Well, hey why not? As long as the product is high quality, its not like its eroding the brand. There was an article not too long about about how Nintendo was branding everything in sight, rather than developing new franchises. I used to be a bit miffed over how they put his face everywhere, but really, how many franchises does Nintendo need? They've got a big enough cast that they can actually make a fighting game out of it with 30 characters from like 10 franchises (star fox, metroid, zelda, kirby, pokemon, mario, fire emblem, f zero, earthbound, donkey kong...). Really, some of those franchises don't do as well as the best, so its natural to help trim the bottom line with some extra income from Mario to help offset the less than stellar F Zero GX sales (still an awesome game and a better buy daily).
Yes, people will buy it because of Mario. Will it be a crappy game because of Mario? Will Mario be crappy because he's in the game? My answers are no and no. Really, whats going on is Nintendo is trying to attract games over to their system, and they're using their own new "seal of approval" to do it. Nintendo gets a popular game on their dominated console (sadly), the game makers get a dose of more sales, and cube owners get more games.
Mario, by and large, will always be judged by the flagship games that bear his name, rather than the one offs with his face on it.
Which distribution would you suggest then, if I want to build a 64bit desktop, and do things like run a scanner, a printer and my wifi card? And if I find a novel 32bit program, such as those Kento Cho shooter games written in D, can I run it?
Toms hardware consistantly favors computer hardware that only pushes above the competition by 1 percent or less. People spend an extra 40 dollars for this performance, and you're not willing to consider that people might like a FREE performance boost of a percent?
Genetic programming is a known quanitity, but if it "wont be running at run time" then its borderline useless. The whole idea is that the genetic algorithm adapts to a workload. Currently, scheduling is about fine tuning at the expense of flexibility. You tune the scheduler for a desktop setup, or for a database server, or whatever you have. Most tweaks yield a few percentage point gains in theory and maybe a single percent in practice for the given problem set, at the expense of about a ten percent penalty in the worst case scenario. Essentially, the genetic algorithm attempts to fine tune the kernel for you in ways.
Its not really that complex, but reasoning about how well it could perform is VERY difficult comparitively speaking. Furthermore, it introduces a much fuzzier notion of fine tuning. Rather than playing with variables like "swappiness" and cpu affinity, you're messing with a fitness function, where minute changes can move you from a percentage gained over a stock kernel to a percentage (or worse) loss. Certainly, I'm impressed that he's managed to make an improvement over stock with it, which puts interested kernel developers on a good first step. Nobody wants to chase technology that doesn't even show a hint of unlocking performance potential.
You kid right? Typically, Linux is treated as a lepper colony by the University administrations. They usually get a good deal on Solaris systems so that's what they use for mainframes. Certainly a few CS/IS professors may use Linux as a testing ground for some sort of reserarch, perhaps something like extensions to MOSIX. But those workstations running linux probably already have eight local vulnerabilites. Just try pulling up a shell and running uname -a on it. How out of date is your kernel?
But what do you want them to do? Kick you off every time a kernel exploit is found so they can patch it? That can't be fun.
Except that developers are even afraid of releasing anything dangerous into Sid, lest they break developer's guis or something. Hence the "experimental" distro. I've used sid for a long time and the only real problem I ever had was a bumpy transition from gnome 1.4 to the 2.x series. Even then, I still had something like twm to fall back on, and everything i needed still worked. I think that Debian really needs a push in the back to shake things up a bit. Part of the problem is that a lot of Debian is simply Done, and the Debian Developers mostly focus on scratching their own backs. For example, compare debian.org to alioth. I think a lot can be learned from the success of gentoo, but Debian leadership typically wants to handle the status quo. Basically, Debian needs better marketing in general.
Valve's new TF game was supposed to be on shelves in 2000 or so. Remember, Halflife is now five years old. I guess they spent to much time and effort on their HL mod community (Counter-Strike, Day of Defeat, Gunman Chronicles (wtf)) that it just wasn't as profitable to do something with TF2.
The thing to remember is that K&R C is no longer with us, in the traditional sense. It's been discredited in favor of the C we're all familiar with. Its largely the same concepts, but for those familiar with the sats Wealth of nations:Future of Money::K&R C:Today's C++ . Mostly the same as what we think about today, but much more cryptic, and missing a few insightful concepts.
