Bad analogy. What adacemics have to do is akin to writing more lines of code that are reviewed by 4 or 5 other distinguished academics. There are a few bad journals out there, but most venues work very hard to get good reviewers and accept the best papers. You can write all you want, but only the good papers get published. (There's a certain amount of randomness about which borderline papers get accepted, but the exceptional ones are not overlooked.) And not all venues count the same either.
It depends. Just because the university has required them to increase their research output doesn't mean the peer-reviewed venues are going to lower their standards. They won't get too far by just doing MORE research. They have to do HIGHER QUALITY research, so that they are able to publish in higher-tier venues. One of those counts a lot more than 10 publications in low-ranked journals that no one reads or cites.
I'm a nobody grad student, yet I have papers in MICRO and HPCA. Mind you, I have a good advisor, but he's new with little funding, and the paper topics were my ideas. If I can do it, then faculty with more experience can do it better.
I'm not sure what transistor geometry Clairmont is manufactured at, but for really small transistors (e.g. 32nm), process variation is a serious problem, making it hard to scale voltage down that low. The results are unpredictable performance from die-to-die and within die and major reliability problems. Static RAMs are hit the hardest, because they use the smallest transistors. "http://www.cse.ohio-state.edu/~millerti/parichute-camera.pdf" is an example of a paper that explores the consequences of ultra-low voltage SRAMs and tries to solve it with forward error correction.
Since I'm not counterfeiting money, is this something that should be of concern to me? It sounds like there's some invisible info printed there. Say I printed something and then handed that to a nefarous person. Would they be able to look under a microscope and find easily decodable information that would allow them to steal my identity? Break into my house?
I looked that up, and I'm not sure if it means what you think it means. To quote "http://www.chacha.com/question/what-is-a-bull-simple-in-the-grapes-of-wrath", "Bull simple,in the movie "Grapes of Wrath" means that too many cops have been pushing him around."
There's been tons of research on alternative technologies, including phase-change memory (http://en.wikipedia.org/wiki/Phase-change_memory) and magnetic tunneling junction (http://drl.ee.ucla.edu/index.php?page=research&function=sttram) memory. Obviously commercializing them is expensive, but some progress has already been made there. I'm sure other competing technologies will be developed in that time as well.
If you were doing profiling, you'd never pick me. I'm a white guy who doesn't get enough sun, and I sound like I'm from central Ohio. Yet absolutely every time I depart through the Columbus airport, I'm asked to be scanned. I don't get it. No other airport has done this to me. Only Columbus. Also, my wife gets asked every time too. Same place. Do they have some kind of quota for locals so they can balance out the profiling?
Anyhow, I always politely ask to opt out, "I'm sorry; I would like to opt out, please," and when I'm being patted down, I chit-chat with the TSA agent. Since I'm not especially body-conscious, the pat-down doesn't bother me. I'm also pretty good at behaving in a compliant manner when I want to. There's a trick to acting slightly confused but quick to follow explicit orders that makes authority figures feel they have control, and that mollifies them. The pat-down always goes smoothly and efficiently. I suspect when they get belligerent patrons, they drag their feet.
We already know that keytones have an ameliorating affect on Alzheimer's. This is because brain cells seem to have trouble absorbing gluclose, but they can get energy from keytones, which they can still absorb. So this skin cancer drug, may help brain cells absorb glucose, or it helps convert something to keytones, or it itself is made of keytones and is nourishing the brain.
As far as I can tell, Apple is NOT taking a chunk out of the carrier's monthly fees. But to sell the more expensive phones, they have to give bigger contractual discounts. With the fees they're charging, it might take six months for the monthy fees to accumulate to be equal to the subsity. It may take the entire contract period for the profits from the monthy charges to be equal to the subsidy. That may be a long time, but it's only a quarter of the contract period, and meanwhile, they're raking in better volume profits because the iPhone sells so damn well. Of course, more phones means more network congestion, but the congestion is just a result of the carriers cheaping out on infrastructure. It amazes me how whiny the carriers are given the huge profits they report every quarter.
You're right that NOT ALL women "smell nicer." I was referring to a statistical norm and my personal experiences. I believe what I was doing is called Synecdoche.
The only way in which ALL black people are different from whites is that their skin tone is darker. And when suntans and albinos get into the mix, even that isn't necessarily true. All other characteristics are completely individual. However, there are certain common trends. For instance, africans tend to have different muscle tone. And sociologists have pointed out that the average black person has better innate social ability than the average white person. I'm sure you can dig up a few other trends that are statistically significant, although we would be foolish to judge individuals on those bases. You ignore the differences when it comes to judging someone as a person, their skills, etc., but those trends are all very interesting anthropologically.
