And as was discussed in the original "airplane kill switch" thread, the Pentagon wasn't asking for a "kill switch". They wanted a "non-lethal weapon" for stopping airplanes: a much more difficult problem.
> They do apply to software. EULAs in most places at least are pretty much unenforceable > nonsense. If software doesn't do it's job, you can return it for a refund.
I'm not talking about EULAs. I agree that they are largely unenforceable. As far as I know in the US what is warranted is the tangible goods: the physical copy. If the CD is scratched you can insist that tney make you whole by replacing it with a good one or refunding your money but the software can be buggy as hell as long as the copy arrives on your computer intact.
No. The UCC is about commerce, and consequential damages can be and usually are disclaimed. If I make a free gift to you of a copy of my software and it turns out to be buggy how can you sue me for selling something defective? On the other hand if you sell a copy of my (GPL) software to someone else they should be able to sue you if it proves to be defective.
You might want to read up on merchantability, implied warranty, and fitness for use. These legal concepts apply to cars and other tangible goods but not to software. They should.
I predict that if you ask the same person the same question over and over again even at wide intervals the answers will converge but not necessarily to the correct value (of course, that value may often be "You asked me that before. Bugger off!")
They didn't say that the second answer was better. They said that the average was better. It would be interesting to know if the second answer was, on average, better than the first.
> Or you could just keep your passport in a ziploc antistatic bag (silver not pink) and > not worry about it...
That won't work at all. The conductivity of the pink plastic is very low: just barely adequate to drain off static charges. It will have no measureable effct on rf.
Just make yourself a duct-tape wallet and line it with aluminum foil.
> How the hell does my ordinary desktop PC survive the hellish fury of the RFID scanner > sitting next to it?
Devices for use in the home are subject to more stringent FCC EMI restricitions (Class B) than are devices intended for use in business (Class A). While the intent is to keep rf in, the shielding will work to keep it out as well. Medical devices, not being for use in the home (usually) will be designed to meet Class A but not Class B.
> Having boxes with hundreds of RFID tags rolling down the hallways of a hospital doesn't > seem so safe now!
Please read the original article, not the erroneous Slashdot headline. The tags are passive. The interference comes from the "readers" which actually transceivers.
> Medical devices usually don't have "rugged" in their specs.
Life safety devices have "fail-safe" in their specs. If "rugged" is what it takes to achieve that for a particular machine then it damn well better be "rugged".
> It's also hard to build a device that is at one hand very, very sensitive and able to > pick up minuscle electronic signals (like brainwaves or heartbeats) and at the same time > completely impervious to strong electronic signals and immune to EM interference.
Pumps and ventilators (the examples in the article) don't deal in miniscule signals. Even machines that do can be designed to be reasonably resistant to interference and to fail safely when overwhelmed.
> Perhaps the medical device manufacturers have a more critical classification where they > have the "right" not to be interfered with (unlike say, your iPod which must accept any > outside interference and generate none of its own) and designed their systems around > this assumption of legal protection by device classifications?
Even if there was such a classification relying on it to the detriment of patient safety would border on criminal negligence. Designers know that people often break such rules (often unknowingly: how many people know which of the gadgets around them emit rf at times?)
Rf is a normal part of the human environment these days. Anywhere you find people you are likely to find radio transmitters. Life safety and life support equipment must be designed in such a way that lives are not put at risk when someone keys a transmitter even when they do so against the rules.
> the implication from the limited text is that they were using the same reader (although > this is not confirmed) but the difference in tags did change the issue rate, so the tags > do share part of the problem.
No they don't. The passive tags are powered by the rf they absorb from the reader so they require the reader to put out more rf. They would have gotten the same results if there had been no tags present at all and they had just done "dummy" readings. The tags themselves emit orders of magnitude less rf than the reader.
The machines that suffered dangerous faults should be recalled and repaired. Keeping them away from RFID readers and other sources of rf will not suffice. The fact that rf interference could cause dangerous faults means that they contain design defects such that component failures or other sorts of damage or interference could also cause dangerous faults.
And yes, I have designed medical life support equipment, though not in this century.
> The LHC's collisions between two particle streams with equal and opposite momentum could > create things that are more or less at rest with respect to the Earth
The particles are going to be colliding at very nearly the speed of light. In that context "more or less at rest with respect to the Earth" will still be many, many times escape velocity.
Slashdot Mirror
> to clarify my post - the difference with facebook is that you have to use your real name
Just as a matter of idle curiousity, how do they know your real name?
I bet these guys believe in the MMPI, too.
Who counts the votes? Who decides who gets on the ballot?
And as was discussed in the original "airplane kill switch" thread, the Pentagon wasn't asking for a "kill switch". They wanted a "non-lethal weapon" for stopping airplanes: a much more difficult problem.
