I think it had more to do with the record labels allowing it. Remember, when iTunes Plus came out, it was basically just EMI artists that were available DRM-free. After EMI didn't go out of business, and the other labels decided to allow Amazon to sell DRM-free tracks (to break Apple's hold on the digital market,) only to realize they didn't go out of business, they finally gave up and let Apple do it too.
The motivator behind DRM in music was the labels, not the distributors. I think that the same thing will eventually happen in e-Books as well -- unfortunately you can't rip a paper book to a digital format nearly as easily as you could rip CDs to MP3s, so if you want to get more than small selection of e-books legally and DRM-free, until then, you're SOL.
Actually, minimum energy to rendezvous is around these close approaches, but isn't terribly dependent on the actual distance to the Earth. In order to rendezvous with a minimum of propellant, you launch about 2 months before the close approach and rendezvous about 5 months afterwards. The rendezvous Delta-V using this intercept trajectory is around 3 km/s, instead of the nearly 6 km/s you'd require during the close approach. Whether you do it during this very close pass, or during further out ones in 2013 or 2021 is inconsequential.
Also, landing is a very difficult problem on an asteroid, since you have very little gravity to depend on. For science you're probably better off maneuvering around it and taking images, using your flybys to determine mass, etc. The main thing you could get out of landing is to be able to physically take samples -- however, the material composition is a matter of percentages of known materials, so we can probably determine it from a standoff distance. Also, the act of taking the samples would be comically hard -- attempting to dig or drill would probably push the probe off of the asteroid before it could dig in at all. The best solution I've heard to do it right is to use a chemical laser to melt a part of the surface, land the probe and let the surface harden around it to keep it attached, something thats probably far more difficult than its worth.
There have been proposals to launch this kind of mission. I'm currently working on one that would study the asteroid, and also use the opportunity to demonstrate that we can move it via gravity tractor and mitigate any potential threat, as a sort of dress rehearsal (we prefer archetype mission). Interestingly, the equipment required to study it is almost exactly the same equipment required to move it, so you may as well do it all at once.
Ah. Probably the most useful thing, and the most computationally intensive, is going to be doing aerodynamics/fluids, whether for aeronautics research, reentry work, or planetary atmosphere science. Unfortunately, most CFD codes require a lot of talking back and forth between nodes, which isn't very good for a distributed computational network, since its bound to be communication limited. An alternative that works well for low-density flows (and thus has potential for modelling re-entry) is the Lattice Boltzmann method, where the hard math is done at each node independently, and so requires much less significant communication between nodes. Structural analysis isn't too bad computationally as far as I know.
I'm really more of a controls/estimation and sensing/systems engineering guy myself though, so I can't say for certain. I did some lattice boltzmann work as an undergrad, but really just enough to convince myself that I didn't want to do fluids.
If I remember correctly that one was very small, so that if it landed at all it would probably mess up someones house, but was far more likely to airburst and be completely unnoticed on the ground.
99942 Apophis, while not a 'planet-killer', is still big enough that it would cause massive regional destruction, probably 10 times more than the Tunguska event, which we were very fortunate to have happen in Siberia and not over a populated region. Also the danger doesn't come from this 2029 close approach, but rather from the following approach in 2036. If it goes through the wrong region of space (a so-called keyhole) in 2029, it will come back around for an impact the next time.
Actually, the discoverers claim that during the initial days when it was a 1/300 impact risk, a god of destruction seemed like a good name. However, it also turns out that they were SG-1 fans...
It is certain not to hit the moon on this pass, just as its guaranteed not to hit the Earth. Uncertainty of the asteroid's position is within 10s of kilometers, more than enough to make sure theres no risk of that.
If it were to impact the moon, we can determine the relative Delta-V it would apply. The velocity of the asteroid relative to the Earth moon system upon entry is approximately 5.9 km/s, according the JPL NEO page, and has a mass of ~2.7e10 kg. The Moon is moving at ~1 km/s and has a mass of 7.3e22 kg. Assuming an inelastic collision, and that the impact is along the velocity vector (where it will have the largest impact), and applying conservation of momentum, you get a whopping 1.8 nm/s velocity change. So basically, the asteroid is far too small to have any kind of noticeable effect on the moon. Looking at the surface these kind of events happen all the time (cosmologically).
In 2029 the asteroid 99942 Apophis will pass well within the GEO satellite belt (36,000 km), but will not impact the Earth. The video simulates this trajectory and as the Earth approaches for a few moments it appears that an impact is likely. However, this is an illusion where the Earth merely dominates the field of view and the in-plane relative velocity is much larger than the horizontal relative velocity.
