Yes and no. A page or cell failure will result in I/O errors if there are no more spares, and if it occurs during a read cycle, it -should- result in I/O errors for all subsequent reads from that cell or page until it gets rewritten to a new cell or page. If it doesn't work that way, then the device is fundamentally violating the contract between the device and the OS to report all nonrecoverable errors that result in data loss.
Also, while a multi-chip design reduces the probability of a device failing outright, it dramatically increases the probability of a failure. First, using a separate controller significantly increases the probability of failure because instead of having interconnect traces on a slab of silicon that (electromigration notwithstanding) almost never change or fail if they work from the factory, you have solder joints exposed on a circuit board. Solder joints are the most common cause of circuit failure in my experience.
Even ignoring the increased risk of having extra solder joints between the controller and flash parts, the odds of failure are still much worse for multi-chip devices. Remember your RAID MTBF theory. The MTBF of a collection of devices is equal to the MTBF of one device divided by the number of devices. If you have one part, the MTBF on that slab of silicon and associated solder joints might be a year. If you have five parts, the MTBF is now 73 days. That's an extreme example, but sadly, I've seen flash sticks with large numbers of failures in the first month, so that's not nearly as gross an exaggeration as you might think.... And whether one part fails or the whole thing fails, you still lose data.
Also, a controller failure is still likely to cause all flash parts to be inaccessible whether it is integrated into a flash chip or is driving eight discrete flash chips. It's not like you're going to use a separate flash controller per flash part. And I -think- that a device showing zero capacity is probably caused by the flash controller being unable to communicate with the flash parts. If so, then that is much more likely to be caused by a failed connection between the two than by a failed flash controller (unless there are problems with interconnects inside the flash controller chip package failing due to overzealous compliance with ROHS rules).
The original poster also failed to mention the most common failure mode, bar none: poor solder joints or other physical interconnects getting broken by physical force. This is very common among cheap flash drives. I wouldn't expect the same with SSDs, of course---you don't normally carry a SSD in your pocket---but at least in my experience, this one cause of failure is easily an order of magnitude more frequent than any other single cause, and is in all likelihood greater than all the others put together. And that's not even counting actual abuse (washing machines, run over by cars, and so on).
My Lexar JumpDrive Secure flash drive suddenly stopped working, and I talked to my mother, whose entire university class was using that same model of drive. Turns out that between us, we had experienced close to a 50% failure rate on those things within the first month or so, having seen somewhere around 14 or 15 failures. The failure was interesting. Mine failed suddenly, but worked if you tipped the connector at an angle... at least for a couple of seconds once or twice. This told me pretty conclusively that the failure was caused by poor hardware design. As best I can tell, when you carry the drive in your pocket, the cap puts pressure on the USB connector. Over time, this gradually causes solder joint or trace failure (I never cut one open to figure out which) at or near the USB connector.
Since then, I only buy flash devices with mechanisms where the USB connector retracts into a solid housing. Sure, you have an elevated risk of gunk from your pocket getting into the connector because it isn't covered, but at least you don't have the flexing problem. Gunk can be cleaned with a flat toothpick and alcohol. Failed solder joints requires disassembly and SMT soldering skills....:-)
This story usually contains the phrase "It's on a flywheel UPS circuit" and ends with "The boot drive experienced a head crash and is now completely dead (it won't even spin up), but the application drive is still working fine, so as long as the power doesn't fail, it should be okay. It has been running this way for three years." And then, with every retelling, the number of years increases by one.... Eventually, it starts to add "interesting" causes for the head crash, e.g. the senior admin slamming the intern's head repeatedly into it after he inadvertently switched off the main UPS that powers the mission-critical servers. Yeah. Good times.
Well, what's interesting is that IIRC the study suggests that Lipitor reduces the risk of heart attacks more than the other statins even if the other drugs reduce cholesterol levels to similar levels.. This at least hints that some other side effect of Lipitor causes cholesterol to not accumulate as much, and that lowering cholesterol levels is a red herring in most of the studies to date. This is further supported by a study by University Hospital in Toronto that showed that among heart attack patients, the probability of a second heart attack did not vary significantly based on cholesterol levels. (Citation needed.) If high serum cholesterol levels were really the root cause, one would expect a significantly higher risk of a second heart attack in those with higher serum cholesterol levels. The lack of continued correlation within this reduced subset of the population suggests to me that high cholesterol levels are likely a frequent (but not guaranteed) symptom of the real underlying problem, rather than the actual cause.
They exist already. For example, the RazorBook. They run Wince... I mean Win CE.
40 hours seems quite plausible. If you figure an iPhone battery has a capacity of about 5.18 Watt hours (1400 mAh * 3.7V) according to ipodbatteryfaq.com and it handles computation at blast for several hours on a charge, ignoring the extra power for a larger screen for the moment, if it had a battery the size of a MacBook (62.4 Watt hours according to System Profiler's battery stats on mine), it would last on the order of 60 hours on a charge even running at full tilt. Doing lighter work, I could easily see that extended by as much as a factor of three. So when you factor in the bigger display, yeah, I could see 40 hours being possible, assuming good power management. And that is definitely a machine I would buy in an instant. I think 20 hours is probably more realistic given the size constraints of a netbook, though.
That said, the battery life on the RazorBook is reportedly only on the order of 4 hours. Given that the CPU is comparable in its power consumption, this tells me that either the screen backlight is an unholy pig or Win CE power management is absolutely terrible. Neither would be much of a surprise. No idea how the Linux version of the RazorBook does on power.
