I've noticed that the internet tends to exacerbate the problems on recognizing science in a sea of opinions. I've been sharing my research in materials science on an internet forum and posting about other topics in technical forums. I've noticed that the most voluminous and most vociferous speakers tend to believed by the masses on the internet far above people who actually have credible knowledge. The knowledgable tend to explain their facts once or twice and answer questions. The unknowledgable tend to continuously post the same drivel. If you google the question you end up with 100 posts of that drivel and 2 with the correct information. Pagerank tends to lead people to bad information in the hard sciences.
Laypeople who base their knowledge on google concensus rather than fact get a very skewed answer to certain questions.When these same laypeople google the climate debate, I suspect all they find is the right leaning and left leaning conspiracy theories. Good information is replicated once because to those in the know, facts stand on their own and need not be repeated. Stupid information is posted again and again because belief in non-factual theories (in keeping with ones own minimally informed beliefs) is essentially a form of religion. Until scientists continuously post the correct information with the same religious fervor, the masses will continue to believe what they want to since its all they hear.
I think that Enemies the air force considers for projects like this are taxpayers and congress. The only way the bus design is of interest to actual enemies (likely ignored in the analysis) IMHO is that it might work better than what they were going to use. Neither congress nor taxpayers are too keen on watching 200 million dollars get spent reinventing the wheel. If the system design is so bad that you have to hide it to make it "secure" then the design has already failed some important milestones.
Space is a challenging environment and there are relatively high numbers of bit errors there compared to earth. You can't assume that processors execute instructions correctly each time, that memory contains what you think it does, or that a bus transfers the data you told it to transfer. We can take all of these things for granted on earth for the most part. Such a space bus likely would require a much higher degree of error correction and preferably forward error correction than would a garden variety earth bus.
I have a sager np5950v laptop with two nvidia quadro FX 2500 cards. The laptop display is attached to the second card rather than the first one. Suse's video configuration always tries to put video by default on the first card. I had to screw with bus ID's and xorg.conf in suse 10.0 and 11.1. It never detects it correctly. I also find that every time I do a video driver update I have to copy the old xorg.conf over the one autogenerated by the configuration tools since they don't get it right. I've contributed to parts of the yast2 video stuff before in suse so I can work around the failures but it is annoying and definitely the kind of problem to google for. It's also annoying that the nvidia drivers don't recognize this configuration for SLI though perhaps they have fixed it since I last update drivers.
I once was subcontracted by a friend to examine source code provided to him under court seal about what I believe had been a software patent case but might have been misappropriation of trade secrets or something similar. The source code was not released publicly, it was just released to my friend who was serving as a consultant to the plaintiff who we will call party A. In my opinion, party A had patented linear interpolation and the whole thing should have been thrown out but that wasn't what the case was about. What the case was about was whether party B had copied party A's obscurely implemented form of linear interpolation. The details of the algorithm were so wacky that it was obvious that party B had copied A's implementation since no sane person would have done it the way A did with the same variable names. What I couldn't figure out in the whole thing is why anybody would copy something like that. B wouldn't have gotten into trouble if they had derived the entire thing from first principles and used the obvious textbook implementation. Instead, they used one that looked almost certainly like it had been stolen directly from Party A's source code.
At any rate, I have seen with my own eyes that code is often released under Non Disclosure Agreement and court seal to neutral parties representing an adversarial position for the purpose of resolving a dispute. I think that this Apple request will likely generate source code which will be reviewed by an expert. It certainly won't result in GNU-iPhone 1.0 unless someone royally screws up. Were that screw up to occur and the code was accidentally released, whoever was responsible had better move somewhere that doesn't have an extradition treaty with the U.S.
I remember hearing about building a one instruction computer back in engineering school. The one I heard about was based on Subtract and Branch if Not Equal. My roommate at the time figured it ought to be a way to get a very high clock rate. It seems like he found a proof in a hoary old book that such a computer was in fact Turing complete. I'm sure I'll get flamed for posting a vague recollection but. . . here it is.
I worked on a low cost military training system that used some older civilian GPS hardware. Our GPS's provided instantaneous velocity at 1Hz and instantaneous position at 1Hz. The velocity tended to be much better than the position in noisy GPS conditions. You can also use the velocity to kalman filter the position leading to increased position accuracy. It's hard to tell what a GPS is displaying but internally, the velocity measurement is very accurate but at too low a time resolution for some situations involving moving vehicles. If the GPS in the article was logging at 30 second intervals, it would be very difficult to know anything about the instantaneous speed of the vehicle in question. That my $.02
I was advised by my company's IP lawyer about some work I did on a sabbatical that the easiest way to protect yourself by not patenting is to file a provisional patent and then abandon the application. The fees and hassle are lower that the full patent and the provisional patent is good for a year. Finally, one can also file a statutory invention disclosure.