To be clear, the orbit of the asteroid as it enters the Earth's sphere of influence is a very high-energy hyperbolic orbit, where the closest point of approach is much further than 6,400 km (the Earth's radius).
Additional computing power isn't really needed for this problem. JPL already has the Standard Dynamic Model they use to model all bodies in the solar system accurately, and the current hardware is perfectly capable of handling the problem.
What is needed to refine and understand the trajectory is more observations. Radar range and range-rate measurements, along with optical angle measurements are fed together to estimate the current position and velocity, and using estimation techniques you can estimate your uncertainty as well. In order to bring down the uncertainty, we need more measurements that give a better statistical sample and allow you to have more confidence in your averages. Sadly most people don't have radio telescopes are large enough optical telescopes (20"+ preferably) to really make a good observation. For that reason, it will probably take till 2013, the next close approach, to get a new set of data that will make it easy to determine whether there is a 2036 impact risk.
While the headline should have explicitly stated "Oklahoma State University" rather than simply "OSU," I'm not sure how Ohio gets the right to the acronym. Growing up in Oklahoma and attending a Big 12 School, my assumption is that OSU refers to Oklahoma State.
And don't forget those kids in Oregon and their amusingly named mascot -- especially when playing USC.
Actually this pass (the 2029 close approach) is not a concern at all. The error brackets are brought in well enough that we know it will not impact the Earth, but will pass well within the GEO belt. What we don't know, and when the actual 1/250000 impact risk is, is the next pass, in 2036. If the asteroid passes through what is known as a 'gravitational keyhole' in 2029, the effect of Earth's gravity will actually swing the asteroid back around on an impact path in 2036.
Right now we can predict where Apophis will be in 2029 fairly well, within a few 10s of kilometers I believe. When you're talking about hitting the Earth, a thousand kilometers or more is good enough precision. The problem is that that during that flyby in 2029, any small uncertainties magnify by a few orders of magnitude, so when you carry it through another 7 years of orbits the uncertainty is 10s of Earth radii instead. Add in uncertainties about the effects of solar wind and the Yarkovsky effect and it just gets more complicated.
I realize that a lot of it is just good-hearted ribbing, but there seems to be a lot of discussion of PHP coders being bad coders, which is something I'm not sure I get. My background is as an aerospace engineer who ends up writing a lot of code, but I wouldn't call myself a developer, or even a *real* programmer by any stretch; I prefer to work in MATLAB, as its an industry standard and has built-in functions for much of what I need, but I've also written code in C, C++, C#, Objective C, Python, BASIC, JavaScript, probably some others, and yes, PHP.
I've been using PHP for a couple of web-based projects I'm working on, and I've enjoyed using it. It has some rough edges, particularly with inconsistent function naming, and 'duck' typing does lend itself to some poor practices if you're not careful, but its been great for many of the things I'm trying to do. It integrates well with Apache and is dead-simple to work with on Linux and OSX. It lets you do OO stuff where appropriate but doesn't force it down your throat (sometimes you just need a simple conditional to display a chunk of html). You can call class names from variables making it easy to do automated form generation without running through tedious case statements. It facilitates access to 'GET' and 'POST' data and has adaptable data structures and good string manipulation functionality. And of course it makes it easy to interface with the database.
Obviously, all of this leaves some room for abuse, and its certainly not perfect, but it doesn't mean its impossible to write decent PHP code. I like to think that with my projects I've done a decent job of keeping the code straightforward, maintainable and efficient -- moving database access into data model objects, keeping repeated segments of code, headers, etc. in distinct files so they only have to be modified once, using descriptive subroutine names so its obvious what the code is doing, planning out the overall program structure so I know how its all laid out in advance, and maintaining a MVC architecture where appropriate. The same kind of things you'd do to make the code 'decent' in any language.
There's also less of a reason to deny it. Solving climate change involves significant lifestyle changes (at least in the short term) and a lot of investment by society. This is really why its so controversial. People want to keep driving their SUVs without feeling guilty, and switching all our power plants off of coal is a massive prospect. Solving an asteroid impact is a one-time cost of a few hundred million for a small one, or a few billion for a large one, and then people can go on about their lives as normal.
Actually, no. The *asteroid* 99942 Apophis is 200-300 meters in diameter and is well understood to be a threat of regional destruction, not world-wide devastation.
The real problem I see with this analysis is simply that our ability to track doesn't tell you where its going to hit, it will simply tell you theres a 25% of it hitting the Earth at all, and that one spot might be at the peak of the probability distribution (bell curve). To be sure it wouldn't hit you have to move it by many Earth-radii anyway, so you probably would never know that you just had it aimed at Russia.