So there's nothing wrong with what you say. There are always exceptions to the statistical norm. In fact, what I was implying is that we should DEFEAT the statistical norm and learn from each other's cultural conditioning. The women I have worked with have brought aspects to the workplace that have made it more pleasant. "Pleasant" is my value judgement. Statistically, what women will bring to the workplace is merely "different." Men can do those same things if they learn to, and I have known a few who did. Not enough, though.
But to sum up what I was really trying to get across is that by embracing diversity, we generally improve our ability to solve problems, because we can leverage more perspectives.
I'll have to do better homework on Islam. Your anecdote contradicts other anecdotes, and I need to wipe away all the anecdotes and read some real history.
But is it a stereotype to notice that women are culturally conditioned to pay better attention to their appearance? It's a fact. Women are TRAINED to do that. This is evident in the fact that more women wear makeup than do men. I'm not judging women for putting effort into their appearance. I'm saying that that's how EVERYONE should behave. Men CAN dress better, bathe better, and put two seconds of thought into making a communal office environment more visually appealing. But more often than not, they don't.
To put it another way, women and men are conditioned differently. This is a fact. I'm sure there are families out there that work very hard to avoid gender-based differential training, but it's almost impossible to avoid the instant you turn on the TV. And what I AM SAYING is that men should open their eyes and learn a few things from the women they work with. I'm sure it goes both ways. There's nothing physioligical that prevents women from taking more positions of leadership in the world. Most of the barriers are cultural, and perhaps a few more women would rise a little higher if they could get past their training to be "ladylike", which usually just means passive and submissive.
Another thing that makes men and women different is that women are the ones who get pregnant. It's because of the NEED for things like maternity leave that unfairly bar married women from certain positions, because employers don't want workflow disrupted. Unmarried women rise higher in the workforce because they're at lower risk of getting pregnant. Married men rise higher because they're judged as being more stable. That's bullshit. Both men AND women should get materity/paternity leave, and leeway should be made to allow professional mothers to have children and still do their jobs. Most pregnant women work right up to their due dates anyhow. The fathers should do the same and then they both get time off from work for several weeks. This all comes from this bizarre idea that women are somehow more responsible for babies than men are. Fathers need to take just as much responsibility for their kids.
A few people have pointed out that we should not emphasize Ms. Bergeron's gender. In an ideal world, this would be true. But in THIS world, it's a big deal, simply because women in high IT positions are so rare. Having worked with some very talented female engineers, I can say that as skilled engineers, they brought more to the work environment than just engineering. Women are not inferior, but there are differences, some biological, some sociological, and those differences have huge benefits when it comes to working in a creative environment. Since they don't think exactly like men, they solve problems differently, and this means we as teams can solve more problems more effectively. So I say congratulations to the Fedora Project for bringing greater practical diversity to their team.
Many of our scientific advances came from Arabs. Then Islam came along, and they fell behind. This is not just because they subjugated half their workforce but also because they suppressed half their variety in ways of thinking.
Women also wear better looking clothes, smell nicer, and have a penchant for adding things to the physical environment that make the workplace more pleasant. How can you beat that?
(Yes, I will favor a good-smelling slightly less talented applicant over one who is very skilled but smells like ass. Tough shit for those of you with bad diets and poor hygiene habits who can't manage to take an interview seriously enough to BATHE before they show up. People have to work next to you.)
What Facebook does is all perfectly above-board, because Facebook's ownership of everything you put there is right in the Terms of Service that you agreed to when you signed up. While some such agreements have been overturned in court, most of them are legally binding. As long as Facebook stays within the bounds of their side of the contract, then there's nothing you can do legally about it.
Where does Facebook make their money? Let's see
- Matching you against advertizers so that the ads you see are more likely to be clicked. Ad clicks are revenue. - Datamining "anonymized" information about all their users to sell to companies that want statistics about people. - Kickbacks from leading users to paid services.
Facebook started out as a social networking site. That is what Zuck had in mind. But when he had to turn in into a business that made money, the obvious thing to do was to use the information people put there. Facebook's three primary engineering trusts are (a) improving their site so as to keep you addicted to their service, (b) improving their site so as to maximize the value of the datamining output, and (c) minimizing the cost of providing those services.
In fact, this is little different from what Google does. They keep cookies about what your searches have been and use that to match you against advertizers. If you combine Google searches with gmail, Google+, and Google Docs, you have the same amount of information, a vast treasure trove from which to learn general things about people and to profile individuals in order to match the against ads.
This is the nature of all free web services. But even paid services like Netflix, Newegg, and Amazon mine your searches and purchases and compare you with other people in order to do a better job of recommending things you'll like. Netflix had a million-dollar prize dedicated to this. Amazon always recommends products bought by others looking at what you're looking at. And I regularly get emails from Amazon telling me about products I might like, based on what I've bought in the past. Is this an invasion of privacy? It's hard to say, because it's not clear where the ethical line is between helpful recommendation systems and scouring every detail of your life.
