Until we can do something about an asteroid that is going to hit us, there is really no point in spending a fortune (that can be spent on useful projects) on watching out for them.
That's chicken and egg logic. You watch to see if something is coming knowing full well that we've been hit in the past and will be in the future.
So maybe you can't do anything if you only get a month's warning like we did this time. But if you find something that just missed this time but will hit us on the next orbit, say 3 years from now, well then... you just might be able to do something about it.
I can't think of a more useful government activity than figuring out when and where the next Tunguska is going to be.
...and then what do you do with the heat the peltier adds to the mix? All a pelt buys you is a cold plate to chill your cpu with but it costs btu's to do that.
Eventually, you're going to have to dump the heat to the room - water does it better than fans + heatsinks. Quieter as well.
...that security will suffer when you make an os too easy to use. It's an age-old tradeoff: security vs. ease of use. Moreover, with more features comes more complexity and with more complexity come more security holes.
Don't want to check to see if there's a patch needed for your OS? Don't worry, we'll have the OS check for you. We can't guarantee that your computer will be talking to our servers when it downloads the patches but hey! it'll be automatic! Come to think of it, we can't even secure our own servers so we're not too sure what you'll be downloading even if you are talking to our servers but hey! - it's automatic!
I can't think of a better argument for limiting the services an os provides than this fiasco.
GE is not selling the home unit - they're just talking about it.
The GE unit is made by Plug Power and has been on GE's web site for close to a year now. Evidently, they've hit some snags. The fact Plug Power recently laid off almost 1/4 of their work force and their press releases talk more about financial than technology milestones doesn't bode well.
...till you feed the lawyers
on
This is IT?
·
· Score: 2
A chunk of the 3k is going to liability insurance.
The NY Times article references both Andy Grove and John Doerr's concerns that it will spark a feeding frenzy among the contingency fee lawyers.
You still need rockets to get off earth. What maglev buys you is the first 2~3,000 mph with no associated fuel mass. It doesn't get you into orbit or beyond.
But consider, if the first 3k is "free" in terms of mass, then it takes a much smaller rocket to get the same amount of material into orbit.
Flinging robots with maglev is easier simply because you don't need as long an acceleration ramp - they can stand much higher g-loads than we can.
Transportation won't be the limiting factor. If you're not trying to shoot humans to the moon and back, you can use maglev to launch robots to the moon and return the goods to earth. You can accelerate equipment to much higher velocities with maglev because equipment can handle much higher g-loads than we can.
You shoot maglev-ramp-building robots to the moon, they build the return ramp on the moon and you've handled the transport cost issue. The maglev ramp on the moon is used to fling the ore back at us ala Heinlein's "The Moon is a Harsh Mistress."
The biggest problem with mining the moon will be pollution. Just because the mining is happening on the moon won't mean we'll end up with no pollution here. The stuff is coming at us at a high speed and has to be decelerated without ablating into the atmosphere or cratering somewhere. If it ablates, you end up filling the atmosphere with ore dust. So somehow, the ore has to be gently brought back to earth.
Is there some sort of hook for cvs to parse the code diffs in Excel? Excel Basic is embedded in the spreadsheet in some non-textual format and it isn't clear to me how it can be handled by cvs.
Search google for "sandwell chicxulub cretaceous." It'll point you to a gravity map of the crater. I'd give you the link but slashcode keeps mangling it.
The book is a lot of fun. As usual, moderate up your favorite questions, and I'll pass on the highest scoring ones to Bruce so that he can reply to 8 or so. Bruce has a website for you to check it. It's got a FAQ and bio for you to examine. As usual, the highest moderated questions get passed on to Bruce, of which he will answer 8 or so.
One of the symptoms of alzheimers is repeating yourself. One of the symptoms of alzheimers is repeating yourself.
If you're relying on your equipment to give you reproducable results, you're going to have measure what it's actually capable of and not rely on published specifications.
I stopped relying on spec sheets when I discovered they weren't very accurate. I've seen variances as high as 50% off spec.
Re:I still wouldnt get an Athlon or any AMD chip.
on
AMD And THG update
·
· Score: 5, Insightful
Intel stuff is just better engineered. That statement is not supported by most benchmarks I've seen (excluding quake demos).
I have both Intel and AMD based machines and I'm not seeing any discrepancy in stability as measured by crash rates - Windows will as happily crash on the Pentium as it will the Athlon. As long as you cool the AMD properly and feed it clean power, it does just fine. Cheapo cooling and/or power supplies will break any cpu.