Its a big deal because we're just now reaching the point where we know how to avert a potential impact. Since its a relatively cheap thing to do (~$300M for an Apophis-sized asteroid) it would be irresponsible not to consider the possibilities and have plans in place to handle the event. It's not a choice of one or the other.
There are a *LOT* of people in this world, all doing a lot of things. Having people work on finding solutions to a potential asteroid impact doesn't take effort away from solving global climate change. I work with NEA's, and while I'm not a climate change denier, if I weren't doing this I wouldn't be working on green technology instead. My skills are in spacecraft mission design, so if I weren't working on asteroids I'd be working on other missions instead -- and with unemployment at 10% we're not in a labor shortage right now either (believe me, this is hitting just graduated engineers as much as anyone else). The miniscule amount of money involved in NEA research would do nothing as far as climate change goes -- climate change is a political problem (requiring potential lifestyle changes to fix) rather than a monetary one anyway.
If you've got enough time, the gravity tractor provides an easier method as well. Since you're not landing you don't have to worry about the rotation (or the complex control systems required to land), and its just as efficient as far as propellant is concerned -- though it does prevent using the asteroid material as propellant.
Obviously, the main problem is that you're limited in the amount of force you can apply by the mass and standoff distance -- for small asteroids with plenty of warning its a no-brainer, for big asteroids the margins are tighter, but it can still be viable.
No need to be so cynical. You can deflect a smaller, Apophis-sized asteroid for around $300M with 15-20 years of warning. A planet killer could probably be handled with a couple of billion. In the scheme of space programs' budgets, this is a large but not unreasonable budget, and is exactly what they exist to do. Governments have sent missions to completely non-threatening asteroids for study already, so doing it with a save-the-Earth mission attached is a no-brainer (and the ability to alter trajectory via gravity tractor adds minimal complexity and weight above a pure exploration mission).
The harder part is finding and tracking the objects so that you have enough warning -- this is not nearly as sexy, but even then the US congress has already directed NASA to do so and provided at least some funding to support it. It helps that its cheaper. Though I can't say for sure I'd assume that other countries have similar programs.
While a gyro is necessary to actually do full 6-DOF position tracking (otherwise you must assume that you're holding a specific orientation... this can still be good for something like an in-the-air mouse), the Wii system still wouldn't be good for detecting absolute motion without the sensor bar as well.
The problem is that while the sensors are fairly precise as far as measuring the accelerations (if they're anything like the iPhone sensors they're around 0.02g precise), when you try and integrate them twice to get the position, things start to fall apart. Imagine you do a simple up-and-down motion. You get a sinusoidal acceleration curve that when you integrate it once gives you an offset sinusoid to represent your velocity, and a second integration gives a third one to represent your position. However, at the end, your integration to the velocity level comes out to be not quite zero, because those small acceleration errors will mostly cancel out, but not perfectly. This is still a pretty good velocity estimate, since its close to zero. However, as far as your position is concerned, close to zero and actually zero are very different, so you get a constant, growing drift in your position from a small velocity error. The same things apply to gyros, although the math is a little more complex.
Basically if you want to use a sensor as a double integrator it has to be extraordinarily precise, and even then you're going to get some drift that you have to remove every once in a while, or have an absolute position value to keep it in check (kalman filters do a great job of interpreting data from multiple sensors). What the sensor bar and IR sensors do is give you an incomplete but useful reference on position and orientation that you can use to keep that drift in check. Adding the gyros definitely helps a lot too, but you still need the sensor bar to keep drift in check.
And forgetting my high school science classes, averaging gives you precision, not accuracy. Accuracy is a whole other issue, but its not very different between cheap and expensive sensors, and calibration can eliminate the issues.
In my experience, doing some sensor systems with cheap sensors and expensive sensors, the difference is that cheap ones can be fast or accurate, while good ones can be both. Because you can average data and apply statistical methods to eliminate noise, a long integration time can get you very good precision (as long as your not doing an integrator... using an iPhone as a position sensor won't work, since you can't average the acceleration to get it).
In this case, a cheap sensor is going to work quite well unless your trying to collect usable data faster than about 1 Hz. Its really a matter of knowing what you need. In many cases a cheap sensor works really well.
Then I would argue that the failure mode is poor management and schedule rush -- definitely things not included in whatever safety numbers were quoted when the shuttle was being designed. The point is that whenever those 1 in 1000 numbers are pulled out they are almost meaningless -- the failures that did occur weren't included in those.
