I think you're doing it wrong. I'm getting a masters degree in aerospace engineering, and I definitely feel like I can learn and do other things. I spend a lot of time on classwork and research, but I value my free time and make sure I have some. Currently I'm working on my cooking skills and learning to do web development, in addition to other time consuming activities like keeping up with current events, working with some non-profits, and spending time with my girlfriend and other friends. This doesn't include hobby astronomy I'm also doing, because its got some crossover with my research so its hard to say whats work and what isn't. I've also found that having a 'real job' (i.e. 8-5) doesn't make it harder to keep up an outside life as well (both from internships and from the fact that I come from a family of engineers).
Additionally, I like to think that my well-roundedness will make my career better. Some will directly: keeping up with current events and modern astronomy makes me better able to understand the political and scientific considerations behind projects I'll work on, and my non-profit work is great for networking and getting to know whats out there. Others will indirectly: things I enjoy like spending time with friends/girlfriend, cooking and reading make my life more enjoyable, and quite frankly give me a better attitude that means I'll be more productive at work.
Even in upper-level undergrad, which seems to be the single most miserable and time-consuming part of any engineering career, I always was able to make the time to do other things with my life -- and I'm definitely one of those engineers who loves doing it, so its not like I'm desperate to get away from it either. Most engineers I know who's work I'd actually trust and like to work with also tend to be fairly well-rounded.
What about weeklies (Time, Newsweek, the Economist, etc.)?
I'm the type who prefers the quick fix of internet news to see if anything big has happened during the day, and a paying for a daily paper doesn't make sense for me, because I don't necessarily have time every day for anything more than a quick skim. However, I have subscriptions to the those three magazines, and I can usually find the time to read through those throughout the week.
If you separate the time-sensitive immediate news, provided by wire services on the internet, from the deeper analysis and real journalism from weekly publications, I think you may have a workable model that provides the benefits of both.
$17B a year is not going to make a dent in the economy or in poverty or homelessness, or climate change or anything else. Those are the results of human nature and/or normal cycles, and fixing them is a matter of political will and good policy, not a few extra dollars.
Spending a small amount on space exploration is EXACTLY what the government exists to do -- do things that require large amounts of money (for an individual or group) with high risks and low immediate reward, but that have the potential for great reward for all of society.
And if you think $17B a year with increases less than inflation and ever new directives and goals are 'endless resources' I think you need to take a look at the scale of the federal budget.
Of course it looks the same. Physics and engineering principles are the same now as they were 50 years ago. Stacked stages and capsules are simple and effective. This is meant to deliver astronauts to orbit, and the Ares V and other heavy lift concepts are meant to deliver large amounts of mass to orbit. The shuttle was meant to take everything and everyone up all at once, and be able to bring it back and do very large cross-range landings -- these all made the shuttle what it was. With the first two requirements being separated, and the last two being proven unnecessary, it makes since to go back to something simpler.
This isn't to say its all the same as Apollo era technology. New manufacturing methods, better computers, and sheer experience mean that any well-designed new vehicles are going to be more flexible, cheaper, safer and lighter, even if they look the same on the surface.
Thanks for that quote. As a Christian who does astronomy* it has a lot of appeal to me -- quotes like that from respected religious figures (as opposed to from say, Einstein) are good to know.
* I'm an engineering student setting up and using astronomical equipment, so while I get paid to do astronomy, my scientific work is at the dedicated amateur level, so I can't exactly claim to be an Christian astronomer.
The entire point and purpose of this launch is to receive telemetry from the large number of instruments on the vehicle. Being able to compare the data to the models is the single most valuable thing that will come from this launch, since otherwise its mostly a publicity stunt (everything that is not a shuttle SRB is a dummy component) and theres a good chance that Ares 1 will be canceled.
With clouds and the potential to build up an electrical charge on the surface, there's a risk of losing data along the way. The cost of postponing the launch is far less than the cost of wasting this $450M launch.
While the locations of development and construction are based largely on political reasons, the selection of launch sites is not. When selecting a launch site, you want something as far as south as possible with water along the direction of travel. You want to go south because you get more speed from the Earth's rotation and you can get into lower inclination orbits (you can't go lower than the launch site latitude. You want to fly over water so that if there's an accident you're not raining flaming debris on people. Downrange is quite far so ocean is still preferable to desert.
