I have never had a problem getting an automated response for a sign up verification. I get maybe 5-10 unsolicited spams a day (all of which go directly to a junk mail folder) and 20+ solicited spams (email lists, tech groups, companies I deal with, etc...) emails a day (once again, it all goes to junk mail).
So while other user's may have problems, I guess I'm just lucky and I've never really had a problem with Hotmail. To the extent that it has been my primary email provider since '97 (pre-MS days).
The problem is that the vast majority of petition signers are not willing to perform action in association with the petition. I wish I had the cites handy to back that statement up, but that paper was from a few years ago.
In short, 165,000 users signing a petition, but still paying their bills means nothing to AT&T.
10,000 users canceling their service and citing the PP as the primary reason will likely invoke a reaction.
You are correct, it is about critical mass. 10 dollar-voters will not make a difference, but 10,000 petition-signers will also not make a difference. 10,000 dollar-voters will likely make a difference, as would 10,000,000 petition-signers. Is it easier to come up with 10 million people to threaten to vote with their dollar, or 10 thousand people who will vote with their dollar?
It would be kinda interesting to be able to visit someone's blog and be able to talk with other visitors and the author live time. And it could also be handy if you keep a "speed dial" list on your startup page, no need to dig out numbers, just hit the home button and click the link to your favorite relative.
Now, this could be a heck of a performance hit, but... What if a company supplied a local VOIP call-in bank. As in POTS copper lines that you could dial into on a modem. That VOIP call could then be secured and piped to some other country. The other end of the VOIP call would then be answered by a modem bank which spits out into some foreign ISP.
It would be slow as tar, but it should get you a connection that isn't being directly reviewed by the NSA.
The other problem is that even those of us who don't sign AT&T's privacy agreement can still be monitored as packets bound for us may travel over their backbone pipes.
The best way to force AT&T to change their game is to vote with your all-mighty dollar. A single dollar-voting customer is worth any number of petitions and angry letters.
I bought an '06 VW Golf TDI (turbo diesel) a few months back, and so far I am averaging low 40's mileage (worst ~39.5, best ~43.5). I hit 600 miles a tank with no problem. True, it's no lean burning, single passenger, 25mph, wheel chair, but it does have 170ft/lbs of torque, and cruises beautifully at 80mph.
"Three: It may (possibly) be a deterrent for the teams to not cop out and buy a super duper-efficient experimental engine from some no-name company and call it as their own. "
And what's wrong with that? If a team wins using some start-up company's new experimental engine, the company with the engine gets advertising and investment, and the team gets a win. Not to mention the team winners will likely have a great shot at getting a job with that company.
The IBM chip is silicon at extremely low temperatures (~4.5K). This story is about carbon nanotubes being used in transistors. Two completely seperate and unrelated technologies.
So, proper engineering it is. First, use a non-conductive cooling fluid. That will immediately reduce the critical nature of a leak. Second, get a plastics manufacturer to make a 1/2U tray with a slant to force any spilt fluid to one specific corner. Next use a GFCI style system on the flow, measure the amount of coolant going into the line, and the amount being returned, if they don't match, use a UPS interface to send a shutdown command to the server, notify the staff, and once the machine is down close the electric valve for that server.
The non-conductive cooling fluid is a bit more expensive, but in a closed system you wouldn't need to add too much over time, and the safety it brings is well worth the cost. The 1/2U trays could be manufactured for pennies after the initial tooling. If you are HP and selling a rack integrated water cooling systems (so that all water plugs are modular and built into the rack) the extra pennies from those trays are a drop in the bucket. The GFCI and UPS interface is a bit more costly and has to work at the server level. And with the cooling fluid is less of a concern, but is would still be good to have an electric system to sound the alarm for a leak.
As for the building cooling system, it won't work. HVAC cooled water is too cold, it would cause condensation inside the server which IS conductive and would do bad things. It would be much better to have a cooling fluid-water or cooling fluid-air interchange where the water system for the servers can release heat to the HVAC system. You would have to set up a thermostat and control the water temp to make sure it is cool, but not cold enough to condense water from the air.
They just upped my rates again. I pay $115 for a 3Mb connection, Digital cable, and the premium movie channels. If I drop the premium channels and down to a lower connection, my bill would actually increase to $120. As much as I hate over paying, I hate paying a company that I know is packet forwarding to the NSA and attempting to get a QoS billing scheme even more.
