That's pretty much in line with my experience though; my grandfather died of cancer (lung cancer- he was a big smoker), one grandmother died of cancer, my hamster died of cancer, a colleague at work she died of cancer, a family friend died of cancer. My aunts dog recently died of cancer.
But I live in the UK.
I'm not saying that there aren't more cancers around Chernobyl, particularly thyroid cancers, but I am saying that there's a heck of a lot of cancer everywhere anyway.
Basically, every day, every cell in my and your body has atleast one strand of DNA break. These get patched up pretty well most of the time. But some of those breaks can eventually lead to cancer.
If they were asked a {day, week, month, year} later, what percentage would be retained?
If they knew they needed to retain it for years they can certainly do so; basically, they would have to revise it at the end of the day, a week later, and a month later; and then they would know it pretty much forever.
Is this merely a good parlor trick, or a useful mechanism for real-world use?
Somewhere between the two. They need a mnemonic scheme for each kind of thing they need to learn, but, for example, remembering the names of hundreds of people is useful.
Can their techniques be used to retain multiple unrelated data sets simultaneously?
Yes. You can link mnemonics together and hierarchically arrange the data sets that way.
Cancer is a really common disease anyway. It's the biggest cause of death, bar heart disease, in other countries. Around Chernobyl, most people will naturally blame Chernobyl for all of the cancers, when the vast majority of cancers were/are naturally induced.
The patent is on 'multiplayer' versions of solitaire- in other words the players can message each other and there would be a high score table etc.
There's even a faint chance that this is a 'valid' patent, but I wouldn't bet on it. I mean either there's some prior art (there pretty much has to be, maybe even minitel in France did something like this?), or else it may not be valid because it's too obvious; patents have to novel.
Any of the big ISPs can do this. They just tell everyone in advance- do this by date X, or we blackhole you. There's nothing that forces anyone to carry anyone else's traffic ever.
The liquid fueled rockets that nasa uses today use liquid hydrogen and liquid oxygen in the reaction:
2 H2 + O2 -> 2 H2O
Nope. That's the stochiometric ratio, nothing like that is ever used. Actually it's more like:
4 H2 + O2 -> 2H20 + 2 H2
(Actually, it's much messier than that, you really get a bunch of HO's O's H's H2O2's but that's the gist of it).
The point is rockets run very fuel rich, because that gives a much higher exhaust velocity (the hydrogen has less places to hide energy than complex molecules- you want as much energy as possible to be in kinetic form), the scramjet would do the same thing. So your fuel/oxidiser ratio is way off.
Yeah, but a scramjet can't go much beyond mach 8- it burns up. Scramjets might be able to go from mach 3 to mach 8 and gain some altitude, but beyond that you're going to need a rocket fella.
There have been some credible airbreathing designs for reaching orbit with a space plane, but *none* of them involve a scramjet.
You are both wrong. In a Liquid Hydrogen/Liquid Oxygen rocket, 8 times as much oxygen mass is needed compared to hydrogen mass.
4H + O2 => 2H2O...twice as many hydrogens as oxygen but
Oxygen is 16 times more massive.
Ahem. That's what's called the stochiometric ratio, which is NOT used for rockets. Rockets burn very fuel rich; because (to oversimplify somewhat) the light, hot, hydrogen gives better thrust than the water (less places to hide the energy in simple molecules like hydrogen- rockets want the energy to go into kinetic energy of the molecules rather than internal vibration modes of the molecules).
The upshot is that rockets typically run with about 2/3 by mass of the propellent oxygen, with the rest made up of the fuel.
IRC This translates in the case of Apollo as 6 moles of hydrogen per mole of oxygen- but 8 gives a higher exhaust velocity, but hydrogen tanks are too heavy so the *vehicle* optimises at a lower ratio.
I dont know how you define efficiency but in my aproximation having to lift 20x the payload mass because of extra fuel is an inefficency.
