The thing is, you can do most of this from a user account. Users have access to a lot of stuff that can be just as bad, and once you have access to a user account, it's fairly easy to get root.
Once someone with malicious intent can access your machine, you can pretty much kiss it goodbye. You can never really be sure that root wasn't compromised without an extensive investigation.
You have a command line, emacs, vi, the gcc suite, perl, clisp and sbcl. What more could you possibly want?
As long as there's a terminal available and gcc, you just can't complain about lack of power user features in Linux.
He complains about the multiple package management programs. There's no problem here, since they all use the same underlying database, and a newbie would never know about the command line ones, and wouldn't need to.
A new user will get along just fine with the simplicity of Ubuntu on the desktop. A power user will hit the command line and have no problems.
It seems like this guy knows just enough about Debian to be dangerous, and is now cranky that Ubuntu is slightly different.
I'm not a materials scientist either, but I did take a structural engineering class and sleep in a Holiday Inn express last night.
There are many classifications of materials that could be interpreted as "brittle." Brittle is much too general a term to be used in engineering, so you have to be suspicious of the news article.
You can measure tensile strength, which is a measure of how much something can bend until it break. There's another measurement where you find how much something can bend until it permanently deforms, so that it won't go back to its original state. Each of these could be called "flexibility" but that doesn't tell you the whole story.
Carbon fiber when it fails may fail explosively and shatter, while a soft metal would simply deform slowly when bent far enough. This could be called "brittleness" but it really has little to do with the actual engineering problem, since if you design the carbon fiber component to high enough tolerances, you're not worried about it breaking, since the force required to break it would be so huge you'd have other, much bigger problems besides the breaking of the part. (Like, how do we get the people out of the broken plane when Godzilla is about to eat it?)
It would be easy to criticize the engineering of the plane on the news, because nobody is going to sit there for three months to check everything out -- they'll watch the demo of a small piece of carbon fiber breaking and think, "Oh my god, that could be the wing of my plane!"
Complaints can be valid all day long, but that doesn't necessarily make them helpful.
When you're managing a project, usually you have to make decisions that are going to piss some people off. Those people can whine about it forever or simply realize that the decision had to be made and shut up about it. If they feel a bad decision is THAT big a deal, then it's time to put up or shut up, and show everyone else how wrong they are. That's productive and helpful, complaining isn't.
I find OpenOffice to be really good software, and it's improving rapidly. I don't see the problem in the grand scheme of things.
I'm not sure what depth of statistical analysis you're talking about, but I've found statistical analysis to be exceptionally difficult when applied to Electrical Engineering, and I know I'm not alone here -- it was by far the most difficult post-graduate class at my school.
I can confidently say that nobody who's graduating with me has a complete grasp of all of the statistical tools we were taught. Enough to get by, yes, but most of the things are extremely counterintuitive and easily misused, and this is by people who are really good at math.
I have to laugh when I hear statistics in the news now, because it's all such propaganda. Any time you hear "has been linked to" in an article you know you're about to wade through a steaming pile of bullshit.
So in Linux, when you want to swap out the scheduler to one that devotes most of the CPU time to the process owning the current focused window in X, it's just a user level module away, since the scheduler is modular and can be changed at run time?
"And in my opinion, anyone who claims to be a university graduate should be required to take and understand calculus."
Well, I think that everyone should have to take up through Calculus III, since it's been around almost as long, and you can't understand weather reports in great detail unless you know how gradients work.
And differential equations, linear algebra, and advanced probability and statistics, too. You're just not well educated unless you can compute a covariance matrix in your head.
You should also know about the fundamentals of electromagnetic theory and how electrical waves propagate through a transmission line. And if you don't know how to do a link-budget analysis of a satellite link and also calculate where that satellite will be in relation to a ground station, then you can't really call yourself educated.
And how many people who graduate today actually know anything about state space control systems equations? Precious few, and yet I think it's very important since it's used all the time.
Not only that, but people only learn one foreign language, and it's usually something that's only come into fashion recently like French. All courses should be taught in Latin, or better yet, Aramaic. It worked for many centuries, why shouldn't it work today? We graduated plenty of priests who were a lot more knowledgeable than the people who call themselves "graduates" today.
