The point is, proving correctness is possible in a microkernel, but next to impossible in a large monolithic kernel. It's a problem that increases exponentially, not linearly, with the amount of code you have to analyze.
Every function added to a monolithic kernel could possibly alter the behavior of every other function, because they all share the same address space. This is not true of a microkernel.
Once you've proved a microkernel is correct, you're done. You can then go on to prove that other small pieces of the system are correct. Because they all exist in separate address spaces, as long as a piece doesn't change, you can rest assured they will remain correct. But proving the correctness of a constantly changing 6 million lines of code is not feasible.
Thinly veiled insult. Latin words from Freshman Debate 101!!
Self-congratulatory explanation of logical fallacies of above argument. Arrogant insinuation that I am smarter than you. More big words, many in italicized non-English, attempting to display my advanced education.
It sounds like a good idea, but think of all of the money that would leave the health care system entirely if you remove the profit motive.
This is one of those things where having the intention to do the right thing may end up being counterproductive. I'm not saying you're wrong, just that fixing things might not be that simple.
You've got a bunch of good recommendations, but nobody's mentioned QT so far.
I'm a big fan of QT on windows, and if you don't mind a bit of setup, the free GPL'd version works well with MinGW. The documentation is excellent, which probably would be helpful for someone learning to program GUI stuff.
And it can be compiled statically so that distribution is as easy as copying an.exe file.
>- New, faster install system (no more text-mode 'copying files') >Again, whee.
Whee? WHEE? This is the most promising feature! I spend at least %30 of my time in Windows reinstalling it. Hopefully we can get that down to about %20 with Vista.
Python has a lot of modules that would really help you do what you want.
Perl and Java, also. I'd probably recommend them in that order.
They all have fairly comprehensive reference material and some good tutorials on the web.
My favorite development environment is still emacs with the vi key bindings, but IBM's Eclipse is also very good if you're doing Java, and you have a fast enough machine to run it.
In fact, trying Java out is much easier with Eclipse, as it fills in a lot of the code for you, and lets you know immediately when you're making a mistake. I've found it very helpful while learning Java, but later on the usefulness of that diminishes and I want my good text editor back.
Finally, I love C and gtk+ for unix apps. If you're interested in the "guts" of the machine, you really can't beat that combo, and it's very powerful.
If police would do their jobs instead of sitting on their asses at speed traps, we wouldn't need cell phone laws, or studies like this.
Inevitably, anyone on a cell phone is breaking about 15 other driving laws because they can't concentrate. The drunkenness or cell-phone conversation is not the problem -- the swerving and going 20 miles an hour under the speed limit in the passing lane is. Pull them over for those things, and the idiot cell phone holding driver would quickly become a thing of the past.
Of course it would be typical of a Bush supporter. But it does not make a Kerry supporter more informed, just correct by chance. There's a difference.
It would be more typical of a Kerry supporter to think the NRA supports John Kerry because he's a hunter. That's wrong, but it doesn't imply that ignorant Bush supporters are more informed. Both would answer that question not based on facts they know, but what they believe.
And in this case, you can hand pick the questions to favor the beliefs of one side over the other.
It's fine to say that a lot of Bush supporters are ignorant. That's a fact. To say that this implies that Kerry supporters are better informed and smarter is illogical.
>In this case, it's absolutely more correct that Kerry supporters have got more going on in the brain-use department than Bush supporters.
No it's not. The fact that you can publish two opposing views and have half the people agree with you does not imply that half of the people are informed while the other is not. Both sides might agree with you based on ideology, not on facts.
If I flip a coin, and hide the result, it's a fact that the coin is either heads or tails. If you polled a population and asked them to guess what side the coin landed on, half would get it right, and half would get it wrong. Neither side is more informed, even though one side happened to stumble on the right answer.
Just because the issue is politically charged and one side tends to answer one way and the other another way does not change the underlying logic.
Is it any surprise that the vast majority of people are uninformed and have a view of reality based on ideology rather than fact? NO!
Does this make one bit of difference when trying to decide who to vote for? NO!