Its like judging the safety of a car on whether or not a freak string of events is likely to blow up the car on any given trip (or the brake lines fail, or your toyota accelerates without your command), when everyone knows that the most likely reason you're going to die in the car is because you or someone else screws up. While you want to do your best to keep the freak accidents from happening, more time needs to be spent making mistakes less damaging, and training people to avoid them.
To you the ultimate goal is scientific knowledge and resources.
To me the ultimate goal is human settlement beyond Earth.
To congresspeople the ultimate goal is getting reelected.
What you see as nationalistic chest thumping I see as (admittedly often poorly done) continued development of technology to support frontier development. They of course see it as jobs for their district. Conversations about how we should do things first require an agreement on the goals.
Three out of five flights, and the order matters significantly. The two that have been successful are significantly different than the three that failed. They added baffles to the tanks, improved the control algorithms, changed materials. They *FIXED* all of the issues that caused the early failures. Also, if its cheaper to blow up a few unmanned rockets than it is to design it perfectly the first time, then that sounds like the right way to do it. I'd consider the reliability of the Falcon 1 the same as any vehicle with a 2-0 record. Still not too reliable yet, but showing promise.
And those safety numbers are in so many ways bogus, since they only consider known failure modes. Everything thats ever killed an American astronaut was an unknown failure mode. Since Falcon 9 is intended for human use as well, with the same safety goals, and is in a further state of development than Ares 1, I can't help but be shocked by the sheer price of *just* Ares 1-X. If it were the entire Ares 1 program that had cost so much so far I'd say it was pretty reasonable and even cheap -- but no, just the aerodynamic and structural test cost that much.
No, only a few years, but its pretty clear that this is not in the best interest of furthering space exploration, but rather in keeping jobs in a few congressional districts -- namely Huntsville, Alabama. Marshall Space Flight Center stands the most to lose if Ares falls through, but MSFC is in many ways a dinosaur of the Apollo era and hasn't transitioned to being a leaner, more efficient group.
Consider this: for the cost of building Ares 1-X, the test-flight that consisted of a shuttle SRB with some dummy mass on top and made up to look like an Ares 1, what was essentially the worlds largest model rocket, cost $450M -- SpaceX, has developed one working rocket and has almost completed a larger one for around the same cost. While obviously the Ares program will cost more than what a company like SpaceX will spend, since they're building bigger rockets to do riskier things, there is something wrong when a mere model costs that much.
The problem with micromanaging NASA through congress is that the only districts where its an issue that can make a difference in an election are the ones where they want to maintain the status quo, which is not working well. Everyone else who sees it and disagrees with its handling probably aren't going to swing their vote based on it, since there are a myriad of other, more immediate things to consider as well.
In my experience even the ones with GPS require you to still know a bit -- the whole align by centering on the bright star thing never works as well as you'd like, at least in my experience. Magnetic anomalies, mount misalignments, or whatever else sometimes leave me wondering what star it is in the first place... this is particularly frustrating because Meade Autostar doesn't actually tell you what star it is you're looking for. My experience with the Meade 16" I work with has made me a little jaded on GOTO systems.
As far as dealing with sidereal tracking, while annoying, its not as bad on a manual equatorial mount as on an Alt-Az, because you can simply tell them to use the RA knob to keep it in view. On something with as wide a field of view as a 5" or 6" f/5 like the one I pointed out, the movement isn't too fast. This is more likely to be a bigger problem on big Dobsonian. Also, you can get a RA drive for that particular Celestron, which helps the issue.
I guess really in the end it comes down to a combination of factors -- budget, knowledgability of the parent, what you're hoping to see, patience of the kids, whether you expect to expand to more complex things like astrophotography, etc. etc.
Exactly right. Of course that doesn't necessarily mean it has to be expensive. A big (~10") Dobsonian like the one someone else mentioned is nice.
Personally, what I have is a 5" newtonian, the Celestron AstroMaster 130EQ, $180 on Telescopes.com, that I really love, even though I have access to a massive 16" Meade monster for my job. Its small enough to carry easily, but big enough to give you pretty good views of planets, clusters, some of the brighter nebulae, and affordable even on a grad student's stipend. It won't show you as much as a big Dobsonian, but its on a manual equatorial mount, so its a lot better for learning how to find your way around the sky. Also, you can expand it to use the RA motor drive ($40) and a CCD to do astronomical imaging (either use a DSLR or some $300 passively cooled CCDs) -- not as accessible to beginners, but potentially more rewarding in the long run.
Definitely go with a manual mount though, theres something rewarding about finding what you're looking for yourself over just punching things into a keypad. Its cheaper too, so you can get more aperture for the money.