None of these services sell your personal details in an identifiable way. Besides the fact that they'd get into all sorts of consumer protection trouble, Amazon does not want Barnes & Noble to know your purchase history! Same with regard to Netflix and Blockbuster. On Facebook, every tiny piece of info that appears on your page is something you or one of your friends chose consciously to put there. Mind you, that can go wrong, when someone puts up a photo of you that they didn't have permission to put up, but be careful who you're friends with, eh? But everything else is really under your control. It gets really creepy when you get an ad popping up related to something you mentioned in a chat session. I think that's going a bit far. But again, you chose to use Facebook (rather than, say, a telephone or jabber) to communicate that info, and you already know that Facebook owns it. Creepy but completely above board and legal.
Facebook is like the way the devil is described in some religions. He doesn't force you to sin. He simply provides you with many irresistible temptations. Facebook plays on human psychology and this weird combination we have of being introverted (many of us) and wanting to connect with other people. Facebook is designed by experts at addicting people and making them WANT to expose their deepest secrets. The temptation is so great that we consciously choose to walk naked through the streets, knowing full well that many nefarious eyes have really good binoculars. Going well beyond creepiness, Facebook's unfathomable privacy settings make it ripe for identity theft, even newborns who grow up to find out that they have credit card dept of mysterious origin.
And yet Facebook, much like the devil, always follow
I suppose there are some chances that some alien life would superficially resemble life on earth. But not likely. Any significant similarity would be more easily explained by a common ancestor. If there is no common ancestor, all bets are off, and since the mutation and selection pressures on another world would be completely different, the resulting life forms would be completely different. Sure, it's more likely than not that they'd be made of carbon, oxygen, hydrogen, etc., and it's likely (because it's such a huge advantage) that they'd use something LIKE DNA to encode information, but it's not necessarily the case that they would use the same chemicals.
This idea of moving compute into the RAM has been around a long time. Papers have proposed everything from adding simple ALUs to the DRAMs to fully functional microprocessors. Most assume that these are "accelerators" for common vector operations and such, while the heavy lifting is done by beefier cores, but the idea if doing all the compute embedded in a DRAM has been proposed and evaluated before.
One thing we've learned in the past few decades is that modern processors are limited by memory latency and bandwidth. A Sun engineer (talking about Rock) pointed out that a modern out-of-order processor performs a race between last-level cache misses. When you have to go out to DRAM, the CPU instruction window fills up with as much dependent work as possible, before it completely stalls because everything is dependent on that one miss. When that data finally arrives, the CPU blasts through that work really fact, and then soon stalls out again on another miss. OOO processors resolve this (somewhat) by the instruction window, while Rock solved it by speculative execution. One of the reasons for Sandy Bridge's excellent performance is the very large instruction window that can absorb more of the LLC miss stall time.
And so, although these processors have other advantages, OOO processors dedicate a huge amount of logic just to dealing with the cache miss latency. If there were no such latency, then they could get the same performance with a hell of a lot less hardware. Although I haven't seen the figures, my suspicion is that for general computation, TOMI will blow the doors off of whatever else we've got in both performance AND energy efficiency. Only when you have a specialized compute kernel whose working data fits in the cache can you comparatively benefit from something like Sandy Bridge. (I realize that's an overly strong statement, because lots of general purpose workloads have good locality, but nevertheless main memory is a major bottleneck for most workloads.)
I'm not strictly interested whether or not they can find the answers in 5 minutes. I've asked questions even I didn't know the answers to, and for even the simple ones, smart people can draw a blank because they're tired and nervous from interviewing. I'm looking to determine if the candidate can reason at all and has good background knowledge. If someone tries to BS their way through, I don't like them, because this suggests that they don't care that much about reasoning and they aren't willing to be honest about their ignorance. If someone gets mad at themselves because they can't think clearly enough to work it out, I like them, because it's a hint to me that if they had the time to work on it with less pressure, they'd figure it out. (And there's a huge difference between interview pressure and deadline pressure.) I've had a few who were determined to get it, whom I stopped early because I learned what I needed to know, and I could see they were on the right track anyhow. (I explained that being able to answer the question right then and there was not the factor that was going to affect my hiring recommendation.)
Very often, I'd get a vibe about someone right when I first meet them. I would go through the interview process just to verify objectively that my intuition was right. And I did generally turn out to be right. Mind you, I also listened carefully to the the impressions of other interviewers, and sometimes reinterpreted what I had observed. I could spot someone with high intelligence and/or really good engineering skills, but on a few occasions, characteristics of maturity, persistance, and initiative got by me.
For the supercapacitors to do much, they'd have to be on-chip, which is impossible. Expensive voltage regulators already have big capacitors, which really keep the EXTERNAL voltage very smooth. But there's a fair amount of inductive decoupling between the pins and the silicon that makes a lot of internal voltage fluctuations invisible on the outside.