Do you vote for Bush because some ignorant guy off the street was given cigarettes to vote for Kerry? Do you vote for Kerry because some gun nut has a Bush bumper sticker? You shouldn't.
This "research" into popular ignorance is what political science has sunk to. "Ignorant Bush supporters more ignorant than ignorant Kerry supporters." What does this contribute to political discussion? NOTHING. It's the best a group of professors and grad students who don't have one original idea of their own to contribute can do.
Do you honestly believe that every single one of Kerry's supporters is voracious news hound and super-informed political junkie, or do you think it's possible that their chosen ideology happens to jive with the facts this week? If we found WMD in Iraq tomorrow, I'm sure 50% of people polled, blinded by ideology, would still get it wrong. Then we could claim that Kerry supporters are vastly more ignorant than Bush. This proves NOTHING.
This is propaganda. Read what intelligent and informed people on both sides have to say, and then make a decision. Don't fall for this crap disguised as research.
Holding the machine up to a zero-defect standard is the logical fallacy.
Any voting system has potential for fraud, mistakes, etc. The issue should not be whether this system has that potential, but whether it's any better or worse than the current system, which isn't that great. You can't tell me that it's not within the realm of possibility for a purely mechanical machine to malfunction, or be rigged.
If something bad doesn't happen with these new machines, I'd be very surprised. That would worry me far more than hearing about a perfect election with no mistakes, as that's evidence that someone is truly incompetent or lying.
From the title, I thought the article was going to be about how you can easily rack up 250 rupies for the blue ring in Zelda by going to the money making game and hitting reset only when you lose.
Bush: I think it's important to blah. We've spent a lot of money to encourage blah. We need to restrain our spending except for blah.
Kerry: What Bush said, but not what Bush said. More money needs to be spent on blah than he says. I'm just the man to do it, because I fought in a war to defend blah.
Nader: Don't listen to them. They are in bed with EEEEEEVIL corporations who are trying to steal your children's college fund. They do this by fear-mongering. They will kill us all with unsafe cars if you don't elect me.
I'm a fiscal conservative. I'll give up everything except for national defense and highway construction, which the federal government is, and always has been, constitutionally mandated to provide for.
Slashdot Mirror
You can pick which one, too. The War of the Worlds one or the MST3K one. Both rock.
The point is, proving correctness is possible in a microkernel, but next to impossible in a large monolithic kernel. It's a problem that increases exponentially, not linearly, with the amount of code you have to analyze.
Every function added to a monolithic kernel could possibly alter the behavior of every other function, because they all share the same address space. This is not true of a microkernel.
Once you've proved a microkernel is correct, you're done. You can then go on to prove that other small pieces of the system are correct. Because they all exist in separate address spaces, as long as a piece doesn't change, you can rest assured they will remain correct. But proving the correctness of a constantly changing 6 million lines of code is not feasible.
You mean Jim Carrey? The SNL crew?
This is just more evidence that outsourcing, while it may seem appealing at first, just doesn't work over time.
>And none (not even emacs) can do it with so few keystrokes (that does make the learning curve on Vi rather steep).
Not so! That's why there's Viper.
I was firmly on the vi side of the argument until I realized that emacs includes everything you could possibly want to edit code, including vi.
You skipped rain-forest deforestation suffocating us all. That was after we were burned by ozone holes.
I mean, we had bake sales and everything for the rain forest. Why can't we stop global warming the same way?
Thinly veiled insult. Latin words from Freshman Debate 101!!
Self-congratulatory explanation of logical fallacies of above argument. Arrogant insinuation that I am smarter than you. More big words, many in italicized non-English, attempting to display my advanced education.
It sounds like a good idea, but think of all of the money that would leave the health care system entirely if you remove the profit motive.
This is one of those things where having the intention to do the right thing may end up being counterproductive. I'm not saying you're wrong, just that fixing things might not be that simple.
I had the same problem.
You need this mouse.
It's pretty large and high so it fills up large hands and keeps them off the table a bit.
