What part of that 14 minutes improvement -- is actually the result of a speedy RAM, improvements to the processor-memory bus interface? If we strip it down to its bare, I have a high confidence that the cores added only a fraction of improvement.
Totally off base. First thing I tried when I got a new quad core dev box at work was to try the standard build with 1, 2, 3, and 4 cores. Imagine my total lack of surprise when the 2 core build ran in about half the time, the 3 core build in about a third the time, and the 4 core build in about a quarter of the time. How shocking.
No, it's not perfect, as in T(N) = T(1)/N, but it's certainly much stronger a correlation than "only a fraction."
Hell, I'd love to see it turned on for all the official build machines, but it's just a little bit scary, since I have seen it fail (as in the build driver crashing) about 1 in a 100 times. But for daily dev work, it's awesome.
Yes, the prices came back up but MONEY CHANGED HANDS. The people who owned in the morning are not the same people who own now. I suggest you check your mutual funds.
...an IT department should, ideally, refuse to honor the request. You mean, just like Terry Childs did?
Oh fucking get real, it's totally different.
"Your honor, we request you throw the book at this guy for refusing to implement a system that could have been used to produce massive amounts of child porn."
I don't trade stocks, but I do have a 401(k). With a 100% employer match up to 6%, I'd be RETARDED not to. With a 100% match, the market could drop 50% and I still wouldn't be out any of my OWN money. Go ahead, try to tell me I'm crazy for taking that deal.
I expect variation in the market, even very large and protracted variation from time to time. But what happened here is purely the effect of automated trading systems and DIDN'T HAVE TO HAPPEN. You bet I'm pissed off.
Wow, where was your brain? You should have been buying buying buying.
I don't trade on the real market, but I play virtual stock games. For the last four months, this has been the entirety of my strategy: go to a financial site. Look at shit that's dropped more than 10% since yesterday. Buy the fuck out of it. Next day, sell it. It works because none of this shit is real.
It's almost like I am buying a turd under the pretense that someone else will be stupid enough to come along and buy that turd for more money.
Congratulations, you figured it out.
Well, it's not QUITE as simple as that. Sometimes the stupid "someone else" is a company which wants to acquire a smaller fish. Typically, the stock will be bought out for a nice premium, or converted to parent shares at a desirable rate.
Yes, you really are just gambling that one of those two things will happen.
Now, if you hold a LOT of shares, you hold voting power over the board of directors and certain other corporate decisions. This only really comes into play for big holders though.
If moon is doing the tide then how come there's high tide 2 times a month.
I assume you meant two times per day? That's because the moon raises tidal bulges on both the near and far sides of the earth. Thus, as the earth turns, you pass under these bulges roughly twice per day. That's essentially the exact definition of a "tidal effect."
It's not like the moon pulls water away on the near side and does nothing on the far side. The tidal effect is caused by the differential of the moon's gravity across the earth's volume. This has a "squashing" effect.
Seriously? You think termites could dam the Colorado River? While it's interesting to note that termites build structures that, were the termites human sized, would be incredibly large... they are not our size.
Some small animals can move at incredible speeds relative to their own size, jump incredible distances, etc... Your point?
And they will know it's encrypted how, exactly? Yes, your typical encrypted data stream looks like random bits, but so does well compressed data.
So either you have to block all data that looks compressed or encrypted, which is a nice way to fuck yourself as a bandwidth provider since people will stop compressing shit to get past your filter, or you have to actually attempt to decompress and look inside any high-entropy data stream. How many reasonably well-deployed compression methods are there? Well, I'd guess about a hundred, if you include various audio and video codecs. So you need to run a number of decompression attempts just to distinguish compressed data from encrypted data. And you really have to DECOMPRESS IT, not just scan for magic numbers or certain headers, because hell, I'll just throw those on there for good measure to confuse you.
