Try dealing with time when you're actually doing rocket science! Leap seconds, measurable light travel time, and variations due to relativity make for incredible confusion.
Sadly I don't think most people realize how complicated timekeeping is even in terrestrial settings. The fact that Apple and MS face timing bugs is no surprise at all.
That requires Apple to have a monopolistic hold over a market. While right now they do on the tablet market, its only because we have only begun to see worthy Android competitors. I'm pretty sure Android on the tablet will compete with the iPad as well as Android on the phone competes with the iPhone.
The risk for Apple isn't antitrust, its News Corp et. al. being able to get better deals on other platforms. This seems good for everyone -- its that mythical market at work.
Well, as long as the mass of the object is negligible compared to the body its orbiting (in this case, the asteroid is absurdly smaller than the sun), then the mass of the body has no effect on its orbit. More specifically, the force is proportional to the mass so that the acceleration is constant regardless of mass.
The only time it will make a difference is for non-gravitational forces like solar radiation pressure or atmospheric drag. Then the surface area of the spacecraft will be much smaller than the asteroid, so it won't be a big deal, unless you're trying to change it on purpose.
Also, you never want to try and orbit a small asteroid like this. Their gravity fields are way too weird and unpredictable and you can end up crashing after a few revs of what starts out looking stable. Preferably you either land, get far enough away that the gravity is just a perturbing force, or do the standoff required for gravity tractoring.
Those analyses are relatively easy to do, and you're not going to change its trajectory significantly unless you really mean to.
In order to do it with a ~500 kg spacecraft you have to hover about 200 meters away from the asteroid with your engines thrusting essentially continuously for around a year. If you're not close enough to have to worry about hovering, or if you're trying to do a landing, no reasonably sized spacecraft is going to make a difference.
We have a good idea of the composition of asteroids in general, from meteorites, planetary formation models, etc.
We believe Apophis is chondritic because based on its apparent brightness and the way that brightness varies, we have a decent estimate of its size and albedo. If it had a different albedo it would indicate a different composition.
Of course, as with all remote observations based on a lot of educated guesses, there is a chance its wrong. However, if it is its probably a pathological case we could never gotten right, and that would make it even more interesting to visit.
Statistically certain is shorthand for saying the likelihood of the event happening is below some arbitrarily small level. At least its well understood to mean that amongst those I work with, and everything we do has a quantified uncertainty attached to it.
Actually, from what i gathered from TFS, this specifically doesn't require it. It simply provides an extra bit of warning if it's able to detect something bad about to happen. It still requires the driver to pay attention.
Based on that, i would say that until nearly every car has it will definitely increase safety. If you can't depend on it to work with every car, you have to pay just as much attention, but it will still give warnings on occasion.
Of course once you get to the point of having it in most every vehicle, you get to the point where truly automated cars are possible, which would theoretically be even safer. This seems like a good thing to me (although caution is as always warranted).
That was my thought to, but after thinking some more, there's a pretty easy solution.
Whenever the car starts, generate a random ID that's statistically certain to be unique. All the short term benefits of a unique identifier without the long term privacy risks.
Of course the devil would be in the details and I am way not qualified to speak on what it would take to ensure privacy with that kind of system. But it seems possible.
No one is proposing that NASA disappear. Well, a few people are, but they're mostly ignored as the fringe.
Commercialization means that for a potential Mars mission, or Asteroid mission, or anything else, most of the lifting from the surface to LEO would be done by commercial providers where possible. Some people still think a customized NASA-specific heavy lift vehicle would be necessary, so I'll go with that. However, imagine if you could just launch all the heavy stuff on that, and then put the people up in a light and cheap Falcon-9/Dragon combo, or Dreamchaser/EELV, or Orion-lite/EELV, whichever is cheaper for the number fo people you need. Thats the vision.
I'm heavily involved in groups that support the commercialization process, but I also work at JPL, and as such I recognize that there are some things (well-defined, profitable, with quantifiable risks) tasks that fixed-price commercial contracts are better for, while others (unprofitable, expensive, very risky, and ground-breaking) that government development is necessary for. A manned Mars mission definitely falls in the second category -- but it can be helped along by doing some parts, namely launch-to-LEO, using systems in the first category.
No one is stopping funding space exploration. Commercialization in this sense means purchasing launch vehicles for people from commercial providers, just as we do for unmanned vehicles.
This should in theory free up *MORE* money to allow real exploration, technology development, and all the things NASA is good at. The commercialization policies proposed by the administration included an overall increase in the NASA budget.