For most satellites you want to launch East, so you get the speed boost from Earth's rotation, so the east coast of Florida is pretty ideal. For polar orbits (Earth science primarily) you launch north or south, so Vandenberg on the south end of California is a good choice. Weather is a problem in Florida, but I don't think flying rockets over New Mexico and Texas is going to be a viable strategy.
The Augustine commission offered the administration about 10 options, some of which continue Ares V development, while others don't. All options that remain within the current budget (not the extra $3B required to do anything impressive according to the report) continue Ares V development.
However, all of the options presented push for a heavy lift capability. Other options include - 'Ares V Lite': a lower-performance version of Ares V that would be human rated and could potentially reduce development costs primarily by eliminating the need for Ares 1 - Shuttle-derived: Either a sidemount cargo vehicle (probably requiring something like an Ares 1 for crew launch), or a top-mount shuttle derived design like Jupiter. These would be less capable than Ares V, but still powerful and potentially cheaper -- you could achieve a lunar mission with 2 or 3 launches. - EELVs: Creating larger Evolved Expendable Launch Vehicles from the Delta or Atlas family. These would be the least capable. These are also the biggest question mark because cost savings would come in a large part from a restructuring of rocket development to a DoD style model, where contractors are given requirements, not designs.
All of these, in combination with various targets and schedules were analyzed by the committee. None of the options comes out as a clear winner as cheaper or better, since Ares V has some considerable sunk costs that make its cheaper relative to the others, while designing even a sidemount cargo pod is more expensive than some probably think. Personally I like EELVs because it forces a change in the way business is done, but thats me.
There are a lot of steps between Ares 1-X, and an actual Ares 1 + Orion that can take people to orbit. Augustine and crew say 2017 before that happens, and they seem to have a good idea of what they're talking about.
Whats on the pad now is largely a publicity stunt -- especially with the future of Ares 1 itself in doubt. Its a 4-segment SRB with a dummy 5th segment, and a dummy second stage and Orion capsule. The fact that the SRB is different means it doesn't represent the vibrations and harmonics of the actual vehicle well. The upper parts are still under development as well.
Of course, its not a worthless test. Its ridiculously well instrumented, thus why weather matters even though its a suborbital lob, and it has also been modeled extensively. Being able to compare models to actual data on this scale is quite valuable. Probably not worth the $450M this launch will cost in total, but probably worth the remaining cost in the recent decisions to go ahead and continue the launch even with Ares 1 in doubt.
Thats a valid view, but voting with your wallet has one big flaw that I see. I don't think the failure of the Kindle would have told publishers that DRM for e-books is a loser -- it would have told them that e-books are losers. And like the poster you're responding to, I find e-books very convenient.
Personally, I'm hoping that competition and publisher discomfort with a dominant distributor will eventually bring an end to DRM here, just as it did for digital music.
As someone who never bought a DRM-laden piece of music, but buys plenty of stuff for my Kindle (but was never one to rant much about it), the reason is simply one of practicality.
I'm in grad school, have a small room, move a lot, and tend to fulfill some of those 'digital nomad' stereotypes, so the benefits of e-books are pretty strong for me -- however, there is no way to purchase DRM-free e-books without extremely limiting my choices. I figure that by purchasing and using the device, as its useful for me and I feel informed what the DRM implies, I can help to show that there is a market, and that more competition will force more openness, as it did in the music industry.
Music had two critical differences to me. One was that I could purchase a CD and rip it with little effort (I still prefer to purchase music by album, so single-serve songs meant little to me) -- this made it easy to get most of the benefits without the DRM (plus ripping to FLAC). The second is repeatability and cost/length: buying a new copy of an album every year just to relisten to is absurd, while if I were to decide to reread a book 5 years from now, it doesn't seem as ridiculous to rebuy it, thus making the DRM-associated risk less.
That said, first DRM-free e-book store that appears with a comparable selection, I'll jump to immediately, just as I started using the Amazon MP3 store as soon as it appeared.
I mean, not to say you don't have a right to be annoyed with some features, but your complaints strike me as odd.
- What's wrong with having a documents folder? It seems to me to be the best way to do things (having all your work in one easy to back up folder), and as far as I can tell the only difference between XP and Vista is that they took out the 'My'. Also OS/X and presumably other linux distros do the same thing -- what OS has a better model?
