I dunno - this looks like an intentional bomb hoax to me - regardless of what the person claims. This person put lives in jeopardy - including her own and those around her - when she created a situation that would for all intents and purposes appear to be a life-and-death shoot on sight suicide bombing incident. This also tied up police forces in dealing with a hoax that could be better employed watching out for real threats.
She should be punished - probably not with 30 years in prison, but some jail time and a heavy fine would probably be appropriate. It seems inconceivable to me that anybody would not expect the type of reaction that she received.
Now, if it turns out that she is mentally retarded and not some college student with a chip on her shoulder then I could see a little leniency being in order.
I'm sorry - I'm against TSA overreaction and the way that police tend to use storm trooper tactics where they are unnecessary and put lives in danger. However, a person standing in a crowded area with an apparent explosive device warrants exactly these sorts of tactics. The whole problem with suicide bombers is that they need to be stopped BEFORE they carry out their intended crimes. The police apparently showed tremendous restraint, running into a situation that could have very well ended their lives while using tremendous care in the use of lethal force. The hoaxer is very fortunate that she wasn't shot. If she were shot I'm sure there would have been an uproar nonetheless and as a result she needs to be made an example of before other college students get the same idiotic idea into their heads...
Uh, from what I've read M1 tanks aren't all that transportable by just about any standard. They're mainly designed to be transported by ship. That's the whole reason for light vehicles like the striker which don't need the golden gate bridge to cross a small river.
M1 tanks are unstoppable on the battlefield - the trick is just getting them there. It is an essential component of a foreign-soil army, but not really part of a paratroop squad...
They might make sense for a self-defense force - you'd be fighting on your own soil most of the time so transport isn't as much of a problem - trains/trucks work fine. They're really just a deterrant to keep some 2-bit neighbor from deciding you're easy pickings, unless you want to project power.
Unless they can out-turn a missile I doubt that aircraft maneuvering will have much impact in wars in the future. Sure, missile-dependency killed the US in Vietnam, but look at the missles and sensors back then! And the ROEs for that matter which tended to eliminate the advantages of missiles.
If I wanted a state-of-the-art air-force I'd build lots of UAVs that fell into a few classes:
1. Recon craft - long durability, slow speed, linger over targets. 2. Scout craft - decent range, high speed, stealthy enough to get close, lots of sensors. 3. CAP craft - high speed, decent range, half-decent stealth, lots of missile payload, great targetting/tracking, and cheap as possible - maneuvering not necessary 4. Bombers - heavy payload, med speed, good range, good stealth 5. Missile - absolute state of the art air-to-air missiles
None of these craft would be nearly as expensive as the best-of-all-worlds US aircraft.
In a situation where anything flying can be presumed hostile I'd send in CAP craft followed by bombers. In a situation where you need to be careful about what you shoot I'd send in a scout in advance - it would be expendible. If it gets shot at the CAP craft behind it would take out everything in their path. If it isn't shot at then it can get a picture of the situation and the CAP craft can act accordingly.
Nothing would be permitted to get close enough to any warcraft to require dogfighting. All engagements would be at range. Modern missiles can outmaneuver just about any aircraft.
I'd also need not nearly as many pilots. The aircraft would fly themselves for the most part, with humans stepping in at critical decision points. They would be on normal shifts and not pumped up with drugs to keep them awake on long flights. Autopilot would handle the boring parts of missions. The computer would recommend targets/actions, and people would confirm them. In large engagements the computer would get a weapons-free order and would assign targets appropriately - with expended craft automatically pairing up and heading home.
Warplanes these days are basically just cargo craft to haul smart weapons within range of the target. The machines do all the hard work already - so why continue to force humans to be missile haulers in the face of enemy fire?
Sure, my aircraft wouldn't be the best on the planet - but dollar-for-dollar I'd have a LOT of them. And I'd spare no expense in the critical parts - such as the weapons themselves.
Actually, people caught blowing things up not wearing the uniform of a foreign army are not enemy combatants or POWs under the definition of the Geneva Convention. It wasn't he intent that they be treated as such by the people who framed the conventions, and it might not be obvious why.
Part of the goal of the Geneva Conventions was to prevent total warfare - ie a situation where you shoot anybody who isn't on your side. In order to make this work you have to indicate who the people you should shoot at are. If you're a combatant you wear a uniform, if you aren't you don't.
When people make war without wearing a uniform they encourage the opposing force to target civilians - putting their own people in harm's way. It isn't like the opposing army is going to surrender simply because they can't figure out who the enemy is. Instead, they'll shoot first and ask questions later. The Geneva Conventions were written to put a stop to these practices.
A person attacking an army not wearing a uniform is not considered an enemy combatant subject to the limitations of the Geneva Conventions - they are considered a criminal. Therefore they are subject to the capturing army's criminal justice system. Soldiers captured in uniform CANNOT be tried in a court of law, but must be held pending release at the cessation of hostilities. This has been the application of the Conventions throughout history - just look at the treatment of spies as an example - they were often hung after a tribunal when captured on the battlefield (something that I don't think is appropriate today unless the capturing force cannot transport them to a civil court - which is an unlikely condition in the US army).
