How effective would one of these vehicles be in an urban setting?
The word in the robotics community is that there will be another DARPA Grand Challenge, this time focused on urban driving. This should be a significantly harder problem, as if the first one wasn't hard enough!
It should not be forgotten when convenient that the Supreme Court's role is limited to combing through the constitutionality of a past ruling.
That's not true. In fact, the Constitution does not explicitly grant this power to the Supreme Court. The Constitution is actually very vague on exactly what the Supreme Court is for (Article III, Section 1):
The judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish.
The precedent for judicial review was not established until Marbury v. Madison in 1803, when the Marshall court established the right of the court to declare laws unconstitutional based on Article VI, which declares the constitution to be the supreme law of the land.
tell them to pop up a "you've been infected, moron" window
I had a friend in our security group who needed to notify somebody at another university that their machine had been cracked and was attacking others. She sent a note to their local printer - I can only imagine the look on their face when a "You have been hacked, please clean your machine" note printed at their desk!
The moral of the story being, the guys who run these programs are not necessarily all that bright, nor do they have that much background in science and engineering.
Not so. Many if not all of the DARPA program managers are very accomplished academics who take an extended sabbatical (~2 yrs) to take the DARPA job. For them this is a way to influence policy and funding in their discipline and get a broad view of what is out there, and for DARPA it is an opportunity to have real experts handling their programs. These people are far from stupid and inexperienced, but are encouraged to develop programs that are risky and potentially high-value.
Sorry folks there ain't no draft and it isn't a mystery that the US war machine is a "tad" corrupt. you sign up for the military because you want to profit from the misery of others. That is unless you sign up for the military to do something outside of being a grunt [e.g. doctor, engineer, etc]. Then you're ok.
Or maybe, like my dad, they grew up very poor in a part of the country that doesn't offer young people a real chance to advance. His only chance in life (to stay out of the coal mine) was to enlist in the army to get a steady job with decent pay. The Army provided vocational and educational training, and enough money to put his younger brother through college. He eventually got out and was prepared to work in jobs that never would have been open to him with a second-rate high school education and no experience, and was able to provide for a family in FAR better circumstances than he endured growing up.
Judging by the former military and reserve folks I was with in engineering school, I think my dad's story is far from unique; in fact it's probably typical. For many young people in the military joining up is perhaps the only path to a life better than their parents had. To say that that means they want to "profit from the misery of others" is ignorant, snobbish, and just plain stupid.
The pixel size (what most probably think of as resolution) is really 30-60 cm, enabling scientists to resolve features around a meter in size with a few pixels, so "1-meter resolution" is a little misleading. For more information on the camera see the mission web site.
Are you sure the pics in that Pop. Sci. article were from orbit? Many very impressive "spy satellite" pictures out there actually came from U-2 spy planes. I don't think we had that kind of resolving power from orbit 25 years ago.
Where a mission depends on preprogrammed intelligence to orbit safely, the success was really decided months ago when they sealed it in the launcher. It's good that humans are driving it to the last second, they put their energy into it, and error checked and corrected during production.
This isn't quite true. All missions of this type do multiple software patches in flight (even complete rebuilds). Things are by no means locked in at launch. The sequences to actually execute a critical event like orbit insertion are probably uploaded sometime in the week before the event, and changes are sometimes made mere hours ahead of time. This is particularly true for deep space missions like this where no spacecraft just like it has ever flown before, because controllers typically learn critical lessons about how the bird performs in flight that need to be incorporated into the critical even sequences.
Orbital burns of unmanned spacecraft (having implemented a few) are always computer controlled - I have no idea why the post claims this "adds to the risk." Robotic spacecraft are NOT piloted by joystick - carefully generated sequences of commands are sent up hours or days ahead of time. This is so things can be carefully checked by teams of people and simulated, and because joystick control simply isn't feasible past the Moon or so due to light time issues.