PA-RISC is one of those things that will still be around after the nuclear holocaust destroys nearly all life as we know it. Two things will still be alive: a cockroach and a PA-RISC server looking forlorn alone in a corner somewhere wondering why nobody is talking to it anymore.
We'll leave it as an exercise to the reader to determine the precise nature of the word "alive" in this context, as either definition could be deeply disturbing in its own right.
When we talk about sinus infections, we are usually talking about bacterial infections. If an infection isn't bacterial, it's really just a head cold or flu. The symptoms of a sinus infection are different from symptoms caused by viral agents. With bacteria, you have a high fever, while colds rarely cause significant fever in adults. Although influenza can cause a fever, it is accompanied by other symptoms that are not localized to the head (random muscle aches, extreme fatigue, chills, etc.) that you do not get with a bacterial infection. With a sinus infection, you get a fever (sometimes high), your head is clogged, and you may feel somewhat run down after a few days of it, but you basically don't feel sick outside of your head (and possibly throat) unless it moves into pneumonia.
It also sounds like you mistakenly believe that strep throat is just another term for a sore throat. By definition, strep is always bacterial. The term "strep" is short for "streptococcus", which is a genus of bacterium. It's not that strep "can be bacterial", but rather that it must be. If you use it to refer to anything else, you're using the term incorrectly.
If this doctor had said that you shouldn't give antibiotics for most sore throats or sinus congestion, I would have agreed; most stuffy heads and sore throats are viral, will go away on their own, and are not affected by antibiotics. However, making such an argument for sinus infections (a term usually reserved for bacterial infections) or strep (a term for a specific class of bacterial infections that can turn into a much more serious and sometimes fatal form if left untreated), the argument crossed a line from being perfectly sensible advice to being just plain reckless.
Good idea. I started working on fireworks last night. To run it, type "./boom.sh" from within the directory. Right now, it just goes through each of the different sizes/styles of explosion in order at a fixed position.
Right now, it is just a bunch of ASCII image maps built and routines to draw each of them at a given pair of screen coordinates. I still need to write code to randomly choose a launch angle/speed, calculate trajectory, and actually draw the rocket dot while in flight. I might add code to split the line arrays into individual elements so that things can overlap somewhat or I might draw things over and over in the main program loop so that conflicts will alternate. I might add some sane color maps (randomly chosen) to allow the explosions to have mixed colors instead of one per burst. I might use multiple background jobs to do the drawing so I can spin off an arbitrary number of drawing tasks or I might do it with an array and a deadline scheduler. There are a lot of directions I could go with it....:-)
I would not be surprised if this article were wrong about that conclusion. The article is definitely wrong about the conclusion regarding sinus infections.
Ever since I was a kid, I've been prone to sinus infections. Every couple of years, I'll have one. They have never---and I do mean never---gone away on their own. I've left them for well over a week just to see. Never happens. By contrast, when I take antibiotics, whether that is after four days of symptoms or eight, the symptoms decrease dramatically within just a few hours after the start of the first round. Even if the infection would have gone away after a few more days (and I'm not convinced it would have), taking an antibiotic still means you get three or four days of your life back, and that alone is sufficient reason to consider such treatment.
Further, the doctor's view is dangerous. Doctors don't prescribe antibiotics for strep throat because of the symptoms. They prescribe antibiotics for strep throat because if left untreated, strep infections can turn into scarlet fever (which can cause kidney damage) or rheumatic fever (which has a 2-5% mortality rate). Failing to prescribe antibiotics if strep is suspected is madness.
Even in the case of sinus/ear infections, his few is dangerous. My ear infections have gotten painfully close to rupturing my ear drum. If I did not take antibiotics, there is a very good chance that I would have hearing damage today. As a musician, that would be a bad thing.
Now if you want examples of bad medicine, take a look at statin drugs. Heart doctors notoriously prescribe these to people with even moderate cholesterol levels to bring their levels down. While clinical studies do show that in patients with severely elevated cholesterol, heart attack risk can be reduced by some statins, other statins have been shown to increase the risk of heart attacks and strokes.
Also, the side effects are much more common than they lead you to believe and are severe and debilitating. Both my grandparents have experienced textbook adverse reactions to statin drugs (peripheral neuropathy in both, sleep issues in one, confusion/anxiety/cognitive disruption in one). These side effects are very real.
Worse yet, studies show that these drugs don't reduce overall deaths because for every death they prevent from heart disease or strokes, you get an increase in deaths from other causes, including cancer. So basically it reduces quality of life while failing to actually increase the typical length of life....
You're confusing capacity costs with marginal costs. While they can be related (in the case of power, for example, capacity must exceed demand or else you get brownouts that could pose safety risks), they are not the same thing.
The cost of the network is not based on the amount of data the users download at all. It is based on the amount of money the ISP chooses to spend to deliver an average level of bandwidth that they choose. If everybody started downloading twice as much content and the pipes remained saturated 24x7 but the ISP did not increase the number of outgoing pipes, the cost to provide service would still be roughly the same as it is now. If the ISP kept the capacity the same, but the usage dropped in half, the cost to provide the service would still be roughly the same as it is now, but half the capacity would be wasted. Therefore, subject to the assumption that the ISP is unlikely to change capacity if they can help it (which at least appears to be true more often than not), the marginal cost of additional use is zero, at least until it gets so bad that users start switching to a competing service due to the perception of poor network performance. Since a significant portion of cable modem customers have no other choices, even that condition is largely moot.