Okay, so now that you've established that the data is a compressed stream, you need to look inside the decompressed data to see if that itself looks like its encrypted. Sure, it's boneheaded to compress encrypted data, since it's already such high entropy, but how can you know? Especially when there are people like me trying to get around your filter? You can't, unless you try the whole process again recursively. Obviously, at some point you'll give up. Say you set the bar at two levels of nesting -- at that point it's just too expensive to keep analyzing. Well, that's going to have a shitload of false positives, because people do stupid shit like zip up a video file, which doesn't really gain you that much but is certainly widely done, and would trigger your "give up" signal -- at that point, do you fail open or fail closed? Do you reject a huge amount of traffic that's not encrypted, pissing everyone off and rendering your own service unusable and therefore worthless -- or do you throw your hands up and let the data stream through?
Yeah, sure. They'll just "throttle any encrypted traffic." Good luck with that.
Many animals modify their environment, but pretty much only Homo sapiens is capable of modifying the environment in ways that are impossible to achieve through the use of our natural physiology. The Hoover Dam could not have been built by hand. We exceed the animals by the capacity to create tools, and to use those tools to create even more complex tools, and on and on.
There are animals, primates mostly, that have been observed using inanimate objects as tools. That's true tool use, but they do not make improvements to their tools, nor do they fabricate tools using other tools. Only we can do that.
You would think that given the gravity of their findings, the seriousness they attribute to the situation, the huge nature of the changes they propose, the affect the actions will have on everyone, the potential devastation to the world economy, etc. etc., that they would have bothered to fucking hire a few professional statisticians . Shoddy and careless is what this is.
Par for the course is what it is. Most scientists believe they know statistics, whether they actually do or not. And anyway, you're blowing this out of proportion -- the finding was "correct conclusions, but suboptimal methods." The conclusion isn't "totally fucking wrong and misleading."
Even if you wanted to hire a "professional statistician," there aren't really a lot of people with that exact title or training.
For another thing, there's this saying out there, perhaps you've heard it: "There are lies, damn lies, and statistics." Hiring a professional stats wrangler could actually make you look even worse -- "Hey, these guys are cooking the books, and the evidence is that they've hired a professional liar (i.e. a statistician)"
Newsgroups can be moderated and have been moderated for ages and if you really want to you can just as easily put up a usenet server that requires the users to register an account.
Quite a few people, myself included, find moderated groups offensive. It has a ring of "Papers, please" to it. Yeah, there are a lot of spammers out there. But assuming I'm a spammer until you can prove otherwise? Fuck you.
I miss being able to just read my subscriptions, along with using a scorefile/killfile. Now I have to create accounts on dozens of web pages and monitor them all separately, without being able to rank based on what I'm interested in. Each web page has its own formats and options. Yes, there are rss feeds, but that doesn't help much if you are an active poster in the community.
We've gone backwards.
You call it going backwards, I call it an opportunity for somebody to step in and provide a software solution. We have these computer thingies and programming languages. With these tools I figure we can do something to your liking.
The question is, did those ancient Roman and Indian physicians actually know about germs, or were they just making a lucky guess?
But you can apply such an argument to anybody who doesn't have all the facts. Did Robert Boyle know, for certain, that gases were composed of minute particles, the kinetics of which could be used to derive his "Boyle's Law?" He did not. In the same sense as you are now implying, he made a "lucky guess." A guess which turned out to be correct, and his name has survived in history even though, in modern terms, he didn't know what the fuck he was talking about.
If somebody posits that minute organisms are the ultimate cause of disease, then I give that person props. I really don't give a shit that he cannot prove whether he's right. That fact is, he IS right. You attitude smacks of the bitterness of a person who has perpetually sought success but never achieved it. Whatever.
Yes you can. OSS code is owned by somebody. If it was not, how could the license requirements be enforced? You buy the code, you just realize that you can't stop anybody else in the world from using it -- that's the whole point. You do own it though.
Until Obama's first year, where he outspent Bush's 8 years in just his first.
That doesn't mean *over* spending. Pretty much every economist on the planet was in agreement that if the bailout didn't happen the global economy would COLLAPSE. I don't think you would have seen the alternative.
