Actually, there is many more questions about the methodology. First off, is a number of 42 articles representative of the universe of all articles in both encyclopaedias? For what I know, there are millions of articles in them, 42 random articles seems to be a rather small sample.
Moreover, what is the standard deviation and mean? Maybe on average, wikipedia has less errors, but a wikipedia article may contain so many errors that it becomes worthless. That's a piece of information I'd like to see.
Finally, there should be some sort of quality rating for the articles. A small uninformative article with no errors is much worse than a big article with lots of information but some inconsistencies.
Actually, it depends mainly on the application. Telecom companies for one, run their commuting centrals in fault-tolerant hardware/software, that work simultaneously processing the same data. In this way, if one of the components (disk, processors, boards, etc.) fails, the mirror can assume immediately. However, those companies buy very a specific hardware and software from (usually) a single vendor, which are necessary to provide their specific services. Of course vendors charge an obscene price for them, but in this case it is worth to pay it anyway, because a cluster solution would be very complex and is not available off-the-shelf.
On the opposite side, if you not running an aplication that requires hardware/software so specific, a cluster may be a good solution. It is extensible, customizable, and you can choose from multiple vendors and technologies.
I used to have Mandrake installed, which took care of all the gory details for me, and now I've just decided to give Gentoo a try. So far, here are the bad aspects of gentoo compared to my previous distribution:
* You have to be willing to spend a big amount of time tweaking with your system. There's a lot of gotcha's (the USE flag, for instance), and not everything works as you think they would.
* the portage directory (which contains data for the portage system, the package manager of Gentoo) takes up a lot of space. If you running out of space, or have a small hd, or have a dual boot system with a limited space for a linux installation, that's something to consider
* Some aspects of the system (such as the boot file organization and scripts) are set differently from other linux distributions. It's not particulary hard to grasp, but you have to learn it anyway.
On the bright side, what I see is what many have said here before, such as up-to-date packages, good forums, learning opportunities, etc.
My advice is, if you've got the time to spare and the willingness to learn, install it. On the other hand, if you just want a more friendly distro and get things working right away, maybe Gentoo won't fit you that well.
He may not be trying to get rid of the old woman, but apparently he has not replaced her wheel chair as well. But even he is planning to, would you trust someone who set up a jet engine in your whell chair while you were out in the weekend?
I belive you are mistaken, patents actually are one of the greatest inventions to incentivate intellectual production. According to economists, and even empirically, is easy to observe that unless you have a strong incentive to produce and the rights to your production assured, its very unlikely that you will move a finger to do anything at all. That's what happened for instance in the russian revolution, when all excedent grains were confiscated by the government. The farmers stopped planting, and millions starved.
You can arque that in the free software movement, one doesn't get nothing for the work he is willing to do. That's not true, there is still strong incentives for some contributors, like peer recogntion for instance. Unfortunatelly, profit is still the most appealing incentive to production, and patents are the mechanism that guarantee that the creator can profit from his invention
It is not the patent idea that is wrong, is the process of granting them that should be improved and revised to refrain this kind of stupid thing from happening again
Although they may eventually reach some interesting results, it seems very unlikely that this research will change the way diagnostics are made nowadays. I don't see how someone could replace or question physiological exams based on a source of information so unreliable and subject to noises as this.
Anyway, these guys have already prooved that, in some situations, is very hard to get useful information throught sound, even when you know what you may be looking for.
"This sounds like a fun thing to play around with, but I certainly don't see myself using it as a normal web browser. I'll most likely stick with my keyboard"
Well, while you probably have the option to pick your keyboard, there are many handcapped people in the world that would find amazing just surf the Web all by themselves. This will be much more than a toy for them.
Just like in the case of drugs, the vast majority of the population is against it, but as long as there are economic incentives for drugdealers (in other words, demand), the problem will never be solved.
In fact, is even harder to convince spam consumers that they are doing a bad thing, because spam is considered harmless: there is no violence involved, no family breakdowns, no deaths. For all the Sotos in the world, it will always be thought as just another hobbie.
It's not different than tires on cars
That's not quite accurate. In engineering, you are dealing whith quantifiable measures, in the sense that you know exactly what kind of conditions and stresses the materials are going to be exposed to.