The bottom line for me is stability. The bottom line for me is stability and performance. In my business (I grade children's arithmetic tests), I need all the fpu performance the Athlon can deliver and I'm not particularly inclined to pay 3 times the price for equivalent performance. Nor am I about to rewrite my code to take advantage of proprietary sse2 instructions to get that performance either.
By compiling, I'm presuming you mean "screw j-code interpreters - we're going native mode?"
If that's what you meant, what was the payoff? How much faster did office run natively rather than as an interpreted program?
With the cost of RAM being cheaper than shipping charges (unless you buy from Crucial), compiling is one of the perfect applications for RAM disks. It's hard to beat seek time = zero and latency = zero. You don't care about power losses because the only thing in the ram disk is scratch files but boy... does it speed up compiles.
Which is exactly what Western Digital did with their 100 Gig caviar drive. They've taken advantage of cheap dram to pump their cache up to 8MB from the usual 2 MB. The result is their 7200 rpm drive is outrunning 10k rpm drives and is quieter as well.
More info on the Western Digital drive is available at storage review.
Debka.com (right side column, 1/2 way down) has been reporting for the past couple of days that China is moving troops to support the Taliban. The report goes on to mention that Taiwan is now exposed to a Chinese takeover as we reposition our carriers to attack Afghanistan.
In evaluating the news, be aware that debka is based in Israel and is about as reliable as Drudge - sometimes is, sometimes not.
General Electric has been advertising a 7KW home fuel cell for over a year now at their homegen website The unit is ostensibly being built for GE by Plug Power but apparently they've run into some difficulties. The product was supposed to be on market by this past summer - in fact New Jersey Power has been touting the fuel cell for delivery.
Unfortunately, the latest word is next summer at the earliest. Plug Power reported a $30 mil loss as of their past fiscal year and their press releases talk more about financial transactions rather than actual sales or product delivery so things aren't looking all that great for GE or Plug Power's offering right now.
What's worse for Plug Power is their initial offering doesn't take advantage of the fact that the fuel cell produces hot water as a waste product. Were they to design the unit to feed the hot water to a water heater, the fuel cell efficiency would be greater than 70%. Supposedly, the water capture feature won't appear until the second generation offering which makes you wonder who would buy the first one - especially at $15k a pop.
By coincidence, Chevron Oil in San Ramon, CA fired up their 200 KW unit today for the first time. That puppy set them back $850,000 or around $4,250 per KW. More info is available at
SF Chronicle.
Notice the odd ratios - The Chevron unit that's real and online cost about twice what GE's not-available unit is supposed to come in at. Maybe there's a hint there as to why Plug Power can't deliver.
Just as there are a myriad number of reasons why a person runs for office, a politician embodies a desire to help people. An open source programmer has the same instinct but expresses it by coding instead of politicing. Open source is a manifestation of altruism.
Cheaper and less reliable. Transrapid may not need superconductors but it needs very close tolerances on the track and the control circuitry.
As a Transrapid vehicle approaches the track due to uneven surfaces or somebody turning over in their sleeper car, the vehicle is more strongly attracted to the rail. The control circuitry has to step in quickly to prevent the train from grinding into the rail. Conversely, if the train moves away from the rail, the control circuitry has to ramp up the current to pull the train back towards the rail lest the train fall away. Heaven help you if the train loses power.
Another advantage superconducting maglev has is that it can carry freight. As the vehicles sink on the railbed due to loading, the repulsion strength automatically compensates and the levitating force increases. With a properly configured superconducting maglev train, you could compete against steel-on-steel rail. The major drawback to this technology is the huge initial expense.
Airplanes spend most of their power just pushing air out of the way - their drag rises as the cube of their airspeed. An alternative to trying to push faster through air is to build an evacuated tube between New York and Los Angeles. Put in a superconducting maglev train similar to what the Japanese have and let her rip. Since the tube's evacuated, you're not moving air out of the way so the majority of the fuel is used for acceleration and deceleration - the train coasts for most of the trip.
The maglev train's inventors have posted a proposal for a mach 3 train that would get you coast to coast in an hour and a half. Make the tube ultra straight and you can make the same trip in 45 minutes.
A Swedish engineering firm recently built the world's longest tunnel through hard rock for less than $10 million/mile. If the trans-continental tube came in at around that cost, it'd run $22 Billion. The trains themselves are estimated to cost around $5 million per car - a lot cheaper, and faster, than a $80 Million Gulfstream V.
I've quit upgrading due to noise. The fans needed to cool a 1.2 ghz Athlon are too noisy as it is. I ended up water cooling my machine, not to get it to overclock but just to get it to shut up.