Now, regardless of whether the Geneva Conventions apply to terrorists/spies/etc, in general they should still be subject to Habeas Corpus and due process. It really isn't appropriate for any government to just detain people without limit without any process whatsoever. Military tribunals are at least some form of process (when they are even used), although they're a poor substitute for the civil courts. When non-US-citizen-non-lawful-combatants are captured they should probably be arraigned in US courts as soon as they can be conveniently shipped to the US, or they should be rendered to their home government.
Uh, the build usually fails if you do just about anything. I've never seen a more tempermental code base. Forget code-should-not-generate-warnings - let's start with less than 100 fatal errors per run.
In any case, even after two syncs the digest is failing on the file I downloaded...
In grad school I sat through many a class in which 95% or more of the population had no idea what was being explained after the first class or two in a course. (Those who did know probably had prior experience.) Mind you I was generally considered to be at the top of my class. However, as a rule nobody asked for the professor to go back and explain the basics of the material - it would be admitting ignorance, which is unforgivable.
I've sat down in meetings with scientists watching them go on about something which I couldn't even remotely understand (again, was at top of class in a top-10 institution) - this happens easily in fields where there are only a handful of people who are conversant. In casual circumstances I might ask lots of questions out of a genuine desire to learn, but in any type of "interview" situation (whether formal or just chatting with somebody you need to impress) I'd just nod and converse in a way that suggested I was following along fine (most scientists can at least feign understanding reasonably well if they don't get quizzed at the end). I knew that if I didn't feign understanding somebody else would, and the guy doing the talking was too concerned about talking to consider that it was possible that NOBODY he ever talked to could understand him.
Science needs more mentorship and less oneupmanship as you have suggested. Even in non-academic settings this culture is pervasive. I would never take the advice of a MD/PhD on something of practical importance just because of their qualifications - I'd need to listen to their thinking so that I could evaluate whether it is sound. I know somebody who has needed a lot of medical attention and it seems like the advice of doctors basically amounts to advising whatever everybody else advises - mostly so that they can't be sued. It seems like only rarely do they have any rational reason for their advice.
As somebody else pointed out I'm not sure this is a matter of lack-of-science as lack-of-reason. Who needs thinking when you can memorize. As a chemist it nauseates me any time I hear about a friend of my kids mention that they have to memorize the periodic table complete with numerous atomic weights/numbers/etc. Why don't we make our computer science curricula more rigorous and make people memorize the x86 instruction set (complete with opcodes in all indexing modes - who needs compilers anyway?). We waste so much time making people memorize useless junk when we could be educating them to actually be able to solve problems! (And don't get me started on named reactions!)
Ah, the beauty of a 95% confidence level. Find at least 20+ hypothesis remotely related to a dataset and you're reasonably likely to find that one of them is supportable with a 95% confidence level...
Ok, I'll confess that I might be missing some finer element to the statistics here, but the principle is sound. Evaluate enough statements against enough data points and sooner or later one which is totally false will be supportable. And the set of unknown principles in medicine/etc certainly has plenty of degrees of freedom to allow this to happen...
Don't forget to send reminder emails 2 weeks in advance, 1 week in advance, day of, 5 minutes before, and 10 minutes into the meeting when attendance is low.
It isn't like that functionality would be better implemented in a calendar application or anything like that...
Well, I think this is mostly a moot point. Ok, so 95% of the linux kernel is GPL, and 5% is BSD. You can feel free to copy out the lines that were originally BSD if you'd like to. Or you could just copy them from wherever the linux devs copied them from. Same effect.
You can certainly release your own code under a different license as a derivative work. Provided, of course, that the original license doesn't prohibit using the original code in derivative works that use a different license (like the GPL does).
For as much as you claim that there is mass infringement going on with license changing, it doesn't seem like many lawyers in the FOSS world seem to agree. Can you point to any authoritative sources in the FOSS world that agree with you (FSF, EFF, key spokesmen, etc?).
Well, the US policy isn't entirely a bad one. The logic of the US is that it is better to not do business in certain regions than to do certain things which are REALLY BAD - and in itself that isn't a bad thing. The problem comes when it enforces these kinds of laws in domains where the laws are over the top.
I don't have any problem with any company that does business in the US being forbidden to exploit children anywhere in the world. I imagine most EU nations would have similar policies. Ditto for slavery/etc.
Suppose a local law requires you to spy on people for political purposes (think Yahoo/Google in China)? Is it better to be anti-freedom, or anti-local-government? I'd think that it is just better not to do business in some places than be required to behave unethically to make a buck. Sure, there are lots of things we can disagree on, but there is also a lot of stuff that is universally accepted as being morally wrong in most of the civilized world, and you can certainly find governments that have laws that differ.
Ironically, you can't even do that. It has the advertising clause from what I've heard. That makes it incompatible with most OSS licenses.