- Requiring root priveleges to perform administrative tasks was a primary feature of UNIX system from the beginning, long before the abbreviation UAC was ever uttered. It also proves to be good security, since it prevents code that can alter the core system from being run without user intervention (preventing many viruses and trojans).
- Defaulting to the desktop is a reasonable default. Its easy to find and use. If you want something different you can change it yourself. I also don't see how its copying Windows, as far as I know Firefox was the first place I saw that particular behavior.
I guess what I feel is that while some of the defaults may be designed to make it easier for Windows users to switch, the real power comes from the fact that you have a lot more range to try new things with Linux.
Amusing since the report does explicitly discuss safety. Particularly it focuses on two topics: it considers astronaut safety as sine qua non, and says that its impossible to predict 'infant mortality' safety of any launch vehicle.
The first means that they simply refuse any plan that can't be done with a strong expectation of safety.
The second means that its impossible to analyze the initial safety of a paper rocket, and personally I'd love to see the Senator try. You can analyze safety from a PRA perspective (analyzing likelihood of failed components), as opposed to unexpected failure modes, but the report points out that PRA failures have never brought down a manned spacecraft. Unexpected failure modes are just that, unexpected, and as such can't be analyzed properly.
If you consider the full 300M population* and only the cost of human space flight (about half of the total NASA budget) then the numbers line up pretty well.
Of course, considering that the polls the report refers to refer to both manned and unmanned exploration, this seems slightly dishonest, but not ridiculously out of line.
* If you're going to be restricting it to those paying taxes, you should probably also consider the income distribution of those paying taxes, and that the median tax load is going to be less than the average tax load, as far as I know
The first time its proving the science of the fundamental principals at incredible risk, and is a prime fit for government development. The second time its just trying to engineer it better and cheaper -- a better job for competitive enterprise.
You don't need anything as complicated as a genetic algorithm. You have a defined control (thermostat), a defined state (temperature), some external but relatively predictable variables (outside temperature and server load), and a decent model (should be a straightforward ODE) that defines the relationship between those. Define a cost function, balancing the need to keep both temperatures and energy costs low, and you've got a very straightforward optimal control problem.
Because its continuous, not particularly chaotic, and has a limited number of states and controls, basic Lagrangian optimal control methods should be very effective -- they also are little more than solving a set ODEs, and with low precision requirements its computationally easy. GAs are great for very complex problems with discrete components or highly non-linear chaotic components, but are overkill for this kind of problem.
Despite the FUD, you can do that with a Kindle as well. PDF is an issue, because it really has to reflow the text, or its painful to read -- thus why it relies on the email-based kludge (if B&N can come up with a good solution, more power to them).
Other than that, take a.txt,.rtf, drm-free epub file, etc. and drop it onto the Kindle disk (it shows up like any other flash driver), and its there to be read on the home screen.
A PDF is a fixed-page format, not a reflowable format. Since most of those pages are fixed at 8.5x11", and the Kindle screen is about a third the size, has marginal DPI, and a slow refresh rate (making scrolling difficult) its not really a good choice. I've gotten a couple of PDF conversions that go through as pure images, and its not pleasant. From what I've heard using PDFs on Sony's reader with a similar format isn't especially pleasant either. This is also why its enabled on the Kindle DX (bigger screen).
Should Amazon have put it as an option? I think so, but the existence of a PDF would not make up for a copy of a work in a good reflowable format (ePub or even plain text).
As with the Kindle, if you buy from a DRM-based store, then you'll have DRM'd material, and if you get it from somewhere without DRM, you can read those too without any problems. This does seem to allow PDF reading, but I wonder how they deal with the fact that its really hard to read a letter sized page on a screen that small with that DPI (the reason amazon decided to can that feature in the first place). This is just like how you could use the iPod pr any PlaysForSure compatible player with music ripped from any source, it was simply the various music stores that brought DRM into the picture.
Certainly looks like a nice device, the idea of the touch-screen instead of the keyboard is particularly interesting. However, in the end, its going to be content thats critical -- they claim millions of titles, and if thats true, and if its things people want it will succeed. On the DRM front, the store is again critical -- the existence of 'lending' implies theres some form of content control (and I won't be surprised if Amazon tries to emulate it, if its not patent encumbered).