Slashdot Mirror
I think it had more to do with the record labels allowing it. Remember, when iTunes Plus came out, it was basically just EMI artists that were available DRM-free. After EMI didn't go out of business, and the other labels decided to allow Amazon to sell DRM-free tracks (to break Apple's hold on the digital market,) only to realize they didn't go out of business, they finally gave up and let Apple do it too.
The motivator behind DRM in music was the labels, not the distributors. I think that the same thing will eventually happen in e-Books as well -- unfortunately you can't rip a paper book to a digital format nearly as easily as you could rip CDs to MP3s, so if you want to get more than small selection of e-books legally and DRM-free, until then, you're SOL.
Actually, minimum energy to rendezvous is around these close approaches, but isn't terribly dependent on the actual distance to the Earth. In order to rendezvous with a minimum of propellant, you launch about 2 months before the close approach and rendezvous about 5 months afterwards. The rendezvous Delta-V using this intercept trajectory is around 3 km/s, instead of the nearly 6 km/s you'd require during the close approach. Whether you do it during this very close pass, or during further out ones in 2013 or 2021 is inconsequential.
Also, landing is a very difficult problem on an asteroid, since you have very little gravity to depend on. For science you're probably better off maneuvering around it and taking images, using your flybys to determine mass, etc. The main thing you could get out of landing is to be able to physically take samples -- however, the material composition is a matter of percentages of known materials, so we can probably determine it from a standoff distance. Also, the act of taking the samples would be comically hard -- attempting to dig or drill would probably push the probe off of the asteroid before it could dig in at all. The best solution I've heard to do it right is to use a chemical laser to melt a part of the surface, land the probe and let the surface harden around it to keep it attached, something thats probably far more difficult than its worth.
There have been proposals to launch this kind of mission. I'm currently working on one that would study the asteroid, and also use the opportunity to demonstrate that we can move it via gravity tractor and mitigate any potential threat, as a sort of dress rehearsal (we prefer archetype mission). Interestingly, the equipment required to study it is almost exactly the same equipment required to move it, so you may as well do it all at once.
Ah. Probably the most useful thing, and the most computationally intensive, is going to be doing aerodynamics/fluids, whether for aeronautics research, reentry work, or planetary atmosphere science. Unfortunately, most CFD codes require a lot of talking back and forth between nodes, which isn't very good for a distributed computational network, since its bound to be communication limited. An alternative that works well for low-density flows (and thus has potential for modelling re-entry) is the Lattice Boltzmann method, where the hard math is done at each node independently, and so requires much less significant communication between nodes. Structural analysis isn't too bad computationally as far as I know.
I'm really more of a controls/estimation and sensing/systems engineering guy myself though, so I can't say for certain. I did some lattice boltzmann work as an undergrad, but really just enough to convince myself that I didn't want to do fluids.
If I remember correctly that one was very small, so that if it landed at all it would probably mess up someones house, but was far more likely to airburst and be completely unnoticed on the ground.
99942 Apophis, while not a 'planet-killer', is still big enough that it would cause massive regional destruction, probably 10 times more than the Tunguska event, which we were very fortunate to have happen in Siberia and not over a populated region. Also the danger doesn't come from this 2029 close approach, but rather from the following approach in 2036. If it goes through the wrong region of space (a so-called keyhole) in 2029, it will come back around for an impact the next time.
Actually, the discoverers claim that during the initial days when it was a 1/300 impact risk, a god of destruction seemed like a good name. However, it also turns out that they were SG-1 fans...
It is certain not to hit the moon on this pass, just as its guaranteed not to hit the Earth. Uncertainty of the asteroid's position is within 10s of kilometers, more than enough to make sure theres no risk of that.
If it were to impact the moon, we can determine the relative Delta-V it would apply. The velocity of the asteroid relative to the Earth moon system upon entry is approximately 5.9 km/s, according the JPL NEO page, and has a mass of ~2.7e10 kg. The Moon is moving at ~1 km/s and has a mass of 7.3e22 kg. Assuming an inelastic collision, and that the impact is along the velocity vector (where it will have the largest impact), and applying conservation of momentum, you get a whopping 1.8 nm/s velocity change. So basically, the asteroid is far too small to have any kind of noticeable effect on the moon. Looking at the surface these kind of events happen all the time (cosmologically).
In 2029 the asteroid 99942 Apophis will pass well within the GEO satellite belt (36,000 km), but will not impact the Earth. The video simulates this trajectory and as the Earth approaches for a few moments it appears that an impact is likely. However, this is an illusion where the Earth merely dominates the field of view and the in-plane relative velocity is much larger than the horizontal relative velocity.