Some FPGA programming tools require a USB driver to be installed for programming devices, and when Vista was out, there were no drivers for anything other than XP. It's been a while, but I'm afrait to break anything. Also I only have Windows in VMs, so I wouldn't benefit from any new features. I mostly use Windows for certain limited kinds of development work.
The reason you can overclock without raising the voltage is that there is a voltage "guard band," which is like a safety margin. Some of my research has been about finding ways of reducing that guard band, because it's wasted energy. But that guard band is there for a good reason. Typically the critical paths in the chip (those with the longest propagation delay, which limit the safe clock speed) are a bit faster than the clock period. But that's only true whe the voltage is stable. If the voltage droops, then the propagation delay of those paths will increase, possibly too much, and you get incorrect computation. Voltage droops occur when circuits suddenly start switching a lot, demanding more current, or in other words, the effective impedence of the circuit drops, and by V=IR, for the current being supplied by the voltage regulator at that instant, the voltage inside the chip will drop. The regulator cannot respond instantly, so a guard band is provided so that the maximum droop never brings the instantaneous voltage below a certain margin. If you overclock without raising voltage, then your CPU will work fine most of the time, but certain workloads will cause wide swings in current demand, and if you execute one of those, you may crash your system.
This is why memory tests are worthless for stability testing, because due to cache miss latency, the current demand is relatively low and stable. Prime number generators are also not so great, because their current demand is relatively high and stable. I know that some of the SPEC and PARSEC benchmarks have some wild behavior, like FFT, for instance, or anything that has a lot of barrier synchronization. For the regular user, what's likely going to happen is that you'll get random such events where variation in cache hits and vector computation phases will cause significant spikes in current, and your game will crash.
We're going around in circles. Mathematics being unpatentable is not self-evident. It is an aspect of patent law, although I can see that I may have conflated the ethical and legal arguments a bit in my last comment. (Which was posted to slashdot, which has a tendency to suck the thought out of anything anyone says here.)
This is somewhat arbitrary, but it's a matter of tradition and a sense that mathematics is a fundamental aspect of reality. The standard model of quantum physics is also not the sort of thing you could (or should) patent either.
The difference between mathematics versus the sciences and especially the engineering disciplines is that mathematics is the only one that is actually about "truth" (given certain axioms). (Sciences are about models that fit the facts, but they may not be always exactly "true", and engineering is about inventing entirely new things whose "truth value" may be philosophically irrelevant.) In math and science, you do not invent. You DISCOVER. Mathematics doesn't solve a problem. It is a tool used to solve problems.
In some ways, this distinction is purely philosophical. It is this way because people believe it is so, not because it necessarily is by any objective measure. And we can be sure that plenty of math has been patented when it shouldn't, because the patent examiner didn't see through the very thin application.
And as I did say, effort does not imply invention. However, good, novel invention does require effort. Don't get the implication backwards.
I'm not sure if I fully understand your point regarding salary. There is a legal term "consideration" that is required for a contract to be valid, where each party gives something to the other in the exchange. Royalties and work compensation aren't quite the same thing, but just to make the analogy, I believe that authors are entitled to ask for "consideration" in exchange for their work. That consideration may be in the form of the GPL, or it may be royalties, or it may just be a salary. If they're just paid a salary, then this is something they agreed to when they got hired.
But you may be using "salary" metaphorically, in that people who hold copyrights seem to often get perpetual royalties on something they did only once. I thought I addressed this, but it may have been in another comment. I think it's total crap that people can do this. They should be able to profit FOR A REASONABLE TIME, and then their stuff goes into the public domain.
A stab in the air: Patents should be 10 years for engineering, 5 for something that is purely software. Copyrights, perhaps a bit longer. Maybe 10 for software and 20 for more traditional artistic things and literature. Of course, if this were the case, half the GPL stuff we use would be in the public domain now.
I wasn't talking about pure mathematics. If you develop a new theorem in mathematics, it is not patentable. If you use mathematical theorems to solve a difficult engineering problem, then it is. I wasn't talking about inventing mathematics but USING math to do other things.
You surely have SOME mods who are on crack, but there isn't a great conspiracy among them to mod down any particular users or topics. That being said, people who post on slashdot (myself no exception) are typically a bit loco.
In general, interesting stuff is modded up, and stupid stuff is modded down. It's only the mediocre stuff that might get modded the wrong way. Sadly, those who are mediocre are typically unaware that they are mediocre. (Actually, to be fair, people at all points on the spectrum are commonly unaware of their exact competence level.)
