Well, one of the most satisfying things a person can do is help another human being. This may be wired into the human brain, if not it is wired into the human condition. Moral arguments against altruism are like moral arguments against sex: they can only be taken so far before they run up against human nature.
One has to be careful with generalizations, of course, but any successful person who does not turn his attention to good works is as much to be pitied as one who spends his life a virgin because he thinks God gave men penises to lead them into damnation.
Waste is an important requirement of human progress. Waste is to civilization what compost is to gardening.
Take the 19th century British railroad bubble. Loads of people lost their shirts investing in rail companies, the result was that Britain gained an excellent rail network at little public expense.
Take the Internet bubble of the 90s, where many quixotic ventures started with a hasty powerpoint presentation and a VC desperate to get some kind of stake in the land rush. Vast fortunes where burned on hopeless enterprises, and many a personal fortune made off of anticipated cash flows with far less NPV. Vast amounts of money was wasted, but as a result the infrastructure of the Internet was greatly extended and made the foundation of new, more productive enterprises.
Take the most useless human activity of all: war. War is widely acknowledged as a major driver of technological and scientific progress.
Things would be great if human beings could sit down and set their minds on creating progress, but that's not what people are like. Progress comes from people attempting to adapt to altered circumstances. An expanded view of the universe and a reduced view of man's place in it certainly qualifies as altered circumstances. For some it is a calamity.
Well, not burning gasoline isn't a hobby either -- unless you're a hypermiler.
One of the definitions of "religion" is "A cause, principle, or activity pursued with zeal or conscientious devotion." Some people may be atheists in that they just don't believe. They're like people who telecommute; telecommuters don't use gas because they're doing other things.
Others espouse atheism, and have a mania for correcting the metaphysical errors of others that is quite comparable in zeal to any other sort of evangelical. They are proselytizers for the Null God. Those atheists are like the hypermilers, who have more in common with people who like to drive fast on the track than people who prefer not to drive at all. It's all about different kinds of performance.
Well, I'm not claiming this thing works, but your explanation of why the thing can't possibly work is broken.
Liquids are incompressible -- a least to a fix approximation. So the molecules can't get "crammed back together" very much. It's true that the properties you establish in a flow are going to be altered past recognition when you squeeze that flow through a narrow space, but that is not the claim here. The claim is not an alteration of the flow, but an alteration in the liquid's physical properties. Namely, the claim a change in viscosity, which if achieved would certainly alter the behavior of the liquid, most of all in a restricted flow.
The argument that devices resembling this and trying to achieve the same outcome have been junk only means that we have to look at how the claims are based very, very carefully. After all, we've known that heavier than air flight is possible forever, because birds do it; most machines that attempted it had no chance, not because the result was impossible, but because it couldn't be achieved that particular way.
If you could show that the claims of this device involve something like extracting more work from the gasoline than is thermodynamically possible, you'd have an airtight argument against it. For now, the strongest argument against this device is that it's claims haven't been replicated in controlled experiments mimicking real world conditions.
Personally, I'm not going to go out and restructure my investment portfolio because of the claims made for this device. There have been lots of physically possible claims that either didn't work, or were not economical because of their complexity or impact on engine wear.
Well, I've been watching this project for years. I've wanted to use it, but it doesn't run on anything you can buy new except iPods. Given that the iPod already has pretty good firmware, I'd be spending a lot of money to potentially brick a perfectly good music player.
If it'd run on the current generation Sansa players I'd give it a go; if I <em>owned</em> one of those players and didn't like it, I'd be even more motivated.
As its stands, its only practical appeal is for experimental or hack value. Some people will be interested in Ogg, I suppose, but most people can live with high bitrate MP3 for a portable music player. The attraction is tinkering with a device to get it to do different things. I'm not insensible to that, but I'm not going to haunt eBay looking for an old music player somebody's dumped. I've got better uses for my limited hacking time. If there's a player I can get off of Amazon for less than a hundred bucks which supports the majority of the interesting features in RockBox, I'd be there.
It'd be really nice if some contract manufacturer somewhere figured out that supporting RockBox would be a cheap way to sell lots of players over outlets like Amazon.
Of course what I'd really like to see is some mid range media player turned into a viable application platform. If you could port Android to the Zune, that would be really cool.
Well, I think telemetry is much more interesting than control. Instead of just HTML, you could have various values stored in NV RAM accessed through something like RSS.
