I imagine that FDA medical device directives would have rules for data security and for RF emissions as well. I am a bit more familiar with CE medical device directives, where there are different classes of compliance - a device that filters your blood has stricter rules than an exercise machine - but besides protecting patient data, I assume a computer or network device in a medical environment would have to have have low RF emissions, so that it doesn't interfere with other medical devices. When your microwave oven interferes with your cordless phone or your wifi network at home, it might be annoying, but a similar situation in a hospital would be a bigger problem.
Problems of an aging and stroke-prone population cross international boundaries. "Who can afford this?" and "Will insurance pay for this?" are good questions, and the answers are different from country to country, and from year to year. Note also that hospitals and insurance companies are slow-moving organizations. If robotic science was a clearly safe magic pill that cured strokes, I assume we would find someone to pay for that cure. But with cures that provide only some degree of improvement, the treatments go through the normal course of medical research, and if the treatments are found to have sufficient and lasting efficacy, the medical and insurance fields eventually adjust to incorporate the new treatments.
As it is, I've seen research that shows repeatable quality-of-life improvements from our robotic therapy, and I've been at clinics and hospitals where patients and their families have given me heartfelt thanks for my work, which, while very gratifying, does not count as a controlled repeatable verifiable research result.
then you are saying that exercising stroke vitcim's affected limbs doesn't improve their mobility. Which is kind of a dumb thing to say.
Yes, that's almost what I'm saying - that it doesn't necessarily improve mobility. I work on stroke therapy robots that can move people's limbs around in whatever way we feel makes a difference. Through long research, we have found that some ways make a difference, and other ways do not make a difference.
Our researchers have been working on the problem for 25 years - that is, we have research published back to the mid-1980s, for example:
This is a seminal paper by one of our researchers, that spawned the field of rehab robotics. I have already posted other links to research earlier in this thread.
Note well, I'm not a researcher, I write software to control rehab robots. But I know that in our researchers' papers, they do experiments where one group of patients gets beneficial rehabilitative exercises on our robots, and a control group gets non-beneficial "fake" exercises on our robots, where the patient's limbs are moved by the robots, but not in a beneficial way.
So be careful when you use phrases like "It's so obviously likely to be useful."
You imply that I might have a problem with successful results from a competing method, but it's not so. I'm saying there's a difference between "shows promise" and research results, especially from research that has been going on for 25 years, and going on in earnest (it took a while to create and refine the robots) for more than 10 years.
I agree, show me the research. I work in the field of rehabilitation robotics for stroke, and I am not aware of science that says that simply assisting someone's movement will improve their neural/muscular function.
I've been working on this problem for 10 years (as a software designer, not a neuroscience researcher) and researchers who use our robots have many studies that show patient improvement, but this comes from providing controlled rehabilitation exercises, not just by driving their limbs with an exoskeleton. I think research indicates that the rehab benefit comes from having the patients work to control their own limbs (with assistance and guidance if necessary from a robot or therapist) rather than by just driving the limbs without the patient working the neural paths.
Man pages are plain text files, but not just text files. They are also not just ?roff files. They are written using the "man page macros" for [gnt]roff. These man page macros were written for UNIX in the 1970's, and survive pretty much unchanged. You can find doc for them on your Linux system by typing: man 7 man or http://www.unix.com/man-page/Linux/7/MAN/
If Kerrisk is the keeper of man 7, and he was supposed to cover man 7 in his book, then yes, this would be an oversight.
The ?roff language was pretty much "assembler language" for typesetting - you weren't supposed to write your documents in raw ?roff. In those days, before word processors like msword and oo, and before TeX and LaTeX, anyone who wrote docs for UNIX systems was well versed in the different macro packages, including some of man, ms, mm, me, and others.
Another option - a corporation is suing a politician. If the politician loses the lawsuit, the politician's ability to pay the penalty may be contingent on his being gainfully employed (as a politician). Therefore it is in the plaintiff's interest that the defendant remains employed. So the plaintiff (the newspaper) endorses the defendant (the politician).