The code base to GCC is huge and complex, not to mention sparsely documented. Going through the entire thing to make sure it's perfectly secure is likely a harder task than starting from scratch.
It's not like a C compiler is a very difficult thing to write. GCC is much more than than just a C compiler. But if a C compiler is all you need...maybe GCC is overkill.
QT is licensed under the GPL, not the LGPL. It also compiles on FreeBSD just fine. By language portability, I meant there are more bindings for GTK+ and they're easier to keep up to date since it's plain old C and not C++ with a Meta Object Compiler.
Well, before there was fork(), Unix (well, Multics) had to make do with chopsticks(), which worked fine, but didn't check if there was enough memory to actually create the new process, so a lot of brand new processes would get dropped almost instantly, especially if the user wasn't adept at using chopsticks() and didn't realize what was going on. Another drawback was that you had to start a new version of chopsticks() every time you rebooted and type in all of the required command line arguments.
Fortunately, fork() came along and fixed all of this, so there are now very few dropped processes and fork() boots automatically with the system.
Apparently, Irix used an advanced version of fork() called spork(), which was even more stable and performed the functions of fork() AND bind() at the same time.
I'd rather the "no" instead of the "maybe". You can't plan anything on "maybes."
You're assuming that my whole world revolves around providing entertainment for people with cell phones. It does not, as much as they like to think it does.
And I find the implicit "Well, sure, unless something better comes along" that goes along with the "maybe" to be insulting. There's something to be said for making a committment and keeping it, especially among friends.
"And that is why I believe profit-above-all-else mentalities must be destroyed."
I agree, but you can't legislate morality. It really depends on the individual to make the decision to do the right thing.
U.S. society has become increasingly legalistic, to the point where a good number of people see nothing immoral about their actions provided they're not illegal. There are two answers to that -- we either need to make the laws better, or people need to start to realize the moral implications of their actions.
The problem I think a lot of people have is that they believe that with just the right set of rules, everyone will act properly. I don't think that's ever the case, I think people will do what they want to regardless. If that weren't true, there wouldn't be crime.
"Corporations lack this failsafe of greed."
This is simply not true, unless you are talking about forced labor. The Chinese worker still has a choice -- he can keep subsistence farming, or he can make $0.25 a day, which is wealth beyond his wildest dreams, such that he'd nearly kill to get and keep that job.
It sounds crazy to you, but that's because you're used to prices and wages as they are in the developed world. If you were offered $1,000,000 a day to work in a factory run by aliens from Planet X, and you were happy to take it, how would you feel about someone else from Planet X who wanted to shut the factory down unless the owners were forced to pay you $2,000,000 a day, which you know would end up in you getting fired and the factory moving back to Planet X instead?
"What have we got to lose except a massive trade deficit?"
The Chinese hold a significant amount of U.S. currency as a backing reserve for their own. They could announce that they are selling off all U.S. currency reserves, which would sink the value of the dollar almost overnight.
This would send the U.S. into a severe inflation-based depression that would make the deflation-based on in the thirties look like a mild recession, since the Fed is seemingly only capable of inflating the currency even more in a vain attempt to fix economic crises.
This would hit us at the time of a record setting downturn in the real-estate market which has not quite hit the bottom yet, and also with record setting government debt as far as the eye can see that could no longer be paid for by printing more money, forcing many people who depend on social security and medicare to, well, just die.
This would easily cause mass civil unrest, starvation, and rioting as the younger people who are left lose their homes and jobs and can no longer afford to pay for luxuries like heat and food.
So, yeah, other than that, we've got nothing to lose.
But it's ok, we're counting on the Chinese government not to do that because they'd also experience a recession and starvation, and history tells us that concern for the public welfare is always first and foremost when the Chinese government makes decisions.
"You mean like the iPhone? What is it, now, $399? Do you know how much it will cost consumers if manufacturing moves to Mexico? $399."
And you're basing this assertion on what, exactly?
I'll tell you what, it's significantly cheaper and easier to ship things from Mexico than it is from China. The reason that the iPhone is manufactured in China is because even given the shipping cost and hassle, it's still cheaper for Apple.
If manufacturing moved to Mexico, do you know what another price of the iPhone could be? $0, because it wouldn't exist. Just as if Apple had to manufacture it here in the U.S., a modest increase in manufacturing costs might make it unprofitable to even produce the iPhone in the first place.