Missing digits in the display is similar to one of the reasons for the Apollo 13 disaster. The cryotank that exploded was monitored by a temperature sensor with too small a range. During ground testing, the tank had become hot enough for the insulation to burn off of the wiring. Unfortunately, the annunciator on the temperature sensor had no way to record a temperature much above room temperature. If this annunciator had been designed better, engineers may very well have noticed the insane temperature, detected damage caused to that tank on the ground, and repaired it before the launch.
The demo seemed to show that they actually had this down and were smart enough not to make those mistakes. The lead engineer certainly emphasized that they were cognizant of these kinds of problems and made every attempt to avoid them. Once it's out of beta we can all see.
These guys briefed at a company meeting the other day and offered a private beta to those of us with i-phones. Their tool allows you to submit natural language queries for things that involve transactions. You can tell your Siri enabled phone to order you a pineapple pizza and it will find pizza restaurants with web ordering API's and then show you the prices for what you asked for and offer to let you buy them. In the case of pizza during the demo, it showed pizza Hut and Dominoes. They're working towards an interface that would allow you to say "Book me on the next flight to chicago!" You can tell siri, "Get me a copy of $bookname" and it will search amazon and other services with buy online API's and offer to purchase the book for you.
The bottom line to me is that it looked powerful and scary at the same time. It most definitely isn't clippy.
Pay for a consultation with a licensed engineer. Make clear that this is a work for hire situation. Make him/her sign an NDA. At least here in Alabama, complaining to the professional engineering board if the engineer steals your idea after will likely get his/her license suspended and an embarassing writeup in the yearly bulletin from the State Board of Engineers.
This was disproved by Claude Shannon of Bell Labs fame using information theory (a branch of math that looks dangerously like computer science). In short, the amount of energy required to encode the demon's single bit of information about each molecule is exactly balanced by the amount of energy that could be extracted by the work of Maxwell's demon.
Take a look at www.getinthecockpit.com. These guys sell a set of VR goggles and the gear to put a camera in the cockpit of a model aircraft and beam live video back over the RF link. DIsclaimer: it's a distant cousin of mine's company.
Google Therac 25. It was a radiation treatment machine with software control and no mechanical interlocks. It killed circa 100 people in Latin America by exposing them to vastly higher doses of radiation than it should have. One of the worst glitches was a user interface lockup which left the shutter to the radiation source open.
They said it was sensitive information, not classified information. While the rules for military and contractor handling of classified material differ slightly, sensitive information is not protected by the same rules that govern classified information. Leak a classified drive: a bunch of people go to jail. Leak a sensitive drive and everybody goes to slashdot and gets sent to bed without a cookie.
The mistake in thinking that it's a bad thing to never have the data be the same is roughly speaking part of how the Germans lost WWII. The British broke the Enigma cypher by figuring out that a given letter was _NEVER_ encoded as the same letter. That tiny blip in the probability function allowed breaking many coded messages if they could get a small amount of cleartext such as the weather report.
They don't post an individual subscription rate. There have to be a lot more than 60 institutions that subscribe. I'd also bet that almost all subscribing institutions have a deal where they get more than just that journal on a package deal. Since much of this research was funded by the government, I argue that we are paying for the information twice. I'd be happier myself if the information was free and they charged for the indexing. If you magnify the $23,000 a year subscription fee by the thousands of journals in a research library, there seems to be a huge cost to society involved.
I went to ye olde library today to get copies of 2 Articles from the Journal of Applied Polymer Sciences, a Wiley Interscience Publication. Xeroxing the articles under fair use from the library was free for me.
The Whiskey Tango Foxtrot Moment came when I checked online to find out how much it would cost to subscribe to the journal. I thought someone misplaced a decimal point: $23,245 a year is the institutional subscription rate! That's about what I paid yearly in college tuition back when I was in college. Even worse, it's almost the value of the lab equipment I'm using in the work I've been doing on my own time.
Slashdot Mirror
I've noticed that the internet tends to exacerbate the problems on recognizing science in a sea of opinions. I've been sharing my research in materials science on an internet forum and posting about other topics in technical forums. I've noticed that the most voluminous and most vociferous speakers tend to believed by the masses on the internet far above people who actually have credible knowledge. The knowledgable tend to explain their facts once or twice and answer questions. The unknowledgable tend to continuously post the same drivel. If you google the question you end up with 100 posts of that drivel and 2 with the correct information. Pagerank tends to lead people to bad information in the hard sciences.