Although in an ideal world, the capacity should be tied to customer use, in practice, the capacity can be woefully insufficient during peak periods, particularly given that the ISPs would like their infrequent users to experience near maximum theoretical performance during peak periods. Even without the downloaders, this would probably be unlikely. Thus, the ISPs have three choices: increase capacity, convince people to shift bulk usage to off-peak periods, or introduce asinine caps that cause the capacity to be completely wasted for half the day. Two of those solutions work reasonably. Guess which one they picked....
The right way to handle this is to contact customers who download or upload huge amounts of data. Tell them that they are using an excessive amount of bandwidth and request that they do most of their large downloads overnight. Some people will ignore this, but most will comply, and as a result, the average use will be much more balanced and much easier to manage. Another good solution is to queue and delay delivery of packets to/from folks who are pulling or pushing large volumes of data relative to traffic from folks who aren't. In this way, capacity would be used at maximum efficiency at all times, preventing poor service by customers doing casual browsing while not significantly diminishing the bandwidth available to downloaders.
I never said that your bandwidth should be limited based on the assumption you will be using it all the time. I said that the providers need to increase capacity instead of whining that it will cost them money....
As soon as you give in to caps, different plans by different vendors cannot be easily compared. This reduces competition, which invariably hurts consumers. Also, it limits the profitability of upcoming Internet services that use more bandwidth, which has the potential to seriously stifle the future of the Internet.
I did read what you said. I just disagree with your assessment that what you're complaining about is anything more than a minor typographical error. And I'm very glad not everyone thinks the way you do. This is a discussion forum, not a doctoral thesis review. In the context of informal discussion, in my experience, such dogmatic rigor is an impediment to creativity and the free flow of ideas. Making people afraid to post because they are afraid somebody will point out a trivial mistake and blow it out of proportion is the best way to prevent lots of insightful points of view from ever being heard at all.
IMHO, the right response when somebody is sloppy about units or the spelling thereof is to point out the mistake. Once. One sentence, then drop it. it's just not that important in the context of an informal discussion, and blowing such criticism out of proportion is not conducive to open discussion.
So basically if somebody came up with a workable "theory of everything" but spoke English poorly and made a few common spelling errors, you think it would be appropriate to dismiss the theory with a hand wave?
GB vs Gb is almost certainly the most common spelling mistake made on Slashdot. Why is that? The units are just plain wrong. Those units are the only common units I'm aware of where both letters define the magnitude, which in and of itself makes it a confusing unit; the fact that the only difference between adjacent orders of magnitude is capitalization makes it doubly so. Add to this that the SI folks completely changed the meaning of GB from the commonly accepted meaning into something that has an arbitrary base-ten quantity that has little bearing on the real world of computing, which further added to the confusion. Also, in practical networking, the difference between a byte and a bit tends to be closer to a factor of 10 than to 8, which means that both letters are basically defining base 10 orders of magnitude. What a mess.
I could see dismissing somebody's argument if the writing is sufficiently bad that the meaning is obscured. Dismissing it because of pedantry, however, is inane. You might as well complain because somebody failed to use the word kibibyte (which, BTW, my spell checker still says is wrong even to this day).
I also cannot help but wonder for ISPs that are linked to media giants whether there is some line of thinking that says "We're bleeding due to piracy, people are dropping their cable packages, motions against BitTorrent haven't worked, let's find another way to stem the bleeding". If this were a factor it would be putting self protectionism against national infrastructure interests.
We're talking about Time Warner Cable here. They're a cable company. They make almost all of their money by selling pay-per-view movies and providing cable TV service. They see the direction of the market, and it is rapidly headed towards their main market being worthless as more and more people get their entertainment via the Internet. At that point, they will be a trivially replaceable commodity data pipe and their bread and butter dries up. Anything they can do to limit their customers' adoption of alternative ways of getting TV shows online is helping to extend their doomed TV-based business model that much farther.
Indeed, that's why I think the FCC and/or FTC need to step in and say, "No, you will not." This is a very clear example of anticompetitive monopoly abuse by TWC to stifle coming video-on-demand technologies, alternative distribution channels, etc. If the government allows this behavior to continue and to spread to other ISPs, in the long term, everyone loses.
First, they came for the downloaders, but I did not download, so I had nothing to fear. Then, they came for Slingbox, but I didn't own one, so I rolled my eyes. Then, they came for the YouTube users, but I did not use YouTube, so I did not care. Next, they came for the iTunes, Unbox, and Hulu users, but I never watched TV online, so I remained silent. Then, they came for me and there was no one left to speak out.
No, it doesn't. It only invalidates everything else that happens to be dependent on that, and only by a factor of about 8-10. I'm beginning to get rather annoyed with the absolutism on Slashdot---discounting an entire post/idea/thread because of a minor technical error, an exaggeration, etc. That's the best way to ensure that you always remain uninformed about everything. After all, nobody is always right and precise about everything.
You think it is a coincidence your coworkers usually work in dank and dimly lit cubicles with no human contact?
I always assumed it was because they were vampires.... Also explains the lack of sleeping, the pale skin, and the tendency towards being mostly carnivorous.
And those 5% of customers recommend service to the other 95%. Smart companies have long since learned that you do NOT piss off your "pro customers". It eventually comes back to bite you in the backside with such regularity that it almost goes without saying in most industries. The telecoms have near monopoly status, so they can afford to not care. For now. This, too will change.
I cannot see why we get electricity, gas, water, garbage service and other things measured and pay for the amount we use should not also apply to Internet bandwidth.
