Any sufficiently long wire is an antenna, and the leakage from unshielded twisted pair is nontrivial. The carrier that fast ethernet data is modulated onto (125MHz) is of a short enough wavelength in comparison to the wire that it really leaks pretty well.
If you have a aviation band radio, tune it to 125MHz, and listen to the nice sound you get for each transmitted packet on a nearby LAN. I just did the test and heard plenty of noise on mine. Retrieving the bits would probably take a small amount of signal processing, but it's really a pretty conventional reception problem.
If you still believe that bits aren't leaking, I don't know what to tell you.
Another tidbit: switching provides very little security in practice, particularly if you can inject packets. You can man in the middle with ARP requests; you can poison CAM tables. It's not possible to protect against all of these attacks with switch security commands today on any platform I know of.
Cable upstream speeds can be really variable. I know someone on Cox cable and during the day he gets about 4-5 kilobytes per second up over TCP (this is probably about 35-45kbps of raw connectivity). At night he gets as much as 50-60kbytes up.
When I was on SBC/PacBell DSL 384k with a 128kbps upstream, I'd usually only see 10kbytes/sec up. This is nowhere near enough for a 128kbps+ MP3.
The Apple Airport base stations seem to just work, unlike my experience with Dlink and Linksys products.
A comparable box is the Linksys WRT54G; looking through reviews, people have had lots of problems. You can get it for $120-130 compared to $190-$200 for the Airport; and the Linksys doesn't have a USB port for printer sharing.
A $60-$80 premium to have a solution which is dependable is well worth it for me.
This is true only if you get the administrator's account password so that you can run through sudo, or if you can wait for the user to do a privileged operation and hijack it. As long as Joe User doesn't escalate to administrator privileges, it can only run with Joe User's rights. This is a win; I basically only escalate privilege these days for the occasional software installation.
It's a very difficult signal processing problem. When you have a series of snapshots of the positions of "blips" taken 3 seconds apart, how do you find real, faint blips? Figuring out which dots to connect becomes hard, because there's false echoes all over the place in each observation.
There's been a lot of work in this area; probably the most fruitful has been to try and provide support to humans engaging in this type of vigilance task. One system I know of keeps snapshots of radar data and then replays it at higher speed than the observation interval. The human visual system is much better at picking out that kind of blip because it looks more like we're accustomed to having objects move in the real world (smoothly, instead of jumping every 3 seconds).
I'm addressing what you said. All that garbage you said about ballistic trajectory to get attention, was.. garbage.
10kg of explosives is comparable to what a suicide bomber carries, and the pressure wave could be more optimally located by blowing at a slight altitude. This could be pretty effective against a football game and result in hundreds of casualties and tens of fatalities, without having to pick the location that you hit too closely.
On the other hand, I'm not altogether convinced 10kg of explosives would do a whole lot to an airliner on the ground unless you aimed really, really well. Let's think a second-- 380 MPH, let's be really, really generous and say 3 GPS updates a second.. between each update it moves 186 feet! The wingspan of a 747-400 is only 225 feet, with an average width of something like 26 feet... not to mention it'd be hard to get exact coordinates of the 747 in the first place.
If you know as much as you say you do, surely you can compute the pressure wave that 10kg of explosives is going to generate, and compare it to the 670PSI of loading that the wing structure is designed to sustain. The inverse square law is gonna say you're going to have to be really, really close to compromise the wing structure. Probably closer than the RMS error of civilian GPS, not even counting the intricacies involved in getting a missile to express its energy at the right point. I suspect breaching the fuselage is even more difficult. This is completely ignoring the fact that the tanks themselves are reinforced and can take substantial loads. Do the math; the numbers are readily available and it takes less than 5 minutes.
The jet fuel doesn't make that much of a difference because jet fuel doesn't really burn all that well if not atomized first. There aren't many fatal accidents from fires during taxiing or fueling operations, and there have been quite a few fires in those situations. The kinetic energy of a fast-moving airplane is far more effective at atomizing fuel than 10kg of explosives is going to be, and this is why landing accidents and the WTC form such spectacular fireballs.
As an engineer, and a pilot... I can say that if I wanted to to take out an important aircraft.. this is about the last thing in the world I'd try.