Publicly funded space research is going nowhere, don't worry.
You're misinterpreting what commercial space transport means. It doesn't mean that NASA tries to sell what it has to any millionaire looking for a joy ride.
What it means is that rather than designing and using one-off vehicles for its own uses, NASA will instead try to purchase launches from commercial companies where possible. It already does this in fact -- all unmanned NASA missions, as well as all DOD missions, are launched on commercially acquired Evolved Expendable Launch Vehicles, mostly purchased from ULA (i.e. Lockheed/Boeing). Now it is just moving a step further and providing a framework to do the same thing for manned spacecraft. In addition to reducing the abuses inherent to cost-plus contracts, it also opens up some reduced savings by letting other customers subsidize the development costs. For other customers, don't let the 'space tourism' thing get you down. While there may be some of that, the most likely 'other customers' would be other countries looking to do their own research without being as dependent on the whims of NASA.
NASA will continue to be on the forefront of exploration for the near future, funding missions and designing the hardware to do what hasn't been done before. What the commercialization proposals do is try and make the first step (getting to LEO) a little cheaper. Going with your Columbus analogy, he didn't have to design and build the Nina, Pinta and Santa Maria himself, he bought them with the funds provided by the crown, and we can hope this provides NASA with the same opportunity.
Unless you qualify a PID control system as a relatively simple feedback system, particularly when applied to a linearized system (which is what appears to be the case -- extremely fast observation and actuation make it easier to make a small angle assumption). PID controllers are something you can learn in an undergrad control class. While a complete understanding of their behavior and the art of designing one for a given system can be very challenging, the concepts are straightforward. I'd consider a non-linear Lyapunov controller or something with a measure of optimality in it to be a 'complex' feedback algorithm.
And I learned all this at a lowly state school (Texas A&M to be precise). I'm happy that Ms. Public can understand the fundamentals of it.
I work at JPL. The pay is good enough, but it isn't stellar, especially when you look at housing prices around here.
I do it because I really like it. Its still a job, but its one I enjoy most days. If I was in it for the money I'd be working for an oil company in Houston. Thats where many of my classmates are.
I don't feel like it applies to all jobs -- for instance, my job is spacecraft navigation.
That is, I help get probes safely to other planets. And quite frankly, while it might sound fancy, its just a job, and its a lot more straightforward than IT work. Yet when the next mission I work on (GRAIL) starts collecting data in a year, people all over this site will be congratulating the people I work with. And if it doesn't work, unless its something really dumb like the old units thing, people will say 'well, these things happen, what they do is hard.'
And of course, to top it all off, I couldn't do my job without the very well-made and flexible IT system behind everything I do. Obviously, this is a bit of an extreme example, but in my experience IT people and other "infrastructure" jobs get the short end of the appreciation stick (I know I've been guilty of whining when things dont go perfectly myself).
Yes, distribute it by source or distribute it outside the US.
H.264 is an open standard, but patent-encumbered (different things). If you sidestep the patent issue the concerns disappear, and you're free to use it, comfortable in the fact that the definition is well-defined, stable, and supported by players across the industry. This is what open standards imply, not that it's royalty free.
Really we need something that is both open (in the standards sense) and royalty free. Sadly, at present we are presented with a choice between one or the other.
Take the office software segment. You've got open and closed source competitors such as MS Office, LibreOffice, and iWork. Personally, I think that MS Office is nicer, but I use Linux much of the time and don't like to use office software in general, so LibreOffice is my practical choice most of the time. My sister likes iWork. Unfortunately, as we all know, in many cases we are stuck using MS because it's what everyone else uses, and weird compatibility things get in the way (particularly if you're like me and use a lot of equations).
All of these concerns would be eliminated by a proper open standard in office documents. Imagine if OOXML were actually what ti claimed to be, didn't suffer from things like the infamous "like it works in office 95," and weren't so long as to be unwieldy. Then LibreOffice and iWork and any new competitor could implement it and we would be able to select our options based solely on what works best for us, without consideration for what other people use.
Open standards give us the freedom to choose, free of artificial barriers.
Google owns the definition of VP8, while ISO sets the definition of H264. By the definitions of standards, h.264 is open while VP8 is proprietary. Note that an open standard is not the same as open source, or open government, or an open door. Open has different definitions in different contexts.