With the exception of water, all of those other things you mention have very real marginal costs. The more you use, the more the company has to spend. Electricity costs the company proportionally because they have to use more fuel to produce it. With gas, after you exceed a certain amount, somebody has to go out and drill another oil well to capture more. Similarly, with water, if you exceed the natural capacity of the aquifer, the wells eventually dry up and you have to spend money to drill new ones and/or truck in water until the aquifer replenishes itself. Garbage service costs more because you have to hire more people to work more hours if people use it more heavily. There are very real, tangible marginal costs involved with all of those things.
Internet bandwidth is not like that at all. Initial infrastructure costs notwithstanding, the cost of moving a terabyte of data is approximately the same as the cost of moving a gigabyte. Adding lines to increase capacity costs money, but within the limits of the available bandwidth, the wires have to still be maintained and equipment periodically replaced whether you transfer a terabyte or a byte.
Also, all of the things you mention can be conserved and used later. By not using water, you are increasing the levels in the aquifer (to a point) that can be used later when you have a dry spell. By not using electricity, you are causing generators to be taken offline, saving fuel that can be used to produce power later. By not using as much gas, you are leaving gas in an oil field that can be retrieved later or stored in tanks for future consumption. Internet bandwidth, however, cannot be conserved. Once a second has passed, the gigabit you could have transferred in that time was either transferred or it wasn't. If it wasn't, you can't transfer two gigabits in the next second to make up for it.
The marginal cost of providing Internet bandwidth is zero, so the marginal cost to customers should also be zero. Customers should pay for the infrastructure costs amortized over the life of the hardware plus some percentage for profits. Any other scheme is a scam.
The problem is that these companies have lied to consumers for years saying that they can provide X Mbps (for some value of X) to customers in hopes that they would never really use that much, knowing full well that they were massively overselling their capacity to turn a substantial profit. Now, as customers start to do more with that bandwidth, instead of turning around those huge profits to expand the infrastructure, they are looking desperately for ways to continue to turn huge profits without actually improving the infrastructure. After all, it's not enough to break even. They have to make more profit than the year before to add value for their shareholders. At some point, such an economic model breaks down and they have to pay the piper. I think we're to that point, and no amount of tiered bandwidth is going to fix that. If they continue down this path instead of spending the money they need to spend to improve their infrastructure, they will soon be supplanted by disruptive technologies. Maybe that's good, but it certainly won't be good for companies like Time Warner.
Just make sure you provide a copy of the crypto key to the legitimate owner in case the motherboard fails outright.... That would really suck to lose all your data merely because your motherboard blew a couple of filter caps.
Can we mod the parent +1, Sad-but-true? These days, I wouldn't be surprised to see a pharmaceutical ad that says, "Side effects are mild and may include internal bleeding, kidney or liver damage, erectile dysfunction, and brain aneurysms...." Half the drugs out there seem to cause some sort of kidney or liver issues in some people---acetaminophen (e.g. Tylenol), NSAIDs (non-steroidal anti-inflammatory drugs, a common type of pain reliever), bisphosphonates (osteoporosis treatment), gadolinium (used in CT scans), Trasylol (clotting drug), propylene glycol (used in LOTS of over-the-counter medicines), etc. The industry long ago crossed the line from "don't care" into outright intentional negligence territory, IMHO.
Actually, the number would be far smaller. First, the original post mistakenly assumed that capitalization differences are sufficient semantic change to be worthy of copyright. They aren't, generally speaking. Even diacritics usually don't result in different meanings most of the time (though they are still considered spelling errors). Further, commas, periods, and spaces are even less relevant; they actually detract from the amount of content that you can squeeze into 140 characters. As such, they don't increase the number of possible combinations that are relevant; they actually decrease it.
And if you only consider plausibly valid messages in a single language, the number plummets. Ignoring leet speak or other shorthand forms, the average educated English speaker knows about 20,000 words. The average word length is about 5.1 letters per word, but add an additional character for the space in between. Turns out that this is almost exactly 23 words on average. (The extra space after the last word makes the difference between it being under or over 140 characters.) This means that the number of words that are likely to occur is only about 20,000 ^ 23, or 8.388608 × 10^98.
More to the point, though, Twitter messages tend to be conversational in nature, not the sorts of words somebody would use in a term paper on a highly technical subject. The average person only uses about 2,000 words in normal conversations during the course of a typical week. Thus, ignoring highly specialized vocabulary, you could probably cover 99% of Twitter messages with a much more manageable 2,000^23, or 8.388608 × 10^75. Add in word probabilities and grammar rules to reduce this number further.
Still not practically publishable, of course. If you could put one stack of terabyte drives per square foot across the entire U.S., you'd have to stack the drives approximately 4.47473245 × 10^29 light years high....
I'm not sure it's the background apps that are the problem so much as what 99% of background apps do while in the background. The main reason to run an application in the background is to do networking, and that means the radio has to move from GPRS mode into EDGE or 3G mode, which drains significantly more power. It's not at all surprising that it causes a much higher battery drain if background apps keep waking the cellular hardware while it should be idle.
As soon as you bring up the cellular network to get data, you're spending several seconds negotiating with the tower to switch from GPRS mode to EDGE or 3G and obtain an IP number for the interface. Then, your initial DNS lookup, at least based on my experience with AT&T's EDGE network can potentially add another 10-15 seconds in the worst case. Pull even a trivial amount of data and you've probably added another ten or fifteen seconds. At that point, you've spent the better part of a minute with the radio draining significantly more power than it does in its normal GPRS/waiting-for-calls standby state. Do this once a minute, and you almost might as well be talking on the phone to somebody. Okay, so maybe that's a bit of an exaggeration, but it does drain a lot more power....