> At what point does centralizing and/or delegating operational authority over so much of
> our lives become a dangerous practice of its own?
At the very beginning.
> They do apply to software. EULAs in most places at least are pretty much unenforceable
> nonsense. If software doesn't do it's job, you can return it for a refund.
I'm not talking about EULAs. I agree that they are largely unenforceable. As far as I know in the US what is warranted is the tangible goods: the physical copy. If the CD is scratched you can insist that tney make you whole by replacing it with a good one or refunding your money but the software can be buggy as hell as long as the copy arrives on your computer intact.
No. The UCC is about commerce, and consequential damages can be and usually are disclaimed. If I make a free gift to you of a copy of my software and it turns out to be buggy how can you sue me for selling something defective? On the other hand if you sell a copy of my (GPL) software to someone else they should be able to sue you if it proves to be defective.
You might want to read up on merchantability, implied warranty, and fitness for use. These legal concepts apply to cars and other tangible goods but not to software. They should.
I predict that if you ask the same person the same question over and over again even at wide intervals the answers will converge but not necessarily to the correct value (of course, that value may often be "You asked me that before. Bugger off!")
They didn't say that the second answer was better. They said that the average was better. It would be interesting to know if the second answer was, on average, better than the first.
What's offensive about "Read The Fine Manual"?
> Don't feed the trolls - when an AC says something stupid, let it slide.
I'll take that advice (The "A" in "AC" can stand for several things. The "C", however, is constant).
> All you need to have is a simple identifier, which you compare to a database.
Which will be readily available on the Net.
> Or you could just keep your passport in a ziploc antistatic bag (silver not pink) and
> not worry about it...
That won't work at all. The conductivity of the pink plastic is very low: just barely adequate to drain off static charges. It will have no measureable effct on rf.
Just make yourself a duct-tape wallet and line it with aluminum foil.
> How the hell does my ordinary desktop PC survive the hellish fury of the RFID scanner
> sitting next to it?
Devices for use in the home are subject to more stringent FCC EMI restricitions (Class B) than are devices intended for use in business (Class A). While the intent is to keep rf in, the shielding will work to keep it out as well. Medical devices, not being for use in the home (usually) will be designed to meet Class A but not Class B.
> Having boxes with hundreds of RFID tags rolling down the hallways of a hospital doesn't
> seem so safe now!
Please read the original article, not the erroneous Slashdot headline. The tags are passive. The interference comes from the "readers" which actually transceivers.
> Medical devices usually don't have "rugged" in their specs.
Life safety devices have "fail-safe" in their specs. If "rugged" is what it takes to achieve that for a particular machine then it damn well better be "rugged".
> It's also hard to build a device that is at one hand very, very sensitive and able to
> pick up minuscle electronic signals (like brainwaves or heartbeats) and at the same time
> completely impervious to strong electronic signals and immune to EM interference.
Pumps and ventilators (the examples in the article) don't deal in miniscule signals. Even machines that do can be designed to be reasonably resistant to interference and to fail safely when overwhelmed.
Unfortunately most of the news stories Google brings up on this subject make the same error.
> Perhaps the medical device manufacturers have a more critical classification where they
> have the "right" not to be interfered with (unlike say, your iPod which must accept any
> outside interference and generate none of its own) and designed their systems around
> this assumption of legal protection by device classifications?
Even if there was such a classification relying on it to the detriment of patient safety would border on criminal negligence. Designers know that people often break such rules (often unknowingly: how many people know which of the gadgets around them emit rf at times?)
Rf is a normal part of the human environment these days. Anywhere you find people you are likely to find radio transmitters. Life safety and life support equipment must be designed in such a way that lives are not put at risk when someone keys a transmitter even when they do so against the rules.
> the implication from the limited text is that they were using the same reader (although
> this is not confirmed) but the difference in tags did change the issue rate, so the tags
> do share part of the problem.
No they don't. The passive tags are powered by the rf they absorb from the reader so they require the reader to put out more rf. They would have gotten the same results if there had been no tags present at all and they had just done "dummy" readings. The tags themselves emit orders of magnitude less rf than the reader.
The machines that suffered dangerous faults should be recalled and repaired. Keeping them away from RFID readers and other sources of rf will not suffice. The fact that rf interference could cause dangerous faults means that they contain design defects such that component failures or other sorts of damage or interference could also cause dangerous faults.
And yes, I have designed medical life support equipment, though not in this century.
The interference came from the readers not the tags. The tags are passive.
Yes. Next thing you know they will be running competitions will million dollar prizes.
> The LHC's collisions between two particle streams with equal and opposite momentum could
> create things that are more or less at rest with respect to the Earth
The particles are going to be colliding at very nearly the speed of light. In that context "more or less at rest with respect to the Earth" will still be many, many times escape velocity.
Thank you. Very informative.