It's so not as simple as that though- scramjets suffer from incredible heating effects from going mach 8 in the atmosphere, they suffer enormous drag effects, they end up using more *fuel* (as opposed to propellent), and the scramjet engine is bigger and heavier than a rocket engine for the same payload- furthermore engines cost money, whereas LOX and LOX tanks don't. (Fuel is pretty cheap too in fact).
This results in hugely more complex lift vehical, which is... um... huge, and expensive.
No. Perversely perhaps, it's bigger, but cheaper; because 60% of the mass is LOX, and the rocket engines are smaller and cheaper.
That's not entirely correct. The O2 is a third of the mass.
Nope. 2/3 of the mass. And it's cheaper than the rest of the rocket put together times 10.
Keep in mind that in addition to eliminating the weight of the 02, scramjets push such an amazing amount of air out the back that they are far more efficient than rocket engines.
Less efficient of propellent; rockets are more efficient of *fuel*- scramjet's suck more fuel in because they have to overcome airdrag. Fuel is *expensive*, LOX is dirt cheap.
The main problem with space launches is the initial climb and acceleration, when you are pushing forward all of the craft's stages and fuel. By eliminating the 02, it translates into vastly, vastly smaller requirements.
Yeah, smaller and more expensive.
Scramjets are far simpler than rocket engines.
Huh????? Liquid fuelled rockets were built back in the 1930s by Goddard, nobody has built a practical scramjet yet at all. In no sense are they simpler. Rockets are very, very simple- pump propellents into a chamber pretty much anyhow, they burn, they come out through a supersonic nozzle. That's it. You might want to cool everything with water too. A 16 year old could build one. No 16 year old could build a scramjet.
It would be much cheaper to build boosters that use a scramjet as a first stage as opposed to a rocket engine. The fuel savings, the increased payload, and the cheaper cost all make the scramjet a superior option.
First, LOX has ten thousand times more oxygen than air, secondly tanks to hold it are extremely light and cheap, thirdly rockets have enormous thrust to weight ratio- over a hundred, so carrying the extra weight of the LOX is no big deal, whereas a thrust to weight ratio of 15 is more or less unheard of with airbreathing, fourth rockets go out of their way to leave the atmosphere as soon as possible- the atmosphere is nasty, draggy stuff, and when you want to reach mach 25 it's seriously in your way; whereas a scramjet uses more fuel because it has to stay in the atmosphere.
So, scramjets use more fuel, run much, much hotter (mach 8 in the atmosphere...), have much poorer thrust/weight ratio, cost more to build (aerospace materials cost thousands of dollars per kg, LOX costs pennies per kg, LOX tanks are ridiculously lightweight, rockets engines give much more thrust per kg)
Scramjets don't run at zero speed, unlike rockets, and so can't be used to land, unlike jets or rockets (DC-X), and besides who wants a three stage rocket, when normal rockets are two stage?
They're no use for passenger travel either, if Concorde travels at mach 2 is uneconomic, what's a scramjet going to cost?
I'll tell you what scramjets are for- forget space travel, they're for missiles. Why is slashdot supposed to get excited about blowing things up? (Ok, dumb question:-) )
Yeah, but where is the innovation in Microsoft? Where?
Even if Microsoft had it, and whilst certainly true innovation is often a killing business idea, intangibles such as trustworthiness, accuracy, transparency, friendliness, low cost and good quality generally are good reasons to come back to a supplier. I don't think Microsoft usually scores high on any of these- their strengths ultimately relate only to concepts like 'lock-in'.
That works only as long as the lock-in lasts- Linux and other businesses and governments appear to be attacking this reasonably strongly.
But 'google' is the BBC of the internet search engines; high quality, high trust, high transparency. Like the BBC, it's not that google always does the right thing, but when they do the wrong thing, you can often see why they did what they did, and empathise. Nobody really empathises with Microsoft.
VxWorks isn't really hard realtime in that sense anyway- it's a multitasking OS.
Paging in a segment, depends on the processor family, but usually takes rather less than 100 microseconds; which is OK for quite a lot of problems.
Realtime systems are a continum; it just depends where you are on it as to what techniques are appropriate.