Yes, the patent situation makes it more complicated (as I acknowledge), and makes VP8 a better solution for the web in my opinion. However it's incorrect to call h.264 a proprietary codec: it is an open codec encumbered by patent licensing concerns. If the US patent system weren't so screwed up, Mozilla and Google would be free to use x264 and be done with the whole thing.
Sorry, the last paragraph should read "Personally, I worry that the split between VP8 and H.264 is going to kill the tag and leave us with Flash Video for the foreseeable future. If the choice is between VP8 and H.264, I'd pick VP8, but H.264 via is far superior..."
Except VP8 is a proprietary codec (WebM is just a container), while H.264 is an open standard. That is, the definition of VP8 is entirely defined and controlled by Google, while the definition of MPEG4 is controlled by the ISO standards organization.
This is of course complicated by the fact that H.264 is crippled by patents that require licensing fees of anyone wishing to legally distribute in the US, while VP8 is theoretically free of such troubles because Google released all related patents into the public domain. Add to this the debate about whether any marginally useful video codec is truly free of patent concerns, the largely-accepted fact that H.264 is technically superior and has a large installed base for hardware decoding, and lingering concerns over Google becoming too dominant on the web, and you've suddenly got an incredibly complicated and confusing situation.
Personally, I worry that the split between VP8 and H.264 is going to kill the tag and leave us with Flash Video for the foreseeable future. If the choice is between VP8 and H.264, I'd pick VP8, but H.264 via is far superior than H.264 via Flash in my mind. Quite frankly, at this point its a largely intractable problem that can only truly be resolved by patent reform in the US.
Open has many meanings depending on context. For instance, you wouldn't claim leaving a door open is a bastardization of the word, even though the process of making the door maybe be proprietary.
The open referred to here is not in the sense of "open source," but of "open standard.". That is, the difference between.odt and.doc rather than between OO.o and MS Office. Personally i would have a lot less problem with MS Office if it supported a truly open standard, one that could be read and written by other office suites. Having MS Office primarily use the.odt wouldn't make the standard any less open.
The problem here is that while the codec is open in the standards sense (I.e. It's well documented and out there), and there are even open source implementations of it (x264), the openness doesn't make much difference because any commercial implementation faces difficulties in the US because it is patented, and thus requires licensing fees to distribute. While ubuntu can get away with saying "you're not in the US, wink wink nudge nudge, click ok to download things like mp3 and h264 codecs," google really can't because they're targeting an audience that would never dream of installing their own OS. Of course some think WebM will be seen as infringing on those same patents, so it may be a moot point. Maybe this, combined with the patent war over mobile technology will finally bring some sanity to our system? Maybe?
Actually I think thats the point ESMD is trying to make here. Congress mandated that they use SRBs et. al., so ESMD comes back and says "all right, we can do it, but it WILL be late and overbudget."
Because Ares V is not the best design, but the best design that fulfills the requirements of using the existing workforce, SRBs, SSMEs, etc.
This is basically ESMD's way of passing the buck back to congress and saying they can do one of two things: 1. Build an HLV that keeps jobs in all the nice districts... OR 2. Do it on time and on budget.
In other words, congress' requirements are impossible to fulfill, and ESMD is saying it as politely as possible.
I wasn't meaning to make the assumption that colonization must be the goal, but simply that your post assumed what the goal was without stating it.
As far as the debate of whether colonization is a worthy goal: Why can't we work on solving problems here while learning how to live in other places as well? Colonization won't happen unless it can be self-funded, so my question is why shouldn't people pursue it?
You're free not to agree. I was just saying that the parent post talked about manned exploration not being effective without saying what he wanted it to be effective for.
As far as why, the simplest answer is because its what we as people do. Also, a frontier makes the world seem less claustrophobic, and provides a relief valve for stresses in society (but not necessarily for population statistics, I don't think you could get that kind of volume). Aslo, theres the whole backup plan thing.
Most of all though, the cost is not necessarily enormous, we simply don't know. And funding for colonization would not come out of public money anyway -- if its sustainable it will be profitable and privately funded. Colonization would never be run like NASA. Right now we need to keep sending people and figuring it all out, and facilitate the creation of profitable ventures. Basically keep going along the lines we're going now with the current paltry NASA budget.
If the costs to keep looking at it are low, my question is then why not.
The settlers planned to go one-way. The explorers didn't. Columbus went and returned, the Plymouth colony went to stay. Lewis and Clark went and came back, while settlers flowed out after them.
The first mission(s) will probably be two-way missions, but real settlement happens when the one-way trip comes down to the costs of a modest families life-savings.