Laypeople who base their knowledge on google concensus rather than fact get a very skewed answer to certain questions.When these same laypeople google the climate debate, I suspect all they find is the right leaning and left leaning conspiracy theories. Good information is replicated once because to those in the know, facts stand on their own and need not be repeated. Stupid information is posted again and again because belief in non-factual theories (in keeping with ones own minimally informed beliefs) is essentially a form of religion. Until scientists continuously post the correct information with the same religious fervor, the masses will continue to believe what they want to since its all they hear.
Definitely needs Nixie tubes. . .
I think that Enemies the air force considers for projects like this are taxpayers and congress. The only way the bus design is of interest to actual enemies (likely ignored in the analysis) IMHO is that it might work better than what they were going to use. Neither congress nor taxpayers are too keen on watching 200 million dollars get spent reinventing the wheel. If the system design is so bad that you have to hide it to make it "secure" then the design has already failed some important milestones.
Space is a challenging environment and there are relatively high numbers of bit errors there compared to earth. You can't assume that processors execute instructions correctly each time, that memory contains what you think it does, or that a bus transfers the data you told it to transfer. We can take all of these things for granted on earth for the most part. Such a space bus likely would require a much higher degree of error correction and preferably forward error correction than would a garden variety earth bus.
I have a sager np5950v laptop with two nvidia quadro FX 2500 cards. The laptop display is attached to the second card rather than the first one. Suse's video configuration always tries to put video by default on the first card. I had to screw with bus ID's and xorg.conf in suse 10.0 and 11.1. It never detects it correctly. I also find that every time I do a video driver update I have to copy the old xorg.conf over the one autogenerated by the configuration tools since they don't get it right. I've contributed to parts of the yast2 video stuff before in suse so I can work around the failures but it is annoying and definitely the kind of problem to google for. It's also annoying that the nvidia drivers don't recognize this configuration for SLI though perhaps they have fixed it since I last update drivers.
I once was subcontracted by a friend to examine source code provided to him under court seal about what I believe had been a software patent case but might have been misappropriation of trade secrets or something similar. The source code was not released publicly, it was just released to my friend who was serving as a consultant to the plaintiff who we will call party A. In my opinion, party A had patented linear interpolation and the whole thing should have been thrown out but that wasn't what the case was about. What the case was about was whether party B had copied party A's obscurely implemented form of linear interpolation. The details of the algorithm were so wacky that it was obvious that party B had copied A's implementation since no sane person would have done it the way A did with the same variable names. What I couldn't figure out in the whole thing is why anybody would copy something like that. B wouldn't have gotten into trouble if they had derived the entire thing from first principles and used the obvious textbook implementation. Instead, they used one that looked almost certainly like it had been stolen directly from Party A's source code.
At any rate, I have seen with my own eyes that code is often released under Non Disclosure Agreement and court seal to neutral parties representing an adversarial position for the purpose of resolving a dispute. I think that this Apple request will likely generate source code which will be reviewed by an expert. It certainly won't result in GNU-iPhone 1.0 unless someone royally screws up. Were that screw up to occur and the code was accidentally released, whoever was responsible had better move somewhere that doesn't have an extradition treaty with the U.S.
I remember hearing about building a one instruction computer back in engineering school. The one I heard about was based on Subtract and Branch if Not Equal. My roommate at the time figured it ought to be a way to get a very high clock rate. It seems like he found a proof in a hoary old book that such a computer was in fact Turing complete. I'm sure I'll get flamed for posting a vague recollection but. . . here it is.
I'm class of 1997. Good to meet you.
Are you at Harvey Mudd?
I worked on a low cost military training system that used some older civilian GPS hardware. Our GPS's provided instantaneous velocity at 1Hz and instantaneous position at 1Hz. The velocity tended to be much better than the position in noisy GPS conditions. You can also use the velocity to kalman filter the position leading to increased position accuracy. It's hard to tell what a GPS is displaying but internally, the velocity measurement is very accurate but at too low a time resolution for some situations involving moving vehicles. If the GPS in the article was logging at 30 second intervals, it would be very difficult to know anything about the instantaneous speed of the vehicle in question. That my $.02
Maybe embed the antenna on the outer surface of the glass and use the conductive window as a ground plane?