In 15 years of embedded coding I've accidentally twice written memory stompers, one extremely minor, and I've written code that wasn't a memory stomper, until somebody else shrank the stack... trust me, if I can do it, anyone can. IMHO memory management is well worth using, even if it is only at the debugging stage.
And having memory protection only costs you maybe 3% of run speed,
I'd want to see real, hard numbers.
Check out any book on OS design. That's about the typical average cost. The processor's memory manager often has a limited list of memory regions, for example 8 or 16, and this is used as a cache for the active regions that are needed. Whenever the processor attempts to access memory outside those regions, an exception/interrupt is thrown, and then the cache needs to be updated, or an exception thrown to that task. That takes a while to process- the amortised cost is typically around that kind of value.
Clearly there are applications where the N% lost to protection is too much
In my experience, this is exceptionally rare. That final 3% can nearly always be clawed back with subtle changes to the software. If it really can't- then the processor was too slow for the task in hand in the first place- normal feature creep has a much bigger than 3% effect on the system performance.
There's usually no point in memory protection. If the propulsion system walks off the end of a garbage pointer, mission's over. No real use in keeping the guidance system going; it's already on a ballistic uncontrollable arc. If some critical part of the super-smart pacemaker fails (see #1), there's no victory in digging the device out of the corpse and saying, see, this other critical part wasn't affected, thanks to the memory protection! In those cases, memory protection just increases the cost and size of a device, without helping anything.
Garbage, a well designed system would reboot in the middle of thrusting, without affecting the system at all; except maybe anything that was supposed to happen during the reboot would have to wait till after the reboot.
And if a pacemaker didn't kick the heart once- the patient is dead? No. The pacemaker is there to keep the heart running at a particular (often faster) rate, not keep the patient alive, second by second.
I mean, very few computer systems are real-time critical 100% of the time.
And having memory protection only costs you maybe 3% of run speed, but on the plus side, it allow you to find bugs- really nasty bugs 'memory tramplers' for example- that can corrupt the whole system- and you never quite know what that corruption would do- it could do anything at all. Anything.
I remember hearing the very first broadcast radio play- and it was much better than the book (although the books rock in their own way). The sound effects and the music was missing in the book, strangely enough:-), and yet evoked things that the TV program was unable.
I personally don't think that even Peter Jackson could turn this into a movie, but we'll see; oh ok, perhaps he could, but nobody else could.
And what, exactly, would stop them from calculating a prime and sending the same one to all their spammees? Each recipient would check it, and guess- what- it's valid.
Back to the drawing board I think. Atleast the AOL method (probably) works... and with the AOL scheme the mail servers can cache the public keys- with your method every single email in the whole world involves checking with some central authority.
Efficiency isn't the most important metric- usually cost is.
The main potential advantage that ion drives have is that the delta-v is potentially huge (although most current thrusters often wear out fairly quickly- each giving maybe 1 tonne km/s, although you can have multiple thrusters.) In practice, the fuel they can use is often very expensive.
The energetic efficiency that you mention is related to the kinetic energy of the exhaust divided by the supplied energy- that's a different number to the efficiency I'm talking about; you multiply the mission efficiency by the thruster efficiency to get the overall mission efficiency (for a fixed ISP drive).
I don't really see the point in antipathy towards chemical propulsion; ion drives are much too slow for quite a lot of applications- the energetic efficiency ultimately limits their applicability- indeed, somewhat lower ISP, say, 1000-1600 seconds, with good thruster energetic efficiency is probably more desirable.
Interestingly, technically neither efficient, nor fast.
But they are economical of fuel. Jettisoning the exhaust at such high speed means you need hardly any fuel; which is good, but the energy source is an issue.
The reason that they are inefficient is that the exhaust velocity is too high. It turns out that the optimum exhaust velocity for minimum energy is about 2/3 the mission delta-v- and the delta-v to get to the moon is about 4.1 km/s whereas an ion drive exhaust velocity is usually around 30km/s... hugely too high from an energetic point of view.
Ok, big deal- it's only energy right? Wrong. The solar panels end up pretty enormous, and pretty heavy, pretty quickly. Nuclear energy? Power/weight ratio is little better.