XFree86 was abandoned mainly due to such a clause being introduced...
Re:%75 as effective as a prescription 3% the price
on
Science vs. Homeopathy
·
· Score: 1
I'm all for empirical results - half of medicine is governed by them.
But that doesn't mean that you can't use science. You don't need a theory to explain where all the missing mass is in the universe to be upfront about the fact that it seems to be there. Likewise I'd be happy to accept alternative medicine sans mechanism, but only if a controlled study showed that it actually had an effect. Otherwise it is nothing more than anecdote.
I'm not aware of any double-blind studies that show an effect of a homeopathic remedy. Until you have that, you really don't have much to talk about science-wise. How do you posit a mechanism for an effect that you can't even measure?
I think a big problem with psychology is that diagnoses are pretty general. I heard a good analogy from a practicing psychiatrist:
100 years ago a sick person might have been diagnosed with a "cough". If you treated this "cough" with the most modern of antibiotics the patient might die, leading you to conclude that antibiotics must not be very effective in treating "cough". However, another patient might recover overnight.
The problem is that "cough" is really just a symptom of a more particular problem - it could be bacterial pneumonia, or it might be some genetic disorder, or it could be viral, or who knows what else. All of these would have different treatments - all of which would be highly effective for the particular disorder they treat, and not very effective for others.
Likewise, when somebody is depressed there could be a multitude of physical and psychological causes. Some might react very well to particular drugs or other treatments, and others might not. However, based on our current knowledge of the mind all we can do is lump them all in the same treatment group and conclude that most of our treatments aren't terribly effective.
Just think about it - what advice do you give somebody when they tell you that their computer suffers from "slowness"? Do you just tell them to buy a fancy new CPU or $200 worth of RAM?
Re:%75 as effective as a prescription 3% the price
on
Science vs. Homeopathy
·
· Score: 2, Insightful
Hmm - proof that anything can get modded up on/.:)
Uh, the whole point of the double-blind random clinical trial is that it is the only known way to distinguish between drug effects and placebo effects.
What other way would you propose? Tell people to try it and ask them how they feel? The plural of anecdote isn't data. They do precisely that in double-blind trials and guess what - quite a few people report feeling better when in fact they were given only sugar pills...
And the objection isn't to the concept that less dose administered = greater effect. The objection is to the concept that you can take a preparation that is unlikely to contain a single molecule of anything but solvent and have it have any effect at all. If the does contains nothing but water, by what mechanism could it convey any effect at all?
If somebody wanted to rely on the prayers of a minister instead of taking a drug I'd not complain about the minister's actions. He would be completely up-front about the fact that he believes that what he is doing is completely supernatural and is not anything that can be relied on to have any particular outcome beyond whatever some deity intends to have happen.
The problem with homeopathy is that it masquerades as science by asserting that a particular concoction can with some degree of certainty promote a cure for a malady, and it asserts that the effect is somehow natural.
If an effect is natural then it is subject to the laws of nature. It must therefore be testable, and the fact that no effects have been found in suitable experiments forces us to conclude that it has no effect.
Collecting your money after that can be a bit of a pain, but you will get it - they are not a 2bit operation after all.
Actually, it might not even be that much of a pain - in the US at least (and if anything I imagine the UK laws are even more friendly).
In the US you basically ask for the money once you have a judgement. If they ignore you and they are a retail outfit you call up the local sheriff and they will walk into a local store and loot the cash registers until you have your judgement (they'll also go ahead and loot their fee as well). If insufficient funds are available they'll go ahead and stand there and take money from customers as they walk out with goods. If you need a LOT of money they'll go ahead and put the whole store up for auction. Of course, the store will do whatever it has to do the second the police show up regardless of whatever stalling they've done. It isn't good for customer perception to have a police officer explaining to every customer why they're paying him instead of the cashier.
Agreed that there is some potential for ZFS under GPL3, but that isn't reality yet. And neither is linux under GPL3 so we'll have to see. Having both Solaris and Linux under GPL3 would be nice - both projects could improve using code from the other, and competition suits me as a user just fine.:)
I'm not really willing to switch to BSD - support for linux is better just about everywhere, and even FUSE would be less of a traumatic change - especially if it is made to work seamlessly.
As far as JBOD goes - it really isn't an option. I have the better part of a TB of data - mostly myth video/etc. The important stuff (documents, banking records, configuration, etc) gets backed up to other media and is offsite most of the time, but I can really only handle a few GB this way and only about once a week. The remaining data isn't super-critical, but I don't want to lose it all if one of my 6 hard drives fails (not to mention downtime).
ZFS would be a better solution than RAID and would be far less expensive than JBOD and a backup solution sufficient to protect this much data continuously.
Keep in mind the I in raid is inexpensive. There are plenty of alternatives to RAID, but they tend to not be inexpensive.
