Ease of Development: How fast can quality code be turned out by regular programmers?
I don't think it is quite that simple. Different projects have different requirements. For example, Perl can very quickly put out simple quality programs but good luck maintaining and/or extending them.
One really has to consider the project in question. For many projects the most important consideration is whether or not many different programmers can all produce quality code that will fit nicely together into a finished product. Languages like Java appear to be designed to do this. In doing so, sacrifices are made to other aspects of their designs.
So what I'm basically saying is that I agree with your comment. But the point about Ease of Development really needs to take into consideration the project at hand.
You can create a custom sparse disk image that has a maximum size limit - then TimeMachine will know. There are instructions posted on macosxhints.com.
That setup would be good for 20km - slightly overkill for 500m. I would suggest dropping the signal boosters - just another part to break. In my experience, it's the signal boosters that break first.
Cables are generally run using a "Ditch WItch" or some other piece of trenching equipment. You can rent them just about anywhere and they can finish a 500m job in a couple of hours.
on mac os x qt4 looks good to me, there is even an alpha version of qt4 that uses cocoa instead of carbon
The problem is not the look, it's the feel. It is the way the toolbars work, dialog boxes, etc. It is the way all the pieces fit together to provide a user interface. Qt is impressive but it is not native.
there is even an alpha version of qt4 that uses cocoa instead of carbon
Cocoa and Carbon are actually the same thing. Just different APIs used to access the same elements. Qt adds yet another API which is fine but there is no real difference in Qt using Cocoa vs. Carbon. Ok, there is one difference. Carbon will never be 64bit so if Qt wants to be 64bit native then they will have to utilize Cocoa. This is likely why the Qt engineers are looking into Cocoa.
Alaska has more sunlight during the summer. Alaska also lacks fresh produce so growing it there can feed the local population and reduce on shipping. Manitoba has no lack of farms and the resulting produce would have to be shipped to a market (consuming fuel). One last point about Alaska, it's rich in natural gas and if the power plant was powered by natural gas, the resulting CO2 could be captured and pumped underground to replace the extracted natural gas. Oh, Alaska is also better positioned wrt the internet when considering a global market. Manitoba would be great for a Canadian market but not so much for Asia. Of course this all depends on how much latency is acceptable. For low latency applications a data center in the Midwest would be likely be preferable over Manitoba.
Unfortunately, here in Alaska, we're undergoing an energy crisis.
This could be perceived as being a good thing. You see, if Alaska has to invest in new power sources, the costs of building additional generating capacity is actually minor. One just has to build a slightly larger power plant. All the other costs such as securing a fuel source, allocating land, building infrastructure, etc. remain almost constant.
The state would simply require a long term plan. Get work started right away on new generating facilities. If they are not ready in time simply sell power at reduced rates until the new generating facilities are brought online. If the state lacks power, buy if from British Columbia or the Yukon. The benefits of bringing in a new industry are huge - the state can afford the temporary hit to their budget.
I agree - hence my "everyone must be on board" comment. Problem with putting a tax on imported goods is that it is not trivial - too difficult for a province / state. If everyone in North America was working together it would be quite reasonable. Domestic products could be ignored because the carbon tax was paid when manufacturing the goods. Imported products need only be assessed once for any part of North America.
A substantial carbon tax would actually be acceptable if other taxes were reduced. Goods with no environmental impact should have no (or at least minimal) taxes. Destructive goods should be taxed heavily. Overall, it should average out to be the same amount of tax as today. Peoples' habits would change overnight.
Such a solution will only happen when people force their governments to make it happen. This requires awareness of the issues and possible solutions. This is why I'm writing this message even though I've been taken horribly off topic. But your comment was absolutely correct - hopefully some of those European countries can jump-start the whole process.
(posted without karma bonus because it is off topic)
Wish I could say that this is all my idea. But in reality, they have been doing this for some time with power generation in Europe. North America is just behind the times. It's easier to vent the heat into the atmosphere then to design systems to utilize the heat around the power plant (and fyi, ~60%->heat and 40%->electricity). This won't change until the cost of power increases. British Columbia recently introduced a carbon tax - the first step in making this happen. But for any real changes to occur, everyone in North America has to get on board. Adverse effect to domestic industry can be countered by applying import tariffs based on the estimated carbon required to produce those products. This gives an advantage to countries that invest in low carbon power, and incentive to change for those that do not.
