Modern CPU cache lines are typically on the order of tens or hundreds of bytes. Yes, bytes. Not kilobytes. Memory fragmentation doesn't really affect CPU caching. Fragmentation does, however, affect allocation/deallocation performance as well as application memory access speed.
It does seem unlikely that a cellphone would knock out the avionics computer, especially considering that the computer has to deal with all kinds of more exotic radiation sources when it's higher up in the atmosphere. I also tend to think that if the FCC/FAA/etc. truly believed that a cell phone (or other radio transmitting device) inside a plane posed a significant risk to aircraft electronics, they'd confiscate everyone's cell phones before they got on the plane, take out the batteries, and hold them for you until the plane landed at its destination.
Classic example of a question that can't be properly answered by a yes or no: "Do you still beat your wife?" I know what you meant, but just to be pedantic: sure, that question *can* be properly answered with a yes or no (supposing one has, in the past, beaten his wife, if he still currently does, then "yes" is a proper answer; if he now doesn't, "no" is proper).
Regardless of being able to answer a question properly, there's a perfectly valid third answer: "the question is meaningless because it presupposes incorrect information as fact." In that case the "do you still beat your wife?" question would just be unsuitable for a database of factual questions and answers, because it cannot be answered as-is. That's not really a flaw of the data gathering/storage method; it's a flaw of the question and questioner.
Anyhow, I'm still not convinced that yes/no questions about facts aren't useful. "Maybe" as an answer is really a non-answer. You just get a similar case to "the question is meaningless..." in the form of "the question cannot be answered without being made more specific." If the answer to "can birds fly?" is "some of them can, and some of them can't," then the question is just useless. You need, "can pigeons fly?" (yes), "can penguins fly?" (no), etc.
Of course, I'm not really sure where I'm going with this. All I've done here is assert that many (maybe not all) failed attempts at yes/no questions can be 'repaired' by making them more specific. Who knows if that's still useful for AI research purposes.
Sure Apple's to blame. They're responsible for setting requirements for the product and ensuring any parts supplied by manufacturers meet those requirements. If working for a company that outsources some hardware design and assembly has taught me anything, it's this: never trust that what you are given by an outside entity actually does what you expect or want it to do.
Likely Apple just didn't care about supporting odd-shaped or -sized CDs/DVDs. As the owner of an Apple laptop, I don't have a problem with that, though it would be nice if Apple were to warn customers about these things. (And maybe they do; I suck at reading manuals.)
Because TCP and UDP headers aren't of fixed sizes and as such are incredibly difficult to handle in hardware. UDP headers are always 8 bytes long. TCP headers are indeed not fixed-length, but will always be a multiple of 4 bytes, will always be at least 20 bytes, and there's a field in the first 20 bytes that tells how large the header is. All of this can certainly be interpreted by hardware, but, as usual, it's cheaper to do it in software.
Most of us just drive less, drive slower, and drive more efficient vehicles. I have seen no evidence of this whatsoever. I live in southern NH, and over the last year have spent a fair amount of time in NV, ME, MA and NJ. The roads are more congested, driving speeds are higher if anything, and most (75% or more?) of the vehicles on the road around here are goddamned SUVs/pickup trucks/minivans. I think this actually supports the OP's position, if not his evidence for it. Despite rising gas prices, Americans are driving just as much, if not more, and are continuing to buy and drive fuel-inefficient vehicles. If that doesn't scream "Americans don't care that much about the price of gas," I don't know what does.
I tend to do this during heavy traffic. I drive a manual, and shifting in and out of gear over and over gets old real fast. The trick is just to not care. It's hard, I know. Give yourself more than 2 seconds. I tend to go by car lengths, just because traffic conditions vary. In slow traffic (5-25 mph) I'll try to maintain a 10-car-length space or so. This doesn't mean I'm constantly braking and accelerating; if I wanted to do that, I'd just tailgate the guy in front of me like everyone else. If someone cuts in in front of you, let them. They're probably just going to speed up and slam on their brakes when they get to the car in front of you. You can slightly adjust your speed by releasing the gas pedal a little to maintain your space. There's no real hurry -- you have 9 car lengths worth of time to do this. If traffic is heavy and slow enough, you end up with people to the right and left (and behind) you who aren't "cheaters," and people tend to stop cutting in front of you so much. But even if they do, the magic phrase is: "who cares?"
I drive a manual, and I practice this sort of slow plodding during traffic jams all the time. With the really bad traffic days, it can be a problem, but my car can plod along in 1st at 5 mph just fine. Usually the "plod speed is between 10 and 20 mph, though, fortunately. Around 20 mph I can coast in 3rd without having to give it much gas to keep going.
So if you load bad microcode, you simply power-cycle the box to fix it. Unless your microcode contains an HCF op:) I wonder if the Linux kernel has a "CPU is on fire" error message like the old "printer is on fire" message ^_~.