Uh, what do you expect? The procedures around the handling of nuclear weapons HAVE to be classified. That means that everybody who can comment authoritatively in this forum will keep their mouths shut (any serious leaks in this forum would undoubtedly be pursued by VERY serious law enforcement). The only people who are commenting likely know little to nothing about what they're talking about. Some might walk the line - but they're not going to tell you much more than you probably could have guessed on your own...
And regardless of whether it makes sense to nuke Iran I'd hope that the US military has weapons staged appropriately to nuke any point on the planet - I mean, what else would be the point of having them? Somebody else with nukes is going to think that you're ready to blast them at a second's notice, so they're not going to hold back even if your weapons are locked up in some warehouse in Kansas.
I agree that there really is no reason that the US would nuke Iran. About the only thing that might lead to this would be an Iranian first-strike with weapons of mass destruction, and then more as a matter of policy/deterrance than any actual military need (people are more likely to think twice about dirty bombs when you turn cities into parking lots). Nuking Iran BEFORE any such attack on US soil has the opposite effect - if the US is going to nuke you no matter what then you might as well hit them with whatever you have while you have it.
The whole point of a nuclear arsenal is to deter attacks. They're the bluff that nobody dares to call - and that is what they should remain. You keep them ready to go, you tell people what will cause you to use them, and then you just hold onto them until somebody actually does what you told them not to (at which point you wipe them off the face of the Earth). You don't threaten to drag them out every time some leader of a 3rd world nation postures for the TV cameras.
And for deterrance to work everybody has to be convinced that you're ready to go on a moment's notice - that any attack can be immediately countered with devastating effect. So, you don't just keep your nukes in a closet in Topeka...
Actually, that isn't even the question any longer. The question is whether at some point we'll go ahead and repeal term limits so that we can have a madman until he dies in office...
Why is your preference to have US Marines hunting down hostile Japanese soldiers and armed civilians, on their own ground, from house to house, on a heavily fortified mainland, where the population was ready and willing to fight to the death? What is it you LIKE about more people dying?
I've found that a certain segment of the population prefers "fair" conventional warfare because it is more costly to both sides - thus making it less likely to take place. Perhaps the logic is that facing huge casualties the US might have opted to leave Japan alone and just fence them in. That of course would have been unthinkable to just about anybody living at the time, but it is an easier position to take today for some.
There is also a false sense of equality - that it doesn't matter who dies as they're all human beings. It isn't better to kill 10,000 enemies than to suffer 10,000 casualties among your own troops - both are equally wrong. That obviously misses the whole point of warfare, which is of course to achieve one's objectives. If those objectives are sufficiently just and necessary to warrant full warfare then there is no real reason to sacrifice your own soldiers just to prove your commitment to humanitarianism.
Patton said it well - the point of war isn't to die for your country - it is to make the enemy die for his country.
Another factor that many forget is that WWII was largely fought in a state of total warfare (especially in Japan, but also elsewhere). Who WASN'T involved in the war effort in some way? If you weren't fighting you were making bullets, steel for ships, food for rations, or gauze to patch up your fighters to put them back on the field shooting. Nations devoted huge percentages of their GDP to the war effort. In that kind of state there really isn't much distinction between bombing a tank factory or bombing a milk factory - the only thing the milk is being used for is to feed people working in the tank factory...
This isn't meant to be a post in favor of total warfare or anything like that - but more a statement of fact. WWII was essentially fought until the ability to fight was lost - Germany didn't surrender until the tanks on the East and West essentially converged. Japan didn't surrender until it faced complete nuclear annihilation. While military targets always came first (the tank in the field is more dangerous right now than the steamshovel that will dig up coal that will make steel for the next tank), anything that reduced the ability to make war was probably considered fair game.
However, Sun has not been willing to license their ZFS-related patents under anything other than the CDDL, which will prevent eventual inclusion in the kernel. As a FUSE filesystem it will probably never be adopted in a widespread manner (it would be a real pain to boot off of it I'd think). In theory the copyright issues can be circumvented by reimplementing the filesystem, but this won't get around the patent issues.
So, if the court in any way invalidates the various patents or weakens them, it improves the chances of mainstream kernel support for ZFS, or something that is more ZFS-like (depending on what aspects of ZFS are unencumbered).
So, this could have both positive and implications for linux. It would be nice to see a technology like ZFS incorporated in linux - a good percentage of my drive space is unused due to an inability to evenly map it across RAID-5 partitions in a way that won't harm disk access times (my drives aren't all of identical size, and I also have boot/swap/root partitions to squeeze in someplace that won't use RAID-5). With ZFS you don't worry about stuff like that and just allocate whatever space you want to allocate and the OS figures out how to best use it. Plus, the copy-on-write technology makes it a lot faster than RAID-5...
Well, it might except that the ABI changes are probably designed largely to make life as difficult as possible for anybody who distributes binary drivers. Helping them to cope with the changes would sort-of defeat the purpose in making the changes in the first place.
The kernel developers aren't exactly supportive of non-GPL binary modules...