There are variations in autonomy, though. On some spacecraft the burn is completely "open-loop," i.e. the command is to turn to a certain attitude and fire thrusters for XX seconds. On others the command is to provide a certain impulse in a certain direction, and the on-board computer uses accelerometers and gyros, etc. to determine when to start and stop the burn.
One day we may rely on centralized quantum computers that the average Joe won't be able to afford, or maybe won't need to afford. A standardized viewing mechanism could be relied upon, along with keystroke, voice and gesture entry, and a common mechanism for transmitting sound and other media could be defined.
Now if the security were managed correctly, you and I could visit our data from any workstation in the world, because the apps and data are all hosted. You could just identify yourself and take control of whatever data and applications are connected to you from wherever you are. Some folks may opt for better interface equipment (speakers!) but the network and processing power would be the same for everyone (and it would be enough).
Um, hasn't this been basically implemented for years in the form of X? I recently finished working on a flight project at NASA JPL, and all of our operations software (custom stuff) was hosted on a Sun network, and we could log in from anywhere (for example, using X11 on my Mac) and access our data and applications. The only issue was speed if there were a lot of graphics, but even that wasn't so bad and will only get better. It was all wonderfully convenient, and allowed me to work from home (or, I hate to say, on vacation) when emergencies arose (frequently) rather than have to go into work at 2AM.
I understand the point you're trying to make, but I really wouldn't mind having something like this. As a college student, I often have multiple unsynchronized copies of term papers in different places. A service like Writely helps keep everything up-to-date, and in one place.
I agree! As an academic researcher, I frequently work on papers with multiple authors, sometimes spread across the globe. If my co-authors are pretty unix-savvy and we're working in LaTeX, as is usually the case when we're working on an engineering paper, I set up a CVS server to accomplish this. If I'm working in science rather than engineering, though, LaTeX is usually not an option and I really don't want to try to teach anybody CVS, so we end up emailing multiple copies of the paper around and I'm stuck trying to reconcile the various versions.
I am of the opinion that most of the progress to be made on this type of thing is going to be done in the private sector and therefore the technology will be available to our enemies as well.
Just because something is done by the private sector, even with private funding, doesn't mean that the technology will be available to our enemies. The government can still classify it if they really want to keep it hidden, and even without that there's ITAR to deal with. ITAR is so scary that most companies take an extremely conservative line on who they share things with, even if they're not secret at all. The rules are pretty much structured along the same lines of the infamous "I can't define pornography, but I know it when I see it" - there's often no way to tell that you're violating ITAR until they fine you piles of cash (or put you in jail)
Having worked at JPL, I can tell you that it's still part of CalTech. I has, however, a contract from NASA to run NASA's unmanned exploration of space, and all NASA probes and sattilites are run from their.
True. What I meant by "transferred to NASA" was that the oversight was shifted to NASA from the Department of Defense. JPL is "managed for NASA" by Caltech, which is unique among the NASA centers - JPL employees (as I'm sure you know, but others may not) get their paychecks from Caltech rather than NASA. However (having also worked at JPL), NASA still has the opportunity to transfer management responsibilities from Caltech to someone else on a regular basis when the contract is renewed every several years. It's unlikely to happen because of the disruption it would cause, but NASA could take JPL away from Caltech (and other organizations *do* bid to do just that).
Why is the JPL working with solar sails? Aren't they the Jet Propulsion Laboratory? Isn't this a bit out of their department?
JPL never really had anything to do with jets as we know them today. My understanding is that when it was founded by some Caltech faculty and students in the 40's to do rocket research, "rocket" was kind of a dirty word due to lingering memories of German rockets in WWII. Rockets were also commonly called jets until the mid to late 40's. One of JPL's first successes was the development of JATO technology, which stands for "Jet-Assisted Take Off", even though it is really the use of rockets to help large planes take off on short runways.
JPL was transferred to NASA in the 50's and built Explorer I, the US's answer to Sputnik. Since then, JPL has been the prime NASA center for unmanned spacecraft, mostly sent beyond Earth orbit. Since most JPL missions travel pretty far into space, solar sails are a natural thing to be researching.