Slashdot Mirror
Yes and no. A page or cell failure will result in I/O errors if there are no more spares, and if it occurs during a read cycle, it -should- result in I/O errors for all subsequent reads from that cell or page until it gets rewritten to a new cell or page. If it doesn't work that way, then the device is fundamentally violating the contract between the device and the OS to report all nonrecoverable errors that result in data loss.
Also, while a multi-chip design reduces the probability of a device failing outright, it dramatically increases the probability of a failure. First, using a separate controller significantly increases the probability of failure because instead of having interconnect traces on a slab of silicon that (electromigration notwithstanding) almost never change or fail if they work from the factory, you have solder joints exposed on a circuit board. Solder joints are the most common cause of circuit failure in my experience.
Even ignoring the increased risk of having extra solder joints between the controller and flash parts, the odds of failure are still much worse for multi-chip devices. Remember your RAID MTBF theory. The MTBF of a collection of devices is equal to the MTBF of one device divided by the number of devices. If you have one part, the MTBF on that slab of silicon and associated solder joints might be a year. If you have five parts, the MTBF is now 73 days. That's an extreme example, but sadly, I've seen flash sticks with large numbers of failures in the first month, so that's not nearly as gross an exaggeration as you might think.... And whether one part fails or the whole thing fails, you still lose data.
Also, a controller failure is still likely to cause all flash parts to be inaccessible whether it is integrated into a flash chip or is driving eight discrete flash chips. It's not like you're going to use a separate flash controller per flash part. And I -think- that a device showing zero capacity is probably caused by the flash controller being unable to communicate with the flash parts. If so, then that is much more likely to be caused by a failed connection between the two than by a failed flash controller (unless there are problems with interconnects inside the flash controller chip package failing due to overzealous compliance with ROHS rules).
The original poster also failed to mention the most common failure mode, bar none: poor solder joints or other physical interconnects getting broken by physical force. This is very common among cheap flash drives. I wouldn't expect the same with SSDs, of course---you don't normally carry a SSD in your pocket---but at least in my experience, this one cause of failure is easily an order of magnitude more frequent than any other single cause, and is in all likelihood greater than all the others put together. And that's not even counting actual abuse (washing machines, run over by cars, and so on).
My Lexar JumpDrive Secure flash drive suddenly stopped working, and I talked to my mother, whose entire university class was using that same model of drive. Turns out that between us, we had experienced close to a 50% failure rate on those things within the first month or so, having seen somewhere around 14 or 15 failures. The failure was interesting. Mine failed suddenly, but worked if you tipped the connector at an angle... at least for a couple of seconds once or twice. This told me pretty conclusively that the failure was caused by poor hardware design. As best I can tell, when you carry the drive in your pocket, the cap puts pressure on the USB connector. Over time, this gradually causes solder joint or trace failure (I never cut one open to figure out which) at or near the USB connector.
Since then, I only buy flash devices with mechanisms where the USB connector retracts into a solid housing. Sure, you have an elevated risk of gunk from your pocket getting into the connector because it isn't covered, but at least you don't have the flexing problem. Gunk can be cleaned with a flat toothpick and alcohol. Failed solder joints requires disassembly and SMT soldering skills.... :-)
This story usually contains the phrase "It's on a flywheel UPS circuit" and ends with "The boot drive experienced a head crash and is now completely dead (it won't even spin up), but the application drive is still working fine, so as long as the power doesn't fail, it should be okay. It has been running this way for three years." And then, with every retelling, the number of years increases by one.... Eventually, it starts to add "interesting" causes for the head crash, e.g. the senior admin slamming the intern's head repeatedly into it after he inadvertently switched off the main UPS that powers the mission-critical servers. Yeah. Good times.
Well, what's interesting is that IIRC the study suggests that Lipitor reduces the risk of heart attacks more than the other statins even if the other drugs reduce cholesterol levels to similar levels.. This at least hints that some other side effect of Lipitor causes cholesterol to not accumulate as much, and that lowering cholesterol levels is a red herring in most of the studies to date. This is further supported by a study by University Hospital in Toronto that showed that among heart attack patients, the probability of a second heart attack did not vary significantly based on cholesterol levels. (Citation needed.) If high serum cholesterol levels were really the root cause, one would expect a significantly higher risk of a second heart attack in those with higher serum cholesterol levels. The lack of continued correlation within this reduced subset of the population suggests to me that high cholesterol levels are likely a frequent (but not guaranteed) symptom of the real underlying problem, rather than the actual cause.
They exist already. For example, the RazorBook. They run Wince... I mean Win CE.
40 hours seems quite plausible. If you figure an iPhone battery has a capacity of about 5.18 Watt hours (1400 mAh * 3.7V) according to ipodbatteryfaq.com and it handles computation at blast for several hours on a charge, ignoring the extra power for a larger screen for the moment, if it had a battery the size of a MacBook (62.4 Watt hours according to System Profiler's battery stats on mine), it would last on the order of 60 hours on a charge even running at full tilt. Doing lighter work, I could easily see that extended by as much as a factor of three. So when you factor in the bigger display, yeah, I could see 40 hours being possible, assuming good power management. And that is definitely a machine I would buy in an instant. I think 20 hours is probably more realistic given the size constraints of a netbook, though.
That said, the battery life on the RazorBook is reportedly only on the order of 4 hours. Given that the CPU is comparable in its power consumption, this tells me that either the screen backlight is an unholy pig or Win CE power management is absolutely terrible. Neither would be much of a surprise. No idea how the Linux version of the RazorBook does on power.
