As others have pointed out below, while it it true there was a units mix-up, this error wasn't caught due to other, more wide ranging problems. IEEE did a great writeup of this a while back (article link - didn't link to IEEE directly since you probably have to be a member to see the article). Very interesting reading. In summary...
First the spacecraft was asymmetric, causing some issues with the stabilizing flywheels and the onboard thrusters (used for major course corrections). Second, the person doing the calculations for the major course corrections noticed that the burn time (calculated using the bad units) didn't look right compared with previous missions. However, his management made him prove that the calculations were wrong, instead of proving they were right (presumably knowing that they would be different, given the first point about the asymmetries). He didn't catch the units error, and since he couldn't prove they were wrong they went along as if nothing happened. The article was really pointing out that while this was a technical error, the more fundamental issue was a management and culture issue. To me this made for an interesting case study in how to handle unknowns in a mission critical system - assuming things are wrong until proven otherwise, not vice versa.
(I don't seem to have the Spectrum issue with me, but I seem to remember it had some other articles about related management/culture failures).
At the risk of being pedantic and not really contributing to an older thread (hey, it's slashdot, right?), radar originally/typically measures range - that's what the last 'R' stands for (RAdio Detection And Ranging). True, that's evolved to where some modern radars use Doppler for things, but ranging is it's bread and butter.
That being said, I think I agree here - vehicle speed measurements seem like an odd application of LIDAR, given the wavelengths. I haven't worked out the math, but I given the distances/velocities we're talking about here I'd be interested to see how speed error is impacted by pointing "jitter" (that is, the beam moving around b/c a human is pointing the beam at a moving target that has many features that are large compared to the laser's wavelength). It would be even more interesting to compare that against that radar errors, especially when on considers that the radar might have problems localizing the speed when vehicles are close (as others have pointed out). One would think someone did tests like these once upon a time, but.....
Perhaps I should rush to patent a combo device that uses both technologies, claiming that the two pieces of evidence together overcome their individual weaknesses. Apparently you don't even need proof that it works!
While in general you are right, fire departments are quite concerned about how well their radios work (a friend is a firefighter, and they go on test runs of large industrial buildings in their area to make sure they can communicate), in this case I'm not sure it matters.
I believe the passive systems the OP was referring to use their own, local "cell tower", to which all nearby phones try to communicate (since this tower is the closest, by rule that's the one the phones will try to talk to). If this tower isn't connected to the outside world, though, no calls/texts get through. All other RF comms, however, provided they aren't using the cell network, will operate just as they normally would. As someone else suggested, I think there were even provisions to allow emergency calls only pass through.
I'm too lazy to search, but I think there was a slashdot story about French theaters doing something similar. Someone more bored than I am will get some mod points if they can find it.
Ok, I wasn't going to post, but this got modded up as "Insightful" and I couldn't resist - this has to be one of the dumbest posts in this thread...
What do you think that bale of hay is sitting on? Radars recieve ground bounces all the time - even in airborne applications. Usually radar people call that "clutter", since if you're looking for airborne targets it's information you don't want - here it's information you *do* want. Depending on how the radar is mounted, it could create a ground map and trigger alarms when the ground return is either really close or really far away. It really comes down to a sensor fusion problem - by using the combination of radar, lidar, laser range finder (like another posted replied), vision, etc one could determine that there's a large obstruction in the way - either a "postive" one (like your bale of hay) or a "negative" one (like your cliff).
The problem isn't in detecting the drop off - it's in figuring out what to do when you see it. A vehicle that comes the edge of the Grand Canyon is going to have a go a long way to drive around it. This isn't a problem with the sensor, it's a route finding problem. Heck, your sudden drop example is an easier problem - it's probably more difficult to realize you're decending gently in to a canyon that you can't get out of on the other side (again, this is a route finding problem - your route finding software has to be smart enough to avoid this obstacle in the first place, given a map of the area)
Who modded this insightful? Probably someone who's never been to a conference...
Take a look at the header - this was submitted to a conference, *not* a full peer-reviewed journal. Many conferences (I know for sure most IEEE conferences are like this) limit paper submissions to 4 pages. URSI (Union Radio-Scientifique Internationale - they're just like IEEE Antennas and Propigation Society, with mostly the same members and co-host their conference) papers are even limited to 1 page for their conference. *Conference* papers really more discussion points than full blown "proofs". I'd suspect he'd follow this up with an "official" paper in one of their peer-reviewed journals.
The early posts (and the submitter) seem to be missing the point...
