful yet can perform well: normalization. Of course, constraints can only detract from performance, but normalization performs well in comparison with alternative implementations of constraints.
* a declarative query language that tells the database what you want not how to get it. This allows the database engine to optimize and choose a new strategy based on, e.g. the size of a table. No need to rewrite the query in the application or re-code the application to ask for the data in a different way.
I encourage you to read and think about why these things are valuable. The database can be a part of the application development process, and can make radical application changes much simpler. A simple example is that you can almost always avoid painful data format changes when your application needs to meet a new requirement.
If you're an experienced developer, you'll be able to apply the right tools for the problem in a way that interfaces with complex business processes. You'll complete jobs quickly and have a lot of happy customers. However, sometimes customers might become unhappy if your solution doesn't continue to work for their changing needs, or becomes expensive to maintain or alter. You might occasionally surprise clients with additional cost because a tool you used doesn't work well for a new demand.
If you're a computer scientist, you'll be able to solve complex problems in a robust, flexible way. However, you might use the wrong tools, not enough tools, or take longer to accomplish the task initially. Your customers will stay content because your solution will continue to work (although you should learn to equate silence with "thank you"). You are less likely to encounter unexpected extra costs (but that's because you probably wrote too much of the code yourself).
I would say that for short-term projects, lean in the direction of "experienced developer". There's no substitute for a developer that just gets the job done, quickly and under-budget, in a way that works seamlessly with the business needs. However, for long-term projects, lean more in the direction of computer scientist. The computer scientist will be better able to respond to new demands from the system quickly (without costly rewrites or data migration nightmares). Ideally you want someone who is both, and understands when to use which principles.
Both issues [gun rights, abortion] have little relavance to those who scream about it either way.
The right to bear arms isn't just about armed rebellion, it's also a civil right. It's the right to defend yourself. If you depend on the government to defend you, your other rights are meaningless: the government can simply let it be known that they will not protect you. Then, the local criminals will have their way with you.
This has a long history as a government tactic. It happened to blacks in the American South, for instance (although that's far from the only example).
The right to defend yourself is a very important civil right. It's protected by the 2nd Amendment for a reason. And if we let the elected officials deprive us of one right, what's to stop the rest of the Bill of Rights from following? The best thing Americans can do is stand up for the Constitution, even if you disagree with parts of it. If you really think things need changing, you can try to amend the Constitution.
You asked where the money was going to come from.
When?
that'll give them a decent pool of cash to play with
So that's an excuse to make a bad investment? The argument is the same. You have to make choices between alternatives that exist. If you deploy broadband to everyone, even people living in a remote swamp by themselves, you are using resources that might be better spent elsewhere.
Let people who want broadband pay for it. Let people who don't want it invest their money in something more urgent to them. The fact that you can locate expenses that you think aren't worthwhile doesn't mean that we should turn around with "cash to play with" and waste it on whatever you might think is a cool idea. It's not your allowance, it's the tax dollars of everyone in the country.
Free markets only work in ideal conditions, with no transaction costs, no externialities, full information, and localnonsatifiability.
I think what you mean is "free markets only work ideally in ideal conditions". Of course free markets aren't perfect.
However, supply and demand still applies, and it still provides the valuable feedback cycle and information transfer that makes our economy so strong.
You say that people can be satisfied, which can be true, but a large fraction of the people will never be satisfied. There will always be demand for:
(1) better doctors
(2) better medicines
(3) more immediate care
(4) more convenient care
(5) more medical resources
(6) 2nd, 3rd, and 4th opinions
So, who gets those resources? The answer, in socialized medicine, is to use queuing, rationing, and unnavigable bureaucracy to address the shortage.
Some doctors are better than other doctors. Who gets the good doctor? Some medical equipment is expensive. Who gets it? If I want to see a doctor, I want to see a superdoctor, that has graduated from harvard medical school at the top of his class and has a ridiculously high IQ. However, those are in limited supply.
This idea of people being "satisfied" is a myth. Economics deal with scarce resources. If people could be "satisfied" with health care then those people wouldn't even be players in the health economy. But clearly, they are.
Sure, free-market medicine looks bad compared with an ideal. But it looks great compared to socialized medicine.
Disclaimer: I am not suggesting that American medicine is a free market. There are all kinds of non-free elements, like the government not taxing corporations for health benefits. That means that companies pay people with health benefits, which is a really bad idea. We've got a quasi-socialist medical system here already. People are insuring themselves against stuff they know will happen, which is a bad idea (think how stupid it would be to buy gas insurance -- if your tank goes empty you file a claim). But hey, at least if you really need a good doctor you can still get one here in the US.
You have to either add a bunch of anti gells to it or hydrogenate it.
Doesn't hydrogenation increase the freezing temperature? This is a real question, I'm not saying you're wrong. But, as I understand it, hydrogenation is done to make vegetable oils freeze at a higher temperature due to the extra hydrogen bonding.
Consider it an investment. By increasing opportunities in rural or economically disadvantaged areas, by introducing freer information and allowing more people the chance to inform and educate themselves, you only enrich society and create a more vital economy. That $25,000 could result in massive returns.
Lots of things are investments. The question is, is it a good investment. We have to compare it to other things, like the capital and labor we're giving up to deploy the broadband. Maybe it would be a better investment to just let the people keep their money and invest it in capital and labor that's more urgently needed. That capital and labor could instead be used to build housing, protect people with police, or educate people. All of those things are investments, too.
