Contractors, definitely. The Feds needs math types in multiple agencies. The Census Bureau does a lot of stat work, for example. DoD needs math types at the warfare centers scattered around the country. Does she have any interest in Human-Systems Integration or training systems? There are a lot of people trying to find ways to get all those personnel trained up on new systems.
Alternatively, teach at a private school. They have the advantage of being able to select and expel their students and there will be less bureaucracy.
No, no. Since everything appears to cause cancer and other nasty effects in California, the first thing you should consider is whether California is in fact the cause.
You know what else is equally dumb, but has gotten a free pass? Touchscreen interfaces in cars. I make it a point to buy cars with physical controls so that I can do things by touch alone. Plus, the designers always seem to make it a point to bury settings in nested menus; this only makes it worse. 4.6 seconds is probably how long it takes some people to change the station on the radio. And of course, they have to look down at the screen to do it.
People running around and doing the equivalent of tearing down billboards and defacing storefronts. Big whoops. Last I checked, the major players in the global financial network have actual power. And most central/federal governments, too. This guy needs a cold beer and a sense of proportion. Ok, maybe not the latter; we know that's the one thing that no one in the universe can afford to have. Make it two beers, then.
I'd say that burning too much time and money on graphics, sound, FMV, and voice acting at the expense of mechanics, plot and bug-freeness has become a cliche in and of itself.
Obviously the solution is to go back to text-based gaming. OK, fine, EGA and the PC speaker.
Look at Civ or Galactic Civilizations. Those non-FPS/RTS games were turn-based and required thought and planning. Old RPGs are the same way. People like me who grew up with Wasteland and its contemporaries miss the engagement and cleverness. I'm not interested in a fast twitch game and am willing to pay for a game that makes me think and spend time to beat. It's merely a bonus that it's a sequel to one of the all time greats that we're talking about. I'm contributing tonight and then I'm going to fire up my old copy of Wasteland and go see how little firepower I can beat Guardian Citadel with this time.
Exploded blood sausage ftw!
I take issue with DARPA's assertion in TFA that formal verification cannot be scaled up to work on a modern weapon system. My office has done it for very long time and we are very software-intensive. We and our contractors just had to get smarter as the system became more complex and requirements became steeper.
Nonetheless, I would be interested in the potential of such a process to find sneak circuits and latent problems. Use it during the development process prior to integration and verification.
It's no surprise that they can keep a secret. Civilian personnel in defense and intelligence are, by and large, capable of keeping a secret when it counts. They are motivated to do the job and keep such secrets as are necessary to get it done (this does not include fraud, but the Important People do what they want). They know that info getting out could cause soldiers to die and wars to be lost. Speaking for my colleagues, it is not just another job because we know what's at stake.
Now, give classified info to some dummy in Congress...that's scary. Those people get their clearances by virtue of their jobs and not because of their own merits. And the spill procedures that we have to follow don't apply to them. Just like all those other laws and regulations don't apply to them.
As an acquisition guy, I hate this kind of stuff. This is how DoD projects go way over budget and behind schedule. Instead of "give us a long-range recon platform that the enemy is unlikely to notice," you get "give us a long-range recon platform that's stealthy, looks and flies like a bat, weighs this much, is this big, consumes this much power, can do all this other stuff that we think would be cool, etc., and by the way, you'll need to develop technology that's multiple generations ahead of what you have now." The former lends itself to more realistic requirements, useful incremental development, lower costs, and the like. The latter creates expensive messes.
That's the other tricky part. Someone will have to develop an onboard guidance system that can survive the awesome strain. Accomplish that and you will have a truly handy weapon. 200nm in a matter of minutes would be good enough to provide nice support for Marines ashore. It would be better than Tomahawk (albeit with a shorter range and no submunitions capability).
Since this weapon is out of my field, is it practical for anti-ship use? In theory, could you input the known location of an enemy battlegroup (plus a little predicted steaming time), fire some rounds downrange at the area, and then rely on onboard terminal guidance to hit the target? If so, you could replace Harpoon as well.
