And nuclear material is the worst "in-space propulsion" payload to launch from the Earth.
An rtg is the only device to ever survive a launch failure (catastrophic explosion of the lifing rocket), then be recovered, and launched on another mission.
Fear mongers will scream forever about 'nuclear devices' when you mention an rtg, but, history shows, it's one of the SAFEST things to launch, and likely the only thing that'll ever be launched that's capable of surviving a launch accident. this is an especially important detail regarding the shuttle, they do tend to come apart every now and then.
hell, even putting someone is orbit is their sole domain and will be for a long time to come.
In case you haven't noticed, Nasa actually outsources this to Russia. Nasa is incapable of manned flight today.
The fabled 'return to flight' is only a few months away. My money says, he's leaving because the writing is on the wall, and he does NOT want to head the press briefing that announces to the american public, that the shuttle will not fly again. That press conference is only a few months out now.
Years ago, I worked for a megacorp. For the better part of 2 years, that's all I did, was search for prior art to be used invalidating patents held by competitors. Never ceased to amaze me how many patents could be invalidate in less than a day of searches thru journals, and we didn't have neatly cataloged and indexed electronic versions of them, we did it by just going to the paper copies.
Today you can just 'ask the oracle' for all matters related to published information (type it into a google search for those that dont 'get it'). Google knows all, and i'm quite sure that somebody well skilled in using it, could invalidate 80% of patents issued today, in less than an hour, by finding online references to prior art, and validating that reference in the published paper versions of the said journals. It almost makes you wonder, why the USPTO doesn't do exactly that, and pre-filter applications like that. It's kind of trivial.
If you dont want to do the job right, then dont do it. You are free to go find another job. If you are advancing kids because you are scared of the parents, then you ARE the problem. Like it or not, that's the harsh reality, and coddling you for that, is even worse than coddling the kids that are doing bad.
A computer in the workplace infested with spyware can be cleaned, and fixed. An employee in the workplace infested with an ipod will invariably spend more time fussing over thier music than doing thier job. That too is easily fixed, but it normally requires replacement of the person, those that are infested with ipods tend to be unfixable.
If academics were the entrance standard/filter for yale, it would probably have a lot to do with it. But, entrance filters to yale are based on economics, ability to pay, and social circles. Therefore the only correlation between 'smart' and 'yale alumnus' would be due to random chance, and there's (my opinion) a much higher likelihood of finding an inverse correlation in reality.
Actually, I like this technological concept, it will allow me to mark all the free ipod pyramid trolls as 'troll', without even having to lift a mouse. Until that time, sure would be nice to see posts trolling the ipod pyramid schemes get a -1 troll automatically.
This is not news, the technology for this conversion has been around for decades, is widely distributed, and in wide use. It's called a 'grocery store'.
And how is this any different than the american google, which happily excludes all the news sources in the middle east, because the american government wants to keep those publications out of view of americans ?
It's the American People, and our reaction to losing people
That's completely wrong in reality it's all about the spin they fabricate. If losing people was really the issue, then look at the cost benefit equation. A defined risk, 2% probablity of losing the staff, with a 1% probability of not achieving mission (1 launch, one re-entry failure, on just over 100 missions). 4 people required for an hst service mission. 0.02*4 = 0.08 lives the real cost of a hubble service mission in lives.
The american people have demonstrated in Iraq, they are quite happy to kill 3 americans, and an untold number of non americans daily. The HST service mission has a real cost of 0.08 lives, and a potential cost of 4 lives. Americans have no problem watching 4 'heros' get killed, they do it every day, and the majority want to see more of it.
This is all about managing the spin in the press. If they can keep the 'astronaut lives' in the spin, then the real reasons for not going to hubble can be quietly swept under the rug, and the public never knows the real reasons.
The Nasa beurocracy has found a 'perfect' happy equlibrium. Outsource actual launches so there is no real risk. Continue to draw the huge budget on the grounds of 'return to flight'. Billion dollar budgets to do nothing but write reports, a beaurocrats dream come true.
The shuttle will never fly again. There will always be 'one more report showing yet another problem' to prevent that. In the meantime, the budget that used to get spent funding actual launches, is now burned up doing reports justifying why they dont actually launch anything.
Flight risk is an engineering problem, something the management is scared of, and powerless to actually manage. The risk of losing budget because all they do is write reports these days, is manageable beurocratic problem. It's a beaurocrats dream come true, billions of dollars, and only reports required, actual launch hardware can remain mothballed, and all they really need, is more reports to justify keeping it mothballed, yet maintain a 'hope' of return to flight eventually.
Thinking about this some more,/. already has a system where good karma buys you an extra mod point at time of post. A useful addition to that, sig lines that troll for pyramid schemes like freeipod should automagically buy you a -1 troll immediately at posting.