Re:Americans have brought much of this on ourselve
on
Business Under Fire
·
· Score: 1
While I certainly agree with you that libertarian principles are no match for reality (though Seth Finkelstein put it more eloquently), I have a minor quibble. Gatto has argued that the importation of public education from Germany was designed to reinforce the divide between the rich and the poor, and has only recently become a staple among the middle class. Before this, he mentions home schooling as the healthy status quo (although I have no idea how he gets his literacy numbers).
Certainly, public education has come to perform a critical role in today's economy. In fact, in the population center of bleeding red Republican Kansas, we have some of the best school districts out there. Public ones. The only real private schools are the parochial schools for Catholics and the like. These would be the people crying out against evolution in our schools, who have difficulties reconciling their faith with the biological theories that are required reading if their students wish to enter state Universities.
My one question to you is, if public schools have done you so well, why wouldn't you send your children to the same?
Obviously Joel is stressing internships for selfish reasons anyways
No kidding. Every article he writes is either about how great Excel is (and Joel by association)and why, how awesome Fogcreek is to Work or Use and how much more is in the book, or how he'd like you to meet his great friend and their greater idea. Joel is, by practice, in the profession of marketing every bit as much as he is in the software business.
Actually, one of the math guys on there mentioned a different set of basic foundations that proved a theory wrong. And I don't think he was talking about geometry.
Of course, there certainly are things one believes in that you can't prove, even in mathematics. In fact, mathematicians have a name for them: conjectures. The most famous of which is the Pointcare projection, which if I recall correctly, has something to do with the complexity of 3d shapes and spheres. I'm sure as a mathematician you know this better than I.
I'm tempted to say that the difference between mathematicians and theologians is that mathematicians try their damnedest to prove their beliefs, while those caught up in religious ferver simply accept. But as it turns out, even the relgious folk love to hunt for finds that reinforce their beliefs. I think the book "Zen and the Art of Motorcycle Maintanence" brings forth a critical argument that science isn't simply about nature. Science is about reasoning about the world, designing empirical studies to support your position, and measuring the results. Scientific information and theory is what happens when you share your work with the world. The process relies not on individual observers being intellectually honest and impartial, but on the community as a whole to act skeptical and dubious. The pitfall here is that additional information which fits the current hypothesis rarely gets the kind of scrutiney that a contradictory one recieves.
I think a better question would be, "Which is more important in the progress of scientific knowledge: nature or human insight?"
I don't know about you, but my mom was a lying, controlling bitch. By your standards, wouldn't you expect me to be running naked in the night with the devil worshipers?
In theory, pharmacists are indespensable human elements that go over these prescriptions, ensuring that bad combinations of drugs aren't administered (some combinations of drugs have a habit of reacting in the blood stream and forming a precipitate that clogs veins and arteries, among other things). In theory, doctors are supposed to be superhumans as well, and not prescribe these combinations.
In reality, there are no super humans. Its not something the medical profession enjoys admitting. New studies of drug interactions come out regularly, and few can really keep up with the pace. If you were to test a pharmacist and a robot during a month long study, I'd expect that either the robot wins, or the pharmacist winds up being extra dilligent on behalf of the study and ties it for perfection.
You act like its impossible to program in failsafes, like nobody knows exactly how much is too much, let alone poor helpless software engineers. Certainly, lives are put at risk in both avionics and medical computing. In this case, however, one of the core duties is to check exactly for these things, which places extra emphasis on an already important task.
Ironically, most of the talented engineers who design and build on linux have had degrees, and are often in pursuit of advanced degrees. Similarly, Google's founders were in college at the time the envisioned pagerank.
Google is looking for driven, talented people with something to prove to the world. If that's you, then you'll already be striving for this in many ways. Certainly, moving for an advanced degree can be one avenue for building your dream.