I have done structural calculus in college, and belive me, you can literatelly put 3 building on the top of a regular one, just because they introduce all those huge security coefficients (by law and prudence) when figuring out the structure.So, you have known conditions and security coefficients
When it comes to software, you don't have any of these. know A priori, it is very hard to predict all kinds of possible errors and conditions that the system will have to cope to. Security coefficients are also out of question, how can you implement a security measure for a unpredicted situation? God, you can't nearly prove that your implementation is correct!
Only a strong quality assurance and test program can improve the reliability of your program, and of course, it is not unusual that a specific extremelly situation is not foreseen.
How do you figure one could quickly, easily and tracelessly hack a system whose only accessible inputs are things like pressure sensors?
The device has a computer and circuits to monitor and control the mechanical legs movement based on the movements of the pilot. You're right, it's probably very hard to break into the embbebed program, but the point here is not to become "root" of the system, but to break its control capabilities.
This could be done with eletromagnetism for instace, to erase parts of the memory or the code in the ROM (it has been done before in ATMs). God, even water can damage the circuits if they are not properly protected. A shot or nearby explosion can also cause some critical glitches.
Once damaged, it could simply stop, or show a unpredictable behavior, but it certaily will become useless.
It's great to see Napster trying to come back after the "revolution" it stired up, but let's face it, Napster is no longer a cutting-edge technology or concept. They are only trying to catch up and make some dought. This is not bad, it's just not that cool anymore.
The question is not whether someone will or will not turn java down because it is not free, but how much more wildly adopted and improved the language and the VM can become.
Their case is not bullshit, they know that, all in all, there is only battle to win. If they can get the big ones to pay them licenses, the smaller ones will surely follow the same road.
The rationale is quite simple. If large corporations, which have the resources, expertise and time to fight in court againt SCO were defeated, the small ones will look like sitting ducks. No owner of a small business will risk a war that companies much times bigger couldn't win.
On the other hand, SCO knows that they have power now to intimidate and force the little guys to pay licenses, but if these licenses are bought and later on it turns out that SCO did not have the right to charge them, they are going to be sued for so many firms and so many reasons that they can even imagine right now.
Actually, this is not a good example of working monopoly. Sure it is a monopoly, but not one that is meant to last if it goes on like this.
A good monopoly was what the Standard Oil Company (Rockefeller's oil empire) was in the beginning of the century. They sold gas at a relatively low price, customers were pleased, and for a long time, nobody foresee how harmful to the community whole situation would become in the long run.
In fact, it is a good thing that Microsoft is so reckless about IE right now, because this is exactly what is going to open room for its fall.
There are, as Spafford claims, too many people in category 3 who think they belong to the category 4, and that's the primary target audience of the book
People who think they are in category 4 wont buy the book, because they belive they dont need it. So who is the target audience anyway?:-)
When you start receiving money to develop the project, is the company still willing to keep it open sourced? If not, you should consider whether you are going to take the money, or keep on working for free on behalf of the community. Either way, make a conscient decision.
Slashdot Mirror
Mona Lisa is a woman, how can any software possibly tell what she is really thinking?
Moreover, what is the standard deviation and mean? Maybe on average, wikipedia has less errors, but a wikipedia article may contain so many errors that it becomes worthless. That's a piece of information I'd like to see.
Finally, there should be some sort of quality rating for the articles. A small uninformative article with no errors is much worse than a big article with lots of information but some inconsistencies.
Actually, it depends mainly on the application. Telecom companies for one, run their commuting centrals in fault-tolerant hardware/software, that work simultaneously processing the same data. In this way, if one of the components (disk, processors, boards, etc.) fails, the mirror can assume immediately. However, those companies buy very a specific hardware and software from (usually) a single vendor, which are necessary to provide their specific services. Of course vendors charge an obscene price for them, but in this case it is worth to pay it anyway, because a cluster solution would be very complex and is not available off-the-shelf. On the opposite side, if you not running an aplication that requires hardware/software so specific, a cluster may be a good solution. It is extensible, customizable, and you can choose from multiple vendors and technologies.
* You have to be willing to spend a big amount of time tweaking with your system. There's a lot of gotcha's (the USE flag, for instance), and not everything works as you think they would.
* the portage directory (which contains data for the portage system, the package manager of Gentoo) takes up a lot of space. If you running out of space, or have a small hd, or have a dual boot system with a limited space for a linux installation, that's something to consider
* Some aspects of the system (such as the boot file organization and scripts) are set differently from other linux distributions. It's not particulary hard to grasp, but you have to learn it anyway.