I'm convinced you don't know very much about aviation.
The kind of thing we're talking about doesn't have the kind of range necessary to make it an international nuisance. And the types of air defense interrogation zones that have been established for national defense against conventional fighter aircraft provide pretty good defensive measures for countries with capable air forces against internationally launched cruise missiles.
Ballistic missile detection systems typically detect objects going to very high altitudes (>20 miles), or by the infrared signature of large rocket engines. This would have neither. Of course, you could modify it by putting in a ballistic guidance system, a solid or liquid rocket motor, removing the wing area, stressing the airframe for supersonic ascent, characterizing its ballistic coeffecient, etc etc etc... but aren't we talking about building a ballistic missile from scratch at that point?
Do you understand that dumping a bunch of foil somewhere isn't gonna do a whole lot except inconvenience controllers? In fact, the vast majority of the time in aviation, pilots are responsible for avoiding each other without using radar, but just by looking outside for other planes. Controllers provide traffic advisories which are a nice service to have but aren't required.
In fact, I can tell you exactly what would happen if you dumped a ton of foil. It's be a bunch of large, unmoving blips. Because atmospheric conditions change the way radar is bent, the controllers would fiddle for a few minutes with the ground scan settings because they get this kinda stuff all the time when radar is lensed and they get ground echoes. Then the foil would drop to the ground and the blips would go away. Big whoop.
If you're going to use a cruise missile, using it against an aircraft on the ground is a pretty silly idea. LET"S THINK A SECOND. IT MOVES! IT'S RELATIVELY SMALL! THIS MAKES THINGS HARDER FOR NO GAIN. Suddenly you need target acquisition or remote control, and a whole lot more precision. If your goal is terrorism, send it into the stands at a football game; it's easier and it'll do a hell of a lot more damage. The jet fuel doesn't really make a lot of difference in the destructive power in this case.
Y'know, the definition of "cruise missile" means it doesn't fly a ballistic trajectory. Likewise, something with an air-breathing pulsejet isn't gonna stay ballistic very long no matter how much you alter it in other ways.
Sure, anything flying in an ADIZ would worry people. The thing is, if it crosses into an ADIZ and gets detected, it'll get shot down.
It would be hard to retrofit a cruise missile to be a decent surface to air missile. The problems are fundamentally different.
Minor correction-- to discover domestic cruise missile launches. Something that goes up high and fast on a ballistic trajectory will get plenty of attention from USSPACECOM, and even more so from foreign infrared imagery and radars. But a cruise missile just acts like an airplane-- and an airplane flying without talking to ATC in uncontrolled airspace is just fine. And it looks small enough on radars it could be mistaken for a bird or ground reflections.
It wouldn't have a very large radar cross sectional area, and ATC radar picks up all kinds of junk with weak echoes. Small private planes are pretty small blips without a transponder.
There aren't any missile early warning systems set up to discover domestic launches, either.
The fact that triggering chaos from long distances away is becoming easier is worrisome-- whether we're talking about arson by CO2 laser or cheap cruise missiles (heck, I think you could do it for well -under- $5k largely using parts for small model aircraft).
CDs are inherently digital. Audio CD's use 2352 byte blocks of data to store samples and interleaved Reed Solomon correction codes (this is why they tolerate scratches so well).
The actual physical encoding is "eight to fourteen" modulation. It exists to minimize transitions from 1 to 0 and vice versa, to avoid small fragile pits. It encodes 8 input bits to 14 output pit positions.
You're thinking of PWM, pulse -WIDTH- modulation, not PCM.
The new Sledgehammer runs at 1.8Mhz. In a dual-processor combination, it can slaughter the Xeon 3.06Ghz in some applications, and holds it's own in most others.
DAMN that's some instruction level parallelism, if a 1.8MHz part can slaughter a 3.06GHz part. Overcoming a 1700x speed difference!
Laserdisc could carry both analog and digital (PCM) audio.
PCM means you sample audio at regular time intervals (in the case of CD's, 44100 times per second), and quantize the sample to fit into a given number of digits. In the case of CD, the sample is stored into 16 base 2 digits, or bits. So basically, there's an observation of a voltage or magnitude, 16 bits in precision, 44100 times per second per channel. This happens to be the most common WAV format and also what 99.999% of MP3s decode to (other sample rates are allowed to save bandwidth for speech coding and other applications).