Maybe when the 4 ghz chips are out, they'll have figured out how to lower the power requirement so that our computers don't sound like small jet turbines.
If your SAT scores are in the top 15%, you're innudated by colleges urging you to consider them. That happens because ETS sells the data to just about any entity that is willing to pay for it. The data I acquire from the state doesn't tell me how you personally did - the state consolidates it by school. Once they transform the raw data like that, it ceases being private property but is subject to the California public record act.
The state has a very good reason for seeing your scores. The SAT is an external measure of how well education monies are spent. There are schools in California that have average SAT scores in the low 500's as well as one public school that averages in the 1300's. Clearly, there's a huge disparity in the level of service between the affected communities.
If public education is going to reach international standards, it's going to require volumes of good data that's measuring quantitative differences between various teaching methods. Without access to those data, you're flying blind.
Until we can do something about an asteroid that is going to hit us, there is really no point in spending a fortune (that can be spent on useful projects) on watching out for them.
That's chicken and egg logic. You watch to see if something is coming knowing full well that we've been hit in the past and will be in the future.
So maybe you can't do anything if you only get a month's warning like we did this time. But if you find something that just missed this time but will hit us on the next orbit, say 3 years from now, well then... you just might be able to do something about it.
I can't think of a more useful government activity than figuring out when and where the next Tunguska is going to be.
...and then what do you do with the heat the peltier adds to the mix? All a pelt buys you is a cold plate to chill your cpu with but it costs btu's to do that.
Eventually, you're going to have to dump the heat to the room - water does it better than fans + heatsinks. Quieter as well.
That's not what my girlfriend says.
...that security will suffer when you make an os too easy to use. It's an age-old tradeoff: security vs. ease of use. Moreover, with more features comes more complexity and with more complexity come more security holes.
Don't want to check to see if there's a patch needed for your OS? Don't worry, we'll have the OS check for you. We can't guarantee that your computer will be talking to our servers when it downloads the patches but hey! it'll be automatic! Come to think of it, we can't even secure our own servers so we're not too sure what you'll be downloading even if you are talking to our servers but hey! - it's automatic!
I can't think of a better argument for limiting the services an os provides than this fiasco.
GE is not selling the home unit - they're just talking about it.
The GE unit is made by Plug Power and has been on GE's web site for close to a year now. Evidently, they've hit some snags. The fact Plug Power recently laid off almost 1/4 of their work force and their press releases talk more about financial than technology milestones doesn't bode well.
A chunk of the 3k is going to liability insurance.
The NY Times article references both Andy Grove and John Doerr's concerns that it will spark a feeding frenzy among the contingency fee lawyers.
You still need rockets to get off earth. What maglev buys you is the first 2~3,000 mph with no associated fuel mass. It doesn't get you into orbit or beyond.
But consider, if the first 3k is "free" in terms of mass, then it takes a much smaller rocket to get the same amount of material into orbit.
Flinging robots with maglev is easier simply because you don't need as long an acceleration ramp - they can stand much higher g-loads than we can.
Transportation won't be the limiting factor. If you're not trying to shoot humans to the moon and back, you can use maglev to launch robots to the moon and return the goods to earth. You can accelerate equipment to much higher velocities with maglev because equipment can handle much higher g-loads than we can.
You shoot maglev-ramp-building robots to the moon, they build the return ramp on the moon and you've handled the transport cost issue. The maglev ramp on the moon is used to fling the ore back at us ala Heinlein's "The Moon is a Harsh Mistress."
The biggest problem with mining the moon will be pollution. Just because the mining is happening on the moon won't mean we'll end up with no pollution here. The stuff is coming at us at a high speed and has to be decelerated without ablating into the atmosphere or cratering somewhere. If it ablates, you end up filling the atmosphere with ore dust. So somehow, the ore has to be gently brought back to earth.
Is there some sort of hook for cvs to parse the code diffs in Excel? Excel Basic is embedded in the spreadsheet in some non-textual format and it isn't clear to me how it can be handled by cvs.
Search google for "sandwell chicxulub cretaceous." It'll point you to a gravity map of the crater. I'd give you the link but slashcode keeps mangling it.
One of the symptoms of alzheimers is repeating yourself. One of the symptoms of alzheimers is repeating yourself.
Fractional quantum hall effect
If you're relying on your equipment to give you reproducable results, you're going to have measure what it's actually capable of and not rely on published specifications.
I stopped relying on spec sheets when I discovered they weren't very accurate. I've seen variances as high as 50% off spec.
Intel stuff is just better engineered.
That statement is not supported by most benchmarks I've seen (excluding quake demos).