Heck, if they can make it work, they should develop small, less powerful versions for small boat and point defense. Wouldn't even need the onboard guidance at that point. Now you've replaced Tomahawk (maybe), Harpoon, Phalanx, and the Mk 38 gun. That would make the logistics tail a lot easier to handle.
And what happens as soon as the enemy figures out how to take advantage of the the "international laws" of warfare to beat the robots? Because I certainly can't think of any case in which humans have figured out how to predict and use an AI's behavior against it. Or, better yet, wait until the first time a robot on patrol duty shoots some kid in front of his mommy and a cameraman. It will happen because some combination of behavior and circumstances inevitably will combine to create an input that the robot's programming interprets as "kill." What are you supposed to say? Oops, there was a bug in the software? That ought to trigger big time public outcry.
Unless we get some serious leaps forward in AI, I think our best scenario for combat robots is, coincidentally, the best scenario for combat troops: every person you see is known to be hostile and collateral damage is not an issue. Achieve the objective and don't worry about red force casualties or collateral damage. In fact, the more enemy dead the better, because fewer enemy troops will be left to oppose you.
I think the real problem is that Army doesn't have any real experience with developing and sustaining large-scale, integrated, systems of systems. I also bet that the Army program managers in charge of FCS (and the generals/secretaries they report to) don't have a grasp of the advantages of open systems and smaller, competed contracts in terms of avoiding vendor lock-in and increasing competition. The Navy is just now figuring this stuff out and they've been at it for decades.
I think the next biggest problem is that the Army didn't have firm requirements to work toward. They thought not in terms of "what do we need to get done, and what do we need from FCS to accomplish it," but rather "what cool stuff can the new shiny do?" That leads to requirement creep and constant changes because you're basing your development requirements on capability and not on what doctrine says you need. So you start work and three years later someone comes up with a new Cool Thing that the generals are impressed by. Now you have to add that, requiring changes that require other changes and increase costs.
Oh, I don't know, it's like unintentionally creating a compliment machine. Imagine how confused these women will be when the machine thinks they're too young. On one hand, no cigarettes. On the other hand, they're being called young, which is one of the ultimate compliments in a society with women as obsessed with youth is Japan. Traditionally, single Japanese women are considered to be over the hill at 30. Women who are 25 used to be referred to as "Christmas cakes" because their desirability ended soon afterwards.
I have one cousin in her early 30s who the entire family is frantically trying to get married off. Objectively speaking, she's smoking hot (no jokes, come on) and nice to a fault, so I don't understand the problem. But when she turned 30, it's like someone flipped a switch in the minds of my relatives. I think it would be a nice pick me up for her once in a while, even if it's from a vending machine
This is not that much of an improvement for force protection unless procedures change. In the Cole incident, there was no marked "stay back or we'll shoot" perimeter. I doubt, then, that they would have been allowed to shoot without permission from a junior officer who may or may not have been present or in posession of the individual initiative required to shoot on gut feel and risk criminal prosecution/career suicide. This was in a known hostile port, mind you, where there would not have been time to go up the chain for permission, lock and load, bring weapons to bear, and engage before the fast boat hit.
Things could be worse with an automated system - what CO is going to trust that thing to protect his ship autonomously? Even a remotely-piloted system will face procedural problems similar to the ones I mentioned above.
Real benefits come from sending it into harms way in place of a ship. Porter fired on and sank two pirate "skiffs" a little while ago that were trying to hijack a benzene tanker. Rather than bring a lightly-armored DDG that close to a floating bomb and unknown hostiles, I'd rather send an armed robot.
If anything, the big shortage is of people to work with/direct/support the engineers, which in turn hurts growth in engineering jobs and salaries. The Navy is a great example. I work in a large building full of experienced, highly competent engineers. You name it, there's probably a team around here that knows all about it. The problem is that there aren't nearly enough people capable of coordinating all this work, working with the chain of command, and making the judgment calls necessary to keep programs on track. Based on the work I do every day, I don't understand what they did before my position was created (I'm not being arrogant and I wish I were joking).