Your post hints at being a college student; as a recent graduate the most important advice I can give you right now is to seek internships. Think of the greatest companies out there, and think of why you'd like to work for them. Hunt down their HR sites and figure out how to get an internship. This is basically where their new hires come from. I'd call it an apprenticeship system, but its really nothing of the sort. It's a chance to prove you are willing to work hard, know what you're doing and capable of learning quickly. Most often you're provided with little on the job training. Since you can't get an internship without being in school, you're basically stuck, no? I guess you'll have to suck it up, and start participating in class. Its not like a real job as a salaryman is any easier.
Should none of this at least sound reasonable to you, one last position: if you're so badass, why not take the opportunity to prove them all wrong and start your own company?
Its not like the 2004 release date goal was held secret in some clandestine fortress of the FSF. I'd hope the buyers discovered this relatively easy to find fact, and continued anyways. In all likelyhood, this is the first step in releasing the source. By quietly discussing this with the community, the liscencee is in a much worse position to fight a GPL release. Its entirely possible that a month from now they contact Carmack and ask him to release the source under the GPL, and ask him to drop their name or something (once they're ready to show the public their product).
Of course, abstractions have a very important role in crypto-security. We don't yet have a widely used, practical notation for secure networks and protocols. Its easiest to thwart things like man-in-the-middle attacks from the start with abstract reasoning of what handshakes and protocols need to do. Naturally, there will be a dischord between the abstract model and the implementation. Nothing can move us seamlessly from one to the other, but human thought and effort can bring us the closest.
I agree at a fundamental level, it takes effort on all levels and aspects to create a solid security system. To borrow a metaphore from Seven Samurai "Defense is harder than attacking."
Clearly you mistook the context of it. The 350lb man is an FBI agent. Like agent Smith only fatter, and a more accurate shot. Plus, you can't stop time.
Well, he may or may not be "that bright," but he happened to be around the White House when a lot of the Big News was going down. He happened to be a speech writer and lawyer for Nixon (now there's a tough gig).
I'm not so certain you can just change power plants overnight. I hear it takes something like 30 years to decomission a power plant, and who knows when the coal lines will come down. Sure, there's less, but they're also much, much larger and have a goverment buerocracy that borders on monopoly (where it isn't blatantly so).
Firstly, automotive engineers are paid to think about all possible cases, even ones that are unfavorable such as running out of gas or driving into a tree. Most engineers I know would consider it a design defect should a hybrid car refuse to turn on a gas motor when the batteries failed, reguardless of how likely that scenario is. But this is generally a moot point.
I think you misinterpreted my other statement. What I mean is that a significant amount of energy is spent fighting air resistance just to keep at the current speed. The faster you go, and the longer you maintain that speed, the more energy is spent moving air, which regenerative braking won't bring back. Regenerative braking only works on kinetic engergy that would otherwise be discarded. You might say that its about as useful as coasting to a stop, but does the job much quicker (and with a significant loss in efficinecy). There's also some minor gains from regenerating energy that came from regenerative braking already, but I can't tell if that's already factored into recovery statistics or not.
Of course, speed and accelleration will always play a significant factor in the performance of a system like this. My roommate was in shock over just how much more energy it would take to move their car up from 50 to 55 mph. I certainly wonder how the leadfooted driver would perform comparitively. After all, its not exactly going to help fuel efficiency in a normal car either.
Firstly, I'd have to assume that should the batteries have no charge, the gasoline engine would kick in, defeating your "no impact" hypothesis. Similarly, I'd imagine that when you're out of gasoline, the electric would act as a failsafe. I haven't had the opportunity to test these out on a hybrid, but it certainly would make sense.
Secondly, using regenerative braking absolutely does improve milage. When you burn gas to move forward, that's energy spent. When you have to apply the real brakes, that's energy lost in friction. Regenerative braking, on the other hand, takes the kinetic energy of moving forward, and rather than simply throwing it away like a normal car, converts it into battery charge via the electric engine now acting as a generator. The important question is, how much energy from the gasoline is spent fighting air resistance, and how much is thrown away through braking? This quickly turns into a matter of distance per stop, with stop 'n' go traffic playing a larger role.