Just remember, on all E-readers, you're not restricted to the store. If you have DRM-free content, you can use it. DRM only shows up in the equation when you consider the store, and this doesn't appear to do anything to change that. The fact that you can't easily rip a book like you could a CD makes it more of a pain though.
A PhD is not and should not be a requirement to engineering design jobs. I don't know anyone who's ever claimed it was. I know good engineers with a mere bachelor's degree, and bad engineers with a PhD.
What a PhD implies is an ability and proclivity for basic research. For that ability to dig deep into a fundamental problem and push the boundaries of knowledge, and spend years on research that may at times prove to be a dead end. These guys don't seem to accomplish significant things, because they're behind the scenes. But I imagine (CS is not my forte) that Bill Gates and Linus Torvalds would not have done nearly as well if there hadn't been some PhD's working on operating system theory work that was never practical on its own. I know in my field (aerospace controls), all the practical implementations critically depend on work done by PhDs years ago on things like state-space control structures and the Kalman filter.
Finished products, what you see as significant accomplishments, sit at the end of a long process that depends on a lot of people. Some things really require the research experience a PhD implies, and most things that don't won't ask for it. In fact for many design positions PhD candidates are often considered overqualified (i.e. too expensive).
I think reverse applies in this case because its not US-native students and professionals leaving to go to other places, but rather that the visitors aren't staying as often. Not really an ideal descriptions, but it does effectively imply a difference from what you'd first think of with 'brain drain', and it works pretty well if you consider the original 'brain drain' being us taking the best and brightest from other countries..
Of course, whether or not that is a good or bad thing I leave as an exercise for the reader -- my bias is towards more globalism, but I'm an engineer and not an economist, so I'll leave it for more knowledgeable people to discuss.
Go write it then. You've got the journal article to work off of, which should be all the citations you need, since I think this is the definitive work on the subject right now.
Its really all defined by Planck's law (http://en.wikipedia.org/wiki/Planck's_law), where the Sun's temperature is 5700 K. Of course theres some differences where the atmosphere has absorption bands, but the peak energy is centered around 480 nm, and most of the higher wavelengths (UV) are blocked by the atmosphere.
The whole reason visible light is visible to us is that its the most energetic, and thus easiest to see. The fact that the atmosphere blocks is also why UV burns us.
You've identified the main problem -- but its a problem thats solvable. Without any kind of calibration its accurate to around 2 degrees, and with a 6-measurement linear calibration I can get it down to under 1 degree. Calibration is fairly straightforward, optimizing the coefficients to give gravity vector magnitudes of unity in all directions. Unfortunately, at least for my telescope, I need less than half a degree to get it within the field of view with a good wide-field eyepiece. In order to do this I figure a quadratic or cubic calibration is needed, but my matrix inversion routines are very basic and aren't fast enough for 9x9 or 12x12 matrices -- fixing them is where I stopped and need to pick up again. So basically I haven't proven the accuracy is enough, but I think it is.
Precision/repeatability is another issue. A with any sensor, you can improve it by integrating for longer. Of course, with a consumer application you don't want to have to think about it much, so I figured out a way to do a multi-tiered low-pass filter that resets the initialized values when it recognizes that its moving, allowing a fast response but fairly high precision. I also make it so you can reset the filter manually, to make sure you're really where you think you are. I'm personally very happy with the way that part has worked out, and as long as you don't need a fast response and a high precision at the same time, its pretty good (using it as an IMU is basically impossible).
As a note, since you seem to have some idea of the details, up till now I've only been messing with the accelerometer for this, and as such its restricted to use with equatorial mounts -- obviously, a gravity sensor can't detect rotations about the z-axis, and thus is pretty useless on an Alt-Az mount. Hopefully, after integrating magnetometer measurements I'll be able to make it work on all kinds of mounts, and hopefully improve precision at the same time (doubling the number of samples for the same amount of time).
However, I'd also say that its probably a similar sensor to any other consumer level product, so I seriously doubt you couldn't do the same thing on a G1 or any similar phone. Once I get the thing finished, and if there seems to be much interest, I'll put the algorithms out there for others to use.