Not to diss Stallman, but he was not the director of the AI Lab, and it's hard to say he was their most famous hacker at that time - the AI Lab spawned many great hackers, and especially then, during the early years of Symbolics and LMI. The most famous AI Lab hackers were LISP hackers (at that time - remember, it was a AI Lab.) Gerald Sussman, Guy Steele, JonL White, David Moon, et al.
Anyway, there is still a lot potential for improvements since the top 2 detectors use entirely different detection paradigms
This implies that the lower-scoring detectors are less valuable in terms of looking for sources of improvement. That's not true, and that wasn't stated in the paper's "Conclusions" section. If the lowest scoring detector finds 5% of the bad data, and it's a different slice from what the other detectors find, then that's quite valuable.
It would be far simpler with today's technology to give everyone individual steerable, zoomable access to video cameras. I don't expect that to happen, and I don't believe that Airbus will ever build Wonder Woman's plane, the passenger version.
Planes already have under-body forward-view video cameras that you can see on your personal seatback display. I had one on a flight from NYC to Hong Kong (over the north pole), and I got to look at solid white clouds for 15 hours, except for takeoff and landing - hearing which noises synched with the landing gear retracting was interesting. Transparent skin seems silly to me.
Measurement may be constrained by conditions like timing and signal levels (before amplification). Both of these can be solved by using a properly designed sensor that might live in a fancy plastic box, and it wouldn't be the same with a few extra feet of wire. (Would you put a few extra feet of wire between a motherboard and an in-circuit emulator?)
A hacker pal of mine worked at Tektronix in the late 70's, he told me this story. Tektronix made all kinds of oscilloscopes and electronic test gear. Apparently, they had a fancy special-purpose scope (cost maybe $10k/each), that they sold about 20 of each year. Suddenly, one company started ordering 4 or these scopes a month. This was surprising to Tektronix, and they had to change their inventory handling to deal with this change in demand. They decided to call the customer and figure out what they were using all these scopes for.
Turns out the customer was one of the research labs (LANL or something, I forget which). They were measuring nuclear reactions, and using these scopes because they had a particular kind of sensor, but the tests were destructive, and every time they ran the experiment (once a week), they vaporized a scope. I think they figured out a way to sell the customer the sensor without wrapping it in all the fancy scopey packaging.
Slashdot Mirror
which has a half-life of 30 years.
http://en.wikipedia.org/wiki/Zork
Truly sir, 20 mils == 508 microns. Make sure you're protected.
The Plymouth Satellite came a bit after tail fins. It is still cool classic.
I imagine that FDA medical device directives would have rules for data security and for RF emissions as well. I am a bit more familiar with CE medical device directives, where there are different classes of compliance - a device that filters your blood has stricter rules than an exercise machine - but besides protecting patient data, I assume a computer or network device in a medical environment would have to have have low RF emissions, so that it doesn't interfere with other medical devices. When your microwave oven interferes with your cordless phone or your wifi network at home, it might be annoying, but a similar situation in a hospital would be a bigger problem.
As it is, I've seen research that shows repeatable quality-of-life improvements from our robotic therapy, and I've been at clinics and hospitals where patients and their families have given me heartfelt thanks for my work, which, while very gratifying, does not count as a controlled repeatable verifiable research result.
Yes, that's almost what I'm saying - that it doesn't necessarily improve mobility. I work on stroke therapy robots that can move people's limbs around in whatever way we feel makes a difference. Through long research, we have found that some ways make a difference, and other ways do not make a difference.
Our researchers have been working on the problem for 25 years - that is, we have research published back to the mid-1980s, for example:
http://www.jneurosci.org/cgi/content/abstract/5/7/1688
This is a seminal paper by one of our researchers, that spawned the field of rehab robotics. I have already posted other links to research earlier in this thread.
Note well, I'm not a researcher, I write software to control rehab robots. But I know that in our researchers' papers, they do experiments where one group of patients gets beneficial rehabilitative exercises on our robots, and a control group gets non-beneficial "fake" exercises on our robots, where the patient's limbs are moved by the robots, but not in a beneficial way.
So be careful when you use phrases like "It's so obviously likely to be useful."