You make the assumption that Caller ID can be relied on. Totally invalid assumption, but, not surprising coming from somebody that's trolling/. for free ipod pyramid scams. It's a strong indication of intelligence (or lack thereof), and just how much weight to put on an opinion.
This is a typical progression with a north american style government. During stage 1, they convince everybody to register with the 'do not call' list, on the guise it will stop telemarketers from actually calling. In phase 2 of the plan, now they have this list, they are going to rent it out to telemarketers, and create a loophole in the regulations so that folks they rent the list to, are excluded from the 'do not call'.
I wonder what ever happened to the concept of integrity and principles in government ?
Hardware guys are already running at 1-3% profit per machine,
This is what happens in a free market, with enough competitors properly funded to actually work in the market. The production quantities over it's lifetime on a pc chipset are HUGE, typical orders running in the 100K units range, and those orders will repeat. Early in thier run, new chips demand a premium, companies can quickly amortize the development costs off against the early run on a given chip. Once that's done, they can supply into the market with very slim margins, and still be profitable because the numbers get staggeringly huge. Since there are multiple vendors of a 'motherboard chipset', a large volume motherboard producer can, and will, have a bidding war between them to determine which chipset is used on a given motherboard design. You can bet your last dollar, Via, Intel, Nvidia and the rest will all get out the pencil sharpener when Asus comes looking for a quote. A design win with Asus will justify the entire chipset line.
Back in the 80's, there was only one supplier of x86 processors into the pc marketplace. Processors were EXPENSIVE by today's standards. The mac with it's 68K processor was threatening to become a serious player, then the clones in x86 space started to show up. pricing in the x86 market started to reflect cost of production rather than 'what the market will bear'. The rest is history, and now we have single chip integrated systems, because they are ultimately cheaper to produce in quantity, even tho the engineering costs up front are staggering.
On the software side of the equation, this type of competition has not happened, mostly because parts are not interchangeable. You can swap an ati video out of a machine and toss in an Nvidia. You cannot swap a windows program out, and swap in it's linux equivalent. Competetive pricing on software, based on 'cost of manufacture' rather than 'what the market will bear' will only happen when the predominant components can be interchanged. In the hardware world this is done by using common signalling on standard bus. In software, it's only going to happen if there is common and interchangeable api systems, and the common api ends up with the lions share of the market. From a software vendor point of view today, the only common api to work with is the Win32 api.
It's in the hardware vendors best interest to undermine the cost of software. they have known this for years, but they are currently hostage to the Win32 api to sell hardware. Software vendors are in the same boat. The real solution to a free market, is to have an alternative vendor from which to purchase the Win32 api for deployment on new machines, or for both hardware and software vendors to settle on an alternative api. Neither of these are going to happen in the short term. If the Win32 api remains single vendor, single source, it's inevitable the market will migrate to an alternative, but, it may take 2 more generations (people generations, not those 18 month hardware generations). Any time you do a product in the design phase, a major consideration is the risk attached to single source components. In today's market, the risk/reward equations favour the single vendor Win32 solution. Eventually, the market will abandon the single source solution, but, that wont happen till the risk/reward equations come up in favour of the alternative.
Typical flash today is good for a million writes per cell. If you use a file system that's not doing wear levelling, and write on average once a minute, you will see the cells start breaking down in 2 years. But, include wear levelling into the flash (most off the shelf flash drives today actually include it at the hardware level), you can increase that lifetime by a factor of 100, so useful life heads up to around 200 years. if you assume it's turned off for even 8 hours a day, it goes up to on the order of 300 years (based on rewriting the same data once a minute for all the time it's turned on).
Using flash drives is only a problem if you build it without enough ram, and do something stupid like put swap on the flash drive. If you build a system that's not thrashing the swap, and use modern wear levelled flash, the unit will likely outlive the owner (even a typical/. first year college kid) before the flash starts to die from wear.
While it's true, flash does have write limits, they are vastly overrated today. if you are going to compare flash to spinning media, then factor things like bearings into the equation, and write frequency, and possibly even power consumption. Flash with wear levelling, after you factor in bearing failures on traditional spinning media, is actually more reliable than a hard drive. If you are truely paranoid, use a reed-solomon based write methodology so you can recover data after a cell failure from writes, and you are looking at a system with _at least_ an order of magnitude higher reliability ratings (mtbf) than one with spinning media, and that's even before you factor in some 'harsh environment' details, like 'ooops, it got dropped' etc. It doesn't matter what kind of error handling/correction you apply to the spinning media, bearings and motors will give it a useful lifetime that's not in any way tied to read/write cycles, but rather to calendar time and physical handling.
note, i'm comparing reliability here, not cost per bit of storage. Spinning media is still a couple orders of magnitude cheaper for large storage quantities, but that's changing rather rapidly these days too.