Actually, loading bad microcode into an Intel CPU isn't a big deal at all -- the load isn't persistent; you have to reload the microcode every time you boot. So if you load bad microcode, you simply power-cycle the box to fix it.
For values of reasonable only accepted by economists and scalpers.
Er, no, as defined by the 350,000 people in America who bought a Wii last week. Probably more than that, actually, considering that there a likely many people who would have paid the same price as the 350k people did, but were unable to get one due to the shortages.
A (supply/demand driven) economist or a scalper whould set the price higher than Nintendo have. If my supposition is correct, they're actually going cheaper than they strictly have to. From the seller's point of view, I'd think the "ideal" price to sell something at is the price at which you sell every single unit in your inventory at a price that all the actual buyers -- and no one else -- are willing to pay. So you have 350k units, and you find a price that 350k people -- and only 350k people -- are willing to pay. Yes, I just made that up, but it sounds reasonable to me. I'm no economist.
They would then be pretty likely to share it because there is no reason not to do so. Individuals, yes, probably. Organizations? Maybe, maybe not. In my experience, when someone at a company fixes a bug in 'upstream' software, they keep it to themselves[1]. It cost the company money to find and fix that bug, so they figure something like: why should we give that time (money) to our competitors for free?
Not saying I agree or disagree with this attitude... it's just how it is.
[1] Well, except for fixes to GPLed code.
...it's not funny anymore with very few exceptions. Who died and made you the final word on what is and isn't funny? (Our new Slashdot overlord, perhaps?) People have different tastes and appreciate different kinds of humor. Live and let live.
Perhaps, but that's not necessarily a bad thing. If the result of locking things down and policing content means 4% users leave Facebook or use it less because it's less useful, but leaving things as-is only causes a loss 1% of their users, then they are indeed "doing the right thing" in the face of their customers, the advertisers.
Of course, I'm just making up numbers and I don't know which would cause more of a loss. But this is Slashdot; why do I need hard facts?
In my US-centric experience, salaried professionals are always on the clock, even on vacation. When I'm on vacation, I'm on vacation: I don't even check my work email. My boss and those coworkers who have my cell number wouldn't call me unless it was an emergency, and even then I might not answer if it wasn't a good time. (Not to mention I might be in a place where my cell doesn't work.) If there was a situation where I had to work a non-trivial amount of time while on vacation, I would expect something in return, like one or more days of my vacation being 'free', depending on how much I had to work.
It doesn't sound fair, but these positions are often quintessential to a business (or, at least, they should be), and thus they need to be on call if needed. IMO any company which can't do without any particular employee for a couple weeks is doing something wrong. Exceptions might be officers and top executives, and possibly employees at small startups who go into it knowing they're likely giving up their personal time.
Facebook already looks out for the interests of their customers -- their advertisers. Facebook users are not Facebook customers. Facebook users are the product that they sell to advertisers.
Right, because ignoring "other people's problems" worked so well for us in World War 2. Straw man. No one's claiming the US's entrance into WWII was a mistake. The US entered WWII not to bail out Europe, but because failing to do so likely would have lead to a much-stronger Nazi Germany being capable of successfully attacking the US.
Leaving Iraq alone to its own devices is a bit different. I tend to think we should have left them alone -- or at least have done a much better (WWII-era, Japan-style) job of 'introducing' Iraq to democracy -- but there are certainly valid reasons for believing otherwise.
Or do you only turn into a pedantic snobbish asshat when it's convenient to dodge criticism of your preciousssss.... preciousssss... Avoiding the obvious flamebait here... well, yes. It doesn't matter what 'we' think Linux is. The "Linux Driver Project" was set up by a bunch of Linux kernel developers. To them, when they say "I work on Linux," they mean that they work on the kernel (and sometimes nothing more). Really, the only thing that matters here are the intentions of the people running the project, and their intentions are clearly to solicit feedback as to what devices need Linux *kernel* drivers. They even say this on their website, but of course no one knows that since that would require RTFA-ing.
Are they against learning userspace programming? What exactly, is the objection to doing whatever it takes? For an unpaid volunteer, "I'd rather work on X instead of Y" is really the only objection necessary. Not all OSS developers have a "take over the world" mentality. Some just do it because it's fun.
I'm stating that it is wrong for that someone (whether or not he is a provider of 'A') to claim that the 'B' thing that I want is a non-issue simply because he does not want to work on it (irrespective even of the reason why he doesn't). No one's saying it's a non-issue. In fact, the Linux driver project webpage has this to say about printers:
All Linux printer drivers are done in userspace. Contact the Linux Printing Project if you have a printer that you wish to get properly supported under Linux. So not only are they NOT saying it's a non-issue, they're directing people to where they should go for printer issues.