Here it is, it's short and out of context, but it's also the entire quote provided by Theo:
I wanted to update you on the situation with the Univ of Penn. project. As a result of the DARPA review of the project, and due to world events and the evolving threat posed by increasingly capable nation-states, the Government on April 21 advised the University to suspend work on the "security fest" portion of the project.
Now where does it say in that "open source is bad"? Could it be that the government has decided other threats are more immediate to address with DARPA's limited budget? I mean, we know Theo has never stirred up shit for the fun of it. </SARCASM>
You wouldn't happen to know if NASA has really done any in-depth contingency planning of what to do if a shuttle gets stuck in orbit? Ie, venting of OMS/RCS propellant preventing deorbit burn or damage to the tiles?
I've not been able to find any publically available planning papers from NASA on this topic-- and this seems to be a fairly likely failure scenario (things have a couple of weeks to break in bad ways in orbit..) and one that is just horribly bad PR for the space program if it happens, too.
Sure, but when there's a "new" space there's the chance for massive growth. Lots of things like online shopping are spaces that have taken off extremely quickly. If you choose to deliberately grow slowly, you're giving up control of the space to someone else.
On the other hand, if you try and grow beyond your means, your company will die. One needs to understand the limitations of the industry one is in and the company one has built to avoid this fate.
Now is a good time to build a small business. Rent, labor, and equipment costs will be relatively low. But with the difficulty in access to capital, it'll be difficult to build new markets. Those will, for the most part, belong to established players for the next few years.
I notice you mention interest rates-- using debt for capital requires you to sell your soul and take a risk just as large as the infusion of venture capital-- if not larger.
It IS possible to build a business quickly in a growing market in a sustainable way. This was my experience as CTO & cofounder of Recourse Technologies, the third largest software acquisition of 2002 after the bubble burst ($135M).
It's very difficult for a business to grow at a CAGR of 20%/yr without an infusion of capital. If your market is growing faster than that, not having access to capital is effectively condemning you to be a niche player.
I was CTO of Recourse Technologies, which was the third largest software acquisition in 2002 after the bubble burst. We had built the foundations of a sound business in a sound space in which a company can be profitable, AND grew revenues quickly. You can't say that about many dotcoms.
If you assume the normal flight mortality is 2%, and the chance of a fatal accident on this rushed one is five times that... and you know with 90% confidence the shuttle is gonna break up on reentry:
Don't try to rescue them: (.9)(7) = 6.3 mortalities average
Try to rescue them: (.10)(10) = 1 mortalities on average
I don't know what the numbers really are, but wouldn't it be nice to be able to think about this -before- reentry? One could work on multiple of these possible rescue scenarios in parallel. Hopefully NASA's process was not broken, and it was just a mistake in engineering judgment that reentry would be safe. Likewise, hopefully the process for evaluating the safety of the tiles was scientifically sound (and not "we had this weird thing happen with tiles, but the shuttle was OK and we got away with it once so it's safe").
It's hard to fix cultural problems.. Adding things that haven't been considered in engineering analysis in the past to when you do it in the future isn't so hard. So I hope it's not a process problem and just mistaken analysis.
Just because energy has to be dissipated doesn't mean it all has to be dissipated evenly, and doesn't mean that all dissipation modes have equivalent survival chances for the vehicle.
BTW, to respond to your original post.. Probably the best hope of survival was a TAL abort rather than RTLS. Neither RTLS nor TAL achieves full orbital velocity, and TAL loads the airframe less. RTLS is mostly reserved for situations where it's desirable to get the shuttle down quick-- large cabin leaks, for instance; or when there's a main engine failure at a time on an ascent profile that doesn't allow for a successful TAL.
It is possible for the astronauts to hand-fly the reentry sequence. Joe Engle did it on STS-2 (due to incorrect drag parameters for the flight control system). So if a new reentry profile could be designed, it could be used. ALSO, I think the OPS programs that do the actual reentry have numerous parameters that can be modified in orbit.