While I believe your campus is as you described, I do not believe your conclusion is valid. It appears to be a case where the campus was not initially designed to support a high performance computing lab. It is quite typical for these things to be added on after the fact. Even if the building is new, the heating, cooling, water, electrical will likely come from a central source that was designed without thought of the lab. And even if the lab was planned, adding other buildings can still cause the system to adversely effect the lab.
So you point is good. Such a solution would have to be designed appropriately. But a poorly designed campus in the Midwest does not mean that the idea will not work in practice. I have seen many buildings that suffer from the symptoms you described (it reaches -50C here). All is good until it gets real cold then some rooms are freezing while others are like saunas. It all comes down to design - something the described buildings lacked (wrt heating).
I would suggest locating data centers in a cool climate where farming is popular. Pump the waste heat from the data centers into greenhouses that can surround the data center. Now that waste is helping to grow food.
Alaska is actually a good place to implement such a solution. There is a huge amount of sunlight in the summer which, assuming you can avoid frosts, can grow amazing produce. All you need are greenhouses and a heat source. In the winter, when sunlight is no longer plentiful and farming shuts down, the heat can be pumped into local housing. Such a solution would also provide local produce in Alaska - produce that is fresh and doesn't require expensive shipping. One last point about Alaska, it's very central. It might not appear to be when looking at a map, but if you look at a globe you will see that it sits nicely between Asia and North America. I don't know where the current internet pipes are located but if they pass close to Alaska then this idea would be worth some consideration.
But those low pressure sodium lights don't turn on and off instantly. It can take several minutes depending on the temperature. And when in the process of turning on, the efficiency sucks. They are really designed to be turned on and then left on.
If a different form of lighting were used (like LED) then your suggestion would be worth consideration. But such a suggestion would currently require that all street lighting be replaced with an alternative that can quickly turn on/off.
If scripts are brought together from different hosts this could be a problem. Or how about scripts from a local page that reference a foreign host. What happens if the host is compromised in the future?
I realize that when used properly, this is probably not an issue - hence "misuse" in the title. But the language should be designed such that it does not promote dangerous code. We are talking about running code from a remote host - that can always result in problems.
My first thought is that allowing for "signed" scripts could be a good idea. How about only allowing "unsigned" scripts that originate from the same host?
There are lots of potential problems and solutions to "Program Units". I was just hoping someone more knowledgeable on the subject would provide some thoughts on the issues.
Reading the article I found "Program Units" to be interesting. Most importantly, how does the running program know that the downloaded script is safe? At first glance it appears that one could easily inject malicious script via a man in the middle attack. Now I'm sure that the designers have thought about this so my question is, how does JavaScript 2.0 protect against this?
While obviously emotionally charged, the parent post is not flamebait and should not be modded as such. In fact, the parent brings up the interesting points of insurance and shared benefits. Bringing up these points makes this a legitimate counter-argument, hence it is not flamebait. May I suggest a mod of "off topic" instead?
The centralized solutions are typically easier to design and can result in optimal plans. Decentralized techniques have their advantages, but they are harder to design and do not offer optimal plans. Because it is reasonable to have a central computer with sufficient reliability at an airport, the centralized solutions make more sense.
The best solution might be to utilize both techniques. Centralized planning at airports and decentralized planning (ie, p2p) for incase something goes wrong. Decentralized planning would also be a good solution for when planning routes mid-flight where multiple aircraft might be in danger of colliding.
Since when does going to a smaller process increase yields?
Because the die size is reduced at the smaller process. If you normally get 100 dies from a wafer then you will now get 150. Of course this doesn't take into account flaws in the silicon. If you have 20 flaws then you could lose ~20 dies. This results in a loss of 20% at the larger process and 13.3% at the smaller process.
Great comment, but I would like to make one observation.
The biggest failure I can see, given the targeted market segment, is the lack of a cellular connection capability. And given that such a feature would lock one into a particular cellular network, I can understand the omission -- but a space to add such a card at a later time would have been nice.