Slashdot Mirror
Well, hey why not? As long as the product is high quality, its not like its eroding the brand. There was an article not too long about about how Nintendo was branding everything in sight, rather than developing new franchises. I used to be a bit miffed over how they put his face everywhere, but really, how many franchises does Nintendo need? They've got a big enough cast that they can actually make a fighting game out of it with 30 characters from like 10 franchises (star fox, metroid, zelda, kirby, pokemon, mario, fire emblem, f zero, earthbound, donkey kong...). Really, some of those franchises don't do as well as the best, so its natural to help trim the bottom line with some extra income from Mario to help offset the less than stellar F Zero GX sales (still an awesome game and a better buy daily).
Yes, people will buy it because of Mario. Will it be a crappy game because of Mario? Will Mario be crappy because he's in the game? My answers are no and no. Really, whats going on is Nintendo is trying to attract games over to their system, and they're using their own new "seal of approval" to do it. Nintendo gets a popular game on their dominated console (sadly), the game makers get a dose of more sales, and cube owners get more games.
Mario, by and large, will always be judged by the flagship games that bear his name, rather than the one offs with his face on it.
Which distribution would you suggest then, if I want to build a 64bit desktop, and do things like run a scanner, a printer and my wifi card? And if I find a novel 32bit program, such as those Kento Cho shooter games written in D, can I run it?
Toms hardware consistantly favors computer hardware that only pushes above the competition by 1 percent or less. People spend an extra 40 dollars for this performance, and you're not willing to consider that people might like a FREE performance boost of a percent?
Genetic programming is a known quanitity, but if it "wont be running at run time" then its borderline useless. The whole idea is that the genetic algorithm adapts to a workload. Currently, scheduling is about fine tuning at the expense of flexibility. You tune the scheduler for a desktop setup, or for a database server, or whatever you have. Most tweaks yield a few percentage point gains in theory and maybe a single percent in practice for the given problem set, at the expense of about a ten percent penalty in the worst case scenario. Essentially, the genetic algorithm attempts to fine tune the kernel for you in ways.
Its not really that complex, but reasoning about how well it could perform is VERY difficult comparitively speaking. Furthermore, it introduces a much fuzzier notion of fine tuning. Rather than playing with variables like "swappiness" and cpu affinity, you're messing with a fitness function, where minute changes can move you from a percentage gained over a stock kernel to a percentage (or worse) loss. Certainly, I'm impressed that he's managed to make an improvement over stock with it, which puts interested kernel developers on a good first step. Nobody wants to chase technology that doesn't even show a hint of unlocking performance potential.
You kid right? Typically, Linux is treated as a lepper colony by the University administrations. They usually get a good deal on Solaris systems so that's what they use for mainframes. Certainly a few CS/IS professors may use Linux as a testing ground for some sort of reserarch, perhaps something like extensions to MOSIX. But those workstations running linux probably already have eight local vulnerabilites. Just try pulling up a shell and running uname -a on it. How out of date is your kernel?
But what do you want them to do? Kick you off every time a kernel exploit is found so they can patch it? That can't be fun.
Except that developers are even afraid of releasing anything dangerous into Sid, lest they break developer's guis or something. Hence the "experimental" distro. I've used sid for a long time and the only real problem I ever had was a bumpy transition from gnome 1.4 to the 2.x series. Even then, I still had something like twm to fall back on, and everything i needed still worked. I think that Debian really needs a push in the back to shake things up a bit. Part of the problem is that a lot of Debian is simply Done, and the Debian Developers mostly focus on scratching their own backs. For example, compare debian.org to alioth. I think a lot can be learned from the success of gentoo, but Debian leadership typically wants to handle the status quo. Basically, Debian needs better marketing in general.
Valve's new TF game was supposed to be on shelves in 2000 or so. Remember, Halflife is now five years old. I guess they spent to much time and effort on their HL mod community (Counter-Strike, Day of Defeat, Gunman Chronicles (wtf)) that it just wasn't as profitable to do something with TF2.
The thing to remember is that K&R C is no longer with us, in the traditional sense. It's been discredited in favor of the C we're all familiar with. Its largely the same concepts, but for those familiar with the sats Wealth of nations:Future of Money::K&R C:Today's C++ . Mostly the same as what we think about today, but much more cryptic, and missing a few insightful concepts.