Still, it works, but it's not even as efficient as chemical rocketry; chemical rockets can hit 80+% energetic efficiency in fact (it's very high because of the high temperatures used in the combustion chamber, rocket engines are actually classed as heat engines).
Warning people ("potential terrorists") that these kinds of things are illegal under the current anti-terrorism laws will get you locked up in Guantanimo ba...
Yes, of course. If you burn every single molecule then it will no longer be infectious; it's not a homeopathic disease:-).
But experimentally, if you build a little bonfire, burn an infected animal, take the ash and inject it into mouse brains- they get BSE. The protein is incredibly hardy.
Slashdot Mirror
But I live in the UK.
I'm not saying that there aren't more cancers around Chernobyl, particularly thyroid cancers, but I am saying that there's a heck of a lot of cancer everywhere anyway.
Basically, every day, every cell in my and your body has atleast one strand of DNA break. These get patched up pretty well most of the time. But some of those breaks can eventually lead to cancer.
If they knew they needed to retain it for years they can certainly do so; basically, they would have to revise it at the end of the day, a week later, and a month later; and then they would know it pretty much forever.
Is this merely a good parlor trick, or a useful mechanism for real-world use?
Somewhere between the two. They need a mnemonic scheme for each kind of thing they need to learn, but, for example, remembering the names of hundreds of people is useful.
Can their techniques be used to retain multiple unrelated data sets simultaneously?
Yes. You can link mnemonics together and hierarchically arrange the data sets that way.
Cancer is a really common disease anyway. It's the biggest cause of death, bar heart disease, in other countries. Around Chernobyl, most people will naturally blame Chernobyl for all of the cancers, when the vast majority of cancers were/are naturally induced.
'Spirit is willing but the flash is weak' over and over again...
The patent is on 'multiplayer' versions of solitaire- in other words the players can message each other and there would be a high score table etc.
There's even a faint chance that this is a 'valid' patent, but I wouldn't bet on it. I mean either there's some prior art (there pretty much has to be, maybe even minitel in France did something like this?), or else it may not be valid because it's too obvious; patents have to novel.
Any of the big ISPs can do this. They just tell everyone in advance- do this by date X, or we blackhole you. There's nothing that forces anyone to carry anyone else's traffic ever.
Nope. That's the stochiometric ratio, nothing like that is ever used. Actually it's more like:
4 H2 + O2 -> 2H20 + 2 H2
(Actually, it's much messier than that, you really get a bunch of HO's O's H's H2O2's but that's the gist of it).
The point is rockets run very fuel rich, because that gives a much higher exhaust velocity (the hydrogen has less places to hide energy than complex molecules- you want as much energy as possible to be in kinetic form), the scramjet would do the same thing. So your fuel/oxidiser ratio is way off.
There have been some credible airbreathing designs for reaching orbit with a space plane, but *none* of them involve a scramjet.
Ahem. That's what's called the stochiometric ratio, which is NOT used for rockets. Rockets burn very fuel rich; because (to oversimplify somewhat) the light, hot, hydrogen gives better thrust than the water (less places to hide the energy in simple molecules like hydrogen- rockets want the energy to go into kinetic energy of the molecules rather than internal vibration modes of the molecules).
The upshot is that rockets typically run with about 2/3 by mass of the propellent oxygen, with the rest made up of the fuel.
IRC This translates in the case of Apollo as 6 moles of hydrogen per mole of oxygen- but 8 gives a higher exhaust velocity, but hydrogen tanks are too heavy so the *vehicle* optimises at a lower ratio.
I dont know how you define efficiency but in my aproximation having to lift 20x the payload mass because of extra fuel is an inefficency.
It's so not as simple as that though- scramjets suffer from incredible heating effects from going mach 8 in the atmosphere, they suffer enormous drag effects, they end up using more *fuel* (as opposed to propellent), and the scramjet engine is bigger and heavier than a rocket engine for the same payload- furthermore engines cost money, whereas LOX and LOX tanks don't. (Fuel is pretty cheap too in fact).