You can use the Boehm garbage collector to eliminate a huge class of typical memory errors:
http://www.hpl.hp.com/personal/Hans_Boehm/gc/
This isn't necessarily something you'd have to design around, either. You can add it later.
You've got a bunch of good recommendations, but nobody's mentioned QT so far.
.exe file.
I'm a big fan of QT on windows, and if you don't mind a bit of setup, the free GPL'd version works well with MinGW. The documentation is excellent, which probably would be helpful for someone learning to program GUI stuff.
And it can be compiled statically so that distribution is as easy as copying an
>- New, faster install system (no more text-mode 'copying files')
>Again, whee.
Whee? WHEE? This is the most promising feature! I spend at least %30 of my time in Windows reinstalling it. Hopefully we can get that down to about %20 with Vista.
Python has a lot of modules that would really help you do what you want.
Perl and Java, also. I'd probably recommend them in that order.
They all have fairly comprehensive reference material and some good tutorials on the web.
My favorite development environment is still emacs with the vi key bindings, but IBM's Eclipse is also very good if you're doing Java, and you have a fast enough machine to run it.
In fact, trying Java out is much easier with Eclipse, as it fills in a lot of the code for you, and lets you know immediately when you're making a mistake. I've found it very helpful while learning Java, but later on the usefulness of that diminishes and I want my good text editor back.
Finally, I love C and gtk+ for unix apps. If you're interested in the "guts" of the machine, you really can't beat that combo, and it's very powerful.
For those who don't know, Sarah O'Connor is the waitress-turned-military-assasin in the version of Terminator II that was released in Ireland.
If police would do their jobs instead of sitting on their asses at speed traps, we wouldn't need cell phone laws, or studies like this.
Inevitably, anyone on a cell phone is breaking about 15 other driving laws because they can't concentrate. The drunkenness or cell-phone conversation is not the problem -- the swerving and going 20 miles an hour under the speed limit in the passing lane is. Pull them over for those things, and the idiot cell phone holding driver would quickly become a thing of the past.
What does this mean?
Is the height of the stacks tied to population or numbers of voters or to the margin of victory in that county, percentage or absolute?
This isn't particularly self-explanatory. A key would be nice.
Think of all the time it will waste killing people over and over again.
I wonder how many neurons it takes to recognize simple patterns.
Like patterns in. . . say . . . news article headlines.
Of course it would be typical of a Bush supporter. But it does not make a Kerry supporter more informed, just correct by chance. There's a difference.
It would be more typical of a Kerry supporter to think the NRA supports John Kerry because he's a hunter. That's wrong, but it doesn't imply that ignorant Bush supporters are more informed. Both would answer that question not based on facts they know, but what they believe.
And in this case, you can hand pick the questions to favor the beliefs of one side over the other.
It's fine to say that a lot of Bush supporters are ignorant. That's a fact. To say that this implies that Kerry supporters are better informed and smarter is illogical.
>In this case, it's absolutely more correct that Kerry supporters have got more going on in the brain-use department than Bush supporters.
No it's not. The fact that you can publish two opposing views and have half the people agree with you does not imply that half of the people are informed while the other is not. Both sides might agree with you based on ideology, not on facts.
If I flip a coin, and hide the result, it's a fact that the coin is either heads or tails. If you polled a population and asked them to guess what side the coin landed on, half would get it right, and half would get it wrong. Neither side is more informed, even though one side happened to stumble on the right answer.
Just because the issue is politically charged and one side tends to answer one way and the other another way does not change the underlying logic.
Correlation does not necessarily equal causation.
Is it any surprise that the vast majority of people are uninformed and have a view of reality based on ideology rather than fact? NO!
Does this make one bit of difference when trying to decide who to vote for? NO!
Do you vote for Bush because some ignorant guy off the street was given cigarettes to vote for Kerry? Do you vote for Kerry because some gun nut has a Bush bumper sticker? You shouldn't.
This "research" into popular ignorance is what political science has sunk to. "Ignorant Bush supporters more ignorant than ignorant Kerry supporters." What does this contribute to political discussion? NOTHING. It's the best a group of professors and grad students who don't