The number is meaningless, because "duplication" is arbitrary. Where do you draw the line? If duplication means "copying data from one place to another" then data is duplicated every time function parameters are pushed on the stack, every time memcpy() is called, every time something is loaded from disk into RAM. I could write a simple loop that copies a 32-bit quantity from EAX to EBX three billion times per second. If you include all that shit going on, I bet their number would be higher by a factor of a thousand or more. What the fuck is the use of this metric? How do you justify the arbitrary stopping point?
A law is an observation. Something we have observed to always be true. The "law of gravity" for instance. Whenever you observe anything, you see that gravity is there. Thus, it's a law. "Law" doesn't mean, "We've proven it," it means "We know of no counterexample."
Now, as soon as the observation is found to be untrue -- in this case, 18 months pass without a doubling of single-die transistor count -- then the law falls. Until then, "law" is a perfectly good name for it.
Valid point, but if your goal is to ruthlessly acquire wealth, it still makes more sense to go into business, not politics. At least in business we are a bit more socially tolerant of such behavior. My point is that for politicians, payoffs and underhanded dealings are stroking the power trigger, not the money trigger.
I agree that Pleurotus is a most precious species of fungus.
The problem is that the saprophytic fungi species which do so well at consuming raw hydrocarbon cannot survive in sea water. Oyster mushrooms will consume everything from pulp waste to coffee grounds to dead wood to diesel fuel but they need a friendly environment to do it in. Stamets didn't discover this, he was just the first person to test it on a large scale and publicize the results.
Both parties are largely filled with corrupt politicians who want to line their pockets
People always say this. So, where's the list of all Congressmen with net worth of more than a billion dollars? Ten million dollars? If you want to get rich, you go into finance or business, not politics. Even a politician can understand that.
Money and power are interrelated, but let's not get confused here. Politicians are primarily interested in power. Business tycoons are primarily interested in money. Who in their right mind comes up with "Become a US Senator" as a serious strategy for becoming enormously wealthy?
Slashdot Mirror
Penny stocks? Accenture dropped from 40 DOLLARS -- that's 4000 cents -- to 1 cent. Accenture ain't a "penny stock."
Holy shit, would I have loved to ride that one back up again with 40000% profit within 2 minutes.
Fascinating post. Thanks!
What part of that 14 minutes improvement -- is actually the result of a speedy RAM, improvements to the processor-memory bus interface? If we strip it down to its bare, I have a high confidence that the cores added only a fraction of improvement.
Totally off base. First thing I tried when I got a new quad core dev box at work was to try the standard build with 1, 2, 3, and 4 cores. Imagine my total lack of surprise when the 2 core build ran in about half the time, the 3 core build in about a third the time, and the 4 core build in about a quarter of the time. How shocking.
No, it's not perfect, as in T(N) = T(1)/N, but it's certainly much stronger a correlation than "only a fraction."
Hell, I'd love to see it turned on for all the official build machines, but it's just a little bit scary, since I have seen it fail (as in the build driver crashing) about 1 in a 100 times. But for daily dev work, it's awesome.
The market always does that. Ever notice that the Dow, the NASDAQ, and the S&P 500 tend to move up and down in unison?
No shit. But I have NEVER seen a graph shaped like that. It's fucking crazy.
Yes, the prices came back up but MONEY CHANGED HANDS. The people who owned in the morning are not the same people who own now. I suggest you check your mutual funds.
Oh fucking get real, it's totally different.
"Your honor, we request you throw the book at this guy for refusing to implement a system that could have been used to produce massive amounts of child porn."
His Honor: "Dude, WTF are you smoking?"
I'll take the odds on that one.
I don't trade stocks, but I do have a 401(k). With a 100% employer match up to 6%, I'd be RETARDED not to. With a 100% match, the market could drop 50% and I still wouldn't be out any of my OWN money. Go ahead, try to tell me I'm crazy for taking that deal.
I expect variation in the market, even very large and protracted variation from time to time. But what happened here is purely the effect of automated trading systems and DIDN'T HAVE TO HAPPEN. You bet I'm pissed off.
Wow, where was your brain? You should have been buying buying buying.
I don't trade on the real market, but I play virtual stock games. For the last four months, this has been the entirety of my strategy: go to a financial site. Look at shit that's dropped more than 10% since yesterday. Buy the fuck out of it. Next day, sell it. It works because none of this shit is real.
