If the "old" antibiotic is not present in the environment, then the normal mutations may get rid of the resistance without any selection pressure to the contrary. Presumably it may be a quicker process, since acquisition of resistance is very highly selected for, thus it has a high chance of "catching on" once the mutation occurs.
I don't think it's wholly necessary that the life will be more vulnerable to other stressors in presence of "all the chemicals we dump into the ecosystem". This may (or may not) be true on a smaller timescale. I'm sure there's lots of everyday stuff in our environment that would be deadly to organisms from just a few hundred million years ago... It's hard to predict how the life may adapt (or fail to). Overly pessimistic views I'd think are just as bad as overly optimistic ones.
Why is it tyrannical to require doctors to practice medicine properly to retain their licenses?
Because practicing medicine involves patients following directions, too, you know. In case of following treatment prescriptions, patients are as much to blame as anyone.
Yeah, and sometimes this outcolonization of bad bugs can be very important. To my wife, amoxicillin is lethal in 7 days it'd seem, and that's not due to any oversensitivity or allergy. She get's a raging c-diff infection in her gut within days after starting on amoxicillin. So far it seems 100% reproducible, and obviously we won't try no more.
I agree as far as amount of imagery collected and stored on board. They can't transmit all of that in real time, not using the space segment at least, and most likely not using point-to-point links either.
The drones aren't the only ones using the space segment, command posts etc. probably use a lot of it already. All of the drones in one theater using on the order of 500mbit/s on the space segment at once, perhaps 1-1.5gbits/s from all theaters so that it'd be spread across a couple birds -- that I can live with, that's realistic just by looking at what's flying and what the antenna sizes are. But 500Mbps from just one drone, transmitted via space segment -- haha. Perhaps if the other drones transmit nothing besides basic telemetry, or if you're flying just one drone in a theater. I don't even know what sort of an uplink a drone has, it's probably something off the shelf, they likely wouldn't develop a whole new uplink from scratch just for a drone. If they use a radio package that is available for other projects, then it's likely around 64Mbits/s when conditions are good.
Sorry, RF engineering reality is there no matter who pays the bill. And 500 mbits/s downlink just doesn't compute, even if it's line of sight to a control center. They presumably fly many of them at once from one control center...
Even that is probably wrong. Do these things run line-of-sight only? If they do, then it should be pretty damn easy to jam the downlink. If they don't then I wonder how those must clobber whatever space segment is used to relay all the data.
It's neither 500 megabytes/s nor even 500 megabits/s. There is no link capability in the U.S. space communications systems, or even anywhere, that could handle that reliably from just one drone, never mind multiple drones at the same time. That drone would need a big effing antenna to push that much data over a couple dozen thousand kilometers to the space segment. Let's get real: do the/. editors have no sense of magnitude at all?!
With modern parts, solderless breadboards are becoming counterproductive. Any recent chip will have fast edges on the logic outputs and the parasitics of a solderless breadboard will mess things up. You cannot assemble but a fairly low frequency switching power regulator on a solderless breadboard.
Decent sockets may cost more than the chip you're socketing. The only worthwhile, decently performing sockets, with low parasitics, are the individual turned pin sockets that you solder one-by-one directly into the PC board holes. Those are not cheap. Anything else will markedly decrease reliability.
Sigh, the stuff that's on the Pi mainboard is not even available in through hole! Care to explain how you'd package BCM2835 in a through-hole package and still have it perform acceptably? Because it'd be an exercise in futility AFAICT. The leadframe parasitics would make it radiate like hell. Heck, I'm sure there are switching voltage regulators for CPU core voltage that, even if implement in a very compact way using through-hole components, would be an EMI/EMC disaster. Never mind they'd probably take up 25% of current Pi board's real estate. How the heck did Pi mainboard get into any trouble because they didn't do the impossible is beyond me. You must be making stuff up. It's not informative, not that aspect at least.
It's also a pipe dream that through-hole is more mechanically sturdy. It is in some instances, but not generally for mounting ICs. It's not good at all that you have a potentially large stiff structure (think 40 pin DIP) that has pins that sit in relatively fragile metallized holes. Having IC attachment points (pads) separate from vias is a good thing. If you want something that's still easy to assemble for low density circuits, you can always use 1206 discretes and SO packages with their comfortable 50 mil lead pitch. It'll be much faster to assemble, too -- I really like not having to flip the board to do the soldering.