Slashdot Mirror
Bad analogy. What adacemics have to do is akin to writing more lines of code that are reviewed by 4 or 5 other distinguished academics. There are a few bad journals out there, but most venues work very hard to get good reviewers and accept the best papers. You can write all you want, but only the good papers get published. (There's a certain amount of randomness about which borderline papers get accepted, but the exceptional ones are not overlooked.) And not all venues count the same either.
It depends. Just because the university has required them to increase their research output doesn't mean the peer-reviewed venues are going to lower their standards. They won't get too far by just doing MORE research. They have to do HIGHER QUALITY research, so that they are able to publish in higher-tier venues. One of those counts a lot more than 10 publications in low-ranked journals that no one reads or cites.
I'm a nobody grad student, yet I have papers in MICRO and HPCA. Mind you, I have a good advisor, but he's new with little funding, and the paper topics were my ideas. If I can do it, then faculty with more experience can do it better.
I'm not sure what transistor geometry Clairmont is manufactured at, but for really small transistors (e.g. 32nm), process variation is a serious problem, making it hard to scale voltage down that low. The results are unpredictable performance from die-to-die and within die and major reliability problems. Static RAMs are hit the hardest, because they use the smallest transistors. "http://www.cse.ohio-state.edu/~millerti/parichute-camera.pdf" is an example of a paper that explores the consequences of ultra-low voltage SRAMs and tries to solve it with forward error correction.
Since I'm not counterfeiting money, is this something that should be of concern to me? It sounds like there's some invisible info printed there. Say I printed something and then handed that to a nefarous person. Would they be able to look under a microscope and find easily decodable information that would allow them to steal my identity? Break into my house?
I looked that up, and I'm not sure if it means what you think it means. To quote "http://www.chacha.com/question/what-is-a-bull-simple-in-the-grapes-of-wrath", "Bull simple,in the movie "Grapes of Wrath" means that too many cops have been pushing him around."
There's been tons of research on alternative technologies, including phase-change memory (http://en.wikipedia.org/wiki/Phase-change_memory) and magnetic tunneling junction (http://drl.ee.ucla.edu/index.php?page=research&function=sttram) memory. Obviously commercializing them is expensive, but some progress has already been made there. I'm sure other competing technologies will be developed in that time as well.
If you were doing profiling, you'd never pick me. I'm a white guy who doesn't get enough sun, and I sound like I'm from central Ohio. Yet absolutely every time I depart through the Columbus airport, I'm asked to be scanned. I don't get it. No other airport has done this to me. Only Columbus. Also, my wife gets asked every time too. Same place. Do they have some kind of quota for locals so they can balance out the profiling?
Anyhow, I always politely ask to opt out, "I'm sorry; I would like to opt out, please," and when I'm being patted down, I chit-chat with the TSA agent. Since I'm not especially body-conscious, the pat-down doesn't bother me. I'm also pretty good at behaving in a compliant manner when I want to. There's a trick to acting slightly confused but quick to follow explicit orders that makes authority figures feel they have control, and that mollifies them. The pat-down always goes smoothly and efficiently. I suspect when they get belligerent patrons, they drag their feet.
Don't you love it how Bible thumpers like to spout Bible verses at you, but never have the courtesy to quote them for you?
We already know that keytones have an ameliorating affect on Alzheimer's. This is because brain cells seem to have trouble absorbing gluclose, but they can get energy from keytones, which they can still absorb. So this skin cancer drug, may help brain cells absorb glucose, or it helps convert something to keytones, or it itself is made of keytones and is nourishing the brain.
Eat your coconut oil!
As far as I can tell, Apple is NOT taking a chunk out of the carrier's monthly fees. But to sell the more expensive phones, they have to give bigger contractual discounts. With the fees they're charging, it might take six months for the monthy fees to accumulate to be equal to the subsity. It may take the entire contract period for the profits from the monthy charges to be equal to the subsidy. That may be a long time, but it's only a quarter of the contract period, and meanwhile, they're raking in better volume profits because the iPhone sells so damn well. Of course, more phones means more network congestion, but the congestion is just a result of the carriers cheaping out on infrastructure. It amazes me how whiny the carriers are given the huge profits they report every quarter.
You're right that NOT ALL women "smell nicer." I was referring to a statistical norm and my personal experiences. I believe what I was doing is called Synecdoche.
The only way in which ALL black people are different from whites is that their skin tone is darker. And when suntans and albinos get into the mix, even that isn't necessarily true. All other characteristics are completely individual. However, there are certain common trends. For instance, africans tend to have different muscle tone. And sociologists have pointed out that the average black person has better innate social ability than the average white person. I'm sure you can dig up a few other trends that are statistically significant, although we would be foolish to judge individuals on those bases. You ignore the differences when it comes to judging someone as a person, their skills, etc., but those trends are all very interesting anthropologically.