Heating on the wings is even when the vehicle is banked, eh? The fuselage of the vehicle produces no shielding of the "up" wing when atmospheric density is so low? Not to mention that spending more time at bank means that you descend quicker. THere's also "skip" trajectories like many of the Apollo missions flew (these provide two very short windows of extreme heating, as compared to the "moderate" heating of a normal re-entry)Obviously the re-entry profile that is flown affects the degree of stress the orbiter goes through. Are you saying it's impossible to design a reentry profile with different stress characteristics? I'm not sure what profile/loading on the vehicle is ideal for the damage the shuttle suffered (for we don't even know what that damage is), but thermal, mechanical stress, and aerodynamic simulations could establish that.
Shuttle managers said that if they were willing to skip testing, they could have a shuttle in orbit 2 weeks of having it on the pad. Atlantis's prep was finished. The critical thing is the ability to get rid of CO2. Humans produce -much- less CO2 when at rest. Thinking about stretching mission time by 50-75% is not out of the question. You'd have to do EVA to shuttle people between shuttles.
Opening the door does NOT depressurize the entire space shuttle. The lab that was in the cargo bay had provisions to just depressurize the lab to do an emergenecy EVA. Keep in mind that valuable weight is spent on every shuttle mission to be able to manually close the payload bay doors if they stick open.
THe thing is, all of this stuff I'm describing is extremely hazardous stuff, especially to try and pull off in two weeks without practice ahead of time. But if you know you've got no other choice, and that the vehicle is almost certainly lost-- you might want to try something like this.
Re:How does a website spend $80mln?
on
Salon Asks for Help
·
· Score: 2, Insightful
If you're not willing to go public or be acquired, you're not going to get venture money.
And without access to capital of some kind or another, you're not going to get massive growth rates.
Going public is not a death sentence for a company by any means. The capital from the public markets is useful to continue to grow your business, along with the credibility you get with customers. Going public without knowing damn well what your revenues are going to be for the next several years, though.. that's stupid.
I think the question is whether the process NASA followed was sound.
Aerospace is complicated stuff, and engineers will make mistakes and failure chains will occur. The question is whether the process that is in place causes the correct analysis to be attempted (even if the results are faulty) and there are a sufficient level of checks and balances in place.
In the Challenger accident, the process failed. In this case, there may have been bad analysis done by the individual engineers, or there may not have been enough information on the videotape to know (and taking drastic measures based on a guess is in itself dangerous). But as long as we operated on the best knowledge that we had at the time, I don't have a problem with what NASA did post launch.
On the other hand, Richard Feynman's paper on the Challenger accident is very appropriate here. Tiles have been getting damaged for a long time; they were not designed to take damage. Just because you survive a phenomena you don't understand once.. doesn't mean that it's a safe risk to take again. Engineers need to figure out why these things happen, and correct the design. Some steps were taken in the early 90's with the tile adhesive, but were probably not sufficient.
The tiles are a really good ablative shield (the best known, perhaps), but very delicate. And trying to use them on something that is going to experience launch energies is an inherently risky proposition. Managing that risk by adapting to newly observed behavior is the job of the engineering staff at United Space Alliance and NASA. Was it done sufficiently with the tile impact problems? I don't know.
STS-1, 2, and 3 were hand-flown approaches after the retrorocket burn from orbit. I also believe that it's possible to modify parameters to the OPS 304/305 reentry programs on orbit.
It'd be a hard decision to make, to fly a profile that would certainly damage the vehicle beyond repair, and would definitely risk the lives of the crew to take the vehicle to a survivable speed/altitude (50,000 ft) for ditching. But certainly it would be possible to design profiles that put additional loads on one wing to spare the other (e.g. bank one way more often than the other).
An effort to conserve supplies and launch a rescue shuttle is not out of the question, either, but also marginal (what if the same thing happened to Atlantis?)
Still, if it was a problem with the analysis, it'd be better to know so these possibilities could have been examined in depth.
Slashdot Mirror
You're on crack.
Any sufficiently long wire is an antenna, and the leakage from unshielded twisted pair is nontrivial. The carrier that fast ethernet data is modulated onto (125MHz) is of a short enough wavelength in comparison to the wire that it really leaks pretty well.