The $50k reward is just that - a reward. It's not like with the X Prize that the reward covered development costs. It's just an incentive - the *real* reward comes after you win. That's when you secure licensing deals, like Rutan did with Virgin.
Whoa - how is this insightful? Not to nitpick too much, but to say "Defense is excluded" isn't entirely accurate. A quick google search of 'budget cuts defense department' returns this Washington Post article as the *first* hit. The short of it - $55 billion in cuts over 6 years (same timeframe as the $300 billion in cuts the parent mentions), including $5 billion from missile defense.
That might not be a huge chunk of the $300 billion, but during time of war I'd say that's definately more than "excluding" the DoD.
The Diebold one I used looked like this: http://www.diebold.com/dieboldes/accuvote_ts.htm , which is quite different than the one you described (it says Diebold in big letters across the top and is a touchscreen based deal - no LEDs, buttons, or bells). I was actually quite pleased with the interface (but would much rather have it really just print out a scan-tron type sheet to be later read by an optical scanner instead of recording the vote on a smartcard) - it would have been pretty hard to vote for something by mistake. I can't comment on how well they held up over the course of the day, but there didn't seem to be any problems while I was there.
That's still interesting/confusing/disturbing that even though state law requires a paper trail you're not really getting one (like another poster in this thread said - a paper log printed at the end of the day shouldn't count as a paper trail).
I'm not so sure you understand correctly.... My family is from Ohio (Akron) and they didn't use the Diebold machines, but rather old punch card style machines. Friends in the Columbus area said the same thing. I voted on a Diebold machine in Maryland, which did not produce a paper receipt (well, it didn't produce one that I saw, anyway). However, http://www.sos.state.oh.us/sos/hava/index.htmlOhio state law *requires* a paper trail for electronic voting machines. This would seem to imply no Diebold machines in Ohio....
Actually, isn't it really a much higher percentage than that? He needs those 140,000 just to make up the gap, but your numbers ignore the 50,000 remaining ballots. To win Kerry would need at least half of those remaining ballots (assuming Bush gets all the others), so he really needs 165,001/190,000 = 86.8%. That really seems to shore up your core point even more...
Slashdot Mirror
As others have pointed out below, while it it true there was a units mix-up, this error wasn't caught due to other, more wide ranging problems. IEEE did a great writeup of this a while back (article link - didn't link to IEEE directly since you probably have to be a member to see the article). Very interesting reading. In summary...
First the spacecraft was asymmetric, causing some issues with the stabilizing flywheels and the onboard thrusters (used for major course corrections). Second, the person doing the calculations for the major course corrections noticed that the burn time (calculated using the bad units) didn't look right compared with previous missions. However, his management made him prove that the calculations were wrong, instead of proving they were right (presumably knowing that they would be different, given the first point about the asymmetries). He didn't catch the units error, and since he couldn't prove they were wrong they went along as if nothing happened. The article was really pointing out that while this was a technical error, the more fundamental issue was a management and culture issue. To me this made for an interesting case study in how to handle unknowns in a mission critical system - assuming things are wrong until proven otherwise, not vice versa.
(I don't seem to have the Spectrum issue with me, but I seem to remember it had some other articles about related management/culture failures).
At the risk of being pedantic and not really contributing to an older thread (hey, it's slashdot, right?), radar originally/typically measures range - that's what the last 'R' stands for (RAdio Detection And Ranging). True, that's evolved to where some modern radars use Doppler for things, but ranging is it's bread and butter.
That being said, I think I agree here - vehicle speed measurements seem like an odd application of LIDAR, given the wavelengths. I haven't worked out the math, but I given the distances/velocities we're talking about here I'd be interested to see how speed error is impacted by pointing "jitter" (that is, the beam moving around b/c a human is pointing the beam at a moving target that has many features that are large compared to the laser's wavelength). It would be even more interesting to compare that against that radar errors, especially when on considers that the radar might have problems localizing the speed when vehicles are close (as others have pointed out). One would think someone did tests like these once upon a time, but.....
Perhaps I should rush to patent a combo device that uses both technologies, claiming that the two pieces of evidence together overcome their individual weaknesses. Apparently you don't even need proof that it works!
While in general you are right, fire departments are quite concerned about how well their radios work (a friend is a firefighter, and they go on test runs of large industrial buildings in their area to make sure they can communicate), in this case I'm not sure it matters.