Slashdot Mirror
I dunno - this looks like an intentional bomb hoax to me - regardless of what the person claims. This person put lives in jeopardy - including her own and those around her - when she created a situation that would for all intents and purposes appear to be a life-and-death shoot on sight suicide bombing incident. This also tied up police forces in dealing with a hoax that could be better employed watching out for real threats.
She should be punished - probably not with 30 years in prison, but some jail time and a heavy fine would probably be appropriate. It seems inconceivable to me that anybody would not expect the type of reaction that she received.
Now, if it turns out that she is mentally retarded and not some college student with a chip on her shoulder then I could see a little leniency being in order.
I'm sorry - I'm against TSA overreaction and the way that police tend to use storm trooper tactics where they are unnecessary and put lives in danger. However, a person standing in a crowded area with an apparent explosive device warrants exactly these sorts of tactics. The whole problem with suicide bombers is that they need to be stopped BEFORE they carry out their intended crimes. The police apparently showed tremendous restraint, running into a situation that could have very well ended their lives while using tremendous care in the use of lethal force. The hoaxer is very fortunate that she wasn't shot. If she were shot I'm sure there would have been an uproar nonetheless and as a result she needs to be made an example of before other college students get the same idiotic idea into their heads...
Uh, from what I've read M1 tanks aren't all that transportable by just about any standard. They're mainly designed to be transported by ship. That's the whole reason for light vehicles like the striker which don't need the golden gate bridge to cross a small river.
M1 tanks are unstoppable on the battlefield - the trick is just getting them there. It is an essential component of a foreign-soil army, but not really part of a paratroop squad...
They might make sense for a self-defense force - you'd be fighting on your own soil most of the time so transport isn't as much of a problem - trains/trucks work fine. They're really just a deterrant to keep some 2-bit neighbor from deciding you're easy pickings, unless you want to project power.
Unless they can out-turn a missile I doubt that aircraft maneuvering will have much impact in wars in the future. Sure, missile-dependency killed the US in Vietnam, but look at the missles and sensors back then! And the ROEs for that matter which tended to eliminate the advantages of missiles.
If I wanted a state-of-the-art air-force I'd build lots of UAVs that fell into a few classes:
1. Recon craft - long durability, slow speed, linger over targets.
2. Scout craft - decent range, high speed, stealthy enough to get close, lots of sensors.
3. CAP craft - high speed, decent range, half-decent stealth, lots of missile payload, great targetting/tracking, and cheap as possible - maneuvering not necessary
4. Bombers - heavy payload, med speed, good range, good stealth
5. Missile - absolute state of the art air-to-air missiles
None of these craft would be nearly as expensive as the best-of-all-worlds US aircraft.
In a situation where anything flying can be presumed hostile I'd send in CAP craft followed by bombers. In a situation where you need to be careful about what you shoot I'd send in a scout in advance - it would be expendible. If it gets shot at the CAP craft behind it would take out everything in their path. If it isn't shot at then it can get a picture of the situation and the CAP craft can act accordingly.
Nothing would be permitted to get close enough to any warcraft to require dogfighting. All engagements would be at range. Modern missiles can outmaneuver just about any aircraft.
I'd also need not nearly as many pilots. The aircraft would fly themselves for the most part, with humans stepping in at critical decision points. They would be on normal shifts and not pumped up with drugs to keep them awake on long flights. Autopilot would handle the boring parts of missions. The computer would recommend targets/actions, and people would confirm them. In large engagements the computer would get a weapons-free order and would assign targets appropriately - with expended craft automatically pairing up and heading home.
Warplanes these days are basically just cargo craft to haul smart weapons within range of the target. The machines do all the hard work already - so why continue to force humans to be missile haulers in the face of enemy fire?
Sure, my aircraft wouldn't be the best on the planet - but dollar-for-dollar I'd have a LOT of them. And I'd spare no expense in the critical parts - such as the weapons themselves.
Actually, people caught blowing things up not wearing the uniform of a foreign army are not enemy combatants or POWs under the definition of the Geneva Convention. It wasn't he intent that they be treated as such by the people who framed the conventions, and it might not be obvious why.
Part of the goal of the Geneva Conventions was to prevent total warfare - ie a situation where you shoot anybody who isn't on your side. In order to make this work you have to indicate who the people you should shoot at are. If you're a combatant you wear a uniform, if you aren't you don't.
When people make war without wearing a uniform they encourage the opposing force to target civilians - putting their own people in harm's way. It isn't like the opposing army is going to surrender simply because they can't figure out who the enemy is. Instead, they'll shoot first and ask questions later. The Geneva Conventions were written to put a stop to these practices.
A person attacking an army not wearing a uniform is not considered an enemy combatant subject to the limitations of the Geneva Conventions - they are considered a criminal. Therefore they are subject to the capturing army's criminal justice system. Soldiers captured in uniform CANNOT be tried in a court of law, but must be held pending release at the cessation of hostilities. This has been the application of the Conventions throughout history - just look at the treatment of spies as an example - they were often hung after a tribunal when captured on the battlefield (something that I don't think is appropriate today unless the capturing force cannot transport them to a civil court - which is an unlikely condition in the US army).