The problem with self policing in most MMOs is that players have limited resources in dealing with griefers. Atleast in open PvP games anyone can assemble a group and kill a griefer. Although, in an open PvP game griefers have more ability to grief others. In limited PvP games or PvE only games, there is no way to punish a kill stealer, agro puller, or loot thief.
One likely reason for not using HVAC cooled water with out warming it is condensation. Yeah, 40 degree water would really cool the proc down, but it would also cool the air surrounding the pipe and CPU block down too, which would lead to condensation inside the box. I'm not sure how conductive it would be, but it likely wouldn't be good. If instead you use water that is 60-65 degrees, 5 to 10 degrees under room temperature, you'll not have nearly as much of a problem with condensation.
Err, my bust. The article mentions "Jets" numerous times, I mistakenly thought the electric part of the title was referring to the duct (in so form of control) and the jet was referring to the power source. I see now that it article was just craply written. Thanks for the clarification.
If it was the connection between the nipple and the block then it was a manufacturing error. You could atleast get the water cooler replaced under warenty and you may be able to hold the manufacturer liable for the rest of the damages.
All you need to do for it to work on a larger scale is to increase the amount of water in the system (the amount of energy the system can contain) and improve the transfer of that energy from the system. Instead of using an air conditioner to cool hundreds of thousands of cubic feet of air to cool a server room, you can use a chiller to pull the heat out much more efficiently.
As for failure modes, water cooling isn't all cheap little toys with no engineering. With proper design and quality the critical failure rate (ie: leaks) can be reduce to an insignificant amount. We aren't talking about 2 mil cheapo hoses with a rubber band holding them together.
Is a traditional air cooled system safer? Yes. Is a model air plane engine based cooling solution more efficient or reliable? No.
But look at the energy used in the situation. A single 1/2 hp pump could push water through the entire rack and to an external radiator. You stick that radiator in the building HVAC exhaust vent (ie: passive, or parasitic) or into a highly efficient cooling system (active, phase change, peltier, etc..) and you have a pretty low power, high efficiency system. And a single water pump is a pretty reliable device. And for saftey, it wouldn't be hard to put a second redundant pump in place. I can head down to home depot and get a 1/2hp water pump and assorted plumbing items for under $100. Bulk rate on the no-break disconnects would be pretty cheap also. Toss in a hand full of flow sensors and alarms and you're ready to rock.
All of the items involved are time tested and proven to work. As compared to an array of model airplane engines which will cost $300 a piece, require specialize and expensive (compared to electricity) fueling, and are not going to hold up to extended service (24 hours a day, 365 days a year, for multiple years? I just don't see a model airplane engine holding up to that)
Air is such a poor heat transfer medium. Why not build a rack with a water cooling system built in? I have an external water cooled solution on my home PC connected via a set of no-break quick release couplings. So any time I need to pull my PC apart I can pop the coolant lines with out losing a drop of coolant or introducing air into the system.
I can't imaging running a fleet of model airplane engines is going to be quite, cheap, or all that reliable. Especially when compared to an rack integrated water cooling system.
Or maybe MMO's should move in the direction of being skill based with gear/time invested effecting the outcome of combat less. That way, casual players can still compete, and PvPer's don't need to spend 800 hours leveling, farming, and questing to be uber, they just need to practice.
In this case the user is clicking on a hyper link in the IM Conversation which uses a web browser to download an external application. If someone on your buddy list sent you this message, it would come through with no problems. You could click the link and download the file with no problems. It could even execute it's payload while you are wrapped in your GAIM blanket of security. The only thing that it MIGHT not be able to do is to propagate itself to all of the members of your friends list.
Slashdot Mirror
I have never had a problem getting an automated response for a sign up verification. I get maybe 5-10 unsolicited spams a day (all of which go directly to a junk mail folder) and 20+ solicited spams (email lists, tech groups, companies I deal with, etc...) emails a day (once again, it all goes to junk mail).
So while other user's may have problems, I guess I'm just lucky and I've never really had a problem with Hotmail. To the extent that it has been my primary email provider since '97 (pre-MS days).
