As an actual physician working at KP, I can tell you the system actually works. From my end the downtime has been truly minimal, and it is proving to be powerful despite intial misgivings. I don't doubt our IT has messed part of this up, but the system does actually work and I am already seeing some significant advantages when treating my patients....
Everything needs to be taken in context. Acetinobacter is a very uncommon infection, even among ICU patients. A better study would study whether the presence of cell phones actually changed the infection rate, not just colonization rates. Hospitals are colonized by may different bacteria, but that is not necessarily clinically significant. It only makes a difference if people actually get infected at a higher rate.
So, I come from a different world than most of you posting on this thread. From a medical perspective, a Palm is invaluable. Knowledge, especially drug databases, are invaluable to have at your fingertip....especially considering how bulky equivalent paper sources are. 100 percent of young physicians carry handhelds. That's right, everyone. It's a big market, and only going to get bigger. We need and use handhelds on a daily basis. They are currently too unweildy and insecure for patient care, but that should change within the next 5-10 years. Look for official handheld usage at major hospitals within that time frame.....anyway, just giving a different POV.
I copied Moby's newest CD, and am glad I did. It really isn't very good, original, or otherwise worth $15. On the flip side, Moby is correct when he states that techies aren't gonna buy CDs...especially when they aren't good. It is interesting, though, that many people are perfectly willing to pay for games, having realized that programmers have to make money somehow. Perhaps the music industry is just where games where fifteen years ago....
Anyone else feel cheated? Moore's law states a doubling should occur every 18 months. 6 years is four doublings but, instead of 32 GHz, we get 20.....what a crock....;)
On the flip side, a lot of work has already been done in this area. For example, just take a look at the Porsche Stability Management system, which can selectively brake each individual wheel, or any competing system from a variety of manufacturers. High-end cars already come equiped with a variety of sensors that can adjust shocks, brake, and power distribution as needed, on a fully dynamic basis, and adjusted hundreds of times each second. I do agree that integrating the package into F1, a competitive arena instead of a relatively low speed open road, will provide a challenge. Also interesting will be how the company deals with the fact that computer assistance in many aspects of F1 racing, such as ABS braking or traction control, is outlawed, but would obviously compose part of an integrated computer controlling a car. F1 teams currently have vastly sophisticated models of their cars, tracks, and intricate telemetry set up so that, for some tracks, qualifying times nearly perfectly match theoretical times. The challenge to integrate all of this data and produce a working product is huge, though.
As a final point, I would like to point out that some people think that human performance is "magical", and cannot be duplicated. While we may not currently have the ability to create neural nets to match KAsparov's ability, in terms of the physical world, we _can_ create sensors that can detect and react to situations faster than any human possibly could. For example, powerful sensors could detect wheel slippage occurring at such a minute level that a human _could_ not notice, and correct appropriately to better the computer's lap times. Once again, I do not mean to diminish what these F1 drivers accomplish, but it is certainly not magic. They take in cues from the external world and react in the way they have been trained. A computer could do the same.
Slashdot Mirror
As an actual physician working at KP, I can tell you the system actually works. From my end the downtime has been truly minimal, and it is proving to be powerful despite intial misgivings. I don't doubt our IT has messed part of this up, but the system does actually work and I am already seeing some significant advantages when treating my patients....
Everything needs to be taken in context. Acetinobacter is a very uncommon infection, even among ICU patients. A better study would study whether the presence of cell phones actually changed the infection rate, not just colonization rates. Hospitals are colonized by may different bacteria, but that is not necessarily clinically significant. It only makes a difference if people actually get infected at a higher rate.
So, I come from a different world than most of you posting on this thread. From a medical perspective, a Palm is invaluable. Knowledge, especially drug databases, are invaluable to have at your fingertip....especially considering how bulky equivalent paper sources are. 100 percent of young physicians carry handhelds. That's right, everyone. It's a big market, and only going to get bigger. We need and use handhelds on a daily basis. They are currently too unweildy and insecure for patient care, but that should change within the next 5-10 years. Look for official handheld usage at major hospitals within that time frame.....anyway, just giving a different POV.
I copied Moby's newest CD, and am glad I did. It really isn't very good, original, or otherwise worth $15. On the flip side, Moby is correct when he states that techies aren't gonna buy CDs...especially when they aren't good. It is interesting, though, that many people are perfectly willing to pay for games, having realized that programmers have to make money somehow. Perhaps the music industry is just where games where fifteen years ago....
Anyone else feel cheated? Moore's law states a doubling should occur every 18 months. 6 years is four doublings but, instead of 32 GHz, we get 20.....what a crock....;)
E
On the flip side, a lot of work has already been done in this area. For example, just take a look at the Porsche Stability Management system, which can selectively brake each individual wheel, or any competing system from a variety of manufacturers. High-end cars already come equiped with a variety of sensors that can adjust shocks, brake, and power distribution as needed, on a fully dynamic basis, and adjusted hundreds of times each second. I do agree that integrating the package into F1, a competitive arena instead of a relatively low speed open road, will provide a challenge. Also interesting will be how the company deals with the fact that computer assistance in many aspects of F1 racing, such as ABS braking or traction control, is outlawed, but would obviously compose part of an integrated computer controlling a car. F1 teams currently have vastly sophisticated models of their cars, tracks, and intricate telemetry set up so that, for some tracks, qualifying times nearly perfectly match theoretical times. The challenge to integrate all of this data and produce a working product is huge, though. As a final point, I would like to point out that some people think that human performance is "magical", and cannot be duplicated. While we may not currently have the ability to create neural nets to match KAsparov's ability, in terms of the physical world, we _can_ create sensors that can detect and react to situations faster than any human possibly could. For example, powerful sensors could detect wheel slippage occurring at such a minute level that a human _could_ not notice, and correct appropriately to better the computer's lap times. Once again, I do not mean to diminish what these F1 drivers accomplish, but it is certainly not magic. They take in cues from the external world and react in the way they have been trained. A computer could do the same.