Well, guess what, now it runs on x86 hardware... why in the world would you want to limit your market? You have a product, and a good one, but you put an artificial limit of your own on what you will allow it to be used on, which effectively cuts you out of 99% of the market that would potentially purchase your product, and for what reason? Because you don't want to support all that different hardware?
Yes, because you don't want to support all that hardware. Apple has worked hard to build a brand around reliability - Macs just work. While I think a good part of that is that they just have a better OS, another big part is that they have tight control over the hardware so their OS doesn't have to struggle to make whatever crazy hardware setup you happen to have work. The average consumer probably isn't going to realize how hard of a task that is, and when they buy a Mac from Dell and it is no more reliable than their Windows machine was they are going to wonder what all the hubbub was about.
Government agencies are not allowed to hold copyrights. If the images belonged to a third party that NASA contracted out to, then you might have an issue. Fortuanetly, it's usually NASA's mission to get those photos, so they belong to NASA even if NASA contract for the space vehicle to be built by someone else.
Usually this is the case, but not always. Malin Space Systems built the high-res camera for Mars Global Surveyor (among other things), and has been very restrictive about releasing images to either the public or the scientific community - limited numbers of images are released after a long delay so they get to pick over them first and publish the best papers. It's a huge bummer, but they've got the best cameras and so they get to set the terms. They deserve some priority since they put the work into developing the cameras, but I hate seeing the taxpayer pay to take these pictures and then not get to look at them right away.
Compare that to the pictures from the Mars Exploration Rovers, where we all get to see the images practically as they hit the ground.
In the words of John Haddon, "Why build one, when you can build two for twice the price?" We should build two and target both the pole and the equator. Example: two mars landers. Good idea.
It's not always this straightforward. While I haven't worked on Moon missions, I've worked on Mars missions, and on Mars different landing latitudes result in dramatically different requirements. This means different entry speeds (which doesn't apply to the moon, but approach speed would be different), different landing terrain, etc. Power and thermal requirements end up being very different, too, which makes a single design for multiple locations inefficient for either.
Redundancy is always key and it is more efficient to built two highly probably successes than one extremely probably success.
I'm not saying I disagree, but redundancy is a tricky thing. In order for two spacecraft to be redundant to a certain risk, the risk factor has to be independent between the two missions. Two spacecraft give you redundancy to a particular random failure mode like landing on a huge jagged rock that breaks something, but not to a design flaw that would kill both. For example, if we had sent two Mars Polar Landers, we would have been pretty likely to lose both as the failure was due to a design flaw not a random event. Having sent only one, we can now fix the design flaw and send a fixed model.
I've often felt there are two good ways to get a lot of satisfaction out of your job:
1. Have a job that may take a lot of commitment, but provides big rewards in terms of satisfaction. That is, you feel the work is important and meaningful. I spent time as an engineer at JPL, and this definitely fell into this category - the hours were long and the pay not great, but the work itself was amazing. Many people there have very little lives outside of work for years at a time when things get busy, but they do it in exchange for the chance to work on something meaningful.
2. Have a job that pays well, doesn't suck, and generally takes 40 hours or less of your life, freeing you to fill the remaining hours with rewarding hobbies and other interests. I've had friends work as engineers at Boeing, and this was the model for most of their coworkers. Many of them were very happy to leave the job at the door and do all sorts of other things outside of work.
I find I'm happiest with number 1, but many people are different.
I'm sorry but you have absolutely no clue about what you are talking about. Managers do not have to a complete understanding of what their staff members do. The managerial skills they possess are equally important to the an organization's success as your technical skills.
When I was graduating college, the best interview I had was at a small aerospace research center in Palo Alto (run by Lockheed, but not the gigantic Missiles and Space in Sunnyvale). This was a hardcore R&D facility, with something like 80% Ph.D.'s working there. The manager I would have worked for had I not decided to go off to grad school had a "mere" master's degree and little understanding of his underlings's jobs beyond a superficial level. And yet, they all loved him. The reason? He knew what he knew and what he didn't, and he let the talented people under him do their jobs with minimal interference. His job was to find projects for them and make sure the resources they needed were in place so they could focus on the work they had devoted their lives to and not the everyday BS. In my mind this is exactly what a manager of technical people is for.