Given that standard Linux is not a realtime system, and that to do anything worthwhile you have to write device drivers because there's no direct access even to gpio, I'd say that comparing it to vxWorks is like apples to trolleys, not even oranges. I looked at realtime linux and it seems to be a mess. There's no single, well supported implementation, it seems, and it looked like it'd be more of a liability, long-term, to depend on it. In the end we went with TwinCat.
I concur. Even stranded #10 hook-up wire is at the limit of what I'm designing my power electronics for; #8 is comparatively unwieldy. Pulling a cable with two solid #10 conductors and probably a #12 grounding conductor is a royal pain I'm sure. I'd charge more too, were I an electrician.
This should be informative. Wiring (that Arduino uses) is just a C++ framework. You can of course use it and pretend you have something "simple", but it's really not, and once you hit a compiler error, you better knew C++.
There's plenty of low bandwidth peripherals where PC/104 is perfectly adequate and offers lowest cost. It's easiest to design for. To deal with PCI or PCIe in a fully conformant way, you need an FPGA. If all you need is to control a bunch of, say, relays, then nothing beats ISA as the interface is a dollar or two worth of trivial glue logic. There's plenty of automation where I/O is simple like that.
I've got a highest-end (available at the time) Kodak easyshare digital camera a couple years ago. A day of shooting later, it went back to the store. It was useless. Slow saving, abominable UI decisions, noisy pictures, bugs in the firmware. That's my anecdote. That was the only Kodak camera I ever had. Got a powershot G9 instead and my wife is reasonably content.
What a bunch of brouhaha. Excitement will raise your heart rate all right, so you can technically alter the heart rate with nothing but your mind. But raised heart rate is not what makes your heart "active", and your heart won't get "weak" if you keep your heart rate low. Physical exercise changes way more besides just heart rate -- the peripheral circulation flow rates change due to heightened metabolism in the muscles, there's a bunch of endocrine stuff that regulates all that, your breathing goes up, etc. Your mind alone cannot simulate effects of exercise, sorry. Heart itself is a very special organ and it gets exercised all the time -- it's never off. Whatever you claim to be clinically shown is not, it's some bs you're repeating. The "will to live" does not improve outcomes. And when it comes to someone with ALS who can't really move of his own accord, there's very little effect that person's thinking will have on the outcome.
This is just a bunch of misconceptions. A condition is obviously survivable if you've got living examples of its survivability, I mean duh. Those religious people who are quick to call such low-probability survivals "miraculous" are quite silly because they place full, unchecked trust in an MDs opinion (or some statistics published somewhere, or a combination of both). I always have fun politely telling them that hey, perhaps they shouldn't be idolizing the MDs in question, and treating their word as the sole judgment that defines their veneration of something as miraculous...
If there's one person who has survived an "unsurvivable" condition, then this makes such a condition permanently and forever after survivable. You see, it's a black-and-white thing, either it's survivable or it's not. One example is enough to move it into the survivable category. You can then speak of probabilities of survival, and people are notoriously bad at reasoning there. Apparently with ALS plenty of commenters on this story do not know much about demonstrated 30-year survivability -- and that one is not as bad as everyone thinks it is. 5% of people live 20+ years with ALS, that's freaking awesome if you ask me, and nowhere near the doom-and-gloom everyone has in their heads.
If this is a non-critical part, why is it receiving stresses that is causing it to fail?
Because such parts are often attached to other stressed parts, and stresses simply redistribute over the whole assembly. As the non-critical part starts failing, the stresses depart to where they belong.
That's a non-sequitur if there ever was one. Yeah, small cracks have a tendency to become larger under stress. So what. They always did, and they always do, and any plane that's flying out there has plenty of small cracks. This tells us nothing. What we need to know is what is the predicted rate of growth of those particular cracks under the stresses the material at the crack tip, in particular, is subject to. Add in tasty details about expected contributions of structure (will the cracks join like in Tu-144?), corrosion, etc.
Take a close look at the skin of the jet next time you fly. You may be surprised how many metal patches you will find -- patches that repair cracks or dings/dents.
Slashdot Mirror
I don't even think it's reasonable to still consider it an error. It's slang for sure, but I believe I have used it correctly, and consciously, too.
Agreed about overbooking. Hmm, binary orders of magnitude, never thought of these.