This results in hugely more complex lift vehical, which is... um... huge, and expensive.
No. Perversely perhaps, it's bigger, but cheaper; because 60% of the mass is LOX, and the rocket engines are smaller and cheaper.
Nope. 2/3 of the mass. And it's cheaper than the rest of the rocket put together times 10.
Keep in mind that in addition to eliminating the weight of the 02, scramjets push such an amazing amount of air out the back that they are far more efficient than rocket engines.
Less efficient of propellent; rockets are more efficient of *fuel*- scramjet's suck more fuel in because they have to overcome airdrag. Fuel is *expensive*, LOX is dirt cheap.
The main problem with space launches is the initial climb and acceleration, when you are pushing forward all of the craft's stages and fuel. By eliminating the 02, it translates into vastly, vastly smaller requirements.
Yeah, smaller and more expensive.
Scramjets are far simpler than rocket engines.
Huh????? Liquid fuelled rockets were built back in the 1930s by Goddard, nobody has built a practical scramjet yet at all. In no sense are they simpler. Rockets are very, very simple- pump propellents into a chamber pretty much anyhow, they burn, they come out through a supersonic nozzle. That's it. You might want to cool everything with water too. A 16 year old could build one. No 16 year old could build a scramjet.
It would be much cheaper to build boosters that use a scramjet as a first stage as opposed to a rocket engine. The fuel savings, the increased payload, and the cheaper cost all make the scramjet a superior option.
First, LOX has ten thousand times more oxygen than air, secondly tanks to hold it are extremely light and cheap, thirdly rockets have enormous thrust to weight ratio- over a hundred, so carrying the extra weight of the LOX is no big deal, whereas a thrust to weight ratio of 15 is more or less unheard of with airbreathing, fourth rockets go out of their way to leave the atmosphere as soon as possible- the atmosphere is nasty, draggy stuff, and when you want to reach mach 25 it's seriously in your way; whereas a scramjet uses more fuel because it has to stay in the atmosphere.
So, scramjets use more fuel, run much, much hotter (mach 8 in the atmosphere...), have much poorer thrust/weight ratio, cost more to build (aerospace materials cost thousands of dollars per kg, LOX costs pennies per kg, LOX tanks are ridiculously lightweight, rockets engines give much more thrust per kg)
Scramjets don't run at zero speed, unlike rockets, and so can't be used to land, unlike jets or rockets (DC-X), and besides who wants a three stage rocket, when normal rockets are two stage?
They're no use for passenger travel either, if Concorde travels at mach 2 is uneconomic, what's a scramjet going to cost?
I'll tell you what scramjets are for- forget space travel, they're for missiles. Why is slashdot supposed to get excited about blowing things up? (Ok, dumb question :-) )
Even if Microsoft had it, and whilst certainly true innovation is often a killing business idea, intangibles such as trustworthiness, accuracy, transparency, friendliness, low cost and good quality generally are good reasons to come back to a supplier. I don't think Microsoft usually scores high on any of these- their strengths ultimately relate only to concepts like 'lock-in'.
That works only as long as the lock-in lasts- Linux and other businesses and governments appear to be attacking this reasonably strongly.
But 'google' is the BBC of the internet search engines; high quality, high trust, high transparency. Like the BBC, it's not that google always does the right thing, but when they do the wrong thing, you can often see why they did what they did, and empathise. Nobody really empathises with Microsoft.
Paging in a segment, depends on the processor family, but usually takes rather less than 100 microseconds; which is OK for quite a lot of problems.
Realtime systems are a continum; it just depends where you are on it as to what techniques are appropriate.
In 15 years of embedded coding I've accidentally twice written memory stompers, one extremely minor, and I've written code that wasn't a memory stomper, until somebody else shrank the stack... trust me, if I can do it, anyone can. IMHO memory management is well worth using, even if it is only at the debugging stage.
And these super duper blind phreakers are so good they never got cau... um. I'm sure I've got a point here somewhere.
Ok, perhaps not :-)
Why is this news exactly?
I'd want to see real, hard numbers.