For example, instead of developing some kind of special hardware, wiring and protocols to connect all the devices in your factory, you just run ethernet everywhere and slap one of these things on to all your machines. You then write a simple PERL script to fetch certain URIs from all the machines in the address range, storing performance data and relaying exception conditions (mill #5 can't change its cutting tool) to a pager.
It's not that simple. True, you need a supermajority in the Senate to do anything against the will of the minority. However, you don't even need a majority to stop something. Cutting off funds for the war falls under the category of stopping.
The problem was this was a huge game of chicken against a player who is proven to be extremely reckless. There's a good chance that Bush would simply have attempted to muddle through, counting on the inevitable disaster to get Congress to open the purse strings.
Umm.. You do realize the crisis is in private credit, not public spending? The relevant issue is regulatory oversight, which is an executive function. The only legislative way to intervene would require a supermajority in the Senate.
Well, it's not really fundamentally different than academic tracking, is it? Take a student on one track and move him into a more challenging one, and his grades will drop.
What this means is that the grades themselves are somewhat arbitrary. They aren't some kind of ideal measure that we're tampering with. They are there to measure progress. As artificial and somewhat arbitrary measures they are open to interpretation. Possibly there should be some kind of logarithmic scale because the distinction between 50 an 40 is very different than the distinction between 90 and 80; averaging scores on some logarithmic scale would have much the same effect.
The point is, grades are tools. If the tool is discouraging student effort, then the tool can be modified, provided it produces better results.
I'm not for this or against this particular policy in general. If it helps fine, if it hurts then bad. It may be that different schools, even different subjects need different treatment. A student who falls behind in math in general can't catch up. If you completely fail to grasp the first third of the year of geometry, it doesn't matter how hard you work on the last third. A student who falls behind in history is somewhat handicapped, but not to the same degree. Flunking the Pilgrims doesn't preclude a credible effort on the Civil War. In the end an excellent job on the Civil War, while not compensating for a lack of knowledge of the 17th century, could at least demonstrate advancement in critical thinking about history.
Lesson #2, method dealing with the enemy while occupying a strategically disadvantageous position: see lesson #1.
Does anybody believe that the proprietary/free clock will be rolled back to the late 1980s, when printing licenses was like printing currency? Of course not. Open source is here to stay. That doesn't mean there aren't opportunities to make money in software, both in competition with and by using free software. It seems to me the smart business leader chooses the mix of competition and cooperation. Google hasn't done too bad, after all.
Where it's tough is when you have a company with a cash cow. Microsoft. ESRI. Oracle. The cash cow may be doomed, but ever year it is kept alive represents money, a great deal of money.
So it makes sense to position your product, say Windows, against the open source "competition". It really boils down to one thing: compete. Give your customers reasons to keep buying your product and cut prices to keep them from moving away from your products. There are now free as in beer versions of Oracle and SQL Server, just to establish a bulwark on the low end of the product position.
22% of the time wrong? It's much, much, much worse than that. People who procure these systems ought to take a course in Bayesian probability before they're let anywhere near the contracting process.
Now, we don't know anything about the false positive vs. false negative rate. We only have "78%" accurate to go on, so let's take a 22% false result rate.
Now suppose, for purposes of argument, that one in every hundred passengers is a terrorist. If we put a hundred passengers through the system, it would generate 23 positives: one terrorist (on average) and 22 mistakes. Therefore a positive result would only be about 4% accurate.
Now there are probably on the order of ten million passengers who travel by air every day in the US. It would seem reasonable to assume that the number of terrorists who fly in the US on any particular day is considerably smaller than 100,000. Let's say optimistically that only a thousand out of the ten million passenger are terrorists. When we use our magic system to find the terrorist needles in the passenger haystack, it will flag 220,000 innocent passengers as suspicious, 780 terrorists as suspicious, and let 220 terrorists through as benign.
Thus, out of the 220,780 people flagged, only 780 are terrorists. That mean for every 283 people who are flagged, you get only one terrorist. That amounts to an accuracy of only three tenths of a percent.
It's ironic that geeks are supposed to be the ones with poor social skills, when management treats IT workers as extensions of the machines they work with.
The whole MBO thing is a moron's approach to leadership. It's not that objectives aren't a good thing, it's just that the manager who doesn't really understand his people plugs people into the plan as if they were standardized parts. Geeks are different.
I've been in meetings discussing things like establishing bonuses for achieving certain objectives, and this is the point I always make: yeah, bonuses are fine, but if they really make a difference in performance you aren't going to get the best work. Guys who come out of sales just can't get their brains around anything more sophisticated than a financial carrot and stick, because they excelled in a game where it was grab the carrot and leave the people behind to deal with the stick.