I believe the passive systems the OP was referring to use their own, local "cell tower", to which all nearby phones try to communicate (since this tower is the closest, by rule that's the one the phones will try to talk to). If this tower isn't connected to the outside world, though, no calls/texts get through. All other RF comms, however, provided they aren't using the cell network, will operate just as they normally would. As someone else suggested, I think there were even provisions to allow emergency calls only pass through.
I'm too lazy to search, but I think there was a slashdot story about French theaters doing something similar. Someone more bored than I am will get some mod points if they can find it.
While it's true that you used to have to pay for access at MHT, that's no longer the case :)
http://www.flymanchester.com/about/news.php - free wifi announcement down near the bottom
Ok, I wasn't going to post, but this got modded up as "Insightful" and I couldn't resist - this has to be one of the dumbest posts in this thread...
What do you think that bale of hay is sitting on? Radars recieve ground bounces all the time - even in airborne applications. Usually radar people call that "clutter", since if you're looking for airborne targets it's information you don't want - here it's information you *do* want. Depending on how the radar is mounted, it could create a ground map and trigger alarms when the ground return is either really close or really far away. It really comes down to a sensor fusion problem - by using the combination of radar, lidar, laser range finder (like another posted replied), vision, etc one could determine that there's a large obstruction in the way - either a "postive" one (like your bale of hay) or a "negative" one (like your cliff).
The problem isn't in detecting the drop off - it's in figuring out what to do when you see it. A vehicle that comes the edge of the Grand Canyon is going to have a go a long way to drive around it. This isn't a problem with the sensor, it's a route finding problem. Heck, your sudden drop example is an easier problem - it's probably more difficult to realize you're decending gently in to a canyon that you can't get out of on the other side (again, this is a route finding problem - your route finding software has to be smart enough to avoid this obstacle in the first place, given a map of the area)
Who modded this insightful? Probably someone who's never been to a conference...
Take a look at the header - this was submitted to a conference, *not* a full peer-reviewed journal. Many conferences (I know for sure most IEEE conferences are like this) limit paper submissions to 4 pages. URSI (Union Radio-Scientifique Internationale - they're just like IEEE Antennas and Propigation Society, with mostly the same members and co-host their conference) papers are even limited to 1 page for their conference. *Conference* papers really more discussion points than full blown "proofs". I'd suspect he'd follow this up with an "official" paper in one of their peer-reviewed journals.
The early posts (and the submitter) seem to be missing the point... The $50k reward is just that - a reward. It's not like with the X Prize that the reward covered development costs. It's just an incentive - the *real* reward comes after you win. That's when you secure licensing deals, like Rutan did with Virgin.
Whoa - how is this insightful? Not to nitpick too much, but to say "Defense is excluded" isn't entirely accurate. A quick google search of 'budget cuts defense department' returns this Washington Post article as the *first* hit. The short of it - $55 billion in cuts over 6 years (same timeframe as the $300 billion in cuts the parent mentions), including $5 billion from missile defense.
That might not be a huge chunk of the $300 billion, but during time of war I'd say that's definately more than "excluding" the DoD.
The Diebold one I used looked like this: http://www.diebold.com/dieboldes/accuvote_ts.htm , which is quite different than the one you described (it says Diebold in big letters across the top and is a touchscreen based deal - no LEDs, buttons, or bells). I was actually quite pleased with the interface (but would much rather have it really just print out a scan-tron type sheet to be later read by an optical scanner instead of recording the vote on a smartcard) - it would have been pretty hard to vote for something by mistake. I can't comment on how well they held up over the course of the day, but there didn't seem to be any problems while I was there.
That's still interesting/confusing/disturbing that even though state law requires a paper trail you're not really getting one (like another poster in this thread said - a paper log printed at the end of the day shouldn't count as a paper trail).
I'm not so sure you understand correctly.... My family is from Ohio (Akron) and they didn't use the Diebold machines, but rather old punch card style machines. Friends in the Columbus area said the same thing. I voted on a Diebold machine in Maryland, which did not produce a paper receipt (well, it didn't produce one that I saw, anyway). However, http://www.sos.state.oh.us/sos/hava/index.htmlOhio state law *requires* a paper trail for electronic voting machines. This would seem to imply no Diebold machines in Ohio....
Actually, isn't it really a much higher percentage than that? He needs those 140,000 just to make up the gap, but your numbers ignore the 50,000 remaining ballots. To win Kerry would need at least half of those remaining ballots (assuming Bush gets all the others), so he really needs 165,001/190,000 = 86.8%. That really seems to shore up your core point even more...
Talk about getting off topic....