So there's nothing wrong with what you say. There are always exceptions to the statistical norm. In fact, what I was implying is that we should DEFEAT the statistical norm and learn from each other's cultural conditioning. The women I have worked with have brought aspects to the workplace that have made it more pleasant. "Pleasant" is my value judgement. Statistically, what women will bring to the workplace is merely "different." Men can do those same things if they learn to, and I have known a few who did. Not enough, though.
But to sum up what I was really trying to get across is that by embracing diversity, we generally improve our ability to solve problems, because we can leverage more perspectives.
I'll have to do better homework on Islam. Your anecdote contradicts other anecdotes, and I need to wipe away all the anecdotes and read some real history.
But is it a stereotype to notice that women are culturally conditioned to pay better attention to their appearance? It's a fact. Women are TRAINED to do that. This is evident in the fact that more women wear makeup than do men. I'm not judging women for putting effort into their appearance. I'm saying that that's how EVERYONE should behave. Men CAN dress better, bathe better, and put two seconds of thought into making a communal office environment more visually appealing. But more often than not, they don't.
To put it another way, women and men are conditioned differently. This is a fact. I'm sure there are families out there that work very hard to avoid gender-based differential training, but it's almost impossible to avoid the instant you turn on the TV. And what I AM SAYING is that men should open their eyes and learn a few things from the women they work with. I'm sure it goes both ways. There's nothing physioligical that prevents women from taking more positions of leadership in the world. Most of the barriers are cultural, and perhaps a few more women would rise a little higher if they could get past their training to be "ladylike", which usually just means passive and submissive.
Another thing that makes men and women different is that women are the ones who get pregnant. It's because of the NEED for things like maternity leave that unfairly bar married women from certain positions, because employers don't want workflow disrupted. Unmarried women rise higher in the workforce because they're at lower risk of getting pregnant. Married men rise higher because they're judged as being more stable. That's bullshit. Both men AND women should get materity/paternity leave, and leeway should be made to allow professional mothers to have children and still do their jobs. Most pregnant women work right up to their due dates anyhow. The fathers should do the same and then they both get time off from work for several weeks. This all comes from this bizarre idea that women are somehow more responsible for babies than men are. Fathers need to take just as much responsibility for their kids.
A few people have pointed out that we should not emphasize Ms. Bergeron's gender. In an ideal world, this would be true. But in THIS world, it's a big deal, simply because women in high IT positions are so rare. Having worked with some very talented female engineers, I can say that as skilled engineers, they brought more to the work environment than just engineering. Women are not inferior, but there are differences, some biological, some sociological, and those differences have huge benefits when it comes to working in a creative environment. Since they don't think exactly like men, they solve problems differently, and this means we as teams can solve more problems more effectively. So I say congratulations to the Fedora Project for bringing greater practical diversity to their team.
Many of our scientific advances came from Arabs. Then Islam came along, and they fell behind. This is not just because they subjugated half their workforce but also because they suppressed half their variety in ways of thinking.
Women also wear better looking clothes, smell nicer, and have a penchant for adding things to the physical environment that make the workplace more pleasant. How can you beat that?
(Yes, I will favor a good-smelling slightly less talented applicant over one who is very skilled but smells like ass. Tough shit for those of you with bad diets and poor hygiene habits who can't manage to take an interview seriously enough to BATHE before they show up. People have to work next to you.)
Sounds great, but $5/year will lose them almost all of their users. Remember... everyone feels entitled to getting things for free.
What Facebook does is all perfectly above-board, because Facebook's ownership of everything you put there is right in the Terms of Service that you agreed to when you signed up. While some such agreements have been overturned in court, most of them are legally binding. As long as Facebook stays within the bounds of their side of the contract, then there's nothing you can do legally about it.
Where does Facebook make their money? Let's see
- Matching you against advertizers so that the ads you see are more likely to be clicked. Ad clicks are revenue.
- Datamining "anonymized" information about all their users to sell to companies that want statistics about people.
- Kickbacks from leading users to paid services.
Facebook started out as a social networking site. That is what Zuck had in mind. But when he had to turn in into a business that made money, the obvious thing to do was to use the information people put there. Facebook's three primary engineering trusts are (a) improving their site so as to keep you addicted to their service, (b) improving their site so as to maximize the value of the datamining output, and (c) minimizing the cost of providing those services.
In fact, this is little different from what Google does. They keep cookies about what your searches have been and use that to match you against advertizers. If you combine Google searches with gmail, Google+, and Google Docs, you have the same amount of information, a vast treasure trove from which to learn general things about people and to profile individuals in order to match the against ads.
This is the nature of all free web services. But even paid services like Netflix, Newegg, and Amazon mine your searches and purchases and compare you with other people in order to do a better job of recommending things you'll like. Netflix had a million-dollar prize dedicated to this. Amazon always recommends products bought by others looking at what you're looking at. And I regularly get emails from Amazon telling me about products I might like, based on what I've bought in the past. Is this an invasion of privacy? It's hard to say, because it's not clear where the ethical line is between helpful recommendation systems and scouring every detail of your life.