If the "old" antibiotic is not present in the environment, then the normal mutations may get rid of the resistance without any selection pressure to the contrary. Presumably it may be a quicker process, since acquisition of resistance is very highly selected for, thus it has a high chance of "catching on" once the mutation occurs.
I don't think it's wholly necessary that the life will be more vulnerable to other stressors in presence of "all the chemicals we dump into the ecosystem". This may (or may not) be true on a smaller timescale. I'm sure there's lots of everyday stuff in our environment that would be deadly to organisms from just a few hundred million years ago... It's hard to predict how the life may adapt (or fail to). Overly pessimistic views I'd think are just as bad as overly optimistic ones.
Why is it tyrannical to require doctors to practice medicine properly to retain their licenses?
Because practicing medicine involves patients following directions, too, you know. In case of following treatment prescriptions, patients are as much to blame as anyone.
Yeah, and sometimes this outcolonization of bad bugs can be very important. To my wife, amoxicillin is lethal in 7 days it'd seem, and that's not due to any oversensitivity or allergy. She get's a raging c-diff infection in her gut within days after starting on amoxicillin. So far it seems 100% reproducible, and obviously we won't try no more.
That's an oxymoronic way to apply humanity if there ever was any.
I agree as far as amount of imagery collected and stored on board. They can't transmit all of that in real time, not using the space segment at least, and most likely not using point-to-point links either.
The drones aren't the only ones using the space segment, command posts etc. probably use a lot of it already. All of the drones in one theater using on the order of 500mbit/s on the space segment at once, perhaps 1-1.5gbits/s from all theaters so that it'd be spread across a couple birds -- that I can live with, that's realistic just by looking at what's flying and what the antenna sizes are. But 500Mbps from just one drone, transmitted via space segment -- haha. Perhaps if the other drones transmit nothing besides basic telemetry, or if you're flying just one drone in a theater. I don't even know what sort of an uplink a drone has, it's probably something off the shelf, they likely wouldn't develop a whole new uplink from scratch just for a drone. If they use a radio package that is available for other projects, then it's likely around 64Mbits/s when conditions are good.
That sounds more like it. Still they can't do probably more than a dozen at once via satellite or they'll run out of channels...
Sorry, RF engineering reality is there no matter who pays the bill. And 500 mbits/s downlink just doesn't compute, even if it's line of sight to a control center. They presumably fly many of them at once from one control center...
Even that is probably wrong. Do these things run line-of-sight only? If they do, then it should be pretty damn easy to jam the downlink. If they don't then I wonder how those must clobber whatever space segment is used to relay all the data.
It's neither 500 megabytes/s nor even 500 megabits/s. There is no link capability in the U.S. space communications systems, or even anywhere, that could handle that reliably from just one drone, never mind multiple drones at the same time. That drone would need a big effing antenna to push that much data over a couple dozen thousand kilometers to the space segment. Let's get real: do the /. editors have no sense of magnitude at all?!
With modern parts, solderless breadboards are becoming counterproductive. Any recent chip will have fast edges on the logic outputs and the parasitics of a solderless breadboard will mess things up. You cannot assemble but a fairly low frequency switching power regulator on a solderless breadboard.
Decent sockets may cost more than the chip you're socketing. The only worthwhile, decently performing sockets, with low parasitics, are the individual turned pin sockets that you solder one-by-one directly into the PC board holes. Those are not cheap. Anything else will markedly decrease reliability.
Sigh, the stuff that's on the Pi mainboard is not even available in through hole! Care to explain how you'd package BCM2835 in a through-hole package and still have it perform acceptably? Because it'd be an exercise in futility AFAICT. The leadframe parasitics would make it radiate like hell. Heck, I'm sure there are switching voltage regulators for CPU core voltage that, even if implement in a very compact way using through-hole components, would be an EMI/EMC disaster. Never mind they'd probably take up 25% of current Pi board's real estate. How the heck did Pi mainboard get into any trouble because they didn't do the impossible is beyond me. You must be making stuff up. It's not informative, not that aspect at least.
It's also a pipe dream that through-hole is more mechanically sturdy. It is in some instances, but not generally for mounting ICs. It's not good at all that you have a potentially large stiff structure (think 40 pin DIP) that has pins that sit in relatively fragile metallized holes. Having IC attachment points (pads) separate from vias is a good thing. If you want something that's still easy to assemble for low density circuits, you can always use 1206 discretes and SO packages with their comfortable 50 mil lead pitch. It'll be much faster to assemble, too -- I really like not having to flip the board to do the soldering.