Slashdot Mirror
Try dealing with time when you're actually doing rocket science! Leap seconds, measurable light travel time, and variations due to relativity make for incredible confusion.
Sadly I don't think most people realize how complicated timekeeping is even in terrestrial settings. The fact that Apple and MS face timing bugs is no surprise at all.
That requires Apple to have a monopolistic hold over a market. While right now they do on the tablet market, its only because we have only begun to see worthy Android competitors. I'm pretty sure Android on the tablet will compete with the iPad as well as Android on the phone competes with the iPhone.
The risk for Apple isn't antitrust, its News Corp et. al. being able to get better deals on other platforms. This seems good for everyone -- its that mythical market at work.
Well, as long as the mass of the object is negligible compared to the body its orbiting (in this case, the asteroid is absurdly smaller than the sun), then the mass of the body has no effect on its orbit. More specifically, the force is proportional to the mass so that the acceleration is constant regardless of mass.
The only time it will make a difference is for non-gravitational forces like solar radiation pressure or atmospheric drag. Then the surface area of the spacecraft will be much smaller than the asteroid, so it won't be a big deal, unless you're trying to change it on purpose.
Also, you never want to try and orbit a small asteroid like this. Their gravity fields are way too weird and unpredictable and you can end up crashing after a few revs of what starts out looking stable. Preferably you either land, get far enough away that the gravity is just a perturbing force, or do the standoff required for gravity tractoring.
Those analyses are relatively easy to do, and you're not going to change its trajectory significantly unless you really mean to.
In order to do it with a ~500 kg spacecraft you have to hover about 200 meters away from the asteroid with your engines thrusting essentially continuously for around a year. If you're not close enough to have to worry about hovering, or if you're trying to do a landing, no reasonably sized spacecraft is going to make a difference.
We have a good idea of the composition of asteroids in general, from meteorites, planetary formation models, etc.
We believe Apophis is chondritic because based on its apparent brightness and the way that brightness varies, we have a decent estimate of its size and albedo. If it had a different albedo it would indicate a different composition.
Of course, as with all remote observations based on a lot of educated guesses, there is a chance its wrong. However, if it is its probably a pathological case we could never gotten right, and that would make it even more interesting to visit.
Statistically certain is shorthand for saying the likelihood of the event happening is below some arbitrarily small level. At least its well understood to mean that amongst those I work with, and everything we do has a quantified uncertainty attached to it.
Actually, from what i gathered from TFS, this specifically doesn't require it. It simply provides an extra bit of warning if it's able to detect something bad about to happen. It still requires the driver to pay attention.
Based on that, i would say that until nearly every car has it will definitely increase safety. If you can't depend on it to work with every car, you have to pay just as much attention, but it will still give warnings on occasion.
Of course once you get to the point of having it in most every vehicle, you get to the point where truly automated cars are possible, which would theoretically be even safer. This seems like a good thing to me (although caution is as always warranted).
That was my thought to, but after thinking some more, there's a pretty easy solution.
Whenever the car starts, generate a random ID that's statistically certain to be unique. All the short term benefits of a unique identifier without the long term privacy risks.
Of course the devil would be in the details and I am way not qualified to speak on what it would take to ensure privacy with that kind of system. But it seems possible.
No one is proposing that NASA disappear. Well, a few people are, but they're mostly ignored as the fringe.
Commercialization means that for a potential Mars mission, or Asteroid mission, or anything else, most of the lifting from the surface to LEO would be done by commercial providers where possible. Some people still think a customized NASA-specific heavy lift vehicle would be necessary, so I'll go with that. However, imagine if you could just launch all the heavy stuff on that, and then put the people up in a light and cheap Falcon-9/Dragon combo, or Dreamchaser/EELV, or Orion-lite/EELV, whichever is cheaper for the number fo people you need. Thats the vision.
I'm heavily involved in groups that support the commercialization process, but I also work at JPL, and as such I recognize that there are some things (well-defined, profitable, with quantifiable risks) tasks that fixed-price commercial contracts are better for, while others (unprofitable, expensive, very risky, and ground-breaking) that government development is necessary for. A manned Mars mission definitely falls in the second category -- but it can be helped along by doing some parts, namely launch-to-LEO, using systems in the first category.
No one is stopping funding space exploration. Commercialization in this sense means purchasing launch vehicles for people from commercial providers, just as we do for unmanned vehicles.
This should in theory free up *MORE* money to allow real exploration, technology development, and all the things NASA is good at. The commercialization policies proposed by the administration included an overall increase in the NASA budget.