While I certainly agree with you that libertarian principles are no match for reality (though Seth Finkelstein put it more eloquently), I have a minor quibble. Gatto has argued that the importation of public education from Germany was designed to reinforce the divide between the rich and the poor, and has only recently become a staple among the middle class. Before this, he mentions home schooling as the healthy status quo (although I have no idea how he gets his literacy numbers).
Certainly, public education has come to perform a critical role in today's economy. In fact, in the population center of bleeding red Republican Kansas, we have some of the best school districts out there. Public ones. The only real private schools are the parochial schools for Catholics and the like. These would be the people crying out against evolution in our schools, who have difficulties reconciling their faith with the biological theories that are required reading if their students wish to enter state Universities.
My one question to you is, if public schools have done you so well, why wouldn't you send your children to the same?
Obviously Joel is stressing internships for selfish reasons anyways
No kidding. Every article he writes is either about how great Excel is (and Joel by association)and why, how awesome Fogcreek is to Work or Use and how much more is in the book, or how he'd like you to meet his great friend and their greater idea. Joel is, by practice, in the profession of marketing every bit as much as he is in the software business.
Well, you might wanna check out the Wealth of Nations, by Adam Smith. Thats pretty much the economic bible equivelent of K&R C.
Actually, one of the math guys on there mentioned a different set of basic foundations that proved a theory wrong. And I don't think he was talking about geometry.
Of course, there certainly are things one believes in that you can't prove, even in mathematics. In fact, mathematicians have a name for them: conjectures. The most famous of which is the Pointcare projection, which if I recall correctly, has something to do with the complexity of 3d shapes and spheres. I'm sure as a mathematician you know this better than I.
I'm tempted to say that the difference between mathematicians and theologians is that mathematicians try their damnedest to prove their beliefs, while those caught up in religious ferver simply accept. But as it turns out, even the relgious folk love to hunt for finds that reinforce their beliefs. I think the book "Zen and the Art of Motorcycle Maintanence" brings forth a critical argument that science isn't simply about nature. Science is about reasoning about the world, designing empirical studies to support your position, and measuring the results. Scientific information and theory is what happens when you share your work with the world. The process relies not on individual observers being intellectually honest and impartial, but on the community as a whole to act skeptical and dubious. The pitfall here is that additional information which fits the current hypothesis rarely gets the kind of scrutiney that a contradictory one recieves.
I think a better question would be, "Which is more important in the progress of scientific knowledge: nature or human insight?"
I don't know about you, but my mom was a lying, controlling bitch. By your standards, wouldn't you expect me to be running naked in the night with the devil worshipers?
On the other hand, arbitration has been found to favor employers over employees in overwhelming majorities.
In theory, pharmacists are indespensable human elements that go over these prescriptions, ensuring that bad combinations of drugs aren't administered (some combinations of drugs have a habit of reacting in the blood stream and forming a precipitate that clogs veins and arteries, among other things). In theory, doctors are supposed to be superhumans as well, and not prescribe these combinations.
In reality, there are no super humans. Its not something the medical profession enjoys admitting. New studies of drug interactions come out regularly, and few can really keep up with the pace. If you were to test a pharmacist and a robot during a month long study, I'd expect that either the robot wins, or the pharmacist winds up being extra dilligent on behalf of the study and ties it for perfection.
You act like its impossible to program in failsafes, like nobody knows exactly how much is too much, let alone poor helpless software engineers. Certainly, lives are put at risk in both avionics and medical computing. In this case, however, one of the core duties is to check exactly for these things, which places extra emphasis on an already important task.
More like they're finding the people willing to replace themselves with a small shell script, and find a new function within the company, and repeat.
Ironically, most of the talented engineers who design and build on linux have had degrees, and are often in pursuit of advanced degrees. Similarly, Google's founders were in college at the time the envisioned pagerank.
Google is looking for driven, talented people with something to prove to the world. If that's you, then you'll already be striving for this in many ways. Certainly, moving for an advanced degree can be one avenue for building your dream.