To be clear, the orbit of the asteroid as it enters the Earth's sphere of influence is a very high-energy hyperbolic orbit, where the closest point of approach is much further than 6,400 km (the Earth's radius).
Additional computing power isn't really needed for this problem. JPL already has the Standard Dynamic Model they use to model all bodies in the solar system accurately, and the current hardware is perfectly capable of handling the problem.
What is needed to refine and understand the trajectory is more observations. Radar range and range-rate measurements, along with optical angle measurements are fed together to estimate the current position and velocity, and using estimation techniques you can estimate your uncertainty as well. In order to bring down the uncertainty, we need more measurements that give a better statistical sample and allow you to have more confidence in your averages. Sadly most people don't have radio telescopes are large enough optical telescopes (20"+ preferably) to really make a good observation. For that reason, it will probably take till 2013, the next close approach, to get a new set of data that will make it easy to determine whether there is a 2036 impact risk.
While the headline should have explicitly stated "Oklahoma State University" rather than simply "OSU," I'm not sure how Ohio gets the right to the acronym. Growing up in Oklahoma and attending a Big 12 School, my assumption is that OSU refers to Oklahoma State.
And don't forget those kids in Oregon and their amusingly named mascot -- especially when playing USC.
Actually this pass (the 2029 close approach) is not a concern at all. The error brackets are brought in well enough that we know it will not impact the Earth, but will pass well within the GEO belt. What we don't know, and when the actual 1/250000 impact risk is, is the next pass, in 2036. If the asteroid passes through what is known as a 'gravitational keyhole' in 2029, the effect of Earth's gravity will actually swing the asteroid back around on an impact path in 2036.
Right now we can predict where Apophis will be in 2029 fairly well, within a few 10s of kilometers I believe. When you're talking about hitting the Earth, a thousand kilometers or more is good enough precision. The problem is that that during that flyby in 2029, any small uncertainties magnify by a few orders of magnitude, so when you carry it through another 7 years of orbits the uncertainty is 10s of Earth radii instead. Add in uncertainties about the effects of solar wind and the Yarkovsky effect and it just gets more complicated.
I realize that a lot of it is just good-hearted ribbing, but there seems to be a lot of discussion of PHP coders being bad coders, which is something I'm not sure I get. My background is as an aerospace engineer who ends up writing a lot of code, but I wouldn't call myself a developer, or even a *real* programmer by any stretch; I prefer to work in MATLAB, as its an industry standard and has built-in functions for much of what I need, but I've also written code in C, C++, C#, Objective C, Python, BASIC, JavaScript, probably some others, and yes, PHP.
I've been using PHP for a couple of web-based projects I'm working on, and I've enjoyed using it. It has some rough edges, particularly with inconsistent function naming, and 'duck' typing does lend itself to some poor practices if you're not careful, but its been great for many of the things I'm trying to do. It integrates well with Apache and is dead-simple to work with on Linux and OSX. It lets you do OO stuff where appropriate but doesn't force it down your throat (sometimes you just need a simple conditional to display a chunk of html). You can call class names from variables making it easy to do automated form generation without running through tedious case statements. It facilitates access to 'GET' and 'POST' data and has adaptable data structures and good string manipulation functionality. And of course it makes it easy to interface with the database.
Obviously, all of this leaves some room for abuse, and its certainly not perfect, but it doesn't mean its impossible to write decent PHP code. I like to think that with my projects I've done a decent job of keeping the code straightforward, maintainable and efficient -- moving database access into data model objects, keeping repeated segments of code, headers, etc. in distinct files so they only have to be modified once, using descriptive subroutine names so its obvious what the code is doing, planning out the overall program structure so I know how its all laid out in advance, and maintaining a MVC architecture where appropriate. The same kind of things you'd do to make the code 'decent' in any language.
Sorry if I'm missing something.
There's also less of a reason to deny it. Solving climate change involves significant lifestyle changes (at least in the short term) and a lot of investment by society. This is really why its so controversial. People want to keep driving their SUVs without feeling guilty, and switching all our power plants off of coal is a massive prospect. Solving an asteroid impact is a one-time cost of a few hundred million for a small one, or a few billion for a large one, and then people can go on about their lives as normal.
Actually, no. The *asteroid* 99942 Apophis is 200-300 meters in diameter and is well understood to be a threat of regional destruction, not world-wide devastation.
The real problem I see with this analysis is simply that our ability to track doesn't tell you where its going to hit, it will simply tell you theres a 25% of it hitting the Earth at all, and that one spot might be at the peak of the probability distribution (bell curve). To be sure it wouldn't hit you have to move it by many Earth-radii anyway, so you probably would never know that you just had it aimed at Russia.