That's what nobody seems to take into account when discussing trade policy -- restrictions on trade don't just raise prices across the board, they also ensure that some of the things that would have happened (production and sale of the iPhone) don't happen at all.
Whether or not you think the cost to society of these things not happening is worth the trade restrictions is a case-by-case matter of debate, and certainly not black and white, no matter what level of economics course you've taken.
The reason I use Linux and FreeBSD is the reason why most people use Windows. I really like the vast quantity of software that's available for them.
Playing games can be a pain, but every other piece of software I use was designed for a Unix-like system, and work better than their Windows ports.
I did some specialized work with software radio recently too -- as a development platform for new hardware, Linux is great.
I keep WinXP around for games and cross-platform testing, and I agree with you, it's quite decent, provided you start with a fresh install and know a thing or two about maintaining it.
In reality, the similarities between all of these OS's and desktop environments far outweigh the differences. Within the spectrum of possible ways to do things, they all fall on a very small portion. There's no real reason to switch if you're happy enough with one.
I still don't get it. When you have virtual desktops, the whole desktop is an MDI.
You have the 800 window problem if you're using another type of MDI anyway, it's just that they're contained in another window. Same problem, pushed down a level.
And the latest version of the Gimp lets you dock any window that you want, so you can tab between commonly used tools. I find it quite flexible.
The only problem I really see is that there aren't typically shortcuts for everything, so there are extra clicks if you want to see the main toolbox, for example.
Slashdot Mirror
The thing is, you can do most of this from a user account. Users have access to a lot of stuff that can be just as bad, and once you have access to a user account, it's fairly easy to get root.
Once someone with malicious intent can access your machine, you can pretty much kiss it goodbye. You can never really be sure that root wasn't compromised without an extensive investigation.
You have a command line, emacs, vi, the gcc suite, perl, clisp and sbcl. What more could you possibly want?
As long as there's a terminal available and gcc, you just can't complain about lack of power user features in Linux.
He complains about the multiple package management programs. There's no problem here, since they all use the same underlying database, and a newbie would never know about the command line ones, and wouldn't need to.
A new user will get along just fine with the simplicity of Ubuntu on the desktop. A power user will hit the command line and have no problems.
It seems like this guy knows just enough about Debian to be dangerous, and is now cranky that Ubuntu is slightly different.
I'm not a materials scientist either, but I did take a structural engineering class and sleep in a Holiday Inn express last night.
There are many classifications of materials that could be interpreted as "brittle." Brittle is much too general a term to be used in engineering, so you have to be suspicious of the news article.
You can measure tensile strength, which is a measure of how much something can bend until it break. There's another measurement where you find how much something can bend until it permanently deforms, so that it won't go back to its original state. Each of these could be called "flexibility" but that doesn't tell you the whole story.
Carbon fiber when it fails may fail explosively and shatter, while a soft metal would simply deform slowly when bent far enough. This could be called "brittleness" but it really has little to do with the actual engineering problem, since if you design the carbon fiber component to high enough tolerances, you're not worried about it breaking, since the force required to break it would be so huge you'd have other, much bigger problems besides the breaking of the part. (Like, how do we get the people out of the broken plane when Godzilla is about to eat it?)
It would be easy to criticize the engineering of the plane on the news, because nobody is going to sit there for three months to check everything out -- they'll watch the demo of a small piece of carbon fiber breaking and think, "Oh my god, that could be the wing of my plane!"
"It also found that 20 percent said they spend less time having sex because they are online."
It didn't say they have LESS sex, just that they spend less TIME doing it. Obviously, the Internet has made them more efficient.
Probably has to do with the massive hard-ons they can now achieve thanks to e-mail offers. What a truly wonderful age in which we live!
"HOw about a garden rake with one stuck on the end?"
Most people don't want to cauterize the weeds, they want to pull them out of the ground.
Now a light saber hedge trimmer would save a lot of time and effort.
Complaints can be valid all day long, but that doesn't necessarily make them helpful.
When you're managing a project, usually you have to make decisions that are going to piss some people off. Those people can whine about it forever or simply realize that the decision had to be made and shut up about it. If they feel a bad decision is THAT big a deal, then it's time to put up or shut up, and show everyone else how wrong they are. That's productive and helpful, complaining isn't.