Publicly funded space research is going nowhere, don't worry.
You're misinterpreting what commercial space transport means. It doesn't mean that NASA tries to sell what it has to any millionaire looking for a joy ride.
What it means is that rather than designing and using one-off vehicles for its own uses, NASA will instead try to purchase launches from commercial companies where possible. It already does this in fact -- all unmanned NASA missions, as well as all DOD missions, are launched on commercially acquired Evolved Expendable Launch Vehicles, mostly purchased from ULA (i.e. Lockheed/Boeing). Now it is just moving a step further and providing a framework to do the same thing for manned spacecraft. In addition to reducing the abuses inherent to cost-plus contracts, it also opens up some reduced savings by letting other customers subsidize the development costs. For other customers, don't let the 'space tourism' thing get you down. While there may be some of that, the most likely 'other customers' would be other countries looking to do their own research without being as dependent on the whims of NASA.
NASA will continue to be on the forefront of exploration for the near future, funding missions and designing the hardware to do what hasn't been done before. What the commercialization proposals do is try and make the first step (getting to LEO) a little cheaper. Going with your Columbus analogy, he didn't have to design and build the Nina, Pinta and Santa Maria himself, he bought them with the funds provided by the crown, and we can hope this provides NASA with the same opportunity.
Unless you qualify a PID control system as a relatively simple feedback system, particularly when applied to a linearized system (which is what appears to be the case -- extremely fast observation and actuation make it easier to make a small angle assumption). PID controllers are something you can learn in an undergrad control class. While a complete understanding of their behavior and the art of designing one for a given system can be very challenging, the concepts are straightforward. I'd consider a non-linear Lyapunov controller or something with a measure of optimality in it to be a 'complex' feedback algorithm.
And I learned all this at a lowly state school (Texas A&M to be precise). I'm happy that Ms. Public can understand the fundamentals of it.
I work at JPL. The pay is good enough, but it isn't stellar, especially when you look at housing prices around here.
I do it because I really like it. Its still a job, but its one I enjoy most days. If I was in it for the money I'd be working for an oil company in Houston. Thats where many of my classmates are.
I don't feel like it applies to all jobs -- for instance, my job is spacecraft navigation.
That is, I help get probes safely to other planets. And quite frankly, while it might sound fancy, its just a job, and its a lot more straightforward than IT work. Yet when the next mission I work on (GRAIL) starts collecting data in a year, people all over this site will be congratulating the people I work with. And if it doesn't work, unless its something really dumb like the old units thing, people will say 'well, these things happen, what they do is hard.'
And of course, to top it all off, I couldn't do my job without the very well-made and flexible IT system behind everything I do. Obviously, this is a bit of an extreme example, but in my experience IT people and other "infrastructure" jobs get the short end of the appreciation stick (I know I've been guilty of whining when things dont go perfectly myself).
Yes, distribute it by source or distribute it outside the US.
H.264 is an open standard, but patent-encumbered (different things). If you sidestep the patent issue the concerns disappear, and you're free to use it, comfortable in the fact that the definition is well-defined, stable, and supported by players across the industry. This is what open standards imply, not that it's royalty free.
Really we need something that is both open (in the standards sense) and royalty free. Sadly, at present we are presented with a choice between one or the other.
Take the office software segment. You've got open and closed source competitors such as MS Office, LibreOffice, and iWork. Personally, I think that MS Office is nicer, but I use Linux much of the time and don't like to use office software in general, so LibreOffice is my practical choice most of the time. My sister likes iWork. Unfortunately, as we all know, in many cases we are stuck using MS because it's what everyone else uses, and weird compatibility things get in the way (particularly if you're like me and use a lot of equations).
All of these concerns would be eliminated by a proper open standard in office documents. Imagine if OOXML were actually what ti claimed to be, didn't suffer from things like the infamous "like it works in office 95," and weren't so long as to be unwieldy. Then LibreOffice and iWork and any new competitor could implement it and we would be able to select our options based solely on what works best for us, without consideration for what other people use.
Open standards give us the freedom to choose, free of artificial barriers.
Google owns the definition of VP8, while ISO sets the definition of H264. By the definitions of standards, h.264 is open while VP8 is proprietary. Note that an open standard is not the same as open source, or open government, or an open door. Open has different definitions in different contexts.