Non-defense works the same way. How many good engineering teams out there are let down by idiotic management who can't or won't generate a functional business case, get engineers the support they need, take bullets for them when things go bad, etc? If the suits had their act together, then I would think that there would be higher demand for engineers at all price points because more of these companies would be thriving.
I'll use this when I'm IN the office. Now I can barricade people in their offices so that they can't run away before I ask them for things! Or keep them after work until I get what they promised me a week ago. All it needs is hardpoints for small arms and melee weapons and I'll have a perfect workflow enforcement machine. Let's see you blow off my quad chart request now, sucker!
I think you're half right. Leave Wal-Mart out of it and think economics. Think about it this way (one of my favorite antitrust examples from law and economics). Real case, btw, I think it was in San Francisco.
There once was an appliance store that had a showroom, informed sales staff, etc, with prices sufficiently high to pay for that overhead. Next door was a guy in a tiny shop with an appliance catalog. You bought from him, he took down the model number, and had it delivered directly to you. Naturally, his prices were lower. They were the only appliance stores for some suprisingly long distance. So people would go to the full-service store, check out appliances in person, get help from the staff, and then walk next door to buy their stuff. Eventually the full-service store got fed up and offered to place bigger orders with the area distributors if they cut off the other guy. He sued, of course, and won in federal court. Result: both soon went out of business because the bigger store lost too much business and shut down. The little guy then went out of business because he was entirely dependent on the big store for answering product questions, etc. Everyone lost, especially the local consumers.
It was argued (based on the Chicago School's philosophy of law and econ, long may it reign) by my professor that the decision was wrong because the the court should have interpreted the law in such a way so as to maximize economic efficiency. i.e., highest valued use of resources and all that. Consumers were better off with only the big store surviving, with higher prices, than no store surviving, because supply with higher prices is better than no supply at all.
This was before the internet, of course, but let's face it: lots of people still prefer to buy things like refrigerator in person.
Does this mean that China's one child policy is creating a race of Han superchildren? Unless, of course, you fail to have a son the first time, then you can try again.
I'm not surprised by this at all. There's actually an effort within the Navy now to build a massive shared, OSS repository of combat system software components and code for combat systems stuff. Everyone gets to examine code, fiddle with it, pick at it, adapt it, go play. And you're required to submit whatever you come up with to the same scrutiny. It's part of a larger effort to get away from lock-in with Raytheon, LockMart, etc. and get more competition and more small players. The surface warfare centers have experimented with creating their own quasi-incubators for small business industry to get a foot in the door. I've heard of a few neat products so far.
My only fear is that all of our efforts will go for nothing when some doofus admiral says, "Vendor X says he can do it cheaper. Drop everything and go prove that you really know what you're doing." Yup. All of my team's work grinds to a halt for 3 months while we pursue a damn wild goose chase to justify that we're more trustworthy than a retired O-6 who's now a salesman.
I remember watching a BBC newsclip once where one of their reporters was near Murmansk (northwestern Russia), talking to two Russian engineers working in the middle of a field of snow. He told them about the theory of global warming and they both visibly perked up. One asked him, "Really? How can we help?"
Maybe people will actually pay attention during meetings today. It always stuns me when my team has meetings to discuss, say, a plan to realign billions of dollars of procurement funds, and some senior jerkoff is too busy shooting off emails to pay attention to the rather more important matter at hand. I always joke that we need to slip in a random bit of classified info so that everyone will be required to check all their devices at the door.
I don't understand how people can possibly believe that technology make for better schools. Sure, it's nice and all, but things like involved (but not sociopathic) parents, competent teachers, accountability for failure, the ability to punish misbehaving kids, etc., are what makes schools better. None of those things require fancy buildings and the like. I guess it comes down to throwing $$$ at silver bullets instead of actually fixing things.
Slashdot Mirror
Contractors, definitely. The Feds needs math types in multiple agencies. The Census Bureau does a lot of stat work, for example. DoD needs math types at the warfare centers scattered around the country. Does she have any interest in Human-Systems Integration or training systems? There are a lot of people trying to find ways to get all those personnel trained up on new systems.