Now, regardless of whether the Geneva Conventions apply to terrorists/spies/etc, in general they should still be subject to Habeas Corpus and due process. It really isn't appropriate for any government to just detain people without limit without any process whatsoever. Military tribunals are at least some form of process (when they are even used), although they're a poor substitute for the civil courts. When non-US-citizen-non-lawful-combatants are captured they should probably be arraigned in US courts as soon as they can be conveniently shipped to the US, or they should be rendered to their home government.
Uh, the build usually fails if you do just about anything. I've never seen a more tempermental code base. Forget code-should-not-generate-warnings - let's start with less than 100 fatal errors per run.
In any case, even after two syncs the digest is failing on the file I downloaded...
In grad school I sat through many a class in which 95% or more of the population had no idea what was being explained after the first class or two in a course. (Those who did know probably had prior experience.) Mind you I was generally considered to be at the top of my class. However, as a rule nobody asked for the professor to go back and explain the basics of the material - it would be admitting ignorance, which is unforgivable.
I've sat down in meetings with scientists watching them go on about something which I couldn't even remotely understand (again, was at top of class in a top-10 institution) - this happens easily in fields where there are only a handful of people who are conversant. In casual circumstances I might ask lots of questions out of a genuine desire to learn, but in any type of "interview" situation (whether formal or just chatting with somebody you need to impress) I'd just nod and converse in a way that suggested I was following along fine (most scientists can at least feign understanding reasonably well if they don't get quizzed at the end). I knew that if I didn't feign understanding somebody else would, and the guy doing the talking was too concerned about talking to consider that it was possible that NOBODY he ever talked to could understand him.
Science needs more mentorship and less oneupmanship as you have suggested. Even in non-academic settings this culture is pervasive. I would never take the advice of a MD/PhD on something of practical importance just because of their qualifications - I'd need to listen to their thinking so that I could evaluate whether it is sound. I know somebody who has needed a lot of medical attention and it seems like the advice of doctors basically amounts to advising whatever everybody else advises - mostly so that they can't be sued. It seems like only rarely do they have any rational reason for their advice.
As somebody else pointed out I'm not sure this is a matter of lack-of-science as lack-of-reason. Who needs thinking when you can memorize. As a chemist it nauseates me any time I hear about a friend of my kids mention that they have to memorize the periodic table complete with numerous atomic weights/numbers/etc. Why don't we make our computer science curricula more rigorous and make people memorize the x86 instruction set (complete with opcodes in all indexing modes - who needs compilers anyway?). We waste so much time making people memorize useless junk when we could be educating them to actually be able to solve problems! (And don't get me started on named reactions!)
Ah, the beauty of a 95% confidence level. Find at least 20+ hypothesis remotely related to a dataset and you're reasonably likely to find that one of them is supportable with a 95% confidence level...
Ok, I'll confess that I might be missing some finer element to the statistics here, but the principle is sound. Evaluate enough statements against enough data points and sooner or later one which is totally false will be supportable. And the set of unknown principles in medicine/etc certainly has plenty of degrees of freedom to allow this to happen...
Don't forget to send reminder emails 2 weeks in advance, 1 week in advance, day of, 5 minutes before, and 10 minutes into the meeting when attendance is low.
It isn't like that functionality would be better implemented in a calendar application or anything like that...
The irony is that everybody is obsessed with click-throughs, and yet impressions are probably the most important effect of an ad.
How many click-throughs do you get on a super-bowl ad? Yet people pay a fortune for them. There is more to advertising than counting clicks...
Well, I think this is mostly a moot point. Ok, so 95% of the linux kernel is GPL, and 5% is BSD. You can feel free to copy out the lines that were originally BSD if you'd like to. Or you could just copy them from wherever the linux devs copied them from. Same effect.
You can certainly release your own code under a different license as a derivative work. Provided, of course, that the original license doesn't prohibit using the original code in derivative works that use a different license (like the GPL does).
For as much as you claim that there is mass infringement going on with license changing, it doesn't seem like many lawyers in the FOSS world seem to agree. Can you point to any authoritative sources in the FOSS world that agree with you (FSF, EFF, key spokesmen, etc?).
Well, the US policy isn't entirely a bad one. The logic of the US is that it is better to not do business in certain regions than to do certain things which are REALLY BAD - and in itself that isn't a bad thing. The problem comes when it enforces these kinds of laws in domains where the laws are over the top.
I don't have any problem with any company that does business in the US being forbidden to exploit children anywhere in the world. I imagine most EU nations would have similar policies. Ditto for slavery/etc.
Suppose a local law requires you to spy on people for political purposes (think Yahoo/Google in China)? Is it better to be anti-freedom, or anti-local-government? I'd think that it is just better not to do business in some places than be required to behave unethically to make a buck. Sure, there are lots of things we can disagree on, but there is also a lot of stuff that is universally accepted as being morally wrong in most of the civilized world, and you can certainly find governments that have laws that differ.