If you have a aviation band radio, tune it to 125MHz, and listen to the nice sound you get for each transmitted packet on a nearby LAN. I just did the test and heard plenty of noise on mine. Retrieving the bits would probably take a small amount of signal processing, but it's really a pretty conventional reception problem.
If you still believe that bits aren't leaking, I don't know what to tell you.
Another tidbit: switching provides very little security in practice, particularly if you can inject packets. You can man in the middle with ARP requests; you can poison CAM tables. It's not possible to protect against all of these attacks with switch security commands today on any platform I know of.
Cable upstream speeds can be really variable. I know someone on Cox cable and during the day he gets about 4-5 kilobytes per second up over TCP (this is probably about 35-45kbps of raw connectivity). At night he gets as much as 50-60kbytes up.
When I was on SBC/PacBell DSL 384k with a 128kbps upstream, I'd usually only see 10kbytes/sec up. This is nowhere near enough for a 128kbps+ MP3.
The Apple Airport base stations seem to just work, unlike my experience with Dlink and Linksys products.
A comparable box is the Linksys WRT54G; looking through reviews, people have had lots of problems. You can get it for $120-130 compared to $190-$200 for the Airport; and the Linksys doesn't have a USB port for printer sharing.
A $60-$80 premium to have a solution which is dependable is well worth it for me.
This is true only if you get the administrator's account password so that you can run through sudo, or if you can wait for the user to do a privileged operation and hijack it. As long as Joe User doesn't escalate to administrator privileges, it can only run with Joe User's rights. This is a win; I basically only escalate privilege these days for the occasional software installation.
It's a very difficult signal processing problem. When you have a series of snapshots of the positions of "blips" taken 3 seconds apart, how do you find real, faint blips? Figuring out which dots to connect becomes hard, because there's false echoes all over the place in each observation.
There's been a lot of work in this area; probably the most fruitful has been to try and provide support to humans engaging in this type of vigilance task. One system I know of keeps snapshots of radar data and then replays it at higher speed than the observation interval. The human visual system is much better at picking out that kind of blip because it looks more like we're accustomed to having objects move in the real world (smoothly, instead of jumping every 3 seconds).
I'm addressing what you said. All that garbage you said about ballistic trajectory to get attention, was.. garbage.
10kg of explosives is comparable to what a suicide bomber carries, and the pressure wave could be more optimally located by blowing at a slight altitude. This could be pretty effective against a football game and result in hundreds of casualties and tens of fatalities, without having to pick the location that you hit too closely.
On the other hand, I'm not altogether convinced 10kg of explosives would do a whole lot to an airliner on the ground unless you aimed really, really well. Let's think a second-- 380 MPH, let's be really, really generous and say 3 GPS updates a second.. between each update it moves 186 feet! The wingspan of a 747-400 is only 225 feet, with an average width of something like 26 feet... not to mention it'd be hard to get exact coordinates of the 747 in the first place.
If you know as much as you say you do, surely you can compute the pressure wave that 10kg of explosives is going to generate, and compare it to the 670PSI of loading that the wing structure is designed to sustain. The inverse square law is gonna say you're going to have to be really, really close to compromise the wing structure. Probably closer than the RMS error of civilian GPS, not even counting the intricacies involved in getting a missile to express its energy at the right point. I suspect breaching the fuselage is even more difficult. This is completely ignoring the fact that the tanks themselves are reinforced and can take substantial loads. Do the math; the numbers are readily available and it takes less than 5 minutes.
The jet fuel doesn't make that much of a difference because jet fuel doesn't really burn all that well if not atomized first. There aren't many fatal accidents from fires during taxiing or fueling operations, and there have been quite a few fires in those situations. The kinetic energy of a fast-moving airplane is far more effective at atomizing fuel than 10kg of explosives is going to be, and this is why landing accidents and the WTC form such spectacular fireballs.
As an engineer, and a pilot... I can say that if I wanted to to take out an important aircraft.. this is about the last thing in the world I'd try.
I'm convinced you don't know very much about aviation.
