The issue I am trying to state is this: AT&T drop rate is 1.4%. Is that for the entire fleet of phones currently in service? Is that for just the iPhone 3Gs? Is that for the iPhone4? These are significantly different populations to be looking at.
We have two statements: "AT&T drop rate is 1.4% " and "Iphone 4 drops more calls than iphone 3Gs at a rate less than 1 per 100" Unfortunately we do not have a way of determining how these two pieces of information correlate. Without knowing the total fleet drop drop rate and the drop rate of the specific phones we cannot have any clear analysis of the numbers.
The issue is not the phone antenna and not the GPS antenna. The issue is when your hand (or a strip of metal for that matter) bridges between the two antennas. The effective length of the antenna is changed so the ability of the antenna to receive and transmit signals is significantly affected.
Just because issue A has a 1% incidence rate and issue B rate a 1.4% incidence rate does not necessarily mean that A+B = 2.4% incidence rate. How many of those incidence overlap or not?
Because there are less people reading the Times, fewer publicists are directing people to be interviewed at the Times. If you know people are reading the Guardian and not the Times and you want to get your message out, you go to the Guardian because more eyes are going to see your message. That is going to set up a feedback loop where people say "hey, the guardian has more content than the Times does, why am I reading the times." Then fewer people produce content for the Times, fewer people read the Times, etc etc etc.
It is hard to develop a user base when you seem to be actively driving away readers and by extension the people who develop your content.
Previously all the papers used the AP/Reuters because the AP covered issues the local paper couldn't. No one cared that everyone used the AP because people didn't read out of state newspapers.
Now the model has shifted. Everyone can read anyone's newspapers, but everyone is annoyed that all you get from any "local" newspaper is the same AP feed (some who charge for it and some who do not). I can see that small papers dropping the AP feed because it isn't useful to them any more. The bandwidth cost to carry information that everyone else has isn't worth it. Then the paper becomes a "local paper" or a "niche paper" again that can justify charging for its content. It will be able to charge because it is covering things that are locally important that you can't get anywhere else.
The AP on the other hand is going to have a problem: With all the small papers dropping them as a source of revenue, they will have to find another way to support themselves. I don't know what that is but they will have to scramble to get it done.
It appears that the quality of the WSJ reporting has declined since Murdoch took over. Most of the serious economists that want hard data and serious analysis have fled the WSJ and moved to the FT. The reason is simple: The WSJ is no longer providing the material that it used to.
On the other hand I think the Bancroft Family took the best advice for the stock market when selling the paper: Buy Low, Sell High.
The London waste disposal rules are very extensive. I won't even try to list it all.
In short: Everything must be presorted: Metals, papers, cloth, plastics (by type), wood, food waste (multiple types). Failure to presort can result fines.
Depending on your State/County requirements your building will need to be issued a Certificate of Occupancy before you can live in it. Check with your local code enforcement officials before you start this project. Some code inspectors take a very dim view of unusual buildings or systems.
Some of the things you will likely need so you can live in your dream house:
A set of construction drawings with an engineers or architect's stamp. - This is the engineer saying the building is not going to kill someone once it is built. Getting a set of drawings for an existing train car may be difficult.
Code review of the drawings will required before you can start construction. Code review is required to make sure something did not get past the engineer. If the code inspector says you have to do something before he will approve for human habitation, you are going to end up doing it whether you like it or not.
DO NOT piss off the code inspector. The code inspector can make your life a living hell, if just by putting your application at the bottom of the pile and not getting to your application for the next 6 weeks.
Find out in advance what inspections are required, when and in what order. Nothing is more frustrating than having Inspector A come out and then finding out that Inspector B's sign off is required before Inspector A will look at the building. Nothing pisses off the electrical inspector finding all of the cover plates have been installed before he has inspected them.
Proper compliance for water and sewer connections.
Proper compliance for heating and ventilation requirements.
Proper compliance for insulation requirements- walls, roof and windows. From what I know about subway windows, those are going to fail and have to be replaced with code compliant windows.