I've got a unit on my desk here, with a 266 mhz processor, and 1 gig of flash. After bringing up X, i've still got on the order of 600 meg of free flash on it, with a basic set of gui apps isntalled and running. This box is all solid state, no fans, runs on a 19v laptop supply. It's actually quite amazing what can be done with this box if you aren't concerned about stupid games, and just want a basic productivity platform (email, word processing, etc).
Slashdot Mirror
Never underestimate the capacity of a truckload of DVDs. Latency sucks, but thruput is tremendous.
An rtg is the only device to ever survive a launch failure (catastrophic explosion of the lifing rocket), then be recovered, and launched on another mission.
Fear mongers will scream forever about 'nuclear devices' when you mention an rtg, but, history shows, it's one of the SAFEST things to launch, and likely the only thing that'll ever be launched that's capable of surviving a launch accident. this is an especially important detail regarding the shuttle, they do tend to come apart every now and then.
In case you haven't noticed, Nasa actually outsources this to Russia. Nasa is incapable of manned flight today.
The fabled 'return to flight' is only a few months away. My money says, he's leaving because the writing is on the wall, and he does NOT want to head the press briefing that announces to the american public, that the shuttle will not fly again. That press conference is only a few months out now.
they have a school ??? I thought they just grabbed all the dropouts from real schools...
That would require intelligence on the part of the poster, a concept mutually exclusive with /.
Today you can just 'ask the oracle' for all matters related to published information (type it into a google search for those that dont 'get it'). Google knows all, and i'm quite sure that somebody well skilled in using it, could invalidate 80% of patents issued today, in less than an hour, by finding online references to prior art, and validating that reference in the published paper versions of the said journals. It almost makes you wonder, why the USPTO doesn't do exactly that, and pre-filter applications like that. It's kind of trivial.
If you dont want to do the job right, then dont do it. You are free to go find another job. If you are advancing kids because you are scared of the parents, then you ARE the problem. Like it or not, that's the harsh reality, and coddling you for that, is even worse than coddling the kids that are doing bad.
Results from the reading and comprehension tests are due next week...
Ipod makes the person slow and unreliable.
A computer in the workplace infested with spyware can be cleaned, and fixed. An employee in the workplace infested with an ipod will invariably spend more time fussing over thier music than doing thier job. That too is easily fixed, but it normally requires replacement of the person, those that are infested with ipods tend to be unfixable.
If academics were the entrance standard/filter for yale, it would probably have a lot to do with it. But, entrance filters to yale are based on economics, ability to pay, and social circles. Therefore the only correlation between 'smart' and 'yale alumnus' would be due to random chance, and there's (my opinion) a much higher likelihood of finding an inverse correlation in reality.
I'm torn between what's worse, spyware, or ponzi schemes trolling for free ipods...
Actually, I like this technological concept, it will allow me to mark all the free ipod pyramid trolls as 'troll', without even having to lift a mouse. Until that time, sure would be nice to see posts trolling the ipod pyramid schemes get a -1 troll automatically.
And this is different from mainstream press in what way ?
This is not news, the technology for this conversion has been around for decades, is widely distributed, and in wide use. It's called a 'grocery store'.
If the shuttle is the safest spacecraft for the job, why is Nasa using Soyuz equipment for thier manned launches ?
And how is this any different than the american google, which happily excludes all the news sources in the middle east, because the american government wants to keep those publications out of view of americans ?
You are confusing a college diploma/degree with intelligence.
That's completely wrong in reality it's all about the spin they fabricate. If losing people was really the issue, then look at the cost benefit equation. A defined risk, 2% probablity of losing the staff, with a 1% probability of not achieving mission (1 launch, one re-entry failure, on just over 100 missions). 4 people required for an hst service mission. 0.02*4 = 0.08 lives the real cost of a hubble service mission in lives.
The american people have demonstrated in Iraq, they are quite happy to kill 3 americans, and an untold number of non americans daily. The HST service mission has a real cost of 0.08 lives, and a potential cost of 4 lives. Americans have no problem watching 4 'heros' get killed, they do it every day, and the majority want to see more of it.
This is all about managing the spin in the press. If they can keep the 'astronaut lives' in the spin, then the real reasons for not going to hubble can be quietly swept under the rug, and the public never knows the real reasons.
The shuttle will never fly again. There will always be 'one more report showing yet another problem' to prevent that. In the meantime, the budget that used to get spent funding actual launches, is now burned up doing reports justifying why they dont actually launch anything.
Flight risk is an engineering problem, something the management is scared of, and powerless to actually manage. The risk of losing budget because all they do is write reports these days, is manageable beurocratic problem. It's a beaurocrats dream come true, billions of dollars, and only reports required, actual launch hardware can remain mothballed, and all they really need, is more reports to justify keeping it mothballed, yet maintain a 'hope' of return to flight eventually.