PA-RISC is one of those things that will still be around after the nuclear holocaust destroys nearly all life as we know it. Two things will still be alive: a cockroach and a PA-RISC server looking forlorn alone in a corner somewhere wondering why nobody is talking to it anymore.
We'll leave it as an exercise to the reader to determine the precise nature of the word "alive" in this context, as either definition could be deeply disturbing in its own right.
When we talk about sinus infections, we are usually talking about bacterial infections. If an infection isn't bacterial, it's really just a head cold or flu. The symptoms of a sinus infection are different from symptoms caused by viral agents. With bacteria, you have a high fever, while colds rarely cause significant fever in adults. Although influenza can cause a fever, it is accompanied by other symptoms that are not localized to the head (random muscle aches, extreme fatigue, chills, etc.) that you do not get with a bacterial infection. With a sinus infection, you get a fever (sometimes high), your head is clogged, and you may feel somewhat run down after a few days of it, but you basically don't feel sick outside of your head (and possibly throat) unless it moves into pneumonia.
It also sounds like you mistakenly believe that strep throat is just another term for a sore throat. By definition, strep is always bacterial. The term "strep" is short for "streptococcus", which is a genus of bacterium. It's not that strep "can be bacterial", but rather that it must be. If you use it to refer to anything else, you're using the term incorrectly.
If this doctor had said that you shouldn't give antibiotics for most sore throats or sinus congestion, I would have agreed; most stuffy heads and sore throats are viral, will go away on their own, and are not affected by antibiotics. However, making such an argument for sinus infections (a term usually reserved for bacterial infections) or strep (a term for a specific class of bacterial infections that can turn into a much more serious and sometimes fatal form if left untreated), the argument crossed a line from being perfectly sensible advice to being just plain reckless.
Good idea. I started working on fireworks last night. To run it, type "./boom.sh" from within the directory. Right now, it just goes through each of the different sizes/styles of explosion in order at a fixed position.
Right now, it is just a bunch of ASCII image maps built and routines to draw each of them at a given pair of screen coordinates. I still need to write code to randomly choose a launch angle/speed, calculate trajectory, and actually draw the rocket dot while in flight. I might add code to split the line arrays into individual elements so that things can overlap somewhat or I might draw things over and over in the main program loop so that conflicts will alternate. I might add some sane color maps (randomly chosen) to allow the explosions to have mixed colors instead of one per burst. I might use multiple background jobs to do the drawing so I can spin off an arbitrary number of drawing tasks or I might do it with an array and a deadline scheduler. There are a lot of directions I could go with it.... :-)
I would not be surprised if this article were wrong about that conclusion. The article is definitely wrong about the conclusion regarding sinus infections.
Ever since I was a kid, I've been prone to sinus infections. Every couple of years, I'll have one. They have never---and I do mean never---gone away on their own. I've left them for well over a week just to see. Never happens. By contrast, when I take antibiotics, whether that is after four days of symptoms or eight, the symptoms decrease dramatically within just a few hours after the start of the first round. Even if the infection would have gone away after a few more days (and I'm not convinced it would have), taking an antibiotic still means you get three or four days of your life back, and that alone is sufficient reason to consider such treatment.
Further, the doctor's view is dangerous. Doctors don't prescribe antibiotics for strep throat because of the symptoms. They prescribe antibiotics for strep throat because if left untreated, strep infections can turn into scarlet fever (which can cause kidney damage) or rheumatic fever (which has a 2-5% mortality rate). Failing to prescribe antibiotics if strep is suspected is madness.
Even in the case of sinus/ear infections, his few is dangerous. My ear infections have gotten painfully close to rupturing my ear drum. If I did not take antibiotics, there is a very good chance that I would have hearing damage today. As a musician, that would be a bad thing.
Now if you want examples of bad medicine, take a look at statin drugs. Heart doctors notoriously prescribe these to people with even moderate cholesterol levels to bring their levels down. While clinical studies do show that in patients with severely elevated cholesterol, heart attack risk can be reduced by some statins, other statins have been shown to increase the risk of heart attacks and strokes.
Also, the side effects are much more common than they lead you to believe and are severe and debilitating. Both my grandparents have experienced textbook adverse reactions to statin drugs (peripheral neuropathy in both, sleep issues in one, confusion/anxiety/cognitive disruption in one). These side effects are very real.
Worse yet, studies show that these drugs don't reduce overall deaths because for every death they prevent from heart disease or strokes, you get an increase in deaths from other causes, including cancer. So basically it reduces quality of life while failing to actually increase the typical length of life....
You're confusing capacity costs with marginal costs. While they can be related (in the case of power, for example, capacity must exceed demand or else you get brownouts that could pose safety risks), they are not the same thing.
The cost of the network is not based on the amount of data the users download at all. It is based on the amount of money the ISP chooses to spend to deliver an average level of bandwidth that they choose. If everybody started downloading twice as much content and the pipes remained saturated 24x7 but the ISP did not increase the number of outgoing pipes, the cost to provide service would still be roughly the same as it is now. If the ISP kept the capacity the same, but the usage dropped in half, the cost to provide the service would still be roughly the same as it is now, but half the capacity would be wasted. Therefore, subject to the assumption that the ISP is unlikely to change capacity if they can help it (which at least appears to be true more often than not), the marginal cost of additional use is zero, at least until it gets so bad that users start switching to a competing service due to the perception of poor network performance. Since a significant portion of cable modem customers have no other choices, even that condition is largely moot.