OSX can simply utilize a cell phone as a cellular connection via a bluetooth link. Since those who purchase a MBA will most likely already be carrying a cell with them, it's no big deal. One might however want to utilize a USB cable to ensure the cell doesn't run out of juice while in use.
But think about all those business users who want a laptop that will fit into their briefcase. They want a thin laptop with a large footprint (for a larger screen) with enough battery life so that they don't have to carry around a power adapter. This is the target audience of the macbook air. For those that want something even more portable they have the iPhone/iPod. Granted, there is some space for a in-between product but it is likely a smaller market then you think. One thing is for certain, should there be a large demand for such a product Apple will eventually deliver one.
I'm inviting a flame war here, but isn't math - at least in the sense that we teach it - an artificial construct that we invented to describe our observations? Math can contain theories, but I don't think it could ever be classified as one because we actually know what it is.
I agree.
sn't gravity demonstrable? If I understand correctly, there's still a lot of uncertainty in the scientific community about how gravity works exactly, but it's clearly an observable and demonstrable fact that it does.
Gravity is a theory that explains why object fall to the ground when dropped. No one can prove that this will continue to happen in the future - but the theory of gravity predicts that it will.
Isn't biology something we can study that's in front of our faces? We can actually watch plants growing, babies forming in a womb, organs working, cells replicating etc...
We can observe what is happening in front of our faces, but biology is more then that. Biology lets us predict what will happen in the future under different circumstances. No one can be absolutely certain that these predictions will come true - this is why biology is not an absolute fact.
Neither group actually "knows" how these things came to be, they've just adopted a view of it that they are comfortable with. What I don't understand is how the evolutionists, who are supposed to be the more objective and open minded of the two groups, can be so "holier than thou" as to suggest that the creationists' theory doesn't even deserve a place.
The thing is, creationists' do not have a scientific theory. All scientific theories can be proved wrong - but not the creationists' theory. This is why scientists do not appreciate it. Believe it or not, science welcomes evidence that a theory is wrong. It allows science to be improved. But all of this points away from the creationists' theory. Should science ignore all of the hard work that has been done over the centuries and blindly follow a non-scientific theory? Can the theories even be compared? The answer from scientists - no.
I just don't see it being real useful to try and harvest waste heat from an ICE or turbine. If a power-plant turbine had useful exhaust steam, they would already be using it to turn another turbine I expect.
A steam turbine receives stream at a high temperature and releases steam at a low temperature. It should be noted that the H2O is always in the form of steam. But before this steam can go back into the boiler it must be converted into a liquid. This is where all the energy is lost. Changing states requires a huge amount of energy due to the enthalpy of vaporization. This is where the proposed heat engine would be used.
Slashdot Mirror
I don't think it is quite that simple. Different projects have different requirements. For example, Perl can very quickly put out simple quality programs but good luck maintaining and/or extending them.
One really has to consider the project in question. For many projects the most important consideration is whether or not many different programmers can all produce quality code that will fit nicely together into a finished product. Languages like Java appear to be designed to do this. In doing so, sacrifices are made to other aspects of their designs.
So what I'm basically saying is that I agree with your comment. But the point about Ease of Development really needs to take into consideration the project at hand.
You can create a custom sparse disk image that has a maximum size limit - then TimeMachine will know. There are instructions posted on macosxhints.com.
That setup would be good for 20km - slightly overkill for 500m. I would suggest dropping the signal boosters - just another part to break. In my experience, it's the signal boosters that break first.
Cables are generally run using a "Ditch WItch" or some other piece of trenching equipment. You can rent them just about anywhere and they can finish a 500m job in a couple of hours.
It's a development version. Wait for the final version before complaining too much.
The problem is not the look, it's the feel. It is the way the toolbars work, dialog boxes, etc. It is the way all the pieces fit together to provide a user interface. Qt is impressive but it is not native.
Cocoa and Carbon are actually the same thing. Just different APIs used to access the same elements. Qt adds yet another API which is fine but there is no real difference in Qt using Cocoa vs. Carbon. Ok, there is one difference. Carbon will never be 64bit so if Qt wants to be 64bit native then they will have to utilize Cocoa. This is likely why the Qt engineers are looking into Cocoa.