-Rick
The problem is that the vast majority of petition signers are not willing to perform action in association with the petition. I wish I had the cites handy to back that statement up, but that paper was from a few years ago.
In short, 165,000 users signing a petition, but still paying their bills means nothing to AT&T.
10,000 users canceling their service and citing the PP as the primary reason will likely invoke a reaction.
You are correct, it is about critical mass. 10 dollar-voters will not make a difference, but 10,000 petition-signers will also not make a difference. 10,000 dollar-voters will likely make a difference, as would 10,000,000 petition-signers. Is it easier to come up with 10 million people to threaten to vote with their dollar, or 10 thousand people who will vote with their dollar?
-Rick
It would be kinda interesting to be able to visit someone's blog and be able to talk with other visitors and the author live time. And it could also be handy if you keep a "speed dial" list on your startup page, no need to dig out numbers, just hit the home button and click the link to your favorite relative.
-Rick
Now, this could be a heck of a performance hit, but... What if a company supplied a local VOIP call-in bank. As in POTS copper lines that you could dial into on a modem. That VOIP call could then be secured and piped to some other country. The other end of the VOIP call would then be answered by a modem bank which spits out into some foreign ISP.
It would be slow as tar, but it should get you a connection that isn't being directly reviewed by the NSA.
The other problem is that even those of us who don't sign AT&T's privacy agreement can still be monitored as packets bound for us may travel over their backbone pipes.
-Rick
The best way to force AT&T to change their game is to vote with your all-mighty dollar. A single dollar-voting customer is worth any number of petitions and angry letters.
-Rick
I bought an '06 VW Golf TDI (turbo diesel) a few months back, and so far I am averaging low 40's mileage (worst ~39.5, best ~43.5). I hit 600 miles a tank with no problem. True, it's no lean burning, single passenger, 25mph, wheel chair, but it does have 170ft/lbs of torque, and cruises beautifully at 80mph.
-Rick
"Three: It may (possibly) be a deterrent for the teams to not cop out and buy a super duper-efficient experimental engine from some no-name company and call it as their own. "
And what's wrong with that? If a team wins using some start-up company's new experimental engine, the company with the engine gets advertising and investment, and the team gets a win. Not to mention the team winners will likely have a great shot at getting a job with that company.
-Rick
Opened my mouth too soon. After RTF(/.)A I see that this article is specificly about Nanowires, not carbon nano tubes.
-Rick
The IBM chip is silicon at extremely low temperatures (~4.5K). This story is about carbon nanotubes being used in transistors. Two completely seperate and unrelated technologies.
-Rick
Would Linux users running this still need to pay for the CALs to connect to the Exchange server?
-Rick
So, proper engineering it is. First, use a non-conductive cooling fluid. That will immediately reduce the critical nature of a leak. Second, get a plastics manufacturer to make a 1/2U tray with a slant to force any spilt fluid to one specific corner. Next use a GFCI style system on the flow, measure the amount of coolant going into the line, and the amount being returned, if they don't match, use a UPS interface to send a shutdown command to the server, notify the staff, and once the machine is down close the electric valve for that server.
The non-conductive cooling fluid is a bit more expensive, but in a closed system you wouldn't need to add too much over time, and the safety it brings is well worth the cost. The 1/2U trays could be manufactured for pennies after the initial tooling. If you are HP and selling a rack integrated water cooling systems (so that all water plugs are modular and built into the rack) the extra pennies from those trays are a drop in the bucket. The GFCI and UPS interface is a bit more costly and has to work at the server level. And with the cooling fluid is less of a concern, but is would still be good to have an electric system to sound the alarm for a leak.
As for the building cooling system, it won't work. HVAC cooled water is too cold, it would cause condensation inside the server which IS conductive and would do bad things. It would be much better to have a cooling fluid-water or cooling fluid-air interchange where the water system for the servers can release heat to the HVAC system. You would have to set up a thermostat and control the water temp to make sure it is cool, but not cold enough to condense water from the air.
-Rick
"that talk about using traps but what do you do with the toads after you trap them"
mmmmm... Lunch.
-Rick
Like anyone here doesn't need long term internet service ;)
-Rick
They just upped my rates again. I pay $115 for a 3Mb connection, Digital cable, and the premium movie channels. If I drop the premium channels and down to a lower connection, my bill would actually increase to $120. As much as I hate over paying, I hate paying a company that I know is packet forwarding to the NSA and attempting to get a QoS billing scheme even more.