Although in an ideal world, the capacity should be tied to customer use, in practice, the capacity can be woefully insufficient during peak periods, particularly given that the ISPs would like their infrequent users to experience near maximum theoretical performance during peak periods. Even without the downloaders, this would probably be unlikely. Thus, the ISPs have three choices: increase capacity, convince people to shift bulk usage to off-peak periods, or introduce asinine caps that cause the capacity to be completely wasted for half the day. Two of those solutions work reasonably. Guess which one they picked....
The right way to handle this is to contact customers who download or upload huge amounts of data. Tell them that they are using an excessive amount of bandwidth and request that they do most of their large downloads overnight. Some people will ignore this, but most will comply, and as a result, the average use will be much more balanced and much easier to manage. Another good solution is to queue and delay delivery of packets to/from folks who are pulling or pushing large volumes of data relative to traffic from folks who aren't. In this way, capacity would be used at maximum efficiency at all times, preventing poor service by customers doing casual browsing while not significantly diminishing the bandwidth available to downloaders.
I never said that your bandwidth should be limited based on the assumption you will be using it all the time. I said that the providers need to increase capacity instead of whining that it will cost them money....
As soon as you give in to caps, different plans by different vendors cannot be easily compared. This reduces competition, which invariably hurts consumers. Also, it limits the profitability of upcoming Internet services that use more bandwidth, which has the potential to seriously stifle the future of the Internet.
I did read what you said. I just disagree with your assessment that what you're complaining about is anything more than a minor typographical error. And I'm very glad not everyone thinks the way you do. This is a discussion forum, not a doctoral thesis review. In the context of informal discussion, in my experience, such dogmatic rigor is an impediment to creativity and the free flow of ideas. Making people afraid to post because they are afraid somebody will point out a trivial mistake and blow it out of proportion is the best way to prevent lots of insightful points of view from ever being heard at all.
IMHO, the right response when somebody is sloppy about units or the spelling thereof is to point out the mistake. Once. One sentence, then drop it. it's just not that important in the context of an informal discussion, and blowing such criticism out of proportion is not conducive to open discussion.
I can't help but think of this:
Count de Monet: Sir, the peasants are revolting.
King: That's right. They stink on ice.
---History of the World Part I
So basically if somebody came up with a workable "theory of everything" but spoke English poorly and made a few common spelling errors, you think it would be appropriate to dismiss the theory with a hand wave?
GB vs Gb is almost certainly the most common spelling mistake made on Slashdot. Why is that? The units are just plain wrong. Those units are the only common units I'm aware of where both letters define the magnitude, which in and of itself makes it a confusing unit; the fact that the only difference between adjacent orders of magnitude is capitalization makes it doubly so. Add to this that the SI folks completely changed the meaning of GB from the commonly accepted meaning into something that has an arbitrary base-ten quantity that has little bearing on the real world of computing, which further added to the confusion. Also, in practical networking, the difference between a byte and a bit tends to be closer to a factor of 10 than to 8, which means that both letters are basically defining base 10 orders of magnitude. What a mess.
I could see dismissing somebody's argument if the writing is sufficiently bad that the meaning is obscured. Dismissing it because of pedantry, however, is inane. You might as well complain because somebody failed to use the word kibibyte (which, BTW, my spell checker still says is wrong even to this day).
We're talking about Time Warner Cable here. They're a cable company. They make almost all of their money by selling pay-per-view movies and providing cable TV service. They see the direction of the market, and it is rapidly headed towards their main market being worthless as more and more people get their entertainment via the Internet. At that point, they will be a trivially replaceable commodity data pipe and their bread and butter dries up. Anything they can do to limit their customers' adoption of alternative ways of getting TV shows online is helping to extend their doomed TV-based business model that much farther.
Indeed, that's why I think the FCC and/or FTC need to step in and say, "No, you will not." This is a very clear example of anticompetitive monopoly abuse by TWC to stifle coming video-on-demand technologies, alternative distribution channels, etc. If the government allows this behavior to continue and to spread to other ISPs, in the long term, everyone loses.
First, they came for the downloaders, but I did not download, so I had nothing to fear. Then, they came for Slingbox, but I didn't own one, so I rolled my eyes. Then, they came for the YouTube users, but I did not use YouTube, so I did not care. Next, they came for the iTunes, Unbox, and Hulu users, but I never watched TV online, so I remained silent. Then, they came for me and there was no one left to speak out.
No, it doesn't. It only invalidates everything else that happens to be dependent on that, and only by a factor of about 8-10. I'm beginning to get rather annoyed with the absolutism on Slashdot---discounting an entire post/idea/thread because of a minor technical error, an exaggeration, etc. That's the best way to ensure that you always remain uninformed about everything. After all, nobody is always right and precise about everything.
I always assumed it was because they were vampires.... Also explains the lack of sleeping, the pale skin, and the tendency towards being mostly carnivorous.
And those 5% of customers recommend service to the other 95%. Smart companies have long since learned that you do NOT piss off your "pro customers". It eventually comes back to bite you in the backside with such regularity that it almost goes without saying in most industries. The telecoms have near monopoly status, so they can afford to not care. For now. This, too will change.
With the exception of water, all of those other things you mention have very real marginal costs. The more you use, the more the company has to spend. Electricity costs the company proportionally because they have to use more fuel to produce it. With gas, after you exceed a certain amount, somebody has to go out and drill another oil well to capture more. Similarly, with water, if you exceed the natural capacity of the aquifer, the wells eventually dry up and you have to spend money to drill new ones and/or truck in water until the aquifer replenishes itself. Garbage service costs more because you have to hire more people to work more hours if people use it more heavily. There are very real, tangible marginal costs involved with all of those things.