Slashdot Mirror
I think you're doing it wrong. I'm getting a masters degree in aerospace engineering, and I definitely feel like I can learn and do other things. I spend a lot of time on classwork and research, but I value my free time and make sure I have some. Currently I'm working on my cooking skills and learning to do web development, in addition to other time consuming activities like keeping up with current events, working with some non-profits, and spending time with my girlfriend and other friends. This doesn't include hobby astronomy I'm also doing, because its got some crossover with my research so its hard to say whats work and what isn't. I've also found that having a 'real job' (i.e. 8-5) doesn't make it harder to keep up an outside life as well (both from internships and from the fact that I come from a family of engineers).
Additionally, I like to think that my well-roundedness will make my career better. Some will directly: keeping up with current events and modern astronomy makes me better able to understand the political and scientific considerations behind projects I'll work on, and my non-profit work is great for networking and getting to know whats out there. Others will indirectly: things I enjoy like spending time with friends/girlfriend, cooking and reading make my life more enjoyable, and quite frankly give me a better attitude that means I'll be more productive at work.
Even in upper-level undergrad, which seems to be the single most miserable and time-consuming part of any engineering career, I always was able to make the time to do other things with my life -- and I'm definitely one of those engineers who loves doing it, so its not like I'm desperate to get away from it either. Most engineers I know who's work I'd actually trust and like to work with also tend to be fairly well-rounded.
What about weeklies (Time, Newsweek, the Economist, etc.)?
I'm the type who prefers the quick fix of internet news to see if anything big has happened during the day, and a paying for a daily paper doesn't make sense for me, because I don't necessarily have time every day for anything more than a quick skim. However, I have subscriptions to the those three magazines, and I can usually find the time to read through those throughout the week.
If you separate the time-sensitive immediate news, provided by wire services on the internet, from the deeper analysis and real journalism from weekly publications, I think you may have a workable model that provides the benefits of both.
$17B a year is not going to make a dent in the economy or in poverty or homelessness, or climate change or anything else. Those are the results of human nature and/or normal cycles, and fixing them is a matter of political will and good policy, not a few extra dollars.
Spending a small amount on space exploration is EXACTLY what the government exists to do -- do things that require large amounts of money (for an individual or group) with high risks and low immediate reward, but that have the potential for great reward for all of society.
And if you think $17B a year with increases less than inflation and ever new directives and goals are 'endless resources' I think you need to take a look at the scale of the federal budget.
Of course it looks the same. Physics and engineering principles are the same now as they were 50 years ago. Stacked stages and capsules are simple and effective. This is meant to deliver astronauts to orbit, and the Ares V and other heavy lift concepts are meant to deliver large amounts of mass to orbit. The shuttle was meant to take everything and everyone up all at once, and be able to bring it back and do very large cross-range landings -- these all made the shuttle what it was. With the first two requirements being separated, and the last two being proven unnecessary, it makes since to go back to something simpler.
This isn't to say its all the same as Apollo era technology. New manufacturing methods, better computers, and sheer experience mean that any well-designed new vehicles are going to be more flexible, cheaper, safer and lighter, even if they look the same on the surface.
Thanks for that quote. As a Christian who does astronomy* it has a lot of appeal to me -- quotes like that from respected religious figures (as opposed to from say, Einstein) are good to know.
* I'm an engineering student setting up and using astronomical equipment, so while I get paid to do astronomy, my scientific work is at the dedicated amateur level, so I can't exactly claim to be an Christian astronomer.
The entire point and purpose of this launch is to receive telemetry from the large number of instruments on the vehicle. Being able to compare the data to the models is the single most valuable thing that will come from this launch, since otherwise its mostly a publicity stunt (everything that is not a shuttle SRB is a dummy component) and theres a good chance that Ares 1 will be canceled.
With clouds and the potential to build up an electrical charge on the surface, there's a risk of losing data along the way. The cost of postponing the launch is far less than the cost of wasting this $450M launch.
While the locations of development and construction are based largely on political reasons, the selection of launch sites is not. When selecting a launch site, you want something as far as south as possible with water along the direction of travel. You want to go south because you get more speed from the Earth's rotation and you can get into lower inclination orbits (you can't go lower than the launch site latitude. You want to fly over water so that if there's an accident you're not raining flaming debris on people. Downrange is quite far so ocean is still preferable to desert.