The kind of thing we're talking about doesn't have the kind of range necessary to make it an international nuisance. And the types of air defense interrogation zones that have been established for national defense against conventional fighter aircraft provide pretty good defensive measures for countries with capable air forces against internationally launched cruise missiles.
Ballistic missile detection systems typically detect objects going to very high altitudes (>20 miles), or by the infrared signature of large rocket engines. This would have neither. Of course, you could modify it by putting in a ballistic guidance system, a solid or liquid rocket motor, removing the wing area, stressing the airframe for supersonic ascent, characterizing its ballistic coeffecient, etc etc etc... but aren't we talking about building a ballistic missile from scratch at that point?
Do you understand that dumping a bunch of foil somewhere isn't gonna do a whole lot except inconvenience controllers? In fact, the vast majority of the time in aviation, pilots are responsible for avoiding each other without using radar, but just by looking outside for other planes. Controllers provide traffic advisories which are a nice service to have but aren't required.
In fact, I can tell you exactly what would happen if you dumped a ton of foil. It's be a bunch of large, unmoving blips. Because atmospheric conditions change the way radar is bent, the controllers would fiddle for a few minutes with the ground scan settings because they get this kinda stuff all the time when radar is lensed and they get ground echoes. Then the foil would drop to the ground and the blips would go away. Big whoop.
If you're going to use a cruise missile, using it against an aircraft on the ground is a pretty silly idea. LET"S THINK A SECOND. IT MOVES! IT'S RELATIVELY SMALL! THIS MAKES THINGS HARDER FOR NO GAIN. Suddenly you need target acquisition or remote control, and a whole lot more precision. If your goal is terrorism, send it into the stands at a football game; it's easier and it'll do a hell of a lot more damage. The jet fuel doesn't really make a lot of difference in the destructive power in this case.
Y'know, the definition of "cruise missile" means it doesn't fly a ballistic trajectory. Likewise, something with an air-breathing pulsejet isn't gonna stay ballistic very long no matter how much you alter it in other ways.
Sure, anything flying in an ADIZ would worry people. The thing is, if it crosses into an ADIZ and gets detected, it'll get shot down.
It would be hard to retrofit a cruise missile to be a decent surface to air missile. The problems are fundamentally different.
Minor correction-- to discover domestic cruise missile launches. Something that goes up high and fast on a ballistic trajectory will get plenty of attention from USSPACECOM, and even more so from foreign infrared imagery and radars. But a cruise missile just acts like an airplane-- and an airplane flying without talking to ATC in uncontrolled airspace is just fine. And it looks small enough on radars it could be mistaken for a bird or ground reflections.
I'd have to say it wouldn't do very much at all.
It wouldn't have a very large radar cross sectional area, and ATC radar picks up all kinds of junk with weak echoes. Small private planes are pretty small blips without a transponder.
There aren't any missile early warning systems set up to discover domestic launches, either.
The fact that triggering chaos from long distances away is becoming easier is worrisome-- whether we're talking about arson by CO2 laser or cheap cruise missiles (heck, I think you could do it for well -under- $5k largely using parts for small model aircraft).
CDs are inherently digital. Audio CD's use 2352 byte blocks of data to store samples and interleaved Reed Solomon correction codes (this is why they tolerate scratches so well).
The actual physical encoding is "eight to fourteen" modulation. It exists to minimize transitions from 1 to 0 and vice versa, to avoid small fragile pits. It encodes 8 input bits to 14 output pit positions.
You're thinking of PWM, pulse -WIDTH- modulation, not PCM.
The new Sledgehammer runs at 1.8Mhz. In a dual-processor combination, it can slaughter the Xeon 3.06Ghz in some applications, and holds it's own in most others.
DAMN that's some instruction level parallelism, if a 1.8MHz part can slaughter a 3.06GHz part. Overcoming a 1700x speed difference!
Laserdisc could carry both analog and digital (PCM) audio.
PCM means you sample audio at regular time intervals (in the case of CD's, 44100 times per second), and quantize the sample to fit into a given number of digits. In the case of CD, the sample is stored into 16 base 2 digits, or bits. So basically, there's an observation of a voltage or magnitude, 16 bits in precision, 44100 times per second per channel. This happens to be the most common WAV format and also what 99.999% of MP3s decode to (other sample rates are allowed to save bandwidth for speech coding and other applications).