A licensed, insured and bonded electrician. It is almost impossible to get past this requirement. The code inspector will not certify the building if he cannot be assured that it is not going to burn down because of bad wiring.
Installation of Natural Gas/Propane may require a licensed or bonded installer. Check with your local code inspector. There is a reason why the utilities inject the "natural gas" smell into the lines.
As my signature says: With construction you can only compile once.
Consider the drawings and specifications your program. All of the libraries that you load in as well (and must comply with) are the local and state construction codes.
You cannot undo your compile without a significant amount of time, effort and money. Make sure everything is behind the walls before you sheetrock.
The heavier the load on a server the more it costs the server to operate.
Also each router between the server and the target will eventually want a cut so they can support their servers and routers.
I in no way expressed or implied that I was in favor of a fixed price or subscription price for this on going service. I did state I think that this business model will collapse within the next couple of years because the server has to be kept operating while their is no new revenue coming in.
I still cannot mix it into my playlist on my itunes.
This is going back to the subscription service that Walmart had that collapsed several months ago.
For a service like this to endure it has to do the following thing:
1. Sell a product, not a service agreement.
2. This product has to be complete self contained item, not subject to someone elses server getting shut down because the server is needed to operate whenever I want to listen to the music (ie a service agreement).
What happens when the content provider realizes that bandwidth costs money and that the measly $3 is not going to cover the bandwidth charges because someone wants to listen to the same album ten times a day?
I happen to think that this business model is very broken and a money loser. Websites sell advertising to pay for bandwidth. These guys are trying to sell an unlimited amount of bandwidth usage for a fixed price. This is going to collapse within the next 6-60 months leaving a whole lot of people who bought this app wondering where the songs they 'paid for' are.
If the drain traps dry out then the sewer gases (ie methane mixed with other products of sewer rot) can travel back up the pipe. It is the same reason all plumbing fixtures in your house have water traps, otherwise you would be getting sewer smell in your house all the time.
The room does not have to be cold. Just set the thermostat to something livable. Just because the room temperature is set to 70F does not mean the air conditioning can't keep up with the server load.
If you set the temperature to 70F and the AC equipment can't keep up with the load (ie the room temperature is always higher) then you have an issue.
The electricity powering the server room has already been used and is now laying around the space as left over heat. It is just a question now of what you do with the heat: keep it and heat another space or eject it out of the building envelope.
You have to ask the following questions when designing a cooling system for an HVAC system for a server room. The conditions are significantly different than cooling for an office.
1. What is your total kW consumed in the room.
2. What is the cooling strategy you want to employ?
Passive cooling - use a fan to dump the heat out of the building.
Active cooling - Install an air conditioner with an external condensing unit to dump the heat outside.
Alternate cooling - Dump the heat into the rest of the building during winter in order to saving on heating costs.
Any of these options have good and bad points: expense, humidity control, thermostat control, expense of use, required backups.
3. How is your server room arranged?
Is everything just thrown in the room?
Are you running a hot isle/cold isle environment?
Do you have a raised server floor so you can pump cold air into the bottom of the racks with a ceiling return?
Do the racks have fans to draw air from front to back?
4. What is your current cooling capacity that is dedicated to that room?
The last server room I designed was 500 square feet and consumed roughly 14 kw (28 watts/sf). That is roughly 4 tons of sensible cooling. To purchase a system capable of 4 tons of sensible cooling you will need to purchase a system capable of 5-6 tons of total cooling (Skipping the lecture on Sensible vs Latent vs Total cooling). So have you have just spent $4,000 in materials. Assume your costs will double for installation. Plus another couple of grand to have an actual engineer come in and design a system that will serve your specific needs.
The question is not one of getting the heat off the chips. The heat is making its way into the air just fine. You need to get the heat out of the air (and the room) and out of the building. If your room is exceeding 30C I would assume your racks are easily running 5-10 degrees hotter. That is getting into the range where your equipment is going to start shutting down.