You imply that I might have a problem with successful results from a competing method, but it's not so. I'm saying there's a difference between "shows promise" and research results, especially from research that has been going on for 25 years, and going on in earnest (it took a while to create and refine the robots) for more than 10 years.
I've been working on this problem for 10 years (as a software designer, not a neuroscience researcher) and researchers who use our robots have many studies that show patient improvement, but this comes from providing controlled rehabilitation exercises, not just by driving their limbs with an exoskeleton. I think research indicates that the rehab benefit comes from having the patients work to control their own limbs (with assistance and guidance if necessary from a robot or therapist) rather than by just driving the limbs without the patient working the neural paths.
refs:
N Engl J Med 2010; 362:1772-1783 May 13, 2010
http://www.nejm.org/doi/full/10.1056/NEJMoa0911341
http://www.interactive-motion.com/clinical_research.htm
http://www.pbs.org/wgbh/nova/tech/trapped-elevator.html
Covered:
If Kerrisk is the keeper of man 7, and he was supposed to cover man 7 in his book, then yes, this would be an oversight.
The ?roff language was pretty much "assembler language" for typesetting - you weren't supposed to write your documents in raw ?roff. In those days, before word processors like msword and oo, and before TeX and LaTeX, anyone who wrote docs for UNIX systems was well versed in the different macro packages, including some of man, ms, mm, me, and others.
I think this will lead to a rise in the arrest of moonwalking gorillas.
http://www.youtube.com/watch?v=awhaCJlfO0w
Another option - a corporation is suing a politician. If the politician loses the lawsuit, the politician's ability to pay the penalty may be contingent on his being gainfully employed (as a politician). Therefore it is in the plaintiff's interest that the defendant remains employed. So the plaintiff (the newspaper) endorses the defendant (the politician).
Yes, he is their most famous hacker now, in 2010. The context of the discussion is 1985. At that time, he was not their most famous hacker.
Not to diss Stallman, but he was not the director of the AI Lab, and it's hard to say he was their most famous hacker at that time - the AI Lab spawned many great hackers, and especially then, during the early years of Symbolics and LMI. The most famous AI Lab hackers were LISP hackers (at that time - remember, it was a AI Lab.) Gerald Sussman, Guy Steele, JonL White, David Moon, et al.
(Suffer through the short ad, it's worth it...)
http://www.nbc.com/saturday-night-live/video/irwin-mainway/1185611/
This implies that the lower-scoring detectors are less valuable in terms of looking for sources of improvement. That's not true, and that wasn't stated in the paper's "Conclusions" section. If the lowest scoring detector finds 5% of the bad data, and it's a different slice from what the other detectors find, then that's quite valuable.
Planes already have under-body forward-view video cameras that you can see on your personal seatback display. I had one on a flight from NYC to Hong Kong (over the north pole), and I got to look at solid white clouds for 15 hours, except for takeoff and landing - hearing which noises synched with the landing gear retracting was interesting. Transparent skin seems silly to me.
Measurement may be constrained by conditions like timing and signal levels (before amplification). Both of these can be solved by using a properly designed sensor that might live in a fancy plastic box, and it wouldn't be the same with a few extra feet of wire. (Would you put a few extra feet of wire between a motherboard and an in-circuit emulator?)
One sensor to bring them all and in the darkness bind them
Would it be able to use a Tolkien ring protocol?
FTFY
Turns out the customer was one of the research labs (LANL or something, I forget which). They were measuring nuclear reactions, and using these scopes because they had a particular kind of sensor, but the tests were destructive, and every time they ran the experiment (once a week), they vaporized a scope. I think they figured out a way to sell the customer the sensor without wrapping it in all the fancy scopey packaging.
Isn't this kind of like buying a dead guy's shoes?
Haystack boy genius should set up a play date with wall-climbing Spider-Boy genius (who ripped off an idea from a BBC TV show).
Good idea. http://tech.slashdot.org/story/10/09/12/0155259/Google-Caffeine-Drops-MapReduce-Adds-Colossus
http://en.wikipedia.org/wiki/Affirming_the_consequent