Last time I checked, compact discs stored the audio in digital form, and "digital audio extraction" extracts that digital audio.
;P
Where does it become analog again? CDs don't become analog until you play them out speakers-- that's what the DAC is for
Here it is, it's short and out of context, but it's also the entire quote provided by Theo:
I wanted to update you on the situation with the Univ of Penn. project. As a result of the DARPA review of the project, and due to world events and the evolving threat posed by increasingly capable nation-states, the Government on April 21 advised the University to suspend work on the "security fest" portion of the project.
Now where does it say in that "open source is bad"? Could it be that the government has decided other threats are more immediate to address with DARPA's limited budget? I mean, we know Theo has never stirred up shit for the fun of it. </SARCASM>
You wouldn't happen to know if NASA has really done any in-depth contingency planning of what to do if a shuttle gets stuck in orbit? Ie, venting of OMS/RCS propellant preventing deorbit burn or damage to the tiles?
I've not been able to find any publically available planning papers from NASA on this topic-- and this seems to be a fairly likely failure scenario (things have a couple of weeks to break in bad ways in orbit..) and one that is just horribly bad PR for the space program if it happens, too.
Sure, but when there's a "new" space there's the chance for massive growth. Lots of things like online shopping are spaces that have taken off extremely quickly. If you choose to deliberately grow slowly, you're giving up control of the space to someone else.
On the other hand, if you try and grow beyond your means, your company will die. One needs to understand the limitations of the industry one is in and the company one has built to avoid this fate.
Now is a good time to build a small business. Rent, labor, and equipment costs will be relatively low. But with the difficulty in access to capital, it'll be difficult to build new markets. Those will, for the most part, belong to established players for the next few years.
I notice you mention interest rates-- using debt for capital requires you to sell your soul and take a risk just as large as the infusion of venture capital-- if not larger.
It IS possible to build a business quickly in a growing market in a sustainable way. This was my experience as CTO & cofounder of Recourse Technologies, the third largest software acquisition of 2002 after the bubble burst ($135M).
Oh, it IS possible to build a business quickly.
It's very difficult for a business to grow at a CAGR of 20%/yr without an infusion of capital. If your market is growing faster than that, not having access to capital is effectively condemning you to be a niche player.
I was CTO of Recourse Technologies, which was the third largest software acquisition in 2002 after the bubble burst. We had built the foundations of a sound business in a sound space in which a company can be profitable, AND grew revenues quickly. You can't say that about many dotcoms.
If you assume the normal flight mortality is 2%, and the chance of a fatal accident on this rushed one is five times that... and you know with 90% confidence the shuttle is gonna break up on reentry:
Don't try to rescue them: (.9)(7) = 6.3 mortalities average
Try to rescue them: (.10)(10) = 1 mortalities on average
I don't know what the numbers really are, but wouldn't it be nice to be able to think about this -before- reentry? One could work on multiple of these possible rescue scenarios in parallel. Hopefully NASA's process was not broken, and it was just a mistake in engineering judgment that reentry would be safe. Likewise, hopefully the process for evaluating the safety of the tiles was scientifically sound (and not "we had this weird thing happen with tiles, but the shuttle was OK and we got away with it once so it's safe").
It's hard to fix cultural problems.. Adding things that haven't been considered in engineering analysis in the past to when you do it in the future isn't so hard. So I hope it's not a process problem and just mistaken analysis.
My bad; shows me for not previewing after editing my post.
And then your specific issue is what, exactly?
Just because energy has to be dissipated doesn't mean it all has to be dissipated evenly, and doesn't mean that all dissipation modes have equivalent survival chances for the vehicle.
BTW, to respond to your original post.. Probably the best hope of survival was a TAL abort rather than RTLS. Neither RTLS nor TAL achieves full orbital velocity, and TAL loads the airframe less. RTLS is mostly reserved for situations where it's desirable to get the shuttle down quick-- large cabin leaks, for instance; or when there's a main engine failure at a time on an ascent profile that doesn't allow for a successful TAL.
Also, STS-57 tested EVA from SPACEHAB.
Did you even read what I said?