Now onto your problem: $600 budget. Option: Throw a patch of your choice at it. A roll-a-way unit from Wal-mart or target where you can push a heat rejection duct out of the building, for example. This is a patch, not a solution.
Assume you are going to have to pony up $10,000 (USD) to solve this problem.
Sign me up for some of that alien technology goodness.
Slightly more realistic answer:
You are correct, producing helium from hydrogen would give you a very large net energy release. Problem: Last I knew anything that remotely resembles a fusion reactor is about the size of an office block. I don't plan on driving that around anytime soon.
You need initiating energy to start the reaction (to pull apart the CH4). Once the CH4 is pulled apart, it releases energy which breaks up the rest of the CH4 molecule and the O2. Energy is then consumed to form CO2 and H20. Once this is complete, there is some left over energy, normally heat.
Chemical reactions try to descend to their lowest (stable) energy state. Water and CO2 are some of the most stable molecules out there. You need to inject more energy into the molecule break them up then you get out of them:
H20 + Energy = H2 + O2
This is a net energy loss. You will get this energy back after you form water again.
If we could 'burn' (or whatever term you want to use) water to get energy from it, lightning strikes out in the Atlantic would have beaten us to it long ago.
If you are going to drive energy into water into its component elements (O2 and H2) and then attempt to extract energy from it, what is the end point of the reaction going to be if it is not water?
This sounds like All My Sins Remembered
on
Joss Whedon Back on TV
·
· Score: 2, Informative
The series sounds like All My sins remembered by Joe Haldeman. An average citizen is conditioned through hypnotherapy to assume given roles for a time period until the mission is over. After that he returns to base, debriefs and is assigned another mission. Eventually the brain cannot take the strain of repeated wipes and reprogramming and by the end when he is not on mission he is a schizophrenic stew of previous programmed personalities (thus - All my sins remembered).
Slashdot Mirror
The issue I am trying to state is this: AT&T drop rate is 1.4%. Is that for the entire fleet of phones currently in service? Is that for just the iPhone 3Gs? Is that for the iPhone4? These are significantly different populations to be looking at. We have two statements: "AT&T drop rate is 1.4% " and "Iphone 4 drops more calls than iphone 3Gs at a rate less than 1 per 100" Unfortunately we do not have a way of determining how these two pieces of information correlate. Without knowing the total fleet drop drop rate and the drop rate of the specific phones we cannot have any clear analysis of the numbers.
The issue is not the phone antenna and not the GPS antenna. The issue is when your hand (or a strip of metal for that matter) bridges between the two antennas. The effective length of the antenna is changed so the ability of the antenna to receive and transmit signals is significantly affected.
Just because issue A has a 1% incidence rate and issue B rate a 1.4% incidence rate does not necessarily mean that A+B = 2.4% incidence rate. How many of those incidence overlap or not?
Because there are less people reading the Times, fewer publicists are directing people to be interviewed at the Times. If you know people are reading the Guardian and not the Times and you want to get your message out, you go to the Guardian because more eyes are going to see your message. That is going to set up a feedback loop where people say "hey, the guardian has more content than the Times does, why am I reading the times." Then fewer people produce content for the Times, fewer people read the Times, etc etc etc.
It is hard to develop a user base when you seem to be actively driving away readers and by extension the people who develop your content.
Previously all the papers used the AP/Reuters because the AP covered issues the local paper couldn't. No one cared that everyone used the AP because people didn't read out of state newspapers.
Now the model has shifted. Everyone can read anyone's newspapers, but everyone is annoyed that all you get from any "local" newspaper is the same AP feed (some who charge for it and some who do not). I can see that small papers dropping the AP feed because it isn't useful to them any more. The bandwidth cost to carry information that everyone else has isn't worth it. Then the paper becomes a "local paper" or a "niche paper" again that can justify charging for its content. It will be able to charge because it is covering things that are locally important that you can't get anywhere else.