Given that standard Linux is not a realtime system, and that to do anything worthwhile you have to write device drivers because there's no direct access even to gpio, I'd say that comparing it to vxWorks is like apples to trolleys, not even oranges. I looked at realtime linux and it seems to be a mess. There's no single, well supported implementation, it seems, and it looked like it'd be more of a liability, long-term, to depend on it. In the end we went with TwinCat.
I concur. Even stranded #10 hook-up wire is at the limit of what I'm designing my power electronics for; #8 is comparatively unwieldy. Pulling a cable with two solid #10 conductors and probably a #12 grounding conductor is a royal pain I'm sure. I'd charge more too, were I an electrician.
This should be informative. Wiring (that Arduino uses) is just a C++ framework. You can of course use it and pretend you have something "simple", but it's really not, and once you hit a compiler error, you better knew C++.
There's plenty of low bandwidth peripherals where PC/104 is perfectly adequate and offers lowest cost. It's easiest to design for. To deal with PCI or PCIe in a fully conformant way, you need an FPGA. If all you need is to control a bunch of, say, relays, then nothing beats ISA as the interface is a dollar or two worth of trivial glue logic. There's plenty of automation where I/O is simple like that.
I've got a highest-end (available at the time) Kodak easyshare digital camera a couple years ago. A day of shooting later, it went back to the store. It was useless. Slow saving, abominable UI decisions, noisy pictures, bugs in the firmware. That's my anecdote. That was the only Kodak camera I ever had. Got a powershot G9 instead and my wife is reasonably content.
What a bunch of brouhaha. Excitement will raise your heart rate all right, so you can technically alter the heart rate with nothing but your mind. But raised heart rate is not what makes your heart "active", and your heart won't get "weak" if you keep your heart rate low. Physical exercise changes way more besides just heart rate -- the peripheral circulation flow rates change due to heightened metabolism in the muscles, there's a bunch of endocrine stuff that regulates all that, your breathing goes up, etc. Your mind alone cannot simulate effects of exercise, sorry. Heart itself is a very special organ and it gets exercised all the time -- it's never off. Whatever you claim to be clinically shown is not, it's some bs you're repeating. The "will to live" does not improve outcomes. And when it comes to someone with ALS who can't really move of his own accord, there's very little effect that person's thinking will have on the outcome.
This is just a bunch of misconceptions. A condition is obviously survivable if you've got living examples of its survivability, I mean duh. Those religious people who are quick to call such low-probability survivals "miraculous" are quite silly because they place full, unchecked trust in an MDs opinion (or some statistics published somewhere, or a combination of both). I always have fun politely telling them that hey, perhaps they shouldn't be idolizing the MDs in question, and treating their word as the sole judgment that defines their veneration of something as miraculous...
If there's one person who has survived an "unsurvivable" condition, then this makes such a condition permanently and forever after survivable. You see, it's a black-and-white thing, either it's survivable or it's not. One example is enough to move it into the survivable category. You can then speak of probabilities of survival, and people are notoriously bad at reasoning there. Apparently with ALS plenty of commenters on this story do not know much about demonstrated 30-year survivability -- and that one is not as bad as everyone thinks it is. 5% of people live 20+ years with ALS, that's freaking awesome if you ask me, and nowhere near the doom-and-gloom everyone has in their heads.
If this is a non-critical part, why is it receiving stresses that is causing it to fail?
Because such parts are often attached to other stressed parts, and stresses simply redistribute over the whole assembly. As the non-critical part starts failing, the stresses depart to where they belong.
That's a non-sequitur if there ever was one. Yeah, small cracks have a tendency to become larger under stress. So what. They always did, and they always do, and any plane that's flying out there has plenty of small cracks. This tells us nothing. What we need to know is what is the predicted rate of growth of those particular cracks under the stresses the material at the crack tip, in particular, is subject to. Add in tasty details about expected contributions of structure (will the cracks join like in Tu-144?), corrosion, etc.
Take a close look at the skin of the jet next time you fly. You may be surprised how many metal patches you will find -- patches that repair cracks or dings/dents.
Cha-ching, you win the internets for today. Finally a voice of reason.