Its a big deal because we're just now reaching the point where we know how to avert a potential impact. Since its a relatively cheap thing to do (~$300M for an Apophis-sized asteroid) it would be irresponsible not to consider the possibilities and have plans in place to handle the event. It's not a choice of one or the other.
There are a *LOT* of people in this world, all doing a lot of things. Having people work on finding solutions to a potential asteroid impact doesn't take effort away from solving global climate change. I work with NEA's, and while I'm not a climate change denier, if I weren't doing this I wouldn't be working on green technology instead. My skills are in spacecraft mission design, so if I weren't working on asteroids I'd be working on other missions instead -- and with unemployment at 10% we're not in a labor shortage right now either (believe me, this is hitting just graduated engineers as much as anyone else). The miniscule amount of money involved in NEA research would do nothing as far as climate change goes -- climate change is a political problem (requiring potential lifestyle changes to fix) rather than a monetary one anyway.
If you've got enough time, the gravity tractor provides an easier method as well. Since you're not landing you don't have to worry about the rotation (or the complex control systems required to land), and its just as efficient as far as propellant is concerned -- though it does prevent using the asteroid material as propellant.
Obviously, the main problem is that you're limited in the amount of force you can apply by the mass and standoff distance -- for small asteroids with plenty of warning its a no-brainer, for big asteroids the margins are tighter, but it can still be viable.
No need to be so cynical. You can deflect a smaller, Apophis-sized asteroid for around $300M with 15-20 years of warning. A planet killer could probably be handled with a couple of billion. In the scheme of space programs' budgets, this is a large but not unreasonable budget, and is exactly what they exist to do. Governments have sent missions to completely non-threatening asteroids for study already, so doing it with a save-the-Earth mission attached is a no-brainer (and the ability to alter trajectory via gravity tractor adds minimal complexity and weight above a pure exploration mission).
The harder part is finding and tracking the objects so that you have enough warning -- this is not nearly as sexy, but even then the US congress has already directed NASA to do so and provided at least some funding to support it. It helps that its cheaper. Though I can't say for sure I'd assume that other countries have similar programs.
While a gyro is necessary to actually do full 6-DOF position tracking (otherwise you must assume that you're holding a specific orientation... this can still be good for something like an in-the-air mouse), the Wii system still wouldn't be good for detecting absolute motion without the sensor bar as well.
The problem is that while the sensors are fairly precise as far as measuring the accelerations (if they're anything like the iPhone sensors they're around 0.02g precise), when you try and integrate them twice to get the position, things start to fall apart. Imagine you do a simple up-and-down motion. You get a sinusoidal acceleration curve that when you integrate it once gives you an offset sinusoid to represent your velocity, and a second integration gives a third one to represent your position. However, at the end, your integration to the velocity level comes out to be not quite zero, because those small acceleration errors will mostly cancel out, but not perfectly. This is still a pretty good velocity estimate, since its close to zero. However, as far as your position is concerned, close to zero and actually zero are very different, so you get a constant, growing drift in your position from a small velocity error. The same things apply to gyros, although the math is a little more complex.
Basically if you want to use a sensor as a double integrator it has to be extraordinarily precise, and even then you're going to get some drift that you have to remove every once in a while, or have an absolute position value to keep it in check (kalman filters do a great job of interpreting data from multiple sensors). What the sensor bar and IR sensors do is give you an incomplete but useful reference on position and orientation that you can use to keep that drift in check. Adding the gyros definitely helps a lot too, but you still need the sensor bar to keep drift in check.
And forgetting my high school science classes, averaging gives you precision, not accuracy. Accuracy is a whole other issue, but its not very different between cheap and expensive sensors, and calibration can eliminate the issues.
In my experience, doing some sensor systems with cheap sensors and expensive sensors, the difference is that cheap ones can be fast or accurate, while good ones can be both. Because you can average data and apply statistical methods to eliminate noise, a long integration time can get you very good precision (as long as your not doing an integrator... using an iPhone as a position sensor won't work, since you can't average the acceleration to get it).
In this case, a cheap sensor is going to work quite well unless your trying to collect usable data faster than about 1 Hz. Its really a matter of knowing what you need. In many cases a cheap sensor works really well.
Then I would argue that the failure mode is poor management and schedule rush -- definitely things not included in whatever safety numbers were quoted when the shuttle was being designed. The point is that whenever those 1 in 1000 numbers are pulled out they are almost meaningless -- the failures that did occur weren't included in those.