The AP on the other hand is going to have a problem: With all the small papers dropping them as a source of revenue, they will have to find another way to support themselves. I don't know what that is but they will have to scramble to get it done.
It appears that the quality of the WSJ reporting has declined since Murdoch took over. Most of the serious economists that want hard data and serious analysis have fled the WSJ and moved to the FT. The reason is simple: The WSJ is no longer providing the material that it used to. On the other hand I think the Bancroft Family took the best advice for the stock market when selling the paper: Buy Low, Sell High.
The London waste disposal rules are very extensive. I won't even try to list it all.
In short: Everything must be presorted: Metals, papers, cloth, plastics (by type), wood, food waste (multiple types). Failure to presort can result fines.
Reason: Landfill space is expensive on an island.
Depending on your State/County requirements your building will need to be issued a Certificate of Occupancy before you can live in it. Check with your local code enforcement officials before you start this project. Some code inspectors take a very dim view of unusual buildings or systems.
Some of the things you will likely need so you can live in your dream house:
A set of construction drawings with an engineers or architect's stamp. - This is the engineer saying the building is not going to kill someone once it is built. Getting a set of drawings for an existing train car may be difficult.
Code review of the drawings will required before you can start construction. Code review is required to make sure something did not get past the engineer. If the code inspector says you have to do something before he will approve for human habitation, you are going to end up doing it whether you like it or not.
DO NOT piss off the code inspector. The code inspector can make your life a living hell, if just by putting your application at the bottom of the pile and not getting to your application for the next 6 weeks. Find out in advance what inspections are required, when and in what order. Nothing is more frustrating than having Inspector A come out and then finding out that Inspector B's sign off is required before Inspector A will look at the building. Nothing pisses off the electrical inspector finding all of the cover plates have been installed before he has inspected them.
Proper compliance for water and sewer connections.
Proper compliance for heating and ventilation requirements.
Proper compliance for insulation requirements- walls, roof and windows. From what I know about subway windows, those are going to fail and have to be replaced with code compliant windows.
A licensed, insured and bonded electrician. It is almost impossible to get past this requirement. The code inspector will not certify the building if he cannot be assured that it is not going to burn down because of bad wiring.
Installation of Natural Gas/Propane may require a licensed or bonded installer. Check with your local code inspector. There is a reason why the utilities inject the "natural gas" smell into the lines.
As my signature says: With construction you can only compile once.
Consider the drawings and specifications your program. All of the libraries that you load in as well (and must comply with) are the local and state construction codes.
You cannot undo your compile without a significant amount of time, effort and money. Make sure everything is behind the walls before you sheetrock.
windows users need DVD43 in place of libdvdcss.
You can have your battlemech after I get my battlemech. I put down a deposit for my battlemaster last week.
http://www.vegawatt.com/
Also each router between the server and the target will eventually want a cut so they can support their servers and routers.
I in no way expressed or implied that I was in favor of a fixed price or subscription price for this on going service. I did state I think that this business model will collapse within the next couple of years because the server has to be kept operating while their is no new revenue coming in.
If you want to have the semantics argument: the electricity needed to run the servers costs money.
I still cannot mix it into my playlist on my itunes.
This is going back to the subscription service that Walmart had that collapsed several months ago.
For a service like this to endure it has to do the following thing:
1. Sell a product, not a service agreement.
2. This product has to be complete self contained item, not subject to someone elses server getting shut down because the server is needed to operate whenever I want to listen to the music (ie a service agreement).
I happen to think that this business model is very broken and a money loser. Websites sell advertising to pay for bandwidth. These guys are trying to sell an unlimited amount of bandwidth usage for a fixed price. This is going to collapse within the next 6-60 months leaving a whole lot of people who bought this app wondering where the songs they 'paid for' are.
1984 Honda civic. I got over 50 MPG with it a couple of times. Normally averaged over 40 wth it.
If the drain traps dry out then the sewer gases (ie methane mixed with other products of sewer rot) can travel back up the pipe. It is the same reason all plumbing fixtures in your house have water traps, otherwise you would be getting sewer smell in your house all the time.
