I don't think you have ANY clue what you are talking about. One of my dads larger accounts is Timken. Among other things Timken makes bearings, they make lots of bearings for the auto industry, but they also make lots of bearings for the military. Neither customer base is large enough to sustain the enterprise but combined it's enough of a market to make a decent profit when things are going well. Timken is important enough to the military that their bearing plant in Canton Ohio had its own thermonuclear warhead during the cold war. Had GM gone bankrupt and stiffed Timken then Timken would have gone bankrupt and there would be no domestic supplier of a whole host of critical parts. If you WANT the government dependent on some supplier in Malaysia or mainland China for critical parts then you're a fool.
And that's better than having them in the middle of the chassis how? You still have all the complexity of the joints and the added possibility of losing one motor driving the car wildly out of control.
The Electric Focus probably wouldn't have been possible without the Volt, the $300M LG plant being built in Michigan was built to fulfill Volt orders but will also be supplying Ford for the Focus.
GM wasn't bailed out to make the Volt, they were bailed out to save the entire manufacturing base of the US. If GM had gone under then they would have stiffed every supplier and sub-supplier in the chain meaning that the entire manufacturing base in the US would have gone bust. That would have affected every auto company as well as most military contractors. That was not an acceptable outcome, bailing out GM was as much about national security as the strategic petroleum reserve is.
Actually two of the Tesla prototypes had electric generating trailers made for them, it lessened pure electric range by ~15% but enabled the total range to exceed most "normal" cars. Obviously it made parking and maneuvering more difficult, but I think for the enthusiast who would only rarely want to take his roadster on long trips I think it was a cool compromise.
In-wheel motor's don't work due to unsprung weight issues, basically you would end up with the suspension of a tank if you put all that weight into the wheels.
No, the $400 hammer was part of a special silent repair kit made for operating outside of the sonically shielded portion of a $Billions submarine. The kit was put together to a very exacting spec and then only a handful were ordered to go on the small fleet of American submarines. The rather high development cost was spread over a small number of kits. The $600 toilet seat was similar, a long out of production aircraft, the P3-C Orion subhunter (still used by NOAA for hurricane insertions) needed to have the existing toilet seats replaced due to age (25 years old at the start of production) and so a new mold needed to be made to fit the particular size and physical requirements for the aircraft. Anyone who works with plastics or fiberglass knows that the majority of the cost is in setting up the mold, so when you order 63 parts your per-part cost is going to be crazy high. Btw, this happens in industry all the time. When I worked at Cisco we spent several million on the tapeout for a new chip that ended up having a critical flaw that required a design spin and hence new tapeout. The handful of chips that were made with the flawed mask could have been said to be x hundred thousand dollar chips, but it would be just as inaccurate as the people yelling about the hammers and toilets. There's plenty of waste in the US government, finding stupid examples like those just makes you look like a fool.
Even if he lost custody in a court it still doesn't make it that much easier at an emotional level. Being separated from his child with little/no chance of ever seeing them again is going to be hard to deal with whether it's legal or not.
Who said I was cheering? I said I felt for him having experienced a similar/same situation. It doesn't absolve him from guilt and it doesn't mean I at all support his tactics, just that as a human being who has experienced something similar I can feel the pain he must have felt. I also have empathy for other victims of crimes including battered spouses that kill their abuser, that doesn't mean I think they shouldn't face justice for their actions.
As someone who has a family member who lost their child to international kidnapping I have to say I feel for the guy. There's really nothing worse than having your child ripped from you and being physically separated with little hope of ever seeing them again.
It's always worked dcpromo/unatted:unattend.txt, how do you think server core DC's get built? One of the last major obstacles in many environments to running core was the broken Broadcomm teaming tools but that issue has been resolved for a couple months.
But a great GUI is one that teaches a new user to eventually graduate to using CLI.
I couldn't agree more which is why it's really cool that the Exchange GUI tools will show you the exact PowerShell commands they are running if you want to learn how to use the CLI (and in fact that are still things that can only be done in the CLI, though the list is greatly reduced in EX 2010 vs 2007).
No, they would normally pay property tax which is based on the book value of the physical assets at the site. For a large datacenter that's a LOT of capital and hence lots of taxable land value.