Ironically, you can't even do that. It has the advertising clause from what I've heard. That makes it incompatible with most OSS licenses.
XFree86 was abandoned mainly due to such a clause being introduced...
I'm all for empirical results - half of medicine is governed by them.
But that doesn't mean that you can't use science. You don't need a theory to explain where all the missing mass is in the universe to be upfront about the fact that it seems to be there. Likewise I'd be happy to accept alternative medicine sans mechanism, but only if a controlled study showed that it actually had an effect. Otherwise it is nothing more than anecdote.
I'm not aware of any double-blind studies that show an effect of a homeopathic remedy. Until you have that, you really don't have much to talk about science-wise. How do you posit a mechanism for an effect that you can't even measure?
I think a big problem with psychology is that diagnoses are pretty general. I heard a good analogy from a practicing psychiatrist:
100 years ago a sick person might have been diagnosed with a "cough". If you treated this "cough" with the most modern of antibiotics the patient might die, leading you to conclude that antibiotics must not be very effective in treating "cough". However, another patient might recover overnight.
The problem is that "cough" is really just a symptom of a more particular problem - it could be bacterial pneumonia, or it might be some genetic disorder, or it could be viral, or who knows what else. All of these would have different treatments - all of which would be highly effective for the particular disorder they treat, and not very effective for others.
Likewise, when somebody is depressed there could be a multitude of physical and psychological causes. Some might react very well to particular drugs or other treatments, and others might not. However, based on our current knowledge of the mind all we can do is lump them all in the same treatment group and conclude that most of our treatments aren't terribly effective.
Just think about it - what advice do you give somebody when they tell you that their computer suffers from "slowness"? Do you just tell them to buy a fancy new CPU or $200 worth of RAM?
Hmm - proof that anything can get modded up on /. :)
Uh, the whole point of the double-blind random clinical trial is that it is the only known way to distinguish between drug effects and placebo effects.
What other way would you propose? Tell people to try it and ask them how they feel? The plural of anecdote isn't data. They do precisely that in double-blind trials and guess what - quite a few people report feeling better when in fact they were given only sugar pills...
And the objection isn't to the concept that less dose administered = greater effect. The objection is to the concept that you can take a preparation that is unlikely to contain a single molecule of anything but solvent and have it have any effect at all. If the does contains nothing but water, by what mechanism could it convey any effect at all?
If somebody wanted to rely on the prayers of a minister instead of taking a drug I'd not complain about the minister's actions. He would be completely up-front about the fact that he believes that what he is doing is completely supernatural and is not anything that can be relied on to have any particular outcome beyond whatever some deity intends to have happen.
The problem with homeopathy is that it masquerades as science by asserting that a particular concoction can with some degree of certainty promote a cure for a malady, and it asserts that the effect is somehow natural.
If an effect is natural then it is subject to the laws of nature. It must therefore be testable, and the fact that no effects have been found in suitable experiments forces us to conclude that it has no effect.
Actually, it might not even be that much of a pain - in the US at least (and if anything I imagine the UK laws are even more friendly).
In the US you basically ask for the money once you have a judgement. If they ignore you and they are a retail outfit you call up the local sheriff and they will walk into a local store and loot the cash registers until you have your judgement (they'll also go ahead and loot their fee as well). If insufficient funds are available they'll go ahead and stand there and take money from customers as they walk out with goods. If you need a LOT of money they'll go ahead and put the whole store up for auction. Of course, the store will do whatever it has to do the second the police show up regardless of whatever stalling they've done. It isn't good for customer perception to have a police officer explaining to every customer why they're paying him instead of the cashier.
Agreed that there is some potential for ZFS under GPL3, but that isn't reality yet. And neither is linux under GPL3 so we'll have to see. Having both Solaris and Linux under GPL3 would be nice - both projects could improve using code from the other, and competition suits me as a user just fine. :)
I'm not really willing to switch to BSD - support for linux is better just about everywhere, and even FUSE would be less of a traumatic change - especially if it is made to work seamlessly.
As far as JBOD goes - it really isn't an option. I have the better part of a TB of data - mostly myth video/etc. The important stuff (documents, banking records, configuration, etc) gets backed up to other media and is offsite most of the time, but I can really only handle a few GB this way and only about once a week. The remaining data isn't super-critical, but I don't want to lose it all if one of my 6 hard drives fails (not to mention downtime).
ZFS would be a better solution than RAID and would be far less expensive than JBOD and a backup solution sufficient to protect this much data continuously.
Keep in mind the I in raid is inexpensive. There are plenty of alternatives to RAID, but they tend to not be inexpensive.
I imagine that would depend on where Israel got their plutonium. Your question might be less applicable than you think... :)
Uh, what do you expect? The procedures around the handling of nuclear weapons HAVE to be classified. That means that everybody who can comment authoritatively in this forum will keep their mouths shut (any serious leaks in this forum would undoubtedly be pursued by VERY serious law enforcement). The only people who are commenting likely know little to nothing about what they're talking about. Some might walk the line - but they're not going to tell you much more than you probably could have guessed on your own...