If you really want to incent a geek, make this a project one where he can do his best work ever. Or make this a project where he can increase his skill level.
A geek wants to be respected for his skill and honored for his contribution. Yeah everybody does, just like everybody likes a nice bonus check, but it doesn't mean the same things grab everyone's attention. You take care of your geeks. You help them advance their skills. You give them room to do their jobs. You show respect by listening to their concerns and by owning your own mistakes when you have to overrule those concerns. Do those things and you'll be richly rewarded.
I find that often the problem in business is not what we don't know; it's the things we know on an intellectual level, but don't live by. Under stress, people fall back into actions that make them feel comfortable, rather than ones that address the situation. So managers who are in trouble don't communicate with their teams, they dump objectives on them. They don't work with their people to create a realistic plan, they dangle a carrot or wave a stick in front of them in hopes of producing a miracle.
The difference between people and horses is that people are much, much smarter. They can figure out how to get the carrot or avoid the stick without moving anything forward. You've got to make them want to go forward.
Well 4x increase to work or 4x increase to be practical. There's a difference you know. It's one thing to put up a cable that can demonstrate the concept, it's another thing to build one that is capable of supporting useful payload with a sufficient safety margin.
The success rate of use of tethers in space to do interesting things is enough to give one pause. Long tethers can be trouble.
I've heard it claimed that the cable failure modes in space elevators would be safe. I'm not sure I buy that. For example a near ground break would cause the cable to rise into orbit. Oceanographic ships that deploy instruments on steel cables sometimes can't retrieve them because the cable, which stows neatly in a spool on deck, has knotted itself into a tangle several times larger than the ship itself. Imagine the size of knots fifteen thousand miles of cable could form, and the hazards it would present to space vehicles.
Likewise, I'm not fully convinced that a space end break would be so benign either. Even if we stipulate that most of the space portion of the cable burns up, the bottom hundred miles might not fall like a stone dropped from the same height, if the lower cable has to accelerate all that mass above it which is out there at near zero gravity. Even if only the bottom five percent made it to the surface, that could be nearly a thousand miles cable. It might not weigh much per foot, but a tangle formed from a thousand miles of ultra strong cable is going to be a hazard.
I'm not saying the space elevator idea is not practical, but it could be extremely hazardous. If we build such a thing, it had better be stupendously overbuilt, not just within the margin of what is technically possible. A space elevator just capable of demonstrating proof of concept is probably a bad idea.
If I were funding a project like this, it would be more with the eye to developing advances in the materials that could be exploited commercially well before the space elevator was a practical possibility. This might even pay for the rest of the project.
We seem to be dealing with a false dichotomy here. It is not the case that any kind of crime against privacy is justifiable if it finds something. Nor is it the case that an invasion of privacy excuses any improprieties that were found. Logically the two are completely unrelated.
I think Governor Palin's privacy was violated, and that the violations uncovered evidence of impropriety on her part -- specifically conducting state business via a personal account. The rules broken here are to protect the public: from invasion of privacy in the case suspicion only, and from government officials evading accountability by hiding their conduct of public business.
Of the two, I judge the privacy violation more serious, but neither of the violations here are things we want to encourage. Of the two, I'd say the account cracker ought to do jail time or public service, and the governor ought to promise to stop mixing private and public business and voluntarily appoint a trusted third party to, for some period of time, examine her private email correspondence and vouch that she is not conducting state business with it.
Well, let's get out of the realm of the abstract for a moment. Anybody who is reasonably educated knows that an order to do something illegal, say to murder a prisoner, is not valid. So you don't have to feel so injured by misunderstanding. There's always going to be a few or course.
On the other hand, the principle that soldiers should not obey an illegal order is really only good as the ability of a soldier to distinguish between legal and illegal. There isn't always a clear line, say between legal, aggressive interrogation techniques and illegal torture. One of the benefits of ROTC is, hopefully, and officer corps with greater critical thinking skills. Still, by in large troops and the officers who lead them are not lawyers, they have to use their ethical common sense to get them through dilemmas.
The real danger when you give a man a lethal weapon and put him under orders is not particular to the military. It is group think. And don't say that isn't a problem. Every military person I have talked to has plenty of stories of bureaucratic pigheadedness on a massive scale.
I have known many military people over the years, and one thing that I think is fair to say is that good soldiers, marines, airmen and sailors have a can do attitude. That contributes to the both the dynamism and dysfunction of the military. Survival may trump that, but the first response to an order to take a fortified position is to view it as a solvable problem. This takes an implicit trust in the competence and judgment of your superiors, and that habit means going along with things you know are damnably stupid -- so long as they aren't illegal or immediately fatal.