I was advised by my company's IP lawyer about some work I did on a sabbatical that the easiest way to protect yourself by not patenting is to file a provisional patent and then abandon the application. The fees and hassle are lower that the full patent and the provisional patent is good for a year. Finally, one can also file a statutory invention disclosure.
Missing digits in the display is similar to one of the reasons for the Apollo 13 disaster. The cryotank that exploded was monitored by a temperature sensor with too small a range. During ground testing, the tank had become hot enough for the insulation to burn off of the wiring. Unfortunately, the annunciator on the temperature sensor had no way to record a temperature much above room temperature. If this annunciator had been designed better, engineers may very well have noticed the insane temperature, detected damage caused to that tank on the ground, and repaired it before the launch.
In the 4th circuit, this has been overturned. See Veeck vs. the Southern Building Code Congress. http://en.wikisource.org/wiki/Veeck_v._Southern_Building_Code_Congress_Int'l,_Inc.
Mod this one up. noxid has a great video and it's actually a primary source on the subject matter!
The demo seemed to show that they actually had this down and were smart enough not to make those mistakes. The lead engineer certainly emphasized that they were cognizant of these kinds of problems and made every attempt to avoid them. Once it's out of beta we can all see.
These guys briefed at a company meeting the other day and offered a private beta to those of us with i-phones. Their tool allows you to submit natural language queries for things that involve transactions. You can tell your Siri enabled phone to order you a pineapple pizza and it will find pizza restaurants with web ordering API's and then show you the prices for what you asked for and offer to let you buy them. In the case of pizza during the demo, it showed pizza Hut and Dominoes. They're working towards an interface that would allow you to say "Book me on the next flight to chicago!" You can tell siri, "Get me a copy of $bookname" and it will search amazon and other services with buy online API's and offer to purchase the book for you.
The bottom line to me is that it looked powerful and scary at the same time. It most definitely isn't clippy.
Pay for a consultation with a licensed engineer. Make clear that this is a work for hire situation. Make him/her sign an NDA. At least here in Alabama, complaining to the professional engineering board if the engineer steals your idea after will likely get his/her license suspended and an embarassing writeup in the yearly bulletin from the State Board of Engineers.
This was disproved by Claude Shannon of Bell Labs fame using information theory (a branch of math that looks dangerously like computer science). In short, the amount of energy required to encode the demon's single bit of information about each molecule is exactly balanced by the amount of energy that could be extracted by the work of Maxwell's demon.
Take a look at www.getinthecockpit.com. These guys sell a set of VR goggles and the gear to put a camera in the cockpit of a model aircraft and beam live video back over the RF link. DIsclaimer: it's a distant cousin of mine's company.
Any lab used for production or quality control in industry has to meet more stringent standards than an academic research lab. The rules to a certain degree assume qualified personnel and experiments without forgone conclusions. See http://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=standards&p_id=10106
Google Therac 25. It was a radiation treatment machine with software control and no mechanical interlocks. It killed circa 100 people in Latin America by exposing them to vastly higher doses of radiation than it should have. One of the worst glitches was a user interface lockup which left the shutter to the radiation source open.
They said it was sensitive information, not classified information. While the rules for military and contractor handling of classified material differ slightly, sensitive information is not protected by the same rules that govern classified information. Leak a classified drive: a bunch of people go to jail. Leak a sensitive drive and everybody goes to slashdot and gets sent to bed without a cookie.
The mistake in thinking that it's a bad thing to never have the data be the same is roughly speaking part of how the Germans lost WWII. The British broke the Enigma cypher by figuring out that a given letter was _NEVER_ encoded as the same letter. That tiny blip in the probability function allowed breaking many coded messages if they could get a small amount of cleartext such as the weather report.
They don't post an individual subscription rate. There have to be a lot more than 60 institutions that subscribe. I'd also bet that almost all subscribing institutions have a deal where they get more than just that journal on a package deal. Since much of this research was funded by the government, I argue that we are paying for the information twice. I'd be happier myself if the information was free and they charged for the indexing. If you magnify the $23,000 a year subscription fee by the thousands of journals in a research library, there seems to be a huge cost to society involved.
I went to ye olde library today to get copies of 2 Articles from the Journal of Applied Polymer Sciences, a Wiley Interscience Publication. Xeroxing the articles under fair use from the library was free for me.
The Whiskey Tango Foxtrot Moment came when I checked online to find out how much it would cost to subscribe to the journal. I thought someone misplaced a decimal point: $23,245 a year is the institutional subscription rate! That's about what I paid yearly in college tuition back when I was in college. Even worse, it's almost the value of the lab equipment I'm using in the work I've been doing on my own time.