For most satellites you want to launch East, so you get the speed boost from Earth's rotation, so the east coast of Florida is pretty ideal. For polar orbits (Earth science primarily) you launch north or south, so Vandenberg on the south end of California is a good choice. Weather is a problem in Florida, but I don't think flying rockets over New Mexico and Texas is going to be a viable strategy.
The Augustine commission offered the administration about 10 options, some of which continue Ares V development, while others don't. All options that remain within the current budget (not the extra $3B required to do anything impressive according to the report) continue Ares V development.
However, all of the options presented push for a heavy lift capability. Other options include
- 'Ares V Lite': a lower-performance version of Ares V that would be human rated and could potentially reduce development costs primarily by eliminating the need for Ares 1
- Shuttle-derived: Either a sidemount cargo vehicle (probably requiring something like an Ares 1 for crew launch), or a top-mount shuttle derived design like Jupiter. These would be less capable than Ares V, but still powerful and potentially cheaper -- you could achieve a lunar mission with 2 or 3 launches.
- EELVs: Creating larger Evolved Expendable Launch Vehicles from the Delta or Atlas family. These would be the least capable. These are also the biggest question mark because cost savings would come in a large part from a restructuring of rocket development to a DoD style model, where contractors are given requirements, not designs.
All of these, in combination with various targets and schedules were analyzed by the committee. None of the options comes out as a clear winner as cheaper or better, since Ares V has some considerable sunk costs that make its cheaper relative to the others, while designing even a sidemount cargo pod is more expensive than some probably think. Personally I like EELVs because it forces a change in the way business is done, but thats me.
There are a lot of steps between Ares 1-X, and an actual Ares 1 + Orion that can take people to orbit. Augustine and crew say 2017 before that happens, and they seem to have a good idea of what they're talking about.
Whats on the pad now is largely a publicity stunt -- especially with the future of Ares 1 itself in doubt. Its a 4-segment SRB with a dummy 5th segment, and a dummy second stage and Orion capsule. The fact that the SRB is different means it doesn't represent the vibrations and harmonics of the actual vehicle well. The upper parts are still under development as well.
Of course, its not a worthless test. Its ridiculously well instrumented, thus why weather matters even though its a suborbital lob, and it has also been modeled extensively. Being able to compare models to actual data on this scale is quite valuable. Probably not worth the $450M this launch will cost in total, but probably worth the remaining cost in the recent decisions to go ahead and continue the launch even with Ares 1 in doubt.
The rockets are male. The spacecraft that ride on top of them are female.
Thats a valid view, but voting with your wallet has one big flaw that I see. I don't think the failure of the Kindle would have told publishers that DRM for e-books is a loser -- it would have told them that e-books are losers. And like the poster you're responding to, I find e-books very convenient.
Personally, I'm hoping that competition and publisher discomfort with a dominant distributor will eventually bring an end to DRM here, just as it did for digital music.
As someone who never bought a DRM-laden piece of music, but buys plenty of stuff for my Kindle (but was never one to rant much about it), the reason is simply one of practicality.
I'm in grad school, have a small room, move a lot, and tend to fulfill some of those 'digital nomad' stereotypes, so the benefits of e-books are pretty strong for me -- however, there is no way to purchase DRM-free e-books without extremely limiting my choices. I figure that by purchasing and using the device, as its useful for me and I feel informed what the DRM implies, I can help to show that there is a market, and that more competition will force more openness, as it did in the music industry.
Music had two critical differences to me. One was that I could purchase a CD and rip it with little effort (I still prefer to purchase music by album, so single-serve songs meant little to me) -- this made it easy to get most of the benefits without the DRM (plus ripping to FLAC). The second is repeatability and cost/length: buying a new copy of an album every year just to relisten to is absurd, while if I were to decide to reread a book 5 years from now, it doesn't seem as ridiculous to rebuy it, thus making the DRM-associated risk less.
That said, first DRM-free e-book store that appears with a comparable selection, I'll jump to immediately, just as I started using the Amazon MP3 store as soon as it appeared.
I mean, not to say you don't have a right to be annoyed with some features, but your complaints strike me as odd.