-Rick
The problem with self policing in most MMOs is that players have limited resources in dealing with griefers. Atleast in open PvP games anyone can assemble a group and kill a griefer. Although, in an open PvP game griefers have more ability to grief others. In limited PvP games or PvE only games, there is no way to punish a kill stealer, agro puller, or loot thief.
-Rick
One likely reason for not using HVAC cooled water with out warming it is condensation. Yeah, 40 degree water would really cool the proc down, but it would also cool the air surrounding the pipe and CPU block down too, which would lead to condensation inside the box. I'm not sure how conductive it would be, but it likely wouldn't be good. If instead you use water that is 60-65 degrees, 5 to 10 degrees under room temperature, you'll not have nearly as much of a problem with condensation.
-Rick
[sob]I never learned to read![crying]
Err, my bust. The article mentions "Jets" numerous times, I mistakenly thought the electric part of the title was referring to the duct (in so form of control) and the jet was referring to the power source. I see now that it article was just craply written. Thanks for the clarification.
-Rick
If it was the connection between the nipple and the block then it was a manufacturing error. You could atleast get the water cooler replaced under warenty and you may be able to hold the manufacturer liable for the rest of the damages.
-Rick
All you need to do for it to work on a larger scale is to increase the amount of water in the system (the amount of energy the system can contain) and improve the transfer of that energy from the system. Instead of using an air conditioner to cool hundreds of thousands of cubic feet of air to cool a server room, you can use a chiller to pull the heat out much more efficiently.
As for failure modes, water cooling isn't all cheap little toys with no engineering. With proper design and quality the critical failure rate (ie: leaks) can be reduce to an insignificant amount. We aren't talking about 2 mil cheapo hoses with a rubber band holding them together.
Is a traditional air cooled system safer? Yes. Is a model air plane engine based cooling solution more efficient or reliable? No.
-Rick
But look at the energy used in the situation. A single 1/2 hp pump could push water through the entire rack and to an external radiator. You stick that radiator in the building HVAC exhaust vent (ie: passive, or parasitic) or into a highly efficient cooling system (active, phase change, peltier, etc..) and you have a pretty low power, high efficiency system. And a single water pump is a pretty reliable device. And for saftey, it wouldn't be hard to put a second redundant pump in place. I can head down to home depot and get a 1/2hp water pump and assorted plumbing items for under $100. Bulk rate on the no-break disconnects would be pretty cheap also. Toss in a hand full of flow sensors and alarms and you're ready to rock.
All of the items involved are time tested and proven to work. As compared to an array of model airplane engines which will cost $300 a piece, require specialize and expensive (compared to electricity) fueling, and are not going to hold up to extended service (24 hours a day, 365 days a year, for multiple years? I just don't see a model airplane engine holding up to that)
-Rick
Air is such a poor heat transfer medium. Why not build a rack with a water cooling system built in? I have an external water cooled solution on my home PC connected via a set of no-break quick release couplings. So any time I need to pull my PC apart I can pop the coolant lines with out losing a drop of coolant or introducing air into the system.
I can't imaging running a fleet of model airplane engines is going to be quite, cheap, or all that reliable. Especially when compared to an rack integrated water cooling system.
-Rick
Hardly, the new Boss says the old Boss's 5 cent pay cut goes to far, while the old Boss says the new Boss's 5 cent pay cut doesn't go far enough!
-Rick
Agreed! This comparison is simple:
As of right now,
1) you can purchase and play a 360
2) you can not purchase or play a P3
-Rick
Or maybe MMO's should move in the direction of being skill based with gear/time invested effecting the outcome of combat less. That way, casual players can still compete, and PvPer's don't need to spend 800 hours leveling, farming, and questing to be uber, they just need to practice.
-Rick
In this case the user is clicking on a hyper link in the IM Conversation which uses a web browser to download an external application. If someone on your buddy list sent you this message, it would come through with no problems. You could click the link and download the file with no problems. It could even execute it's payload while you are wrapped in your GAIM blanket of security. The only thing that it MIGHT not be able to do is to propagate itself to all of the members of your friends list.
-Rick