I later went on to work for JPL in Pasadena. I had good managers and bad in my time there, and some of the worst have been highly skilled technical people who had then been promoted to management. Although they had the skills and intelligence to do my job, they weren't good at handling the management tasks required of them. Again, some of the best were less technically talented people who were able to recognize that and not interfere, instead working to take care of all the administrative stuff so I didn't have to worry about it while devoting 80+ hour weeks to mission success.
As an answer to the original question, JPL was a place where I certainly did not make nearly the money I could elsewhere, but the opportunity to work with an incredibly talented bunch of people on work that was unique (and I felt, important) was well worth it.
I had lunch with Tilden once, and he rubbed me the wrong way too. He's definitely a smart guy, but he had a lot of P.T. Barnum in him too.
I was impressed, though, with his attitude toward tinkerers and hackers of his toys. They go out of their way to make the toys hackable and interesting to folks to want to take them apart. For example, they used much nicer motors in the bio bugs than really necessary so that hobbyists who were frustrated with the unavailability of cheap, high quality motors could buy the bio bugs just to get the motors (which Wowwee could get cheap since they're big).
But it may do more damage than good by being published.
This article shows, and maybe it's because I work with criminals all day (Public Defenders office), that writing malware pays. Before it was for notoriety or to prove you could or to piss people off, but now it can provide an income source and I think we will be seeing more of it from now on just because people are going to be trying to make a buck off of it.
Seems to me that the folks who are capable of writing (good) malware are likely to already be pretty aware of how well it pays. I'm not too worried about some common criminal reading this news and thinking "holy cow, there's a new career for me!"
Well, except for the fact that MSFT, Google, Apple, and Amazon need the telcos more than the telcos need them. By a wide margin -- and especially true for Google and Amazon (and eBay).
I'm not so sure about that - which internet provider would you choose, the one with or without Google? The telco's could only win such a battle if they colluded, which would probably bring anti-trust lawsuits galore.
All of these issues are ones that confront human drivers, too, often with less than stellar performance. These are all tough problems, but as a person with some experience working with automated driving technology (DARPA Grand Challenge), I'm confident they will be solved (but it will take a while).
You're right, though - a big issue will be that even if the performance is as good as a human's, companies will still get sued for accidents that even a human might not have avoided. No system will ever be successful at never having an accident, but eventually we (or our grandchildren) will all be riding in these things.
Speaking of the DARPA Grand Challenge, fairly reliable rumor (i.e. colleagues involved in the race) has it that there will be another Grand Challenge, this time involving the MUCH harder task of *urban* driving.
Slashdot Mirror
The judicial Power of the United States, shall be vested in one supreme Court, and in such inferior Courts as the Congress may from time to time ordain and establish.
The precedent for judicial review was not established until Marbury v. Madison in 1803, when the Marshall court established the right of the court to declare laws unconstitutional based on Article VI, which declares the constitution to be the supreme law of the land.
Or maybe, like my dad, they grew up very poor in a part of the country that doesn't offer young people a real chance to advance. His only chance in life (to stay out of the coal mine) was to enlist in the army to get a steady job with decent pay. The Army provided vocational and educational training, and enough money to put his younger brother through college. He eventually got out and was prepared to work in jobs that never would have been open to him with a second-rate high school education and no experience, and was able to provide for a family in FAR better circumstances than he endured growing up.
Judging by the former military and reserve folks I was with in engineering school, I think my dad's story is far from unique; in fact it's probably typical. For many young people in the military joining up is perhaps the only path to a life better than their parents had. To say that that means they want to "profit from the misery of others" is ignorant, snobbish, and just plain stupid.
Are you sure the pics in that Pop. Sci. article were from orbit? Many very impressive "spy satellite" pictures out there actually came from U-2 spy planes. I don't think we had that kind of resolving power from orbit 25 years ago.