Too many people seem to think that the money comes from nowhere important, as though the money would just be spent on cocaine otherwise. Everything sounds good when you compare it against nothing. That's the problem with huge income taxes on "the rich," often it doesn't reduce their consumption, but merely drains their investments. That is, the capital they once owned (directly or through a series of investment houses) is sold to China and the workers they employed (directly or indirectly) are laid off.
Says here you get 24x7 web and phone support plus 30 minute emergency response time.
Sun Microsystems offers 24x7 PostgreSQL support.
Eat that, pgsql-bugs
I've always found the mailing lists to be great. I'm sorry you didn't have that experience. By the way, pgsql-bugs is not a typical support channel, you'd be better off in pgsql-general or #postgresql if you have support needs. Unless you have an actual bug, of course.
I think that MySQL could show a little more restraint in the configuration file and other options. It's too easy to have two apps -- both of which support MySQL -- which can't work together on the same instance of MySQL. I think it's better to make changes more slowly and send signals to the users about what the better practices are, even if they change over time.
Also, the fact that a table can behave differently depending on the storage engine is quite worrisome. If a storage engine doesn't support a constraint efficiently, support it inefficiently. Then people can evaluate for themselves how important that constraint is given the alternatives that exist.
In this respect MySQL is like another popular open source project, PHP. Both technologies -- if you'll forgive the overstatement -- have configuration files that look like a FAQ: "How do I get MySQL to do xyz?" - "It's a configurable option". PHP is starting to change the trend and be more forceful to its users about using "good" practices. Not everyone agrees on every detail about what's "good" and what's not, but steering towards accepted practices and away from configureware makes PHP apps more compatible, and have less surprising results and interactions. I think MySQL will need to do the same.
It's easier to provide a configurable option (or a new table type that silently ignores whatever is inconvenient) than to educate users about how to use a pre-exising, consistent system to accomplish their task. I think there are some instances where both PHP and MySQL are guilty of that.
This is a major (if not the major) difference in philosophy between MySQL and PostgreSQL.
That's because the government has it's own set of rules.
Same with Social Security: you couldn't run an investment house like Social Security, if you did you'd go to jail.
The person to whom I replied was showing how latency internal to the drive can cause a throughput problem. However he didn't realize it was a throughput problem, he thought it was still latency.
If he had stopped after he said "This may be true in a very technical, pedantic sense," I would have stopped. However, he continued on with confusion about latency versus throughput.
The way I remember the definition of a "bottleneck", it's basically the lowest performance part that brakes everything else.
Bottleneck is an analogy.
If you have a bottle of beer, you can drink the beer much faster than the neck of the bottle allows (at least in the short term, with normal gravitational forces propelling the beer at normal beer-flowing speeds and accounting for the need to exchange air back into the bottle). If you widen the thinnest part of the neck, the second-thinnest part will then become the point of constriction, and will be the new "bottleneck". This is a throughput problem.
However, something like shampoo (when there's only a little left in the bottle) is a latency problem.
If I continously tell the disk to seek to one extreme and read a cacheful, then seek to the other extreme and read a cacheful, it will neither be waiting for requests nor using its full bandwidth.
The disk is at maximum throughput for the given problem set (the requests you give it). A bandwidth rating on the box is for an ideal problem set, but if you change the problem set the bandwidth decreases.
Latency and throughput are unrelated only if there can be infinitely many requests produced and satisfied in parallel.
I never said they were unrelated, I said unequal.
Latency limits throughput.
In some cases.
A bottleneck is not latency. It's just not. A bottleneck can cause latency, and latency can cause a bottleneck, but they can happen independently as well.
Latency causes idle time on *both* devices. If we play tick-tack-toe by mail, that's a latency problem, and it's latent for both of us. If I write you a letter asking you to send me all the gold you find, that's a bandwidth problem (I can put it in my safe much faster than you can find it and send it to me).
What you say does not make sense in the context of the entire thread. Here is how I understand the discussion (posts are obviously paraphrased):
Post #1: I/O is going to be an even greater bottleneck
Post #2: At large datacenters people often don't demand the full bandwidth of their I/O system
Post #3: I think #1 meant latency, not bandwidth
Post #4 (me): bottleneck refers to a bandwidth limitation, not latency.
Now you're talking about erratic request patterns that can decrease the performance of an I/O system. However, to measure both throughput and latency, you need to hold the requests constant. Latency is the time between when the requests arrive at the disk and the time the disk returns the first useful piece of data. Throughput is the rate at which the data arrives at the destination.
You have an I/O bottleneck (aka throughput problem) when the CPU idle time waiting for the full response to the request is much longer than the CPU idle time spent waiting for the first useful piece of data to arrive. You have a latency problem when the CPU idle time spent waiting for the first useful piece of data is much longer than the CPU idle time waiting for the full response.
I stand by my comment: bottleneck != latency problem. The seeking on a disk affects the throughput as well as the latency. The fact that you're not getting the theoretical maximum throughput for your series of random requests does not mean that you're experiencing a latency problem. A bottleneck can cause latency, and latency can cause a bottleneck (such as seeking within a disk, as you pointed out), but they are not equal.
...if there was a similar competition in the OS market. You wouldn't need these mammoth processors in the first place
What makes you say that? What makes you think that there are the same kind of performance gains possible from an OS? What makes you think there is not competition?