Yes, the patent situation makes it more complicated (as I acknowledge), and makes VP8 a better solution for the web in my opinion. However it's incorrect to call h.264 a proprietary codec: it is an open codec encumbered by patent licensing concerns. If the US patent system weren't so screwed up, Mozilla and Google would be free to use x264 and be done with the whole thing.
Sorry, the last paragraph should read "Personally, I worry that the split between VP8 and H.264 is going to kill the tag and leave us with Flash Video for the foreseeable future. If the choice is between VP8 and H.264, I'd pick VP8, but H.264 via is far superior..."
This is why they have the preview button...
Except VP8 is a proprietary codec (WebM is just a container), while H.264 is an open standard. That is, the definition of VP8 is entirely defined and controlled by Google, while the definition of MPEG4 is controlled by the ISO standards organization.
This is of course complicated by the fact that H.264 is crippled by patents that require licensing fees of anyone wishing to legally distribute in the US, while VP8 is theoretically free of such troubles because Google released all related patents into the public domain. Add to this the debate about whether any marginally useful video codec is truly free of patent concerns, the largely-accepted fact that H.264 is technically superior and has a large installed base for hardware decoding, and lingering concerns over Google becoming too dominant on the web, and you've suddenly got an incredibly complicated and confusing situation.
Personally, I worry that the split between VP8 and H.264 is going to kill the tag and leave us with Flash Video for the foreseeable future. If the choice is between VP8 and H.264, I'd pick VP8, but H.264 via is far superior than H.264 via Flash in my mind. Quite frankly, at this point its a largely intractable problem that can only truly be resolved by patent reform in the US.
Open has many meanings depending on context. For instance, you wouldn't claim leaving a door open is a bastardization of the word, even though the process of making the door maybe be proprietary.
The open referred to here is not in the sense of "open source," but of "open standard.". That is, the difference between .odt and .doc rather than between OO.o and MS Office. Personally i would have a lot less problem with MS Office if it supported a truly open standard, one that could be read and written by other office suites. Having MS Office primarily use the .odt wouldn't make the standard any less open.
The problem here is that while the codec is open in the standards sense (I.e. It's well documented and out there), and there are even open source implementations of it (x264), the openness doesn't make much difference because any commercial implementation faces difficulties in the US because it is patented, and thus requires licensing fees to distribute. While ubuntu can get away with saying "you're not in the US, wink wink nudge nudge, click ok to download things like mp3 and h264 codecs," google really can't because they're targeting an audience that would never dream of installing their own OS. Of course some think WebM will be seen as infringing on those same patents, so it may be a moot point. Maybe this, combined with the patent war over mobile technology will finally bring some sanity to our system? Maybe?
Actually I think thats the point ESMD is trying to make here. Congress mandated that they use SRBs et. al., so ESMD comes back and says "all right, we can do it, but it WILL be late and overbudget."
Because Ares V is not the best design, but the best design that fulfills the requirements of using the existing workforce, SRBs, SSMEs, etc.
This is basically ESMD's way of passing the buck back to congress and saying they can do one of two things:
1. Build an HLV that keeps jobs in all the nice districts... OR
2. Do it on time and on budget.
In other words, congress' requirements are impossible to fulfill, and ESMD is saying it as politely as possible.
I wasn't meaning to make the assumption that colonization must be the goal, but simply that your post assumed what the goal was without stating it.
As far as the debate of whether colonization is a worthy goal: Why can't we work on solving problems here while learning how to live in other places as well? Colonization won't happen unless it can be self-funded, so my question is why shouldn't people pursue it?
You're free not to agree. I was just saying that the parent post talked about manned exploration not being effective without saying what he wanted it to be effective for.
As far as why, the simplest answer is because its what we as people do. Also, a frontier makes the world seem less claustrophobic, and provides a relief valve for stresses in society (but not necessarily for population statistics, I don't think you could get that kind of volume). Aslo, theres the whole backup plan thing.
Most of all though, the cost is not necessarily enormous, we simply don't know. And funding for colonization would not come out of public money anyway -- if its sustainable it will be profitable and privately funded. Colonization would never be run like NASA. Right now we need to keep sending people and figuring it all out, and facilitate the creation of profitable ventures. Basically keep going along the lines we're going now with the current paltry NASA budget.
If the costs to keep looking at it are low, my question is then why not.
The settlers planned to go one-way. The explorers didn't. Columbus went and returned, the Plymouth colony went to stay. Lewis and Clark went and came back, while settlers flowed out after them.
The first mission(s) will probably be two-way missions, but real settlement happens when the one-way trip comes down to the costs of a modest families life-savings.