Trust and a willingness to go along with anything short of illegality are good things in a soldier, but bad things in a citizen and especially a civilian leader. A good citizen has to question the competence and judgment of the leadership. When political mistakes reach the military, it's too late to question. One military saying I've heard is that shit rolls downhill, and it's the military's job to deal with the politicians' shit. A politician's ought to avoid handing the shit down to the military by being skeptical.
Skepticism is not a military virtue, which is not to say anything negative about military service. No profession is the beginning and end of all virtues. One of the problems I see of certain political viewpoints is that they like to promote the military as the entire repository of American virtue because obedience or rather willingness to get behind the mission, is so useful to them.
Look at Colin Powell, a great soldier, a top notch military leader, and a bad Secretary of State. He brought his military values of duty and loyalty into the job, and ended up being a catspaw. It wasn't that he accepted an order to lie; he accepted the mission he was given and took ownership of it, the way good soldiers do. It made him both useful and an object of scorn within the administration. By giving his superiors more than they deserved, he gave his true masters less.
He wants them to attribute when they use the results of the code, not when they use the code.
Personally, I think this wrong-headed.
It seems to me that one of two cases apply: the software in question is critical to reproducing the results presented, in which case it would be mandatory to cite the software in any real peer reviewed paper. Or the software is not critical in reproducing the results, in which case the developers don't merit citation in the paper, although they may be deserving of gratitude and a pat on the pack.
For example, should people who used LaTex or Open Office to prepare printed materials used in the course of their research cite those products? Only if the particular materials produced could only be produced in precise form by that software.
Now suppose you used postscript to produce images used for vitual perception experiments. Well, you'd probably want to publish the routines, and certainly stipulate they ran on such and so a Postscript implementation, if there were any chance at all that different implementations would render those images differently.
Now, in cases where results are produced that are dependent on a particular piece of software, whether that software be proprietary, open source, or in the public domain, it is academically dishonest NOT to cite the software, in my opinion.
Well, that's a matter of being in a hurry go get the program out the door. There's lots of parameter input through the compiler going on. Always has been.
I personally like terse languages, but I think coupling is a bigger issue. Fifty lines of self-contained COBOL are easier to understand than twenty lines of highly coupled Python that depends on assumptions spread far and wide in the system.
One of the reasons that so many COBOL systems remain is that they were written in a day when most tasks ran top to bottom. It was before the "event loop" became a familiar pattern to most programmers. In a sense, it shows how reusability can shoot you in the foot (there's few worthwhile tools that can't be dangerous some of the time). Back in the day the vast majority of programs had well defined input, performed a well characterized calculation on that input, and produced a well defined output. Now consider something that is a component in a framework. It has to be damn well conceived because it's meant to operate in situations the designer has never even conceived of.
So, I'll bet that the COBOL that survives is stuff which does something that is clearly defined, simple, and useful. Why convert it to Java if it works fine and is part of a large body of software that works fine?
I try to avoid anger and fear, because they make one closed minded. I never take the narrative given to me by my side for granted, so while I'm a liberal, you won't hear me buying into the rumor mill about McCain's crashing five planes (not true) or causing the Forrestall fire (not true).
There's been a HUGE movement among the Democratic circle to take a page from the Rove playbook this year, because we've had it up to here with Bush, and we don't want somebody who thinks the Bush program just needs a bit of tweaking. So I am amazed at the cavalier attitude a lot of my party mates are taking towards truth, ethics and decorum. I'll say, "Do you really need to gild the lilly here? Do you really need to take the good arguments we have and mix it with this trash?" The response is, "Well, they do and if we don't do it too we'll lose."
In other words, it's a fear response. The Republicans are scared because they're going to lose a lot in Congress and if they lose the Presidency they'll be weaker than they've been in a generation or more. The Democrats are scared because they think the country will go down the tubes if another Republican gets in.
So there's fear all around. It doesn't make you rational, and it doesn't (I believe) make you more effective.
Slashdot Mirror
Well, one of the most satisfying things a person can do is help another human being. This may be wired into the human brain, if not it is wired into the human condition. Moral arguments against altruism are like moral arguments against sex: they can only be taken so far before they run up against human nature.
One has to be careful with generalizations, of course, but any successful person who does not turn his attention to good works is as much to be pitied as one who spends his life a virgin because he thinks God gave men penises to lead them into damnation.