Alternatively, teach at a private school. They have the advantage of being able to select and expel their students and there will be less bureaucracy.
No, no. Since everything appears to cause cancer and other nasty effects in California, the first thing you should consider is whether California is in fact the cause.
You know what else is equally dumb, but has gotten a free pass? Touchscreen interfaces in cars. I make it a point to buy cars with physical controls so that I can do things by touch alone. Plus, the designers always seem to make it a point to bury settings in nested menus; this only makes it worse. 4.6 seconds is probably how long it takes some people to change the station on the radio. And of course, they have to look down at the screen to do it.
People running around and doing the equivalent of tearing down billboards and defacing storefronts. Big whoops. Last I checked, the major players in the global financial network have actual power. And most central/federal governments, too. This guy needs a cold beer and a sense of proportion. Ok, maybe not the latter; we know that's the one thing that no one in the universe can afford to have. Make it two beers, then.
I'd say that burning too much time and money on graphics, sound, FMV, and voice acting at the expense of mechanics, plot and bug-freeness has become a cliche in and of itself.
Obviously the solution is to go back to text-based gaming. OK, fine, EGA and the PC speaker.
Look at Civ or Galactic Civilizations. Those non-FPS/RTS games were turn-based and required thought and planning. Old RPGs are the same way. People like me who grew up with Wasteland and its contemporaries miss the engagement and cleverness. I'm not interested in a fast twitch game and am willing to pay for a game that makes me think and spend time to beat. It's merely a bonus that it's a sequel to one of the all time greats that we're talking about. I'm contributing tonight and then I'm going to fire up my old copy of Wasteland and go see how little firepower I can beat Guardian Citadel with this time. Exploded blood sausage ftw!
I take issue with DARPA's assertion in TFA that formal verification cannot be scaled up to work on a modern weapon system. My office has done it for very long time and we are very software-intensive. We and our contractors just had to get smarter as the system became more complex and requirements became steeper.
Nonetheless, I would be interested in the potential of such a process to find sneak circuits and latent problems. Use it during the development process prior to integration and verification.
It's no surprise that they can keep a secret. Civilian personnel in defense and intelligence are, by and large, capable of keeping a secret when it counts. They are motivated to do the job and keep such secrets as are necessary to get it done (this does not include fraud, but the Important People do what they want). They know that info getting out could cause soldiers to die and wars to be lost. Speaking for my colleagues, it is not just another job because we know what's at stake.
Now, give classified info to some dummy in Congress...that's scary. Those people get their clearances by virtue of their jobs and not because of their own merits. And the spill procedures that we have to follow don't apply to them. Just like all those other laws and regulations don't apply to them.
As an acquisition guy, I hate this kind of stuff. This is how DoD projects go way over budget and behind schedule. Instead of "give us a long-range recon platform that the enemy is unlikely to notice," you get "give us a long-range recon platform that's stealthy, looks and flies like a bat, weighs this much, is this big, consumes this much power, can do all this other stuff that we think would be cool, etc., and by the way, you'll need to develop technology that's multiple generations ahead of what you have now." The former lends itself to more realistic requirements, useful incremental development, lower costs, and the like. The latter creates expensive messes.
That's the other tricky part. Someone will have to develop an onboard guidance system that can survive the awesome strain. Accomplish that and you will have a truly handy weapon. 200nm in a matter of minutes would be good enough to provide nice support for Marines ashore. It would be better than Tomahawk (albeit with a shorter range and no submunitions capability).
Since this weapon is out of my field, is it practical for anti-ship use? In theory, could you input the known location of an enemy battlegroup (plus a little predicted steaming time), fire some rounds downrange at the area, and then rely on onboard terminal guidance to hit the target? If so, you could replace Harpoon as well.