Check out any book on OS design. That's about the typical average cost. The processor's memory manager often has a limited list of memory regions, for example 8 or 16, and this is used as a cache for the active regions that are needed. Whenever the processor attempts to access memory outside those regions, an exception/interrupt is thrown, and then the cache needs to be updated, or an exception thrown to that task. That takes a while to process- the amortised cost is typically around that kind of value.
Clearly there are applications where the N% lost to protection is too much
In my experience, this is exceptionally rare. That final 3% can nearly always be clawed back with subtle changes to the software. If it really can't- then the processor was too slow for the task in hand in the first place- normal feature creep has a much bigger than 3% effect on the system performance.
Garbage, a well designed system would reboot in the middle of thrusting, without affecting the system at all; except maybe anything that was supposed to happen during the reboot would have to wait till after the reboot.
And if a pacemaker didn't kick the heart once- the patient is dead? No. The pacemaker is there to keep the heart running at a particular (often faster) rate, not keep the patient alive, second by second.
I mean, very few computer systems are real-time critical 100% of the time.
And having memory protection only costs you maybe 3% of run speed, but on the plus side, it allow you to find bugs- really nasty bugs 'memory tramplers' for example- that can corrupt the whole system- and you never quite know what that corruption would do- it could do anything at all. Anything.
I remember hearing the very first broadcast radio play- and it was much better than the book (although the books rock in their own way). The sound effects and the music was missing in the book, strangely enough :-), and yet evoked things that the TV program was unable.
I personally don't think that even Peter Jackson could turn this into a movie, but we'll see; oh ok, perhaps he could, but nobody else could.
Back to the drawing board I think. Atleast the AOL method (probably) works... and with the AOL scheme the mail servers can cache the public keys- with your method every single email in the whole world involves checking with some central authority.
The human gut doesn't break down cellulose. That's what cooking is for; it punches holes in the cell wall.
Idiot. Coward. Jackass.
Nice sig :-)
The main potential advantage that ion drives have is that the delta-v is potentially huge (although most current thrusters often wear out fairly quickly- each giving maybe 1 tonne km/s, although you can have multiple thrusters.) In practice, the fuel they can use is often very expensive.
The energetic efficiency that you mention is related to the kinetic energy of the exhaust divided by the supplied energy- that's a different number to the efficiency I'm talking about; you multiply the mission efficiency by the thruster efficiency to get the overall mission efficiency (for a fixed ISP drive).
I don't really see the point in antipathy towards chemical propulsion; ion drives are much too slow for quite a lot of applications- the energetic efficiency ultimately limits their applicability- indeed, somewhat lower ISP, say, 1000-1600 seconds, with good thruster energetic efficiency is probably more desirable.
The good news is that fusion is only 30 years away, unlike 50 years ago, when it was 50 years away.
But they are economical of fuel. Jettisoning the exhaust at such high speed means you need hardly any fuel; which is good, but the energy source is an issue.
The reason that they are inefficient is that the exhaust velocity is too high. It turns out that the optimum exhaust velocity for minimum energy is about 2/3 the mission delta-v- and the delta-v to get to the moon is about 4.1 km/s whereas an ion drive exhaust velocity is usually around 30km/s... hugely too high from an energetic point of view.
Ok, big deal- it's only energy right? Wrong. The solar panels end up pretty enormous, and pretty heavy, pretty quickly. Nuclear energy? Power/weight ratio is little better.
Still, it works, but it's not even as efficient as chemical rocketry; chemical rockets can hit 80+% energetic efficiency in fact (it's very high because of the high temperatures used in the combustion chamber, rocket engines are actually classed as heat engines).
Yeah, but it hasn't been around longer than hot grits :)
Warning people ("potential terrorists") that these kinds of things are illegal under the current anti-terrorism laws will get you locked up in Guantanimo ba...
But experimentally, if you build a little bonfire, burn an infected animal, take the ash and inject it into mouse brains- they get BSE. The protein is incredibly hardy.
If the protocol is written sensibly, you only need to have one signature for all those 30k domains, although you may choose to have several.