404 not found? In that case, you found something. You found the machine. The machine is not working though. That's useful information.
Waste is an important requirement of human progress. Waste is to civilization what compost is to gardening.
Take the 19th century British railroad bubble. Loads of people lost their shirts investing in rail companies, the result was that Britain gained an excellent rail network at little public expense.
Take the Internet bubble of the 90s, where many quixotic ventures started with a hasty powerpoint presentation and a VC desperate to get some kind of stake in the land rush. Vast fortunes where burned on hopeless enterprises, and many a personal fortune made off of anticipated cash flows with far less NPV. Vast amounts of money was wasted, but as a result the infrastructure of the Internet was greatly extended and made the foundation of new, more productive enterprises.
Take the most useless human activity of all: war. War is widely acknowledged as a major driver of technological and scientific progress.
Things would be great if human beings could sit down and set their minds on creating progress, but that's not what people are like. Progress comes from people attempting to adapt to altered circumstances. An expanded view of the universe and a reduced view of man's place in it certainly qualifies as altered circumstances. For some it is a calamity.
Well, not burning gasoline isn't a hobby either -- unless you're a hypermiler.
One of the definitions of "religion" is "A cause, principle, or activity pursued with zeal or conscientious devotion." Some people may be atheists in that they just don't believe. They're like people who telecommute; telecommuters don't use gas because they're doing other things.
Others espouse atheism, and have a mania for correcting the metaphysical errors of others that is quite comparable in zeal to any other sort of evangelical. They are proselytizers for the Null God. Those atheists are like the hypermilers, who have more in common with people who like to drive fast on the track than people who prefer not to drive at all. It's all about different kinds of performance.
Well, I'm not claiming this thing works, but your explanation of why the thing can't possibly work is broken.
Liquids are incompressible -- a least to a fix approximation. So the molecules can't get "crammed back together" very much. It's true that the properties you establish in a flow are going to be altered past recognition when you squeeze that flow through a narrow space, but that is not the claim here. The claim is not an alteration of the flow, but an alteration in the liquid's physical properties. Namely, the claim a change in viscosity, which if achieved would certainly alter the behavior of the liquid, most of all in a restricted flow.
The argument that devices resembling this and trying to achieve the same outcome have been junk only means that we have to look at how the claims are based very, very carefully. After all, we've known that heavier than air flight is possible forever, because birds do it; most machines that attempted it had no chance, not because the result was impossible, but because it couldn't be achieved that particular way.
If you could show that the claims of this device involve something like extracting more work from the gasoline than is thermodynamically possible, you'd have an airtight argument against it. For now, the strongest argument against this device is that it's claims haven't been replicated in controlled experiments mimicking real world conditions.
Personally, I'm not going to go out and restructure my investment portfolio because of the claims made for this device. There have been lots of physically possible claims that either didn't work, or were not economical because of their complexity or impact on engine wear.
Well, I've been watching this project for years. I've wanted to use it, but it doesn't run on anything you can buy new except iPods. Given that the iPod already has pretty good firmware, I'd be spending a lot of money to potentially brick a perfectly good music player.
If it'd run on the current generation Sansa players I'd give it a go; if I <em>owned</em> one of those players and didn't like it, I'd be even more motivated.
As its stands, its only practical appeal is for experimental or hack value. Some people will be interested in Ogg, I suppose, but most people can live with high bitrate MP3 for a portable music player. The attraction is tinkering with a device to get it to do different things. I'm not insensible to that, but I'm not going to haunt eBay looking for an old music player somebody's dumped. I've got better uses for my limited hacking time. If there's a player I can get off of Amazon for less than a hundred bucks which supports the majority of the interesting features in RockBox, I'd be there.
It'd be really nice if some contract manufacturer somewhere figured out that supporting RockBox would be a cheap way to sell lots of players over outlets like Amazon.
Of course what I'd really like to see is some mid range media player turned into a viable application platform. If you could port Android to the Zune, that would be really cool.
Well, I think telemetry is much more interesting than control. Instead of just HTML, you could have various values stored in NV RAM accessed through something like RSS.
For example, instead of developing some kind of special hardware, wiring and protocols to connect all the devices in your factory, you just run ethernet everywhere and slap one of these things on to all your machines. You then write a simple PERL script to fetch certain URIs from all the machines in the address range, storing performance data and relaying exception conditions (mill #5 can't change its cutting tool) to a pager.
It's not that simple. True, you need a supermajority in the Senate to do anything against the will of the minority. However, you don't even need a majority to stop something. Cutting off funds for the war falls under the category of stopping.