That doesn't surprise me, after all, lawyers make very shitty people, so its not surprising that decent folks make really shitty lawyers.
Thinking about this some more, /. already has a system where good karma buys you an extra mod point at time of post. A useful addition to that, sig lines that troll for pyramid schemes like freeipod should automagically buy you a -1 troll immediately at posting.
You make the assumption that Caller ID can be relied on. Totally invalid assumption, but, not surprising coming from somebody that's trolling /. for free ipod pyramid scams. It's a strong indication of intelligence (or lack thereof), and just how much weight to put on an opinion.
I wonder what ever happened to the concept of integrity and principles in government ?
This is what happens in a free market, with enough competitors properly funded to actually work in the market. The production quantities over it's lifetime on a pc chipset are HUGE, typical orders running in the 100K units range, and those orders will repeat. Early in thier run, new chips demand a premium, companies can quickly amortize the development costs off against the early run on a given chip. Once that's done, they can supply into the market with very slim margins, and still be profitable because the numbers get staggeringly huge. Since there are multiple vendors of a 'motherboard chipset', a large volume motherboard producer can, and will, have a bidding war between them to determine which chipset is used on a given motherboard design. You can bet your last dollar, Via, Intel, Nvidia and the rest will all get out the pencil sharpener when Asus comes looking for a quote. A design win with Asus will justify the entire chipset line.
Back in the 80's, there was only one supplier of x86 processors into the pc marketplace. Processors were EXPENSIVE by today's standards. The mac with it's 68K processor was threatening to become a serious player, then the clones in x86 space started to show up. pricing in the x86 market started to reflect cost of production rather than 'what the market will bear'. The rest is history, and now we have single chip integrated systems, because they are ultimately cheaper to produce in quantity, even tho the engineering costs up front are staggering.
On the software side of the equation, this type of competition has not happened, mostly because parts are not interchangeable. You can swap an ati video out of a machine and toss in an Nvidia. You cannot swap a windows program out, and swap in it's linux equivalent. Competetive pricing on software, based on 'cost of manufacture' rather than 'what the market will bear' will only happen when the predominant components can be interchanged. In the hardware world this is done by using common signalling on standard bus. In software, it's only going to happen if there is common and interchangeable api systems, and the common api ends up with the lions share of the market. From a software vendor point of view today, the only common api to work with is the Win32 api.
It's in the hardware vendors best interest to undermine the cost of software. they have known this for years, but they are currently hostage to the Win32 api to sell hardware. Software vendors are in the same boat. The real solution to a free market, is to have an alternative vendor from which to purchase the Win32 api for deployment on new machines, or for both hardware and software vendors to settle on an alternative api. Neither of these are going to happen in the short term. If the Win32 api remains single vendor, single source, it's inevitable the market will migrate to an alternative, but, it may take 2 more generations (people generations, not those 18 month hardware generations). Any time you do a product in the design phase, a major consideration is the risk attached to single source components. In today's market, the risk/reward equations favour the single vendor Win32 solution. Eventually, the market will abandon the single source solution, but, that wont happen till the risk/reward equations come up in favour of the alternative.
Using flash drives is only a problem if you build it without enough ram, and do something stupid like put swap on the flash drive. If you build a system that's not thrashing the swap, and use modern wear levelled flash, the unit will likely outlive the owner (even a typical /. first year college kid) before the flash starts to die from wear.
While it's true, flash does have write limits, they are vastly overrated today. if you are going to compare flash to spinning media, then factor things like bearings into the equation, and write frequency, and possibly even power consumption. Flash with wear levelling, after you factor in bearing failures on traditional spinning media, is actually more reliable than a hard drive. If you are truely paranoid, use a reed-solomon based write methodology so you can recover data after a cell failure from writes, and you are looking at a system with _at least_ an order of magnitude higher reliability ratings (mtbf) than one with spinning media, and that's even before you factor in some 'harsh environment' details, like 'ooops, it got dropped' etc. It doesn't matter what kind of error handling/correction you apply to the spinning media, bearings and motors will give it a useful lifetime that's not in any way tied to read/write cycles, but rather to calendar time and physical handling.
note, i'm comparing reliability here, not cost per bit of storage. Spinning media is still a couple orders of magnitude cheaper for large storage quantities, but that's changing rather rapidly these days too.
I've got a unit on my desk here, with a 266 mhz processor, and 1 gig of flash. After bringing up X, i've still got on the order of 600 meg of free flash on it, with a basic set of gui apps isntalled and running. This box is all solid state, no fans, runs on a 19v laptop supply. It's actually quite amazing what can be done with this box if you aren't concerned about stupid games, and just want a basic productivity platform (email, word processing, etc).