Heck, if they can make it work, they should develop small, less powerful versions for small boat and point defense. Wouldn't even need the onboard guidance at that point. Now you've replaced Tomahawk (maybe), Harpoon, Phalanx, and the Mk 38 gun. That would make the logistics tail a lot easier to handle.
And what happens as soon as the enemy figures out how to take advantage of the the "international laws" of warfare to beat the robots? Because I certainly can't think of any case in which humans have figured out how to predict and use an AI's behavior against it. Or, better yet, wait until the first time a robot on patrol duty shoots some kid in front of his mommy and a cameraman. It will happen because some combination of behavior and circumstances inevitably will combine to create an input that the robot's programming interprets as "kill." What are you supposed to say? Oops, there was a bug in the software? That ought to trigger big time public outcry.
Unless we get some serious leaps forward in AI, I think our best scenario for combat robots is, coincidentally, the best scenario for combat troops: every person you see is known to be hostile and collateral damage is not an issue. Achieve the objective and don't worry about red force casualties or collateral damage. In fact, the more enemy dead the better, because fewer enemy troops will be left to oppose you.
I think the real problem is that Army doesn't have any real experience with developing and sustaining large-scale, integrated, systems of systems. I also bet that the Army program managers in charge of FCS (and the generals/secretaries they report to) don't have a grasp of the advantages of open systems and smaller, competed contracts in terms of avoiding vendor lock-in and increasing competition. The Navy is just now figuring this stuff out and they've been at it for decades.
I think the next biggest problem is that the Army didn't have firm requirements to work toward. They thought not in terms of "what do we need to get done, and what do we need from FCS to accomplish it," but rather "what cool stuff can the new shiny do?" That leads to requirement creep and constant changes because you're basing your development requirements on capability and not on what doctrine says you need. So you start work and three years later someone comes up with a new Cool Thing that the generals are impressed by. Now you have to add that, requiring changes that require other changes and increase costs.
Trekkie Monster was right! Can't wait for the first time a flight attendant has to ask a customer to stop surfing for porn.
Oh, I don't know, it's like unintentionally creating a compliment machine. Imagine how confused these women will be when the machine thinks they're too young. On one hand, no cigarettes. On the other hand, they're being called young, which is one of the ultimate compliments in a society with women as obsessed with youth is Japan. Traditionally, single Japanese women are considered to be over the hill at 30. Women who are 25 used to be referred to as "Christmas cakes" because their desirability ended soon afterwards.
I have one cousin in her early 30s who the entire family is frantically trying to get married off. Objectively speaking, she's smoking hot (no jokes, come on) and nice to a fault, so I don't understand the problem. But when she turned 30, it's like someone flipped a switch in the minds of my relatives. I think it would be a nice pick me up for her once in a while, even if it's from a vending machine
I can hear it now. Better yet, what if I stand in my module, throw a missile through a second module, and hit you in a third module? Then what?
This is not that much of an improvement for force protection unless procedures change. In the Cole incident, there was no marked "stay back or we'll shoot" perimeter. I doubt, then, that they would have been allowed to shoot without permission from a junior officer who may or may not have been present or in posession of the individual initiative required to shoot on gut feel and risk criminal prosecution/career suicide. This was in a known hostile port, mind you, where there would not have been time to go up the chain for permission, lock and load, bring weapons to bear, and engage before the fast boat hit.
Things could be worse with an automated system - what CO is going to trust that thing to protect his ship autonomously? Even a remotely-piloted system will face procedural problems similar to the ones I mentioned above.
Real benefits come from sending it into harms way in place of a ship. Porter fired on and sank two pirate "skiffs" a little while ago that were trying to hijack a benzene tanker. Rather than bring a lightly-armored DDG that close to a floating bomb and unknown hostiles, I'd rather send an armed robot.
If anything, the big shortage is of people to work with/direct/support the engineers, which in turn hurts growth in engineering jobs and salaries. The Navy is a great example. I work in a large building full of experienced, highly competent engineers. You name it, there's probably a team around here that knows all about it. The problem is that there aren't nearly enough people capable of coordinating all this work, working with the chain of command, and making the judgment calls necessary to keep programs on track. Based on the work I do every day, I don't understand what they did before my position was created (I'm not being arrogant and I wish I were joking).