On the bright side, what I see is what many have said here before, such as up-to-date packages, good forums, learning opportunities, etc.
My advice is, if you've got the time to spare and the willingness to learn, install it. On the other hand, if you just want a more friendly distro and get things working right away, maybe Gentoo won't fit you that well.
Does that mean when you watch porn on the Web it is not safe sex anymore? Damn it!!!
He may not be trying to get rid of the old woman, but apparently he has not replaced her wheel chair as well. But even he is planning to, would you trust someone who set up a jet engine in your whell chair while you were out in the weekend?
If green tea is all that good, just imagine when they try beer!
You can arque that in the free software movement, one doesn't get nothing for the work he is willing to do. That's not true, there is still strong incentives for some contributors, like peer recogntion for instance. Unfortunatelly, profit is still the most appealing incentive to production, and patents are the mechanism that guarantee that the creator can profit from his invention
It is not the patent idea that is wrong, is the process of granting them that should be improved and revised to refrain this kind of stupid thing from happening again
Anyway, these guys have already prooved that, in some situations, is very hard to get useful information throught sound, even when you know what you may be looking for.
Well, while you probably have the option to pick your keyboard, there are many handcapped people in the world that would find amazing just surf the Web all by themselves. This will be much more than a toy for them.
In fact, is even harder to convince spam consumers that they are doing a bad thing, because spam is considered harmless: there is no violence involved, no family breakdowns, no deaths. For all the Sotos in the world, it will always be thought as just another hobbie.
I have done structural calculus in college, and belive me, you can literatelly put 3 building on the top of a regular one, just because they introduce all those huge security coefficients (by law and prudence) when figuring out the structure.So, you have known conditions and security coefficients
When it comes to software, you don't have any of these. know A priori, it is very hard to predict all kinds of possible errors and conditions that the system will have to cope to. Security coefficients are also out of question, how can you implement a security measure for a unpredicted situation? God, you can't nearly prove that your implementation is correct!
Only a strong quality assurance and test program can improve the reliability of your program, and of course, it is not unusual that a specific extremelly situation is not foreseen.
The device has a computer and circuits to monitor and control the mechanical legs movement based on the movements of the pilot. You're right, it's probably very hard to break into the embbebed program, but the point here is not to become "root" of the system, but to break its control capabilities.
This could be done with eletromagnetism for instace, to erase parts of the memory or the code in the ROM (it has been done before in ATMs). God, even water can damage the circuits if they are not properly protected. A shot or nearby explosion can also cause some critical glitches.
Once damaged, it could simply stop, or show a unpredictable behavior, but it certaily will become useless.
It's great to see Napster trying to come back after the "revolution" it stired up, but let's face it, Napster is no longer a cutting-edge technology or concept. They are only trying to catch up and make some dought. This is not bad, it's just not that cool anymore.
The question is not whether someone will or will not turn java down because it is not free, but how much more wildly adopted and improved the language and the VM can become.
The rationale is quite simple. If large corporations, which have the resources, expertise and time to fight in court againt SCO were defeated, the small ones will look like sitting ducks. No owner of a small business will risk a war that companies much times bigger couldn't win.
On the other hand, SCO knows that they have power now to intimidate and force the little guys to pay licenses, but if these licenses are bought and later on it turns out that SCO did not have the right to charge them, they are going to be sued for so many firms and so many reasons that they can even imagine right now.
Their strategy makes perfect sense to me.
A good monopoly was what the Standard Oil Company (Rockefeller's oil empire) was in the beginning of the century. They sold gas at a relatively low price, customers were pleased, and for a long time, nobody foresee how harmful to the community whole situation would become in the long run.
In fact, it is a good thing that Microsoft is so reckless about IE right now, because this is exactly what is going to open room for its fall.
There are, as Spafford claims, too many people in category 3 who think they belong to the category 4, and that's the primary target audience of the book
:-)
People who think they are in category 4 wont buy the book, because they belive they dont need it. So who is the target audience anyway?
When you start receiving money to develop the project, is the company still willing to keep it open sourced? If not, you should consider whether you are going to take the money, or keep on working for free on behalf of the community. Either way, make a conscient decision.