And regardless of whether it makes sense to nuke Iran I'd hope that the US military has weapons staged appropriately to nuke any point on the planet - I mean, what else would be the point of having them? Somebody else with nukes is going to think that you're ready to blast them at a second's notice, so they're not going to hold back even if your weapons are locked up in some warehouse in Kansas.
I agree that there really is no reason that the US would nuke Iran. About the only thing that might lead to this would be an Iranian first-strike with weapons of mass destruction, and then more as a matter of policy/deterrance than any actual military need (people are more likely to think twice about dirty bombs when you turn cities into parking lots). Nuking Iran BEFORE any such attack on US soil has the opposite effect - if the US is going to nuke you no matter what then you might as well hit them with whatever you have while you have it.
The whole point of a nuclear arsenal is to deter attacks. They're the bluff that nobody dares to call - and that is what they should remain. You keep them ready to go, you tell people what will cause you to use them, and then you just hold onto them until somebody actually does what you told them not to (at which point you wipe them off the face of the Earth). You don't threaten to drag them out every time some leader of a 3rd world nation postures for the TV cameras.
And for deterrance to work everybody has to be convinced that you're ready to go on a moment's notice - that any attack can be immediately countered with devastating effect. So, you don't just keep your nukes in a closet in Topeka...
Actually, that isn't even the question any longer. The question is whether at some point we'll go ahead and repeal term limits so that we can have a madman until he dies in office...
Why is your preference to have US Marines hunting down hostile Japanese soldiers and armed civilians, on their own ground, from house to house, on a heavily fortified mainland, where the population was ready and willing to fight to the death? What is it you LIKE about more people dying?
I've found that a certain segment of the population prefers "fair" conventional warfare because it is more costly to both sides - thus making it less likely to take place. Perhaps the logic is that facing huge casualties the US might have opted to leave Japan alone and just fence them in. That of course would have been unthinkable to just about anybody living at the time, but it is an easier position to take today for some.
There is also a false sense of equality - that it doesn't matter who dies as they're all human beings. It isn't better to kill 10,000 enemies than to suffer 10,000 casualties among your own troops - both are equally wrong. That obviously misses the whole point of warfare, which is of course to achieve one's objectives. If those objectives are sufficiently just and necessary to warrant full warfare then there is no real reason to sacrifice your own soldiers just to prove your commitment to humanitarianism.
Patton said it well - the point of war isn't to die for your country - it is to make the enemy die for his country.
Another factor that many forget is that WWII was largely fought in a state of total warfare (especially in Japan, but also elsewhere). Who WASN'T involved in the war effort in some way? If you weren't fighting you were making bullets, steel for ships, food for rations, or gauze to patch up your fighters to put them back on the field shooting. Nations devoted huge percentages of their GDP to the war effort. In that kind of state there really isn't much distinction between bombing a tank factory or bombing a milk factory - the only thing the milk is being used for is to feed people working in the tank factory...
This isn't meant to be a post in favor of total warfare or anything like that - but more a statement of fact. WWII was essentially fought until the ability to fight was lost - Germany didn't surrender until the tanks on the East and West essentially converged. Japan didn't surrender until it faced complete nuclear annihilation. While military targets always came first (the tank in the field is more dangerous right now than the steamshovel that will dig up coal that will make steel for the next tank), anything that reduced the ability to make war was probably considered fair game.
However, Sun has not been willing to license their ZFS-related patents under anything other than the CDDL, which will prevent eventual inclusion in the kernel. As a FUSE filesystem it will probably never be adopted in a widespread manner (it would be a real pain to boot off of it I'd think). In theory the copyright issues can be circumvented by reimplementing the filesystem, but this won't get around the patent issues.
So, if the court in any way invalidates the various patents or weakens them, it improves the chances of mainstream kernel support for ZFS, or something that is more ZFS-like (depending on what aspects of ZFS are unencumbered).
So, this could have both positive and implications for linux. It would be nice to see a technology like ZFS incorporated in linux - a good percentage of my drive space is unused due to an inability to evenly map it across RAID-5 partitions in a way that won't harm disk access times (my drives aren't all of identical size, and I also have boot/swap/root partitions to squeeze in someplace that won't use RAID-5). With ZFS you don't worry about stuff like that and just allocate whatever space you want to allocate and the OS figures out how to best use it. Plus, the copy-on-write technology makes it a lot faster than RAID-5...
There is no legitimate use for boardband internet access. Anything broadband internet access can do UUCP can do better.
So, anybody else have a favorite protocol?
Well, it might except that the ABI changes are probably designed largely to make life as difficult as possible for anybody who distributes binary drivers. Helping them to cope with the changes would sort-of defeat the purpose in making the changes in the first place.
The kernel developers aren't exactly supportive of non-GPL binary modules...