Its like judging the safety of a car on whether or not a freak string of events is likely to blow up the car on any given trip (or the brake lines fail, or your toyota accelerates without your command), when everyone knows that the most likely reason you're going to die in the car is because you or someone else screws up. While you want to do your best to keep the freak accidents from happening, more time needs to be spent making mistakes less damaging, and training people to avoid them.
To you the ultimate goal is scientific knowledge and resources.
To me the ultimate goal is human settlement beyond Earth.
To congresspeople the ultimate goal is getting reelected.
What you see as nationalistic chest thumping I see as (admittedly often poorly done) continued development of technology to support frontier development. They of course see it as jobs for their district. Conversations about how we should do things first require an agreement on the goals.
Three out of five flights, and the order matters significantly. The two that have been successful are significantly different than the three that failed. They added baffles to the tanks, improved the control algorithms, changed materials. They *FIXED* all of the issues that caused the early failures. Also, if its cheaper to blow up a few unmanned rockets than it is to design it perfectly the first time, then that sounds like the right way to do it. I'd consider the reliability of the Falcon 1 the same as any vehicle with a 2-0 record. Still not too reliable yet, but showing promise.
And those safety numbers are in so many ways bogus, since they only consider known failure modes. Everything thats ever killed an American astronaut was an unknown failure mode. Since Falcon 9 is intended for human use as well, with the same safety goals, and is in a further state of development than Ares 1, I can't help but be shocked by the sheer price of *just* Ares 1-X. If it were the entire Ares 1 program that had cost so much so far I'd say it was pretty reasonable and even cheap -- but no, just the aerodynamic and structural test cost that much.
No, only a few years, but its pretty clear that this is not in the best interest of furthering space exploration, but rather in keeping jobs in a few congressional districts -- namely Huntsville, Alabama. Marshall Space Flight Center stands the most to lose if Ares falls through, but MSFC is in many ways a dinosaur of the Apollo era and hasn't transitioned to being a leaner, more efficient group.
Consider this: for the cost of building Ares 1-X, the test-flight that consisted of a shuttle SRB with some dummy mass on top and made up to look like an Ares 1, what was essentially the worlds largest model rocket, cost $450M -- SpaceX, has developed one working rocket and has almost completed a larger one for around the same cost. While obviously the Ares program will cost more than what a company like SpaceX will spend, since they're building bigger rockets to do riskier things, there is something wrong when a mere model costs that much.
The problem with micromanaging NASA through congress is that the only districts where its an issue that can make a difference in an election are the ones where they want to maintain the status quo, which is not working well. Everyone else who sees it and disagrees with its handling probably aren't going to swing their vote based on it, since there are a myriad of other, more immediate things to consider as well.
In my experience even the ones with GPS require you to still know a bit -- the whole align by centering on the bright star thing never works as well as you'd like, at least in my experience. Magnetic anomalies, mount misalignments, or whatever else sometimes leave me wondering what star it is in the first place... this is particularly frustrating because Meade Autostar doesn't actually tell you what star it is you're looking for. My experience with the Meade 16" I work with has made me a little jaded on GOTO systems.
As far as dealing with sidereal tracking, while annoying, its not as bad on a manual equatorial mount as on an Alt-Az, because you can simply tell them to use the RA knob to keep it in view. On something with as wide a field of view as a 5" or 6" f/5 like the one I pointed out, the movement isn't too fast. This is more likely to be a bigger problem on big Dobsonian. Also, you can get a RA drive for that particular Celestron, which helps the issue.
I guess really in the end it comes down to a combination of factors -- budget, knowledgability of the parent, what you're hoping to see, patience of the kids, whether you expect to expand to more complex things like astrophotography, etc. etc.
Exactly right. Of course that doesn't necessarily mean it has to be expensive. A big (~10") Dobsonian like the one someone else mentioned is nice.
Personally, what I have is a 5" newtonian, the Celestron AstroMaster 130EQ, $180 on Telescopes.com, that I really love, even though I have access to a massive 16" Meade monster for my job. Its small enough to carry easily, but big enough to give you pretty good views of planets, clusters, some of the brighter nebulae, and affordable even on a grad student's stipend. It won't show you as much as a big Dobsonian, but its on a manual equatorial mount, so its a lot better for learning how to find your way around the sky. Also, you can expand it to use the RA motor drive ($40) and a CCD to do astronomical imaging (either use a DSLR or some $300 passively cooled CCDs) -- not as accessible to beginners, but potentially more rewarding in the long run.
Definitely go with a manual mount though, theres something rewarding about finding what you're looking for yourself over just punching things into a keypad. Its cheaper too, so you can get more aperture for the money.