- What's wrong with having a documents folder? It seems to me to be the best way to do things (having all your work in one easy to back up folder), and as far as I can tell the only difference between XP and Vista is that they took out the 'My'. Also OS/X and presumably other linux distros do the same thing -- what OS has a better model?
- Requiring root priveleges to perform administrative tasks was a primary feature of UNIX system from the beginning, long before the abbreviation UAC was ever uttered. It also proves to be good security, since it prevents code that can alter the core system from being run without user intervention (preventing many viruses and trojans).
- Defaulting to the desktop is a reasonable default. Its easy to find and use. If you want something different you can change it yourself. I also don't see how its copying Windows, as far as I know Firefox was the first place I saw that particular behavior.
I guess what I feel is that while some of the defaults may be designed to make it easier for Windows users to switch, the real power comes from the fact that you have a lot more range to try new things with Linux.
Amusing since the report does explicitly discuss safety. Particularly it focuses on two topics: it considers astronaut safety as sine qua non, and says that its impossible to predict 'infant mortality' safety of any launch vehicle.
The first means that they simply refuse any plan that can't be done with a strong expectation of safety.
The second means that its impossible to analyze the initial safety of a paper rocket, and personally I'd love to see the Senator try. You can analyze safety from a PRA perspective (analyzing likelihood of failed components), as opposed to unexpected failure modes, but the report points out that PRA failures have never brought down a manned spacecraft. Unexpected failure modes are just that, unexpected, and as such can't be analyzed properly.
If you consider the full 300M population* and only the cost of human space flight (about half of the total NASA budget) then the numbers line up pretty well.
Of course, considering that the polls the report refers to refer to both manned and unmanned exploration, this seems slightly dishonest, but not ridiculously out of line.
* If you're going to be restricting it to those paying taxes, you should probably also consider the income distribution of those paying taxes, and that the median tax load is going to be less than the average tax load, as far as I know
No, its rocket engineering.
The first time its proving the science of the fundamental principals at incredible risk, and is a prime fit for government development. The second time its just trying to engineer it better and cheaper -- a better job for competitive enterprise.
You don't need anything as complicated as a genetic algorithm. You have a defined control (thermostat), a defined state (temperature), some external but relatively predictable variables (outside temperature and server load), and a decent model (should be a straightforward ODE) that defines the relationship between those. Define a cost function, balancing the need to keep both temperatures and energy costs low, and you've got a very straightforward optimal control problem.
Because its continuous, not particularly chaotic, and has a limited number of states and controls, basic Lagrangian optimal control methods should be very effective -- they also are little more than solving a set ODEs, and with low precision requirements its computationally easy. GAs are great for very complex problems with discrete components or highly non-linear chaotic components, but are overkill for this kind of problem.
Despite the FUD, you can do that with a Kindle as well. PDF is an issue, because it really has to reflow the text, or its painful to read -- thus why it relies on the email-based kludge (if B&N can come up with a good solution, more power to them).
Other than that, take a .txt, .rtf, drm-free epub file, etc. and drop it onto the Kindle disk (it shows up like any other flash driver), and its there to be read on the home screen.
A PDF is a fixed-page format, not a reflowable format. Since most of those pages are fixed at 8.5x11", and the Kindle screen is about a third the size, has marginal DPI, and a slow refresh rate (making scrolling difficult) its not really a good choice. I've gotten a couple of PDF conversions that go through as pure images, and its not pleasant. From what I've heard using PDFs on Sony's reader with a similar format isn't especially pleasant either. This is also why its enabled on the Kindle DX (bigger screen).
Should Amazon have put it as an option? I think so, but the existence of a PDF would not make up for a copy of a work in a good reflowable format (ePub or even plain text).
As with the Kindle, if you buy from a DRM-based store, then you'll have DRM'd material, and if you get it from somewhere without DRM, you can read those too without any problems. This does seem to allow PDF reading, but I wonder how they deal with the fact that its really hard to read a letter sized page on a screen that small with that DPI (the reason amazon decided to can that feature in the first place). This is just like how you could use the iPod pr any PlaysForSure compatible player with music ripped from any source, it was simply the various music stores that brought DRM into the picture.
Certainly looks like a nice device, the idea of the touch-screen instead of the keyboard is particularly interesting. However, in the end, its going to be content thats critical -- they claim millions of titles, and if thats true, and if its things people want it will succeed. On the DRM front, the store is again critical -- the existence of 'lending' implies theres some form of content control (and I won't be surprised if Amazon tries to emulate it, if its not patent encumbered).