Google's model of fail in place is the way forward, why pay for expensive service contracts and expensive people when you can just design the algorithms to handle failures gracefully (a necessity at those scales anyways) and just buy an extra 20% initial hardware to account for the losses until you reach you're real expected workload. Heck by the time 20% of your servers in a rack fail it's probably so power inefficient that it's time to replace them with newer components anyways. This doesn't apply to the datacenter I manage, but that's only because we only have ~330 servers instead of the hundreds of thousands someone like Google or Yahoo have.
Actually when I was talking to the head IT guy at a particular large bank in California he said that the Northridge earthquake knocked out everything at that campus except for one POTS line, they had power from the two turbine generators but no telecommunications, no AC for the worker areas, and most importantly no water. It took weeks before things were back to normal.
RIM's sales are growing faster than Apple's (+4.5M vs +3.4M year over year growth for the second quarter). They're just growing slower in percentage terms since RIM had such a large number of units shipped all along. Android is growing mostly at the cost of Symbian.
Your typical car doesn't go 200+...
Exotic's that can go that fast typically do have an internal structure like a roll cage, 5 point harness's are common, and they have airbags which racecars don't have. As an example of survivability only about half of these severe accidents with exotics were fatal.
If something goes wrong at 205 mph? well, in that case you don't have to worry how to cut open the car, because the driver is probably very dead. solves that problem.
Why do people assume all or even a majority of accidents at tripple digits are fatal? NASCAR has several crashes at near 200mph every year and yet it's been a decade since they've killed a driver or even seriously injured one. Granted they have the advantage of designing the obstacles the car can hit, but as an accident earlier this year showed they still end up with some unprotected surfaces that can end with sudden deceleration.
Wallstreet's focus on this quarter's return and the opening of the telecomm market to competition instead of a state granted monopoly where the expense of running Bell Labs was written into the rate structure.
Nobody is going to put 48 cores behind a L2 cache, 4 cores is probably the max for a shared L2, beyond that you go to a L3. Heck the newer Intel designs split L3 cache into two sets to keep the number of cores sharing a cache reasonable.
Heck, any 8 socket server with Nehalem-EX can support up to 128 logical processors (64 cores). I don't think I've heard anyone say their new 8 socket servers aren't capable of sustaining decent workloads.
Slashdot Mirror
Because aluminum is $1.08/lb on the spot market right now vs steel at $.30/lb and aluminum is also much hardware to machine and weld.
I don't think you have ANY clue what you are talking about. One of my dads larger accounts is Timken. Among other things Timken makes bearings, they make lots of bearings for the auto industry, but they also make lots of bearings for the military. Neither customer base is large enough to sustain the enterprise but combined it's enough of a market to make a decent profit when things are going well. Timken is important enough to the military that their bearing plant in Canton Ohio had its own thermonuclear warhead during the cold war. Had GM gone bankrupt and stiffed Timken then Timken would have gone bankrupt and there would be no domestic supplier of a whole host of critical parts. If you WANT the government dependent on some supplier in Malaysia or mainland China for critical parts then you're a fool.
And that's better than having them in the middle of the chassis how? You still have all the complexity of the joints and the added possibility of losing one motor driving the car wildly out of control.
The Electric Focus probably wouldn't have been possible without the Volt, the $300M LG plant being built in Michigan was built to fulfill Volt orders but will also be supplying Ford for the Focus.
GM wasn't bailed out to make the Volt, they were bailed out to save the entire manufacturing base of the US. If GM had gone under then they would have stiffed every supplier and sub-supplier in the chain meaning that the entire manufacturing base in the US would have gone bust. That would have affected every auto company as well as most military contractors. That was not an acceptable outcome, bailing out GM was as much about national security as the strategic petroleum reserve is.
Actually two of the Tesla prototypes had electric generating trailers made for them, it lessened pure electric range by ~15% but enabled the total range to exceed most "normal" cars. Obviously it made parking and maneuvering more difficult, but I think for the enthusiast who would only rarely want to take his roadster on long trips I think it was a cool compromise.
In-wheel motor's don't work due to unsprung weight issues, basically you would end up with the suspension of a tank if you put all that weight into the wheels.