The problem was this was a huge game of chicken against a player who is proven to be extremely reckless. There's a good chance that Bush would simply have attempted to muddle through, counting on the inevitable disaster to get Congress to open the purse strings.
as opposed to what you THINK, then facts become a threat. When ideology is what you think, you can revise your thinking without a threat to your ego.
Fostering brand loyalty is a cost effective way to get repeat customers. But you don't <em>have</em> to be a mindless consumer of political ideology.
Umm.. You do realize the crisis is in private credit, not public spending? The relevant issue is regulatory oversight, which is an executive function. The only legislative way to intervene would require a supermajority in the Senate.
Well, it's not really fundamentally different than academic tracking, is it? Take a student on one track and move him into a more challenging one, and his grades will drop.
What this means is that the grades themselves are somewhat arbitrary. They aren't some kind of ideal measure that we're tampering with. They are there to measure progress. As artificial and somewhat arbitrary measures they are open to interpretation. Possibly there should be some kind of logarithmic scale because the distinction between 50 an 40 is very different than the distinction between 90 and 80; averaging scores on some logarithmic scale would have much the same effect.
The point is, grades are tools. If the tool is discouraging student effort, then the tool can be modified, provided it produces better results.
I'm not for this or against this particular policy in general. If it helps fine, if it hurts then bad. It may be that different schools, even different subjects need different treatment. A student who falls behind in math in general can't catch up. If you completely fail to grasp the first third of the year of geometry, it doesn't matter how hard you work on the last third. A student who falls behind in history is somewhat handicapped, but not to the same degree. Flunking the Pilgrims doesn't preclude a credible effort on the Civil War. In the end an excellent job on the Civil War, while not compensating for a lack of knowledge of the 17th century, could at least demonstrate advancement in critical thinking about history.
Lesson #1: Pick your battles
Lesson #2, method dealing with the enemy while occupying a strategically disadvantageous position: see lesson #1.
Does anybody believe that the proprietary/free clock will be rolled back to the late 1980s, when printing licenses was like printing currency? Of course not. Open source is here to stay. That doesn't mean there aren't opportunities to make money in software, both in competition with and by using free software. It seems to me the smart business leader chooses the mix of competition and cooperation. Google hasn't done too bad, after all.
Where it's tough is when you have a company with a cash cow. Microsoft. ESRI. Oracle. The cash cow may be doomed, but ever year it is kept alive represents money, a great deal of money.
So it makes sense to position your product, say Windows, against the open source "competition". It really boils down to one thing: compete. Give your customers reasons to keep buying your product and cut prices to keep them from moving away from your products. There are now free as in beer versions of Oracle and SQL Server, just to establish a bulwark on the low end of the product position.
22% of the time wrong? It's much, much, much worse than that. People who procure these systems ought to take a course in Bayesian probability before they're let anywhere near the contracting process.
Now, we don't know anything about the false positive vs. false negative rate. We only have "78%" accurate to go on, so let's take a 22% false result rate.
Now suppose, for purposes of argument, that one in every hundred passengers is a terrorist. If we put a hundred passengers through the system, it would generate 23 positives: one terrorist (on average) and 22 mistakes. Therefore a positive result would only be about 4% accurate.
Now there are probably on the order of ten million passengers who travel by air every day in the US. It would seem reasonable to assume that the number of terrorists who fly in the US on any particular day is considerably smaller than 100,000. Let's say optimistically that only a thousand out of the ten million passenger are terrorists. When we use our magic system to find the terrorist needles in the passenger haystack, it will flag 220,000 innocent passengers as suspicious, 780 terrorists as suspicious, and let 220 terrorists through as benign.
Thus, out of the 220,780 people flagged, only 780 are terrorists. That mean for every 283 people who are flagged, you get only one terrorist. That amounts to an accuracy of only three tenths of a percent.
It's ironic that geeks are supposed to be the ones with poor social skills, when management treats IT workers as extensions of the machines they work with.
The whole MBO thing is a moron's approach to leadership. It's not that objectives aren't a good thing, it's just that the manager who doesn't really understand his people plugs people into the plan as if they were standardized parts. Geeks are different.
I've been in meetings discussing things like establishing bonuses for achieving certain objectives, and this is the point I always make: yeah, bonuses are fine, but if they really make a difference in performance you aren't going to get the best work. Guys who come out of sales just can't get their brains around anything more sophisticated than a financial carrot and stick, because they excelled in a game where it was grab the carrot and leave the people behind to deal with the stick.