Just remember, on all E-readers, you're not restricted to the store. If you have DRM-free content, you can use it. DRM only shows up in the equation when you consider the store, and this doesn't appear to do anything to change that. The fact that you can't easily rip a book like you could a CD makes it more of a pain though.
A PhD is not and should not be a requirement to engineering design jobs. I don't know anyone who's ever claimed it was. I know good engineers with a mere bachelor's degree, and bad engineers with a PhD.
What a PhD implies is an ability and proclivity for basic research. For that ability to dig deep into a fundamental problem and push the boundaries of knowledge, and spend years on research that may at times prove to be a dead end. These guys don't seem to accomplish significant things, because they're behind the scenes. But I imagine (CS is not my forte) that Bill Gates and Linus Torvalds would not have done nearly as well if there hadn't been some PhD's working on operating system theory work that was never practical on its own. I know in my field (aerospace controls), all the practical implementations critically depend on work done by PhDs years ago on things like state-space control structures and the Kalman filter.
Finished products, what you see as significant accomplishments, sit at the end of a long process that depends on a lot of people. Some things really require the research experience a PhD implies, and most things that don't won't ask for it. In fact for many design positions PhD candidates are often considered overqualified (i.e. too expensive).
I think reverse applies in this case because its not US-native students and professionals leaving to go to other places, but rather that the visitors aren't staying as often. Not really an ideal descriptions, but it does effectively imply a difference from what you'd first think of with 'brain drain', and it works pretty well if you consider the original 'brain drain' being us taking the best and brightest from other countries..
Of course, whether or not that is a good or bad thing I leave as an exercise for the reader -- my bias is towards more globalism, but I'm an engineer and not an economist, so I'll leave it for more knowledgeable people to discuss.
Go write it then. You've got the journal article to work off of, which should be all the citations you need, since I think this is the definitive work on the subject right now.
Its really all defined by Planck's law (http://en.wikipedia.org/wiki/Planck's_law), where the Sun's temperature is 5700 K. Of course theres some differences where the atmosphere has absorption bands, but the peak energy is centered around 480 nm, and most of the higher wavelengths (UV) are blocked by the atmosphere.
The whole reason visible light is visible to us is that its the most energetic, and thus easiest to see. The fact that the atmosphere blocks is also why UV burns us.
You've identified the main problem -- but its a problem thats solvable. Without any kind of calibration its accurate to around 2 degrees, and with a 6-measurement linear calibration I can get it down to under 1 degree. Calibration is fairly straightforward, optimizing the coefficients to give gravity vector magnitudes of unity in all directions. Unfortunately, at least for my telescope, I need less than half a degree to get it within the field of view with a good wide-field eyepiece. In order to do this I figure a quadratic or cubic calibration is needed, but my matrix inversion routines are very basic and aren't fast enough for 9x9 or 12x12 matrices -- fixing them is where I stopped and need to pick up again. So basically I haven't proven the accuracy is enough, but I think it is.
Precision/repeatability is another issue. A with any sensor, you can improve it by integrating for longer. Of course, with a consumer application you don't want to have to think about it much, so I figured out a way to do a multi-tiered low-pass filter that resets the initialized values when it recognizes that its moving, allowing a fast response but fairly high precision. I also make it so you can reset the filter manually, to make sure you're really where you think you are. I'm personally very happy with the way that part has worked out, and as long as you don't need a fast response and a high precision at the same time, its pretty good (using it as an IMU is basically impossible).
As a note, since you seem to have some idea of the details, up till now I've only been messing with the accelerometer for this, and as such its restricted to use with equatorial mounts -- obviously, a gravity sensor can't detect rotations about the z-axis, and thus is pretty useless on an Alt-Az mount. Hopefully, after integrating magnetometer measurements I'll be able to make it work on all kinds of mounts, and hopefully improve precision at the same time (doubling the number of samples for the same amount of time).
However, I'd also say that its probably a similar sensor to any other consumer level product, so I seriously doubt you couldn't do the same thing on a G1 or any similar phone. Once I get the thing finished, and if there seems to be much interest, I'll put the algorithms out there for others to use.