It's almost like I am buying a turd under the pretense that someone else will be stupid enough to come along and buy that turd for more money.
Congratulations, you figured it out.
Well, it's not QUITE as simple as that. Sometimes the stupid "someone else" is a company which wants to acquire a smaller fish. Typically, the stock will be bought out for a nice premium, or converted to parent shares at a desirable rate.
Yes, you really are just gambling that one of those two things will happen.
Now, if you hold a LOT of shares, you hold voting power over the board of directors and certain other corporate decisions. This only really comes into play for big holders though.
If moon is doing the tide then how come there's high tide 2 times a month.
I assume you meant two times per day? That's because the moon raises tidal bulges on both the near and far sides of the earth. Thus, as the earth turns, you pass under these bulges roughly twice per day. That's essentially the exact definition of a "tidal effect."
It's not like the moon pulls water away on the near side and does nothing on the far side. The tidal effect is caused by the differential of the moon's gravity across the earth's volume. This has a "squashing" effect.
Seriously? You think termites could dam the Colorado River? While it's interesting to note that termites build structures that, were the termites human sized, would be incredibly large... they are not our size.
Some small animals can move at incredible speeds relative to their own size, jump incredible distances, etc... Your point?
And they will know it's encrypted how, exactly? Yes, your typical encrypted data stream looks like random bits, but so does well compressed data.
So either you have to block all data that looks compressed or encrypted, which is a nice way to fuck yourself as a bandwidth provider since people will stop compressing shit to get past your filter, or you have to actually attempt to decompress and look inside any high-entropy data stream. How many reasonably well-deployed compression methods are there? Well, I'd guess about a hundred, if you include various audio and video codecs. So you need to run a number of decompression attempts just to distinguish compressed data from encrypted data. And you really have to DECOMPRESS IT, not just scan for magic numbers or certain headers, because hell, I'll just throw those on there for good measure to confuse you.
Okay, so now that you've established that the data is a compressed stream, you need to look inside the decompressed data to see if that itself looks like its encrypted. Sure, it's boneheaded to compress encrypted data, since it's already such high entropy, but how can you know? Especially when there are people like me trying to get around your filter? You can't, unless you try the whole process again recursively. Obviously, at some point you'll give up. Say you set the bar at two levels of nesting -- at that point it's just too expensive to keep analyzing. Well, that's going to have a shitload of false positives, because people do stupid shit like zip up a video file, which doesn't really gain you that much but is certainly widely done, and would trigger your "give up" signal -- at that point, do you fail open or fail closed? Do you reject a huge amount of traffic that's not encrypted, pissing everyone off and rendering your own service unusable and therefore worthless -- or do you throw your hands up and let the data stream through?
Yeah, sure. They'll just "throttle any encrypted traffic." Good luck with that.
Many animals modify their environment, but pretty much only Homo sapiens is capable of modifying the environment in ways that are impossible to achieve through the use of our natural physiology. The Hoover Dam could not have been built by hand. We exceed the animals by the capacity to create tools, and to use those tools to create even more complex tools, and on and on.
There are animals, primates mostly, that have been observed using inanimate objects as tools. That's true tool use, but they do not make improvements to their tools, nor do they fabricate tools using other tools. Only we can do that.
What type of air pollution are you concerned about? Particulate? NOx? CO2?
You would think that given the gravity of their findings, the seriousness they attribute to the situation, the huge nature of the changes they propose, the affect the actions will have on everyone, the potential devastation to the world economy, etc. etc., that they would have bothered to fucking hire a few professional statisticians . Shoddy and careless is what this is.
Par for the course is what it is. Most scientists believe they know statistics, whether they actually do or not. And anyway, you're blowing this out of proportion -- the finding was "correct conclusions, but suboptimal methods." The conclusion isn't "totally fucking wrong and misleading."
Even if you wanted to hire a "professional statistician," there aren't really a lot of people with that exact title or training.