I have both Intel and AMD based machines and I'm not seeing any discrepancy in stability as measured by crash rates - Windows will as happily crash on the Pentium as it will the Athlon. As long as you cool the AMD properly and feed it clean power, it does just fine. Cheapo cooling and/or power supplies will break any cpu.
The bottom line for me is stability.
The bottom line for me is stability and performance. In my business (I grade children's arithmetic tests), I need all the fpu performance the Athlon can deliver and I'm not particularly inclined to pay 3 times the price for equivalent performance. Nor am I about to rewrite my code to take advantage of proprietary sse2 instructions to get that performance either.
By compiling, I'm presuming you mean "screw j-code interpreters - we're going native mode?"
If that's what you meant, what was the payoff? How much faster did office run natively rather than as an interpreted program?
With the cost of RAM being cheaper than shipping charges (unless you buy from Crucial), compiling is one of the perfect applications for RAM disks. It's hard to beat seek time = zero and latency = zero. You don't care about power losses because the only thing in the ram disk is scratch files but boy... does it speed up compiles.
Which is exactly what Western Digital did with their 100 Gig caviar drive. They've taken advantage of cheap dram to pump their cache up to 8MB from the usual 2 MB. The result is their 7200 rpm drive is outrunning 10k rpm drives and is quieter as well.
More info on the Western Digital drive is available at storage review.
In evaluating the news, be aware that debka is based in Israel and is about as reliable as Drudge - sometimes is, sometimes not.
Unfortunately, the latest word is next summer at the earliest. Plug Power reported a $30 mil loss as of their past fiscal year and their press releases talk more about financial transactions rather than actual sales or product delivery so things aren't looking all that great for GE or Plug Power's offering right now.
What's worse for Plug Power is their initial offering doesn't take advantage of the fact that the fuel cell produces hot water as a waste product. Were they to design the unit to feed the hot water to a water heater, the fuel cell efficiency would be greater than 70%. Supposedly, the water capture feature won't appear until the second generation offering which makes you wonder who would buy the first one - especially at $15k a pop.
By coincidence, Chevron Oil in San Ramon, CA fired up their 200 KW unit today for the first time. That puppy set them back $850,000 or around $4,250 per KW. More info is available at
SF Chronicle.
Notice the odd ratios - The Chevron unit that's real and online cost about twice what GE's not-available unit is supposed to come in at. Maybe there's a hint there as to why Plug Power can't deliver.
Yo! Remember CAPITAL LETTERS? itmakesthingsmucheasiertoread,yaknow?
Just as there are a myriad number of reasons why a person runs for office, a politician embodies a desire to help people. An open source programmer has the same instinct but expresses it by coding instead of politicing. Open source is a manifestation of altruism.
Cheaper and less reliable. Transrapid may not need superconductors but it needs very close tolerances on the track and the control circuitry.
As a Transrapid vehicle approaches the track due to uneven surfaces or somebody turning over in their sleeper car, the vehicle is more strongly attracted to the rail. The control circuitry has to step in quickly to prevent the train from grinding into the rail. Conversely, if the train moves away from the rail, the control circuitry has to ramp up the current to pull the train back towards the rail lest the train fall away. Heaven help you if the train loses power.
Another advantage superconducting maglev has is that it can carry freight. As the vehicles sink on the railbed due to loading, the repulsion strength automatically compensates and the levitating force increases. With a properly configured superconducting maglev train, you could compete against steel-on-steel rail. The major drawback to this technology is the huge initial expense.
I found a single reference to power requirements rising by the cube of the speed - perhaps that's what the maglev site's author meant.
The maglev train's inventors have posted a proposal for a mach 3 train that would get you coast to coast in an hour and a half. Make the tube ultra straight and you can make the same trip in 45 minutes.
A Swedish engineering firm recently built the world's longest tunnel through hard rock for less than $10 million/mile. If the trans-continental tube came in at around that cost, it'd run $22 Billion. The trains themselves are estimated to cost around $5 million per car - a lot cheaper, and faster, than a $80 Million Gulfstream V.
Maybe when the 4 ghz chips are out, they'll have figured out how to lower the power requirement so that our computers don't sound like small jet turbines.
The state has a very good reason for seeing your scores. The SAT is an external measure of how well education monies are spent. There are schools in California that have average SAT scores in the low 500's as well as one public school that averages in the 1300's. Clearly, there's a huge disparity in the level of service between the affected communities.
If public education is going to reach international standards, it's going to require volumes of good data that's measuring quantitative differences between various teaching methods. Without access to those data, you're flying blind.