If you really want to incent a geek, make this a project one where he can do his best work ever. Or make this a project where he can increase his skill level.
A geek wants to be respected for his skill and honored for his contribution. Yeah everybody does, just like everybody likes a nice bonus check, but it doesn't mean the same things grab everyone's attention. You take care of your geeks. You help them advance their skills. You give them room to do their jobs. You show respect by listening to their concerns and by owning your own mistakes when you have to overrule those concerns. Do those things and you'll be richly rewarded.
I find that often the problem in business is not what we don't know; it's the things we know on an intellectual level, but don't live by. Under stress, people fall back into actions that make them feel comfortable, rather than ones that address the situation. So managers who are in trouble don't communicate with their teams, they dump objectives on them. They don't work with their people to create a realistic plan, they dangle a carrot or wave a stick in front of them in hopes of producing a miracle.
The difference between people and horses is that people are much, much smarter. They can figure out how to get the carrot or avoid the stick without moving anything forward. You've got to make them want to go forward.
Well, at least we haven't been operantly conditioned to say "what the hell, go ahead."
Well 4x increase to work or 4x increase to be practical. There's a difference you know. It's one thing to put up a cable that can demonstrate the concept, it's another thing to build one that is capable of supporting useful payload with a sufficient safety margin.
The success rate of use of tethers in space to do interesting things is enough to give one pause. Long tethers can be trouble.
I've heard it claimed that the cable failure modes in space elevators would be safe. I'm not sure I buy that. For example a near ground break would cause the cable to rise into orbit. Oceanographic ships that deploy instruments on steel cables sometimes can't retrieve them because the cable, which stows neatly in a spool on deck, has knotted itself into a tangle several times larger than the ship itself. Imagine the size of knots fifteen thousand miles of cable could form, and the hazards it would present to space vehicles.
Likewise, I'm not fully convinced that a space end break would be so benign either. Even if we stipulate that most of the space portion of the cable burns up, the bottom hundred miles might not fall like a stone dropped from the same height, if the lower cable has to accelerate all that mass above it which is out there at near zero gravity. Even if only the bottom five percent made it to the surface, that could be nearly a thousand miles cable. It might not weigh much per foot, but a tangle formed from a thousand miles of ultra strong cable is going to be a hazard.
I'm not saying the space elevator idea is not practical, but it could be extremely hazardous. If we build such a thing, it had better be stupendously overbuilt, not just within the margin of what is technically possible. A space elevator just capable of demonstrating proof of concept is probably a bad idea.
If I were funding a project like this, it would be more with the eye to developing advances in the materials that could be exploited commercially well before the space elevator was a practical possibility. This might even pay for the rest of the project.
We seem to be dealing with a false dichotomy here. It is not the case that any kind of crime against privacy is justifiable if it finds something. Nor is it the case that an invasion of privacy excuses any improprieties that were found. Logically the two are completely unrelated.
I think Governor Palin's privacy was violated, and that the violations uncovered evidence of impropriety on her part -- specifically conducting state business via a personal account. The rules broken here are to protect the public: from invasion of privacy in the case suspicion only, and from government officials evading accountability by hiding their conduct of public business.
Of the two, I judge the privacy violation more serious, but neither of the violations here are things we want to encourage. Of the two, I'd say the account cracker ought to do jail time or public service, and the governor ought to promise to stop mixing private and public business and voluntarily appoint a trusted third party to, for some period of time, examine her private email correspondence and vouch that she is not conducting state business with it.
Well, let's get out of the realm of the abstract for a moment. Anybody who is reasonably educated knows that an order to do something illegal, say to murder a prisoner, is not valid. So you don't have to feel so injured by misunderstanding. There's always going to be a few or course.
On the other hand, the principle that soldiers should not obey an illegal order is really only good as the ability of a soldier to distinguish between legal and illegal. There isn't always a clear line, say between legal, aggressive interrogation techniques and illegal torture. One of the benefits of ROTC is, hopefully, and officer corps with greater critical thinking skills. Still, by in large troops and the officers who lead them are not lawyers, they have to use their ethical common sense to get them through dilemmas.
The real danger when you give a man a lethal weapon and put him under orders is not particular to the military. It is group think. And don't say that isn't a problem. Every military person I have talked to has plenty of stories of bureaucratic pigheadedness on a massive scale.