I find OpenOffice to be really good software, and it's improving rapidly. I don't see the problem in the grand scheme of things.
"Statistical analysis isn't difficult"
I'm not sure what depth of statistical analysis you're talking about, but I've found statistical analysis to be exceptionally difficult when applied to Electrical Engineering, and I know I'm not alone here -- it was by far the most difficult post-graduate class at my school.
I can confidently say that nobody who's graduating with me has a complete grasp of all of the statistical tools we were taught. Enough to get by, yes, but most of the things are extremely counterintuitive and easily misused, and this is by people who are really good at math.
I have to laugh when I hear statistics in the news now, because it's all such propaganda. Any time you hear "has been linked to" in an article you know you're about to wade through a steaming pile of bullshit.
Really?
So in Linux, when you want to swap out the scheduler to one that devotes most of the CPU time to the process owning the current focused window in X, it's just a user level module away, since the scheduler is modular and can be changed at run time?
That's cool, how do I do that?
After the village comes over my house at 4:00 AM to rock my screaming infant to sleep, it can make decisions about what's best for my child.
But not before.
"And in my opinion, anyone who claims to be a university graduate should be required to take and understand calculus."
Well, I think that everyone should have to take up through Calculus III, since it's been around almost as long, and you can't understand weather reports in great detail unless you know how gradients work.
And differential equations, linear algebra, and advanced probability and statistics, too. You're just not well educated unless you can compute a covariance matrix in your head.
You should also know about the fundamentals of electromagnetic theory and how electrical waves propagate through a transmission line. And if you don't know how to do a link-budget analysis of a satellite link and also calculate where that satellite will be in relation to a ground station, then you can't really call yourself educated.
And how many people who graduate today actually know anything about state space control systems equations? Precious few, and yet I think it's very important since it's used all the time.
Not only that, but people only learn one foreign language, and it's usually something that's only come into fashion recently like French. All courses should be taught in Latin, or better yet, Aramaic. It worked for many centuries, why shouldn't it work today? We graduated plenty of priests who were a lot more knowledgeable than the people who call themselves "graduates" today.
If you learn LaTeX only for the equations, it's worth it.
Open Office's equation editor is the same idea, but a bit more verbose, which slows everything down.
"Linux can't even run "Hello World" without GNU System components like GNU Libc."
I beg to differ.
The code base to GCC is huge and complex, not to mention sparsely documented. Going through the entire thing to make sure it's perfectly secure is likely a harder task than starting from scratch.
It's not like a C compiler is a very difficult thing to write. GCC is much more than than just a C compiler. But if a C compiler is all you need...maybe GCC is overkill.
Damn straight!
Liberals are always preaching about evolution, but they never seem to want to let it happen.
"(or dating) someone who DOES read Slashdot."
+1 Funny
QT is licensed under the GPL, not the LGPL. It also compiles on FreeBSD just fine. By language portability, I meant there are more bindings for GTK+ and they're easier to keep up to date since it's plain old C and not C++ with a Meta Object Compiler.
Well, before there was fork(), Unix (well, Multics) had to make do with chopsticks(), which worked fine, but didn't check if there was enough memory to actually create the new process, so a lot of brand new processes would get dropped almost instantly, especially if the user wasn't adept at using chopsticks() and didn't realize what was going on. Another drawback was that you had to start a new version of chopsticks() every time you rebooted and type in all of the required command line arguments.
Fortunately, fork() came along and fixed all of this, so there are now very few dropped processes and fork() boots automatically with the system.
Apparently, Irix used an advanced version of fork() called spork(), which was even more stable and performed the functions of fork() AND bind() at the same time.
I'd rather the "no" instead of the "maybe". You can't plan anything on "maybes."
You're assuming that my whole world revolves around providing entertainment for people with cell phones. It does not, as much as they like to think it does.
And I find the implicit "Well, sure, unless something better comes along" that goes along with the "maybe" to be insulting. There's something to be said for making a committment and keeping it, especially among friends.
"And that is why I believe profit-above-all-else mentalities must be destroyed."
I agree, but you can't legislate morality. It really depends on the individual to make the decision to do the right thing.