It is possible for the astronauts to hand-fly the reentry sequence. Joe Engle did it on STS-2 (due to incorrect drag parameters for the flight control system). So if a new reentry profile could be designed, it could be used. ALSO, I think the OPS programs that do the actual reentry have numerous parameters that can be modified in orbit.
Heating on the wings is even when the vehicle is banked, eh? The fuselage of the vehicle produces no shielding of the "up" wing when atmospheric density is so low? Not to mention that spending more time at bank means that you descend quicker. THere's also "skip" trajectories like many of the Apollo missions flew (these provide two very short windows of extreme heating, as compared to the "moderate" heating of a normal re-entry)Obviously the re-entry profile that is flown affects the degree of stress the orbiter goes through. Are you saying it's impossible to design a reentry profile with different stress characteristics? I'm not sure what profile/loading on the vehicle is ideal for the damage the shuttle suffered (for we don't even know what that damage is), but thermal, mechanical stress, and aerodynamic simulations could establish that.
Shuttle managers said that if they were willing to skip testing, they could have a shuttle in orbit 2 weeks of having it on the pad. Atlantis's prep was finished. The critical thing is the ability to get rid of CO2. Humans produce -much- less CO2 when at rest. Thinking about stretching mission time by 50-75% is not out of the question. You'd have to do EVA to shuttle people between shuttles.
Opening the door does NOT depressurize the entire space shuttle. The lab that was in the cargo bay had provisions to just depressurize the lab to do an emergenecy EVA. Keep in mind that valuable weight is spent on every shuttle mission to be able to manually close the payload bay doors if they stick open.
THe thing is, all of this stuff I'm describing is extremely hazardous stuff, especially to try and pull off in two weeks without practice ahead of time. But if you know you've got no other choice, and that the vehicle is almost certainly lost-- you might want to try something like this.
If you're not willing to go public or be acquired, you're not going to get venture money.
And without access to capital of some kind or another, you're not going to get massive growth rates.
Going public is not a death sentence for a company by any means. The capital from the public markets is useful to continue to grow your business, along with the credibility you get with customers. Going public without knowing damn well what your revenues are going to be for the next several years, though.. that's stupid.
I think the question is whether the process NASA followed was sound.
Aerospace is complicated stuff, and engineers will make mistakes and failure chains will occur. The question is whether the process that is in place causes the correct analysis to be attempted (even if the results are faulty) and there are a sufficient level of checks and balances in place.
In the Challenger accident, the process failed. In this case, there may have been bad analysis done by the individual engineers, or there may not have been enough information on the videotape to know (and taking drastic measures based on a guess is in itself dangerous). But as long as we operated on the best knowledge that we had at the time, I don't have a problem with what NASA did post launch.
On the other hand, Richard Feynman's paper on the Challenger accident is very appropriate here. Tiles have been getting damaged for a long time; they were not designed to take damage. Just because you survive a phenomena you don't understand once.. doesn't mean that it's a safe risk to take again. Engineers need to figure out why these things happen, and correct the design. Some steps were taken in the early 90's with the tile adhesive, but were probably not sufficient.
The tiles are a really good ablative shield (the best known, perhaps), but very delicate. And trying to use them on something that is going to experience launch energies is an inherently risky proposition. Managing that risk by adapting to newly observed behavior is the job of the engineering staff at United Space Alliance and NASA. Was it done sufficiently with the tile impact problems? I don't know.
A reentry profile that cannot be changed?
Why not?
STS-1, 2, and 3 were hand-flown approaches after the retrorocket burn from orbit. I also believe that it's possible to modify parameters to the OPS 304/305 reentry programs on orbit.
It'd be a hard decision to make, to fly a profile that would certainly damage the vehicle beyond repair, and would definitely risk the lives of the crew to take the vehicle to a survivable speed/altitude (50,000 ft) for ditching. But certainly it would be possible to design profiles that put additional loads on one wing to spare the other (e.g. bank one way more often than the other).
An effort to conserve supplies and launch a rescue shuttle is not out of the question, either, but also marginal (what if the same thing happened to Atlantis?)
Still, if it was a problem with the analysis, it'd be better to know so these possibilities could have been examined in depth.