None of these services sell your personal details in an identifiable way. Besides the fact that they'd get into all sorts of consumer protection trouble, Amazon does not want Barnes & Noble to know your purchase history! Same with regard to Netflix and Blockbuster. On Facebook, every tiny piece of info that appears on your page is something you or one of your friends chose consciously to put there. Mind you, that can go wrong, when someone puts up a photo of you that they didn't have permission to put up, but be careful who you're friends with, eh? But everything else is really under your control. It gets really creepy when you get an ad popping up related to something you mentioned in a chat session. I think that's going a bit far. But again, you chose to use Facebook (rather than, say, a telephone or jabber) to communicate that info, and you already know that Facebook owns it. Creepy but completely above board and legal.
Facebook is like the way the devil is described in some religions. He doesn't force you to sin. He simply provides you with many irresistible temptations. Facebook plays on human psychology and this weird combination we have of being introverted (many of us) and wanting to connect with other people. Facebook is designed by experts at addicting people and making them WANT to expose their deepest secrets. The temptation is so great that we consciously choose to walk naked through the streets, knowing full well that many nefarious eyes have really good binoculars. Going well beyond creepiness, Facebook's unfathomable privacy settings make it ripe for identity theft, even newborns who grow up to find out that they have credit card dept of mysterious origin.
And yet Facebook, much like the devil, always follow
I suppose there are some chances that some alien life would superficially resemble life on earth. But not likely. Any significant similarity would be more easily explained by a common ancestor. If there is no common ancestor, all bets are off, and since the mutation and selection pressures on another world would be completely different, the resulting life forms would be completely different. Sure, it's more likely than not that they'd be made of carbon, oxygen, hydrogen, etc., and it's likely (because it's such a huge advantage) that they'd use something LIKE DNA to encode information, but it's not necessarily the case that they would use the same chemicals.
My research area is computer architecture.
This idea of moving compute into the RAM has been around a long time. Papers have proposed everything from adding simple ALUs to the DRAMs to fully functional microprocessors. Most assume that these are "accelerators" for common vector operations and such, while the heavy lifting is done by beefier cores, but the idea if doing all the compute embedded in a DRAM has been proposed and evaluated before.
One thing we've learned in the past few decades is that modern processors are limited by memory latency and bandwidth. A Sun engineer (talking about Rock) pointed out that a modern out-of-order processor performs a race between last-level cache misses. When you have to go out to DRAM, the CPU instruction window fills up with as much dependent work as possible, before it completely stalls because everything is dependent on that one miss. When that data finally arrives, the CPU blasts through that work really fact, and then soon stalls out again on another miss. OOO processors resolve this (somewhat) by the instruction window, while Rock solved it by speculative execution. One of the reasons for Sandy Bridge's excellent performance is the very large instruction window that can absorb more of the LLC miss stall time.
And so, although these processors have other advantages, OOO processors dedicate a huge amount of logic just to dealing with the cache miss latency. If there were no such latency, then they could get the same performance with a hell of a lot less hardware. Although I haven't seen the figures, my suspicion is that for general computation, TOMI will blow the doors off of whatever else we've got in both performance AND energy efficiency. Only when you have a specialized compute kernel whose working data fits in the cache can you comparatively benefit from something like Sandy Bridge. (I realize that's an overly strong statement, because lots of general purpose workloads have good locality, but nevertheless main memory is a major bottleneck for most workloads.)
I'm not strictly interested whether or not they can find the answers in 5 minutes. I've asked questions even I didn't know the answers to, and for even the simple ones, smart people can draw a blank because they're tired and nervous from interviewing. I'm looking to determine if the candidate can reason at all and has good background knowledge. If someone tries to BS their way through, I don't like them, because this suggests that they don't care that much about reasoning and they aren't willing to be honest about their ignorance. If someone gets mad at themselves because they can't think clearly enough to work it out, I like them, because it's a hint to me that if they had the time to work on it with less pressure, they'd figure it out. (And there's a huge difference between interview pressure and deadline pressure.) I've had a few who were determined to get it, whom I stopped early because I learned what I needed to know, and I could see they were on the right track anyhow. (I explained that being able to answer the question right then and there was not the factor that was going to affect my hiring recommendation.)
Very often, I'd get a vibe about someone right when I first meet them. I would go through the interview process just to verify objectively that my intuition was right. And I did generally turn out to be right. Mind you, I also listened carefully to the the impressions of other interviewers, and sometimes reinterpreted what I had observed. I could spot someone with high intelligence and/or really good engineering skills, but on a few occasions, characteristics of maturity, persistance, and initiative got by me.