No, the $400 hammer was part of a special silent repair kit made for operating outside of the sonically shielded portion of a $Billions submarine. The kit was put together to a very exacting spec and then only a handful were ordered to go on the small fleet of American submarines. The rather high development cost was spread over a small number of kits. The $600 toilet seat was similar, a long out of production aircraft, the P3-C Orion subhunter (still used by NOAA for hurricane insertions) needed to have the existing toilet seats replaced due to age (25 years old at the start of production) and so a new mold needed to be made to fit the particular size and physical requirements for the aircraft. Anyone who works with plastics or fiberglass knows that the majority of the cost is in setting up the mold, so when you order 63 parts your per-part cost is going to be crazy high. Btw, this happens in industry all the time. When I worked at Cisco we spent several million on the tapeout for a new chip that ended up having a critical flaw that required a design spin and hence new tapeout. The handful of chips that were made with the flawed mask could have been said to be x hundred thousand dollar chips, but it would be just as inaccurate as the people yelling about the hammers and toilets. There's plenty of waste in the US government, finding stupid examples like those just makes you look like a fool.
Not in the US, and it's been equivalent in the US since the Clinton administration.
The Army already does this with some FOB's, airlift in prefab sections to build a complex.
Even if he lost custody in a court it still doesn't make it that much easier at an emotional level. Being separated from his child with little/no chance of ever seeing them again is going to be hard to deal with whether it's legal or not.
Who said I was cheering? I said I felt for him having experienced a similar/same situation. It doesn't absolve him from guilt and it doesn't mean I at all support his tactics, just that as a human being who has experienced something similar I can feel the pain he must have felt. I also have empathy for other victims of crimes including battered spouses that kill their abuser, that doesn't mean I think they shouldn't face justice for their actions.
As someone who has a family member who lost their child to international kidnapping I have to say I feel for the guy. There's really nothing worse than having your child ripped from you and being physically separated with little hope of ever seeing them again.
The growth of the routing table.
It's always worked dcpromo /unatted:unattend.txt, how do you think server core DC's get built? One of the last major obstacles in many environments to running core was the broken Broadcomm teaming tools but that issue has been resolved for a couple months.
But a great GUI is one that teaches a new user to eventually graduate to using CLI.
I couldn't agree more which is why it's really cool that the Exchange GUI tools will show you the exact PowerShell commands they are running if you want to learn how to use the CLI (and in fact that are still things that can only be done in the CLI, though the list is greatly reduced in EX 2010 vs 2007).
No, they would normally pay property tax which is based on the book value of the physical assets at the site. For a large datacenter that's a LOT of capital and hence lots of taxable land value.
Google's model of fail in place is the way forward, why pay for expensive service contracts and expensive people when you can just design the algorithms to handle failures gracefully (a necessity at those scales anyways) and just buy an extra 20% initial hardware to account for the losses until you reach you're real expected workload. Heck by the time 20% of your servers in a rack fail it's probably so power inefficient that it's time to replace them with newer components anyways. This doesn't apply to the datacenter I manage, but that's only because we only have ~330 servers instead of the hundreds of thousands someone like Google or Yahoo have.
Actually when I was talking to the head IT guy at a particular large bank in California he said that the Northridge earthquake knocked out everything at that campus except for one POTS line, they had power from the two turbine generators but no telecommunications, no AC for the worker areas, and most importantly no water. It took weeks before things were back to normal.
RIM's sales are growing faster than Apple's (+4.5M vs +3.4M year over year growth for the second quarter). They're just growing slower in percentage terms since RIM had such a large number of units shipped all along. Android is growing mostly at the cost of Symbian.
Your typical car doesn't go 200+...
Exotic's that can go that fast typically do have an internal structure like a roll cage, 5 point harness's are common, and they have airbags which racecars don't have. As an example of survivability only about half of these severe accidents with exotics were fatal.
If something goes wrong at 205 mph? well, in that case you don't have to worry how to cut open the car, because the driver is probably very dead. solves that problem.
Why do people assume all or even a majority of accidents at tripple digits are fatal? NASCAR has several crashes at near 200mph every year and yet it's been a decade since they've killed a driver or even seriously injured one. Granted they have the advantage of designing the obstacles the car can hit, but as an accident earlier this year showed they still end up with some unprotected surfaces that can end with sudden deceleration.
Wallstreet's focus on this quarter's return and the opening of the telecomm market to competition instead of a state granted monopoly where the expense of running Bell Labs was written into the rate structure.
Nobody is going to put 48 cores behind a L2 cache, 4 cores is probably the max for a shared L2, beyond that you go to a L3. Heck the newer Intel designs split L3 cache into two sets to keep the number of cores sharing a cache reasonable.
Heck, any 8 socket server with Nehalem-EX can support up to 128 logical processors (64 cores). I don't think I've heard anyone say their new 8 socket servers aren't capable of sustaining decent workloads.