Besides that, you're not going to tell me that people smart enough to learn how to code a kernel can not learn how to implement a user space printer driver. I'm not saying that they have to do that, but they sure can if the[y] want to. That's probably true, but who are you to tell unpaid volunteers what to do with their time? They're Linux kernel developers. It's what they do. I imagine they spend a lot of their free time on it, and spending their free time on other things (like learning how to reverse-engineer a printer interface and write a CUPS driver for it) would take away from their time working on the kernel. Let them do what they want; they need not feel any obligation to support printers if they don't want to.
Slashdot Mirror
Modern CPU cache lines are typically on the order of tens or hundreds of bytes. Yes, bytes. Not kilobytes. Memory fragmentation doesn't really affect CPU caching. Fragmentation does, however, affect allocation/deallocation performance as well as application memory access speed.
Stating "I'm probably wrong" after giving an incorrect numerical estimate in no way should be construed as a reason to avoid correcting that person.
Regardless of being able to answer a question properly, there's a perfectly valid third answer: "the question is meaningless because it presupposes incorrect information as fact." In that case the "do you still beat your wife?" question would just be unsuitable for a database of factual questions and answers, because it cannot be answered as-is. That's not really a flaw of the data gathering/storage method; it's a flaw of the question and questioner.
Anyhow, I'm still not convinced that yes/no questions about facts aren't useful. "Maybe" as an answer is really a non-answer. You just get a similar case to "the question is meaningless..." in the form of "the question cannot be answered without being made more specific." If the answer to "can birds fly?" is "some of them can, and some of them can't," then the question is just useless. You need, "can pigeons fly?" (yes), "can penguins fly?" (no), etc.
Of course, I'm not really sure where I'm going with this. All I've done here is assert that many (maybe not all) failed attempts at yes/no questions can be 'repaired' by making them more specific. Who knows if that's still useful for AI research purposes.
I'm a counterexample to your unsubstantiated anecdotal argument, though, for your average Mac user, I'd think you're probably correct.
Sure Apple's to blame. They're responsible for setting requirements for the product and ensuring any parts supplied by manufacturers meet those requirements. If working for a company that outsources some hardware design and assembly has taught me anything, it's this: never trust that what you are given by an outside entity actually does what you expect or want it to do.
Likely Apple just didn't care about supporting odd-shaped or -sized CDs/DVDs. As the owner of an Apple laptop, I don't have a problem with that, though it would be nice if Apple were to warn customers about these things. (And maybe they do; I suck at reading manuals.)
That's good news for me. I refuse to spend more than $40-50 on a CPU anyway. AMD it is, then.
I tend to do this during heavy traffic. I drive a manual, and shifting in and out of gear over and over gets old real fast. The trick is just to not care. It's hard, I know. Give yourself more than 2 seconds. I tend to go by car lengths, just because traffic conditions vary. In slow traffic (5-25 mph) I'll try to maintain a 10-car-length space or so. This doesn't mean I'm constantly braking and accelerating; if I wanted to do that, I'd just tailgate the guy in front of me like everyone else. If someone cuts in in front of you, let them. They're probably just going to speed up and slam on their brakes when they get to the car in front of you. You can slightly adjust your speed by releasing the gas pedal a little to maintain your space. There's no real hurry -- you have 9 car lengths worth of time to do this. If traffic is heavy and slow enough, you end up with people to the right and left (and behind) you who aren't "cheaters," and people tend to stop cutting in front of you so much. But even if they do, the magic phrase is: "who cares?"
I drive a manual, and I practice this sort of slow plodding during traffic jams all the time. With the really bad traffic days, it can be a problem, but my car can plod along in 1st at 5 mph just fine. Usually the "plod speed is between 10 and 20 mph, though, fortunately. Around 20 mph I can coast in 3rd without having to give it much gas to keep going.
Actually, loading bad microcode into an Intel CPU isn't a big deal at all -- the load isn't persistent; you have to reload the microcode every time you boot. So if you load bad microcode, you simply power-cycle the box to fix it.
Not saying I agree or disagree with this attitude... it's just how it is.
[1] Well, except for fixes to GPLed code.
Unfortunately, with Access, it's not about the database itself, but about the GUI tools that many people find easy to use...
...it's not funny anymore with very few exceptions. Who died and made you the final word on what is and isn't funny? (Our new Slashdot overlord, perhaps?) People have different tastes and appreciate different kinds of humor. Live and let live.Perhaps, but that's not necessarily a bad thing. If the result of locking things down and policing content means 4% users leave Facebook or use it less because it's less useful, but leaving things as-is only causes a loss 1% of their users, then they are indeed "doing the right thing" in the face of their customers, the advertisers.
Of course, I'm just making up numbers and I don't know which would cause more of a loss. But this is Slashdot; why do I need hard facts?
Facebook already looks out for the interests of their customers -- their advertisers. Facebook users are not Facebook customers. Facebook users are the product that they sell to advertisers.
Leaving Iraq alone to its own devices is a bit different. I tend to think we should have left them alone -- or at least have done a much better (WWII-era, Japan-style) job of 'introducing' Iraq to democracy -- but there are certainly valid reasons for believing otherwise.