I have known many military people over the years, and one thing that I think is fair to say is that good soldiers, marines, airmen and sailors have a can do attitude. That contributes to the both the dynamism and dysfunction of the military. Survival may trump that, but the first response to an order to take a fortified position is to view it as a solvable problem. This takes an implicit trust in the competence and judgment of your superiors, and that habit means going along with things you know are damnably stupid -- so long as they aren't illegal or immediately fatal.
Trust and a willingness to go along with anything short of illegality are good things in a soldier, but bad things in a citizen and especially a civilian leader. A good citizen has to question the competence and judgment of the leadership. When political mistakes reach the military, it's too late to question. One military saying I've heard is that shit rolls downhill, and it's the military's job to deal with the politicians' shit. A politician's ought to avoid handing the shit down to the military by being skeptical.
Skepticism is not a military virtue, which is not to say anything negative about military service. No profession is the beginning and end of all virtues. One of the problems I see of certain political viewpoints is that they like to promote the military as the entire repository of American virtue because obedience or rather willingness to get behind the mission, is so useful to them.
Look at Colin Powell, a great soldier, a top notch military leader, and a bad Secretary of State. He brought his military values of duty and loyalty into the job, and ended up being a catspaw. It wasn't that he accepted an order to lie; he accepted the mission he was given and took ownership of it, the way good soldiers do. It made him both useful and an object of scorn within the administration. By giving his superiors more than they deserved, he gave his true masters less.
Look, you shouldn't attack posters whom it turns you essentially agree with because:
(1) It takes the thread off topic.
(2) It shows you haven't read the post carefully.
(3) You'll burn in Hell.
Thank you. You are most gracious.
He wants them to attribute when they use the results of the code, not when they use the code.
Personally, I think this wrong-headed.
It seems to me that one of two cases apply: the software in question is critical to reproducing the results presented, in which case it would be mandatory to cite the software in any real peer reviewed paper. Or the software is not critical in reproducing the results, in which case the developers don't merit citation in the paper, although they may be deserving of gratitude and a pat on the pack.
For example, should people who used LaTex or Open Office to prepare printed materials used in the course of their research cite those products? Only if the particular materials produced could only be produced in precise form by that software.
Now suppose you used postscript to produce images used for vitual perception experiments. Well, you'd probably want to publish the routines, and certainly stipulate they ran on such and so a Postscript implementation, if there were any chance at all that different implementations would render those images differently.
Now, in cases where results are produced that are dependent on a particular piece of software, whether that software be proprietary, open source, or in the public domain, it is academically dishonest NOT to cite the software, in my opinion.
Binary/polar viewpoints are often plainly wrong as the world isn't binary at all.
I have no idea what you're talking about, although it sounds reasonable.
Well, that's a matter of being in a hurry go get the program out the door. There's lots of parameter input through the compiler going on. Always has been.
I personally like terse languages, but I think coupling is a bigger issue. Fifty lines of self-contained COBOL are easier to understand than twenty lines of highly coupled Python that depends on assumptions spread far and wide in the system.
One of the reasons that so many COBOL systems remain is that they were written in a day when most tasks ran top to bottom. It was before the "event loop" became a familiar pattern to most programmers. In a sense, it shows how reusability can shoot you in the foot (there's few worthwhile tools that can't be dangerous some of the time). Back in the day the vast majority of programs had well defined input, performed a well characterized calculation on that input, and produced a well defined output. Now consider something that is a component in a framework. It has to be damn well conceived because it's meant to operate in situations the designer has never even conceived of.
So, I'll bet that the COBOL that survives is stuff which does something that is clearly defined, simple, and useful. Why convert it to Java if it works fine and is part of a large body of software that works fine?
Yes.
I try to avoid anger and fear, because they make one closed minded. I never take the narrative given to me by my side for granted, so while I'm a liberal, you won't hear me buying into the rumor mill about McCain's crashing five planes (not true) or causing the Forrestall fire (not true).
There's been a HUGE movement among the Democratic circle to take a page from the Rove playbook this year, because we've had it up to here with Bush, and we don't want somebody who thinks the Bush program just needs a bit of tweaking. So I am amazed at the cavalier attitude a lot of my party mates are taking towards truth, ethics and decorum. I'll say, "Do you really need to gild the lilly here? Do you really need to take the good arguments we have and mix it with this trash?" The response is, "Well, they do and if we don't do it too we'll lose."
In other words, it's a fear response. The Republicans are scared because they're going to lose a lot in Congress and if they lose the Presidency they'll be weaker than they've been in a generation or more. The Democrats are scared because they think the country will go down the tubes if another Republican gets in.
So there's fear all around. It doesn't make you rational, and it doesn't (I believe) make you more effective.