There are variations in autonomy, though. On some spacecraft the burn is completely "open-loop," i.e. the command is to turn to a certain attitude and fire thrusters for XX seconds. On others the command is to provide a certain impulse in a certain direction, and the on-board computer uses accelerometers and gyros, etc. to determine when to start and stop the burn.
JPL never really had anything to do with jets as we know them today. My understanding is that when it was founded by some Caltech faculty and students in the 40's to do rocket research, "rocket" was kind of a dirty word due to lingering memories of German rockets in WWII. Rockets were also commonly called jets until the mid to late 40's. One of JPL's first successes was the development of JATO technology, which stands for "Jet-Assisted Take Off", even though it is really the use of rockets to help large planes take off on short runways.
JPL was transferred to NASA in the 50's and built Explorer I, the US's answer to Sputnik. Since then, JPL has been the prime NASA center for unmanned spacecraft, mostly sent beyond Earth orbit. Since most JPL missions travel pretty far into space, solar sails are a natural thing to be researching.
...and I'm still waiting.
1. Have a job that may take a lot of commitment, but provides big rewards in terms of satisfaction. That is, you feel the work is important and meaningful. I spent time as an engineer at JPL, and this definitely fell into this category - the hours were long and the pay not great, but the work itself was amazing. Many people there have very little lives outside of work for years at a time when things get busy, but they do it in exchange for the chance to work on something meaningful.
2. Have a job that pays well, doesn't suck, and generally takes 40 hours or less of your life, freeing you to fill the remaining hours with rewarding hobbies and other interests. I've had friends work as engineers at Boeing, and this was the model for most of their coworkers. Many of them were very happy to leave the job at the door and do all sorts of other things outside of work.
I find I'm happiest with number 1, but many people are different.
When I was graduating college, the best interview I had was at a small aerospace research center in Palo Alto (run by Lockheed, but not the gigantic Missiles and Space in Sunnyvale). This was a hardcore R&D facility, with something like 80% Ph.D.'s working there. The manager I would have worked for had I not decided to go off to grad school had a "mere" master's degree and little understanding of his underlings's jobs beyond a superficial level. And yet, they all loved him. The reason? He knew what he knew and what he didn't, and he let the talented people under him do their jobs with minimal interference. His job was to find projects for them and make sure the resources they needed were in place so they could focus on the work they had devoted their lives to and not the everyday BS. In my mind this is exactly what a manager of technical people is for.
I later went on to work for JPL in Pasadena. I had good managers and bad in my time there, and some of the worst have been highly skilled technical people who had then been promoted to management. Although they had the skills and intelligence to do my job, they weren't good at handling the management tasks required of them. Again, some of the best were less technically talented people who were able to recognize that and not interfere, instead working to take care of all the administrative stuff so I didn't have to worry about it while devoting 80+ hour weeks to mission success.
As an answer to the original question, JPL was a place where I certainly did not make nearly the money I could elsewhere, but the opportunity to work with an incredibly talented bunch of people on work that was unique (and I felt, important) was well worth it.
I was impressed, though, with his attitude toward tinkerers and hackers of his toys. They go out of their way to make the toys hackable and interesting to folks to want to take them apart. For example, they used much nicer motors in the bio bugs than really necessary so that hobbyists who were frustrated with the unavailability of cheap, high quality motors could buy the bio bugs just to get the motors (which Wowwee could get cheap since they're big).
All of these issues are ones that confront human drivers, too, often with less than stellar performance. These are all tough problems, but as a person with some experience working with automated driving technology (DARPA Grand Challenge), I'm confident they will be solved (but it will take a while). You're right, though - a big issue will be that even if the performance is as good as a human's, companies will still get sued for accidents that even a human might not have avoided. No system will ever be successful at never having an accident, but eventually we (or our grandchildren) will all be riding in these things.
Speaking of the DARPA Grand Challenge, fairly reliable rumor (i.e. colleagues involved in the race) has it that there will be another Grand Challenge, this time involving the MUCH harder task of *urban* driving.