Internet bandwidth is not like that at all. Initial infrastructure costs notwithstanding, the cost of moving a terabyte of data is approximately the same as the cost of moving a gigabyte. Adding lines to increase capacity costs money, but within the limits of the available bandwidth, the wires have to still be maintained and equipment periodically replaced whether you transfer a terabyte or a byte.
Also, all of the things you mention can be conserved and used later. By not using water, you are increasing the levels in the aquifer (to a point) that can be used later when you have a dry spell. By not using electricity, you are causing generators to be taken offline, saving fuel that can be used to produce power later. By not using as much gas, you are leaving gas in an oil field that can be retrieved later or stored in tanks for future consumption. Internet bandwidth, however, cannot be conserved. Once a second has passed, the gigabit you could have transferred in that time was either transferred or it wasn't. If it wasn't, you can't transfer two gigabits in the next second to make up for it.
The marginal cost of providing Internet bandwidth is zero, so the marginal cost to customers should also be zero. Customers should pay for the infrastructure costs amortized over the life of the hardware plus some percentage for profits. Any other scheme is a scam.
The problem is that these companies have lied to consumers for years saying that they can provide X Mbps (for some value of X) to customers in hopes that they would never really use that much, knowing full well that they were massively overselling their capacity to turn a substantial profit. Now, as customers start to do more with that bandwidth, instead of turning around those huge profits to expand the infrastructure, they are looking desperately for ways to continue to turn huge profits without actually improving the infrastructure. After all, it's not enough to break even. They have to make more profit than the year before to add value for their shareholders. At some point, such an economic model breaks down and they have to pay the piper. I think we're to that point, and no amount of tiered bandwidth is going to fix that. If they continue down this path instead of spending the money they need to spend to improve their infrastructure, they will soon be supplanted by disruptive technologies. Maybe that's good, but it certainly won't be good for companies like Time Warner.
It does if the software is AGPL, assuming that your definition of "use" includes installing/deploying it on a server....
Just make sure you provide a copy of the crypto key to the legitimate owner in case the motherboard fails outright.... That would really suck to lose all your data merely because your motherboard blew a couple of filter caps.
That sounds like it manages restrictions. It doesn't, though. It manages your rights by adding restrictions on them. Digital Rights Munger, perhaps?
Can we mod the parent +1, Sad-but-true? These days, I wouldn't be surprised to see a pharmaceutical ad that says, "Side effects are mild and may include internal bleeding, kidney or liver damage, erectile dysfunction, and brain aneurysms...." Half the drugs out there seem to cause some sort of kidney or liver issues in some people---acetaminophen (e.g. Tylenol), NSAIDs (non-steroidal anti-inflammatory drugs, a common type of pain reliever), bisphosphonates (osteoporosis treatment), gadolinium (used in CT scans), Trasylol (clotting drug), propylene glycol (used in LOTS of over-the-counter medicines), etc. The industry long ago crossed the line from "don't care" into outright intentional negligence territory, IMHO.
Actually, the number would be far smaller. First, the original post mistakenly assumed that capitalization differences are sufficient semantic change to be worthy of copyright. They aren't, generally speaking. Even diacritics usually don't result in different meanings most of the time (though they are still considered spelling errors). Further, commas, periods, and spaces are even less relevant; they actually detract from the amount of content that you can squeeze into 140 characters. As such, they don't increase the number of possible combinations that are relevant; they actually decrease it.
And if you only consider plausibly valid messages in a single language, the number plummets. Ignoring leet speak or other shorthand forms, the average educated English speaker knows about 20,000 words. The average word length is about 5.1 letters per word, but add an additional character for the space in between. Turns out that this is almost exactly 23 words on average. (The extra space after the last word makes the difference between it being under or over 140 characters.) This means that the number of words that are likely to occur is only about 20,000 ^ 23, or 8.388608 × 10^98.
More to the point, though, Twitter messages tend to be conversational in nature, not the sorts of words somebody would use in a term paper on a highly technical subject. The average person only uses about 2,000 words in normal conversations during the course of a typical week. Thus, ignoring highly specialized vocabulary, you could probably cover 99% of Twitter messages with a much more manageable 2,000^23, or 8.388608 × 10^75. Add in word probabilities and grammar rules to reduce this number further.
Still not practically publishable, of course. If you could put one stack of terabyte drives per square foot across the entire U.S., you'd have to stack the drives approximately 4.47473245 × 10^29 light years high....
Can't it be both?
I'm not sure it's the background apps that are the problem so much as what 99% of background apps do while in the background. The main reason to run an application in the background is to do networking, and that means the radio has to move from GPRS mode into EDGE or 3G mode, which drains significantly more power. It's not at all surprising that it causes a much higher battery drain if background apps keep waking the cellular hardware while it should be idle.
As soon as you bring up the cellular network to get data, you're spending several seconds negotiating with the tower to switch from GPRS mode to EDGE or 3G and obtain an IP number for the interface. Then, your initial DNS lookup, at least based on my experience with AT&T's EDGE network can potentially add another 10-15 seconds in the worst case. Pull even a trivial amount of data and you've probably added another ten or fifteen seconds. At that point, you've spent the better part of a minute with the radio draining significantly more power than it does in its normal GPRS/waiting-for-calls standby state. Do this once a minute, and you almost might as well be talking on the phone to somebody. Okay, so maybe that's a bit of an exaggeration, but it does drain a lot more power....