Non-defense works the same way. How many good engineering teams out there are let down by idiotic management who can't or won't generate a functional business case, get engineers the support they need, take bullets for them when things go bad, etc? If the suits had their act together, then I would think that there would be higher demand for engineers at all price points because more of these companies would be thriving.
I'll use this when I'm IN the office. Now I can barricade people in their offices so that they can't run away before I ask them for things! Or keep them after work until I get what they promised me a week ago. All it needs is hardpoints for small arms and melee weapons and I'll have a perfect workflow enforcement machine. Let's see you blow off my quad chart request now, sucker!
I think you're half right. Leave Wal-Mart out of it and think economics. Think about it this way (one of my favorite antitrust examples from law and economics). Real case, btw, I think it was in San Francisco.
There once was an appliance store that had a showroom, informed sales staff, etc, with prices sufficiently high to pay for that overhead. Next door was a guy in a tiny shop with an appliance catalog. You bought from him, he took down the model number, and had it delivered directly to you. Naturally, his prices were lower. They were the only appliance stores for some suprisingly long distance. So people would go to the full-service store, check out appliances in person, get help from the staff, and then walk next door to buy their stuff. Eventually the full-service store got fed up and offered to place bigger orders with the area distributors if they cut off the other guy. He sued, of course, and won in federal court. Result: both soon went out of business because the bigger store lost too much business and shut down. The little guy then went out of business because he was entirely dependent on the big store for answering product questions, etc. Everyone lost, especially the local consumers.
It was argued (based on the Chicago School's philosophy of law and econ, long may it reign) by my professor that the decision was wrong because the the court should have interpreted the law in such a way so as to maximize economic efficiency. i.e., highest valued use of resources and all that. Consumers were better off with only the big store surviving, with higher prices, than no store surviving, because supply with higher prices is better than no supply at all.
This was before the internet, of course, but let's face it: lots of people still prefer to buy things like refrigerator in person.
And come back safely - NASA needs you.
Does this mean that China's one child policy is creating a race of Han superchildren? Unless, of course, you fail to have a son the first time, then you can try again.
I'm not surprised by this at all. There's actually an effort within the Navy now to build a massive shared, OSS repository of combat system software components and code for combat systems stuff. Everyone gets to examine code, fiddle with it, pick at it, adapt it, go play. And you're required to submit whatever you come up with to the same scrutiny. It's part of a larger effort to get away from lock-in with Raytheon, LockMart, etc. and get more competition and more small players. The surface warfare centers have experimented with creating their own quasi-incubators for small business industry to get a foot in the door. I've heard of a few neat products so far.
My only fear is that all of our efforts will go for nothing when some doofus admiral says, "Vendor X says he can do it cheaper. Drop everything and go prove that you really know what you're doing." Yup. All of my team's work grinds to a halt for 3 months while we pursue a damn wild goose chase to justify that we're more trustworthy than a retired O-6 who's now a salesman.
Wish us luck. We'll bloody well need it.
I remember watching a BBC newsclip once where one of their reporters was near Murmansk (northwestern Russia), talking to two Russian engineers working in the middle of a field of snow. He told them about the theory of global warming and they both visibly perked up. One asked him, "Really? How can we help?"
Maybe people will actually pay attention during meetings today. It always stuns me when my team has meetings to discuss, say, a plan to realign billions of dollars of procurement funds, and some senior jerkoff is too busy shooting off emails to pay attention to the rather more important matter at hand. I always joke that we need to slip in a random bit of classified info so that everyone will be required to check all their devices at the door.
I don't understand how people can possibly believe that technology make for better schools. Sure, it's nice and all, but things like involved (but not sociopathic) parents, competent teachers, accountability for failure, the ability to punish misbehaving kids, etc., are what makes schools better. None of those things require fancy buildings and the like. I guess it comes down to throwing $$$ at silver bullets instead of actually fixing things.