WillyTry going here.
Alaska has more sunlight during the summer. Alaska also lacks fresh produce so growing it there can feed the local population and reduce on shipping. Manitoba has no lack of farms and the resulting produce would have to be shipped to a market (consuming fuel). One last point about Alaska, it's rich in natural gas and if the power plant was powered by natural gas, the resulting CO2 could be captured and pumped underground to replace the extracted natural gas. Oh, Alaska is also better positioned wrt the internet when considering a global market. Manitoba would be great for a Canadian market but not so much for Asia. Of course this all depends on how much latency is acceptable. For low latency applications a data center in the Midwest would be likely be preferable over Manitoba.
This could be perceived as being a good thing. You see, if Alaska has to invest in new power sources, the costs of building additional generating capacity is actually minor. One just has to build a slightly larger power plant. All the other costs such as securing a fuel source, allocating land, building infrastructure, etc. remain almost constant.
The state would simply require a long term plan. Get work started right away on new generating facilities. If they are not ready in time simply sell power at reduced rates until the new generating facilities are brought online. If the state lacks power, buy if from British Columbia or the Yukon. The benefits of bringing in a new industry are huge - the state can afford the temporary hit to their budget.
I agree - hence my "everyone must be on board" comment. Problem with putting a tax on imported goods is that it is not trivial - too difficult for a province / state. If everyone in North America was working together it would be quite reasonable. Domestic products could be ignored because the carbon tax was paid when manufacturing the goods. Imported products need only be assessed once for any part of North America.
A substantial carbon tax would actually be acceptable if other taxes were reduced. Goods with no environmental impact should have no (or at least minimal) taxes. Destructive goods should be taxed heavily. Overall, it should average out to be the same amount of tax as today. Peoples' habits would change overnight.
Such a solution will only happen when people force their governments to make it happen. This requires awareness of the issues and possible solutions. This is why I'm writing this message even though I've been taken horribly off topic. But your comment was absolutely correct - hopefully some of those European countries can jump-start the whole process.
(posted without karma bonus because it is off topic)
Wish I could say that this is all my idea. But in reality, they have been doing this for some time with power generation in Europe. North America is just behind the times. It's easier to vent the heat into the atmosphere then to design systems to utilize the heat around the power plant (and fyi, ~60%->heat and 40%->electricity). This won't change until the cost of power increases. British Columbia recently introduced a carbon tax - the first step in making this happen. But for any real changes to occur, everyone in North America has to get on board. Adverse effect to domestic industry can be countered by applying import tariffs based on the estimated carbon required to produce those products. This gives an advantage to countries that invest in low carbon power, and incentive to change for those that do not.
While I believe your campus is as you described, I do not believe your conclusion is valid. It appears to be a case where the campus was not initially designed to support a high performance computing lab. It is quite typical for these things to be added on after the fact. Even if the building is new, the heating, cooling, water, electrical will likely come from a central source that was designed without thought of the lab. And even if the lab was planned, adding other buildings can still cause the system to adversely effect the lab.
So you point is good. Such a solution would have to be designed appropriately. But a poorly designed campus in the Midwest does not mean that the idea will not work in practice. I have seen many buildings that suffer from the symptoms you described (it reaches -50C here). All is good until it gets real cold then some rooms are freezing while others are like saunas. It all comes down to design - something the described buildings lacked (wrt heating).
I would suggest locating data centers in a cool climate where farming is popular. Pump the waste heat from the data centers into greenhouses that can surround the data center. Now that waste is helping to grow food.
Alaska is actually a good place to implement such a solution. There is a huge amount of sunlight in the summer which, assuming you can avoid frosts, can grow amazing produce. All you need are greenhouses and a heat source. In the winter, when sunlight is no longer plentiful and farming shuts down, the heat can be pumped into local housing. Such a solution would also provide local produce in Alaska - produce that is fresh and doesn't require expensive shipping. One last point about Alaska, it's very central. It might not appear to be when looking at a map, but if you look at a globe you will see that it sits nicely between Asia and North America. I don't know where the current internet pipes are located but if they pass close to Alaska then this idea would be worth some consideration.