For another thing, there's this saying out there, perhaps you've heard it: "There are lies, damn lies, and statistics." Hiring a professional stats wrangler could actually make you look even worse -- "Hey, these guys are cooking the books, and the evidence is that they've hired a professional liar (i.e. a statistician)"
Newsgroups can be moderated and have been moderated for ages and if you really want to you can just as easily put up a usenet server that requires the users to register an account.
Quite a few people, myself included, find moderated groups offensive. It has a ring of "Papers, please" to it. Yeah, there are a lot of spammers out there. But assuming I'm a spammer until you can prove otherwise? Fuck you.
I miss being able to just read my subscriptions, along with using a scorefile/killfile. Now I have to create accounts on dozens of web pages and monitor them all separately, without being able to rank based on what I'm interested in. Each web page has its own formats and options. Yes, there are rss feeds, but that doesn't help much if you are an active poster in the community. We've gone backwards.
You call it going backwards, I call it an opportunity for somebody to step in and provide a software solution. We have these computer thingies and programming languages. With these tools I figure we can do something to your liking.
The question is, did those ancient Roman and Indian physicians actually know about germs, or were they just making a lucky guess?
But you can apply such an argument to anybody who doesn't have all the facts. Did Robert Boyle know, for certain, that gases were composed of minute particles, the kinetics of which could be used to derive his "Boyle's Law?" He did not. In the same sense as you are now implying, he made a "lucky guess." A guess which turned out to be correct, and his name has survived in history even though, in modern terms, he didn't know what the fuck he was talking about.
If somebody posits that minute organisms are the ultimate cause of disease, then I give that person props. I really don't give a shit that he cannot prove whether he's right. That fact is, he IS right. You attitude smacks of the bitterness of a person who has perpetually sought success but never achieved it. Whatever.
You can't buy the code.
Yes you can. OSS code is owned by somebody. If it was not, how could the license requirements be enforced? You buy the code, you just realize that you can't stop anybody else in the world from using it -- that's the whole point. You do own it though.
Until Obama's first year, where he outspent Bush's 8 years in just his first.
That doesn't mean *over* spending. Pretty much every economist on the planet was in agreement that if the bailout didn't happen the global economy would COLLAPSE. I don't think you would have seen the alternative.
The number is meaningless, because "duplication" is arbitrary. Where do you draw the line? If duplication means "copying data from one place to another" then data is duplicated every time function parameters are pushed on the stack, every time memcpy() is called, every time something is loaded from disk into RAM. I could write a simple loop that copies a 32-bit quantity from EAX to EBX three billion times per second. If you include all that shit going on, I bet their number would be higher by a factor of a thousand or more. What the fuck is the use of this metric? How do you justify the arbitrary stopping point?
Moore's is not a law, but an observation!
A law is an observation. Something we have observed to always be true. The "law of gravity" for instance. Whenever you observe anything, you see that gravity is there. Thus, it's a law. "Law" doesn't mean, "We've proven it," it means "We know of no counterexample."
Now, as soon as the observation is found to be untrue -- in this case, 18 months pass without a doubling of single-die transistor count -- then the law falls. Until then, "law" is a perfectly good name for it.
Valid point, but if your goal is to ruthlessly acquire wealth, it still makes more sense to go into business, not politics. At least in business we are a bit more socially tolerant of such behavior. My point is that for politicians, payoffs and underhanded dealings are stroking the power trigger, not the money trigger.
I agree that Pleurotus is a most precious species of fungus.
The problem is that the saprophytic fungi species which do so well at consuming raw hydrocarbon cannot survive in sea water. Oyster mushrooms will consume everything from pulp waste to coffee grounds to dead wood to diesel fuel but they need a friendly environment to do it in. Stamets didn't discover this, he was just the first person to test it on a large scale and publicize the results.
Both parties are largely filled with corrupt politicians who want to line their pockets
People always say this. So, where's the list of all Congressmen with net worth of more than a billion dollars? Ten million dollars? If you want to get rich, you go into finance or business, not politics. Even a politician can understand that.
Money and power are interrelated, but let's not get confused here. Politicians are primarily interested in power. Business tycoons are primarily interested in money. Who in their right mind comes up with "Become a US Senator" as a serious strategy for becoming enormously wealthy?