For the supercapacitors to do much, they'd have to be on-chip, which is impossible. Expensive voltage regulators already have big capacitors, which really keep the EXTERNAL voltage very smooth. But there's a fair amount of inductive decoupling between the pins and the silicon that makes a lot of internal voltage fluctuations invisible on the outside.
Some FPGA programming tools require a USB driver to be installed for programming devices, and when Vista was out, there were no drivers for anything other than XP. It's been a while, but I'm afrait to break anything. Also I only have Windows in VMs, so I wouldn't benefit from any new features. I mostly use Windows for certain limited kinds of development work.
The reason you can overclock without raising the voltage is that there is a voltage "guard band," which is like a safety margin. Some of my research has been about finding ways of reducing that guard band, because it's wasted energy. But that guard band is there for a good reason. Typically the critical paths in the chip (those with the longest propagation delay, which limit the safe clock speed) are a bit faster than the clock period. But that's only true whe the voltage is stable. If the voltage droops, then the propagation delay of those paths will increase, possibly too much, and you get incorrect computation. Voltage droops occur when circuits suddenly start switching a lot, demanding more current, or in other words, the effective impedence of the circuit drops, and by V=IR, for the current being supplied by the voltage regulator at that instant, the voltage inside the chip will drop. The regulator cannot respond instantly, so a guard band is provided so that the maximum droop never brings the instantaneous voltage below a certain margin. If you overclock without raising voltage, then your CPU will work fine most of the time, but certain workloads will cause wide swings in current demand, and if you execute one of those, you may crash your system.
This is why memory tests are worthless for stability testing, because due to cache miss latency, the current demand is relatively low and stable. Prime number generators are also not so great, because their current demand is relatively high and stable. I know that some of the SPEC and PARSEC benchmarks have some wild behavior, like FFT, for instance, or anything that has a lot of barrier synchronization. For the regular user, what's likely going to happen is that you'll get random such events where variation in cache hits and vector computation phases will cause significant spikes in current, and your game will crash.
We're going around in circles. Mathematics being unpatentable is not self-evident. It is an aspect of patent law, although I can see that I may have conflated the ethical and legal arguments a bit in my last comment. (Which was posted to slashdot, which has a tendency to suck the thought out of anything anyone says here.)
This is somewhat arbitrary, but it's a matter of tradition and a sense that mathematics is a fundamental aspect of reality. The standard model of quantum physics is also not the sort of thing you could (or should) patent either.
The difference between mathematics versus the sciences and especially the engineering disciplines is that mathematics is the only one that is actually about "truth" (given certain axioms). (Sciences are about models that fit the facts, but they may not be always exactly "true", and engineering is about inventing entirely new things whose "truth value" may be philosophically irrelevant.) In math and science, you do not invent. You DISCOVER. Mathematics doesn't solve a problem. It is a tool used to solve problems.
In some ways, this distinction is purely philosophical. It is this way because people believe it is so, not because it necessarily is by any objective measure. And we can be sure that plenty of math has been patented when it shouldn't, because the patent examiner didn't see through the very thin application.
And as I did say, effort does not imply invention. However, good, novel invention does require effort. Don't get the implication backwards.
I'm not sure if I fully understand your point regarding salary. There is a legal term "consideration" that is required for a contract to be valid, where each party gives something to the other in the exchange. Royalties and work compensation aren't quite the same thing, but just to make the analogy, I believe that authors are entitled to ask for "consideration" in exchange for their work. That consideration may be in the form of the GPL, or it may be royalties, or it may just be a salary. If they're just paid a salary, then this is something they agreed to when they got hired.
But you may be using "salary" metaphorically, in that people who hold copyrights seem to often get perpetual royalties on something they did only once. I thought I addressed this, but it may have been in another comment. I think it's total crap that people can do this. They should be able to profit FOR A REASONABLE TIME, and then their stuff goes into the public domain.
A stab in the air: Patents should be 10 years for engineering, 5 for something that is purely software. Copyrights, perhaps a bit longer. Maybe 10 for software and 20 for more traditional artistic things and literature. Of course, if this were the case, half the GPL stuff we use would be in the public domain now.
I wasn't talking about pure mathematics. If you develop a new theorem in mathematics, it is not patentable. If you use mathematical theorems to solve a difficult engineering problem, then it is. I wasn't talking about inventing mathematics but USING math to do other things.
You surely have SOME mods who are on crack, but there isn't a great conspiracy among them to mod down any particular users or topics. That being said, people who post on slashdot (myself no exception) are typically a bit loco.
In general, interesting stuff is modded up, and stupid stuff is modded down. It's only the mediocre stuff that might get modded the wrong way. Sadly, those who are mediocre are typically unaware that they are mediocre. (Actually, to be fair, people at all points on the spectrum are commonly unaware of their exact competence level.)