U.S. society has become increasingly legalistic, to the point where a good number of people see nothing immoral about their actions provided they're not illegal. There are two answers to that -- we either need to make the laws better, or people need to start to realize the moral implications of their actions.
The problem I think a lot of people have is that they believe that with just the right set of rules, everyone will act properly. I don't think that's ever the case, I think people will do what they want to regardless. If that weren't true, there wouldn't be crime.
"Corporations lack this failsafe of greed."
This is simply not true, unless you are talking about forced labor. The Chinese worker still has a choice -- he can keep subsistence farming, or he can make $0.25 a day, which is wealth beyond his wildest dreams, such that he'd nearly kill to get and keep that job.
It sounds crazy to you, but that's because you're used to prices and wages as they are in the developed world. If you were offered $1,000,000 a day to work in a factory run by aliens from Planet X, and you were happy to take it, how would you feel about someone else from Planet X who wanted to shut the factory down unless the owners were forced to pay you $2,000,000 a day, which you know would end up in you getting fired and the factory moving back to Planet X instead?
"What have we got to lose except a massive trade deficit?"
The Chinese hold a significant amount of U.S. currency as a backing reserve for their own. They could announce that they are selling off all U.S. currency reserves, which would sink the value of the dollar almost overnight.
This would send the U.S. into a severe inflation-based depression that would make the deflation-based on in the thirties look like a mild recession, since the Fed is seemingly only capable of inflating the currency even more in a vain attempt to fix economic crises.
This would hit us at the time of a record setting downturn in the real-estate market which has not quite hit the bottom yet, and also with record setting government debt as far as the eye can see that could no longer be paid for by printing more money, forcing many people who depend on social security and medicare to, well, just die.
This would easily cause mass civil unrest, starvation, and rioting as the younger people who are left lose their homes and jobs and can no longer afford to pay for luxuries like heat and food.
So, yeah, other than that, we've got nothing to lose.
But it's ok, we're counting on the Chinese government not to do that because they'd also experience a recession and starvation, and history tells us that concern for the public welfare is always first and foremost when the Chinese government makes decisions.
"You mean like the iPhone? What is it, now, $399? Do you know how much it will cost consumers if manufacturing moves to Mexico? $399."
And you're basing this assertion on what, exactly?
I'll tell you what, it's significantly cheaper and easier to ship things from Mexico than it is from China. The reason that the iPhone is manufactured in China is because even given the shipping cost and hassle, it's still cheaper for Apple.
If manufacturing moved to Mexico, do you know what another price of the iPhone could be? $0, because it wouldn't exist. Just as if Apple had to manufacture it here in the U.S., a modest increase in manufacturing costs might make it unprofitable to even produce the iPhone in the first place.
That's what nobody seems to take into account when discussing trade policy -- restrictions on trade don't just raise prices across the board, they also ensure that some of the things that would have happened (production and sale of the iPhone) don't happen at all.
Whether or not you think the cost to society of these things not happening is worth the trade restrictions is a case-by-case matter of debate, and certainly not black and white, no matter what level of economics course you've taken.
The reason I use Linux and FreeBSD is the reason why most people use Windows. I really like the vast quantity of software that's available for them.
Playing games can be a pain, but every other piece of software I use was designed for a Unix-like system, and work better than their Windows ports.
I did some specialized work with software radio recently too -- as a development platform for new hardware, Linux is great.
I keep WinXP around for games and cross-platform testing, and I agree with you, it's quite decent, provided you start with a fresh install and know a thing or two about maintaining it.
In reality, the similarities between all of these OS's and desktop environments far outweigh the differences. Within the spectrum of possible ways to do things, they all fall on a very small portion. There's no real reason to switch if you're happy enough with one.
"Mac users are not special."
Shhh! You're going to hurt their feelings.
I still don't get it. When you have virtual desktops, the whole desktop is an MDI.
You have the 800 window problem if you're using another type of MDI anyway, it's just that they're contained in another window. Same problem, pushed down a level.
And the latest version of the Gimp lets you dock any window that you want, so you can tab between commonly used tools. I find it quite flexible.
The only problem I really see is that there aren't typically shortcuts for everything, so there are extra clicks if you want to see the main toolbox, for example.
Congratulations on your Mac purchase :-)