William
But those low pressure sodium lights don't turn on and off instantly. It can take several minutes depending on the temperature. And when in the process of turning on, the efficiency sucks. They are really designed to be turned on and then left on.
If a different form of lighting were used (like LED) then your suggestion would be worth consideration. But such a suggestion would currently require that all street lighting be replaced with an alternative that can quickly turn on/off.
If scripts are brought together from different hosts this could be a problem. Or how about scripts from a local page that reference a foreign host. What happens if the host is compromised in the future?
I realize that when used properly, this is probably not an issue - hence "misuse" in the title. But the language should be designed such that it does not promote dangerous code. We are talking about running code from a remote host - that can always result in problems.
My first thought is that allowing for "signed" scripts could be a good idea. How about only allowing "unsigned" scripts that originate from the same host?
There are lots of potential problems and solutions to "Program Units". I was just hoping someone more knowledgeable on the subject would provide some thoughts on the issues.
Reading the article I found "Program Units" to be interesting. Most importantly, how does the running program know that the downloaded script is safe? At first glance it appears that one could easily inject malicious script via a man in the middle attack. Now I'm sure that the designers have thought about this so my question is, how does JavaScript 2.0 protect against this?
William
to increase the surface area and ensure that it burns up upon reentry. They didn't want any big chunks falling down.
Willy
While obviously emotionally charged, the parent post is not flamebait and should not be modded as such. In fact, the parent brings up the interesting points of insurance and shared benefits. Bringing up these points makes this a legitimate counter-argument, hence it is not flamebait. May I suggest a mod of "off topic" instead?
The centralized solutions are typically easier to design and can result in optimal plans. Decentralized techniques have their advantages, but they are harder to design and do not offer optimal plans. Because it is reasonable to have a central computer with sufficient reliability at an airport, the centralized solutions make more sense.
The best solution might be to utilize both techniques. Centralized planning at airports and decentralized planning (ie, p2p) for incase something goes wrong. Decentralized planning would also be a good solution for when planning routes mid-flight where multiple aircraft might be in danger of colliding.
Willy
Because the die size is reduced at the smaller process. If you normally get 100 dies from a wafer then you will now get 150. Of course this doesn't take into account flaws in the silicon. If you have 20 flaws then you could lose ~20 dies. This results in a loss of 20% at the larger process and 13.3% at the smaller process.
But then that electricity gets turned into heat so you do not lose heat.
Great comment, but I would like to make one observation.
OSX can simply utilize a cell phone as a cellular connection via a bluetooth link. Since those who purchase a MBA will most likely already be carrying a cell with them, it's no big deal. One might however want to utilize a USB cable to ensure the cell doesn't run out of juice while in use.
WillyBut think about all those business users who want a laptop that will fit into their briefcase. They want a thin laptop with a large footprint (for a larger screen) with enough battery life so that they don't have to carry around a power adapter. This is the target audience of the macbook air. For those that want something even more portable they have the iPhone/iPod. Granted, there is some space for a in-between product but it is likely a smaller market then you think. One thing is for certain, should there be a large demand for such a product Apple will eventually deliver one.
I agree.
Gravity is a theory that explains why object fall to the ground when dropped. No one can prove that this will continue to happen in the future - but the theory of gravity predicts that it will.
We can observe what is happening in front of our faces, but biology is more then that. Biology lets us predict what will happen in the future under different circumstances. No one can be absolutely certain that these predictions will come true - this is why biology is not an absolute fact.
The thing is, creationists' do not have a scientific theory. All scientific theories can be proved wrong - but not the creationists' theory. This is why scientists do not appreciate it. Believe it or not, science welcomes evidence that a theory is wrong. It allows science to be improved. But all of this points away from the creationists' theory. Should science ignore all of the hard work that has been done over the centuries and blindly follow a non-scientific theory? Can the theories even be compared? The answer from scientists - no.
A steam turbine receives stream at a high temperature and releases steam at a low temperature. It should be noted that the H2O is always in the form of steam. But before this steam can go back into the boiler it must be converted into a liquid. This is where all the energy is lost. Changing states requires a huge amount of energy due to the enthalpy of vaporization. This is where the proposed heat engine would be used.