Your post hints at being a college student; as a recent graduate the most important advice I can give you right now is to seek internships. Think of the greatest companies out there, and think of why you'd like to work for them. Hunt down their HR sites and figure out how to get an internship. This is basically where their new hires come from. I'd call it an apprenticeship system, but its really nothing of the sort. It's a chance to prove you are willing to work hard, know what you're doing and capable of learning quickly. Most often you're provided with little on the job training. Since you can't get an internship without being in school, you're basically stuck, no? I guess you'll have to suck it up, and start participating in class. Its not like a real job as a salaryman is any easier.
Should none of this at least sound reasonable to you, one last position: if you're so badass, why not take the opportunity to prove them all wrong and start your own company?
Its not like the 2004 release date goal was held secret in some clandestine fortress of the FSF. I'd hope the buyers discovered this relatively easy to find fact, and continued anyways. In all likelyhood, this is the first step in releasing the source. By quietly discussing this with the community, the liscencee is in a much worse position to fight a GPL release. Its entirely possible that a month from now they contact Carmack and ask him to release the source under the GPL, and ask him to drop their name or something (once they're ready to show the public their product).
Of course, abstractions have a very important role in crypto-security. We don't yet have a widely used, practical notation for secure networks and protocols. Its easiest to thwart things like man-in-the-middle attacks from the start with abstract reasoning of what handshakes and protocols need to do. Naturally, there will be a dischord between the abstract model and the implementation. Nothing can move us seamlessly from one to the other, but human thought and effort can bring us the closest.
I agree at a fundamental level, it takes effort on all levels and aspects to create a solid security system. To borrow a metaphore from Seven Samurai "Defense is harder than attacking."
Clearly you mistook the context of it. The 350lb man is an FBI agent. Like agent Smith only fatter, and a more accurate shot. Plus, you can't stop time.
Except, they don't use VPN tech at all. Its just a bunch of private, unpublished ftp servers connected to one another in clandestine ways.
Well, he may or may not be "that bright," but he happened to be around the White House when a lot of the Big News was going down. He happened to be a speech writer and lawyer for Nixon (now there's a tough gig).
I'm not so certain you can just change power plants overnight. I hear it takes something like 30 years to decomission a power plant, and who knows when the coal lines will come down. Sure, there's less, but they're also much, much larger and have a goverment buerocracy that borders on monopoly (where it isn't blatantly so).
Firstly, automotive engineers are paid to think about all possible cases, even ones that are unfavorable such as running out of gas or driving into a tree. Most engineers I know would consider it a design defect should a hybrid car refuse to turn on a gas motor when the batteries failed, reguardless of how likely that scenario is. But this is generally a moot point.
I think you misinterpreted my other statement. What I mean is that a significant amount of energy is spent fighting air resistance just to keep at the current speed. The faster you go, and the longer you maintain that speed, the more energy is spent moving air, which regenerative braking won't bring back. Regenerative braking only works on kinetic engergy that would otherwise be discarded. You might say that its about as useful as coasting to a stop, but does the job much quicker (and with a significant loss in efficinecy). There's also some minor gains from regenerating energy that came from regenerative braking already, but I can't tell if that's already factored into recovery statistics or not.
Of course, speed and accelleration will always play a significant factor in the performance of a system like this. My roommate was in shock over just how much more energy it would take to move their car up from 50 to 55 mph. I certainly wonder how the leadfooted driver would perform comparitively. After all, its not exactly going to help fuel efficiency in a normal car either.
Firstly, I'd have to assume that should the batteries have no charge, the gasoline engine would kick in, defeating your "no impact" hypothesis. Similarly, I'd imagine that when you're out of gasoline, the electric would act as a failsafe. I haven't had the opportunity to test these out on a hybrid, but it certainly would make sense.
Secondly, using regenerative braking absolutely does improve milage. When you burn gas to move forward, that's energy spent. When you have to apply the real brakes, that's energy lost in friction. Regenerative braking, on the other hand, takes the kinetic energy of moving forward, and rather than simply throwing it away like a normal car, converts it into battery charge via the electric engine now acting as a generator. The important question is, how much energy from the gasoline is spent fighting air resistance, and how much is thrown away through braking? This quickly turns into a matter of distance per stop, with stop 'n' go traffic playing a larger role.