The room does not have to be cold. Just set the thermostat to something livable. Just because the room temperature is set to 70F does not mean the air conditioning can't keep up with the server load.
If you set the temperature to 70F and the AC equipment can't keep up with the load (ie the room temperature is always higher) then you have an issue.
I'd like to avoid that option. Once you start cooking things off like that it starts getting more expensive before it gets cheaper.
The electricity powering the server room has already been used and is now laying around the space as left over heat. It is just a question now of what you do with the heat: keep it and heat another space or eject it out of the building envelope.
You have to ask the following questions when designing a cooling system for an HVAC system for a server room. The conditions are significantly different than cooling for an office.
1. What is your total kW consumed in the room.
2. What is the cooling strategy you want to employ?
Passive cooling - use a fan to dump the heat out of the building.
Active cooling - Install an air conditioner with an external condensing unit to dump the heat outside.
Alternate cooling - Dump the heat into the rest of the building during winter in order to saving on heating costs.
Any of these options have good and bad points: expense, humidity control, thermostat control, expense of use, required backups.
3. How is your server room arranged?
Is everything just thrown in the room?
Are you running a hot isle/cold isle environment?
Do you have a raised server floor so you can pump cold air into the bottom of the racks with a ceiling return?
Do the racks have fans to draw air from front to back?
4. What is your current cooling capacity that is dedicated to that room?
The last server room I designed was 500 square feet and consumed roughly 14 kw (28 watts/sf). That is roughly 4 tons of sensible cooling. To purchase a system capable of 4 tons of sensible cooling you will need to purchase a system capable of 5-6 tons of total cooling (Skipping the lecture on Sensible vs Latent vs Total cooling). So have you have just spent $4,000 in materials. Assume your costs will double for installation. Plus another couple of grand to have an actual engineer come in and design a system that will serve your specific needs.
The question is not one of getting the heat off the chips. The heat is making its way into the air just fine. You need to get the heat out of the air (and the room) and out of the building. If your room is exceeding 30C I would assume your racks are easily running 5-10 degrees hotter. That is getting into the range where your equipment is going to start shutting down.
Now onto your problem: $600 budget. Option: Throw a patch of your choice at it. A roll-a-way unit from Wal-mart or target where you can push a heat rejection duct out of the building, for example. This is a patch, not a solution.
Assume you are going to have to pony up $10,000 (USD) to solve this problem.
Sign me up for some of that alien technology goodness.
Slightly more realistic answer:
You are correct, producing helium from hydrogen would give you a very large net energy release. Problem: Last I knew anything that remotely resembles a fusion reactor is about the size of an office block. I don't plan on driving that around anytime soon.
Until then I'll keep riding my bike.
You need initiating energy to start the reaction (to pull apart the CH4). Once the CH4 is pulled apart, it releases energy which breaks up the rest of the CH4 molecule and the O2. Energy is then consumed to form CO2 and H20. Once this is complete, there is some left over energy, normally heat.
Chemical reactions try to descend to their lowest (stable) energy state. Water and CO2 are some of the most stable molecules out there. You need to inject more energy into the molecule break them up then you get out of them:
H20 + Energy = H2 + O2
This is a net energy loss. You will get this energy back after you form water again.
If we could 'burn' (or whatever term you want to use) water to get energy from it, lightning strikes out in the Atlantic would have beaten us to it long ago.
If you are going to drive energy into water into its component elements (O2 and H2) and then attempt to extract energy from it, what is the end point of the reaction going to be if it is not water?
The series sounds like All My sins remembered by Joe Haldeman. An average citizen is conditioned through hypnotherapy to assume given roles for a time period until the mission is over. After that he returns to base, debriefs and is assigned another mission. Eventually the brain cannot take the strain of repeated wipes and reprogramming and by the end when he is not on mission he is a schizophrenic stew of previous programmed personalities (thus - All my sins remembered).
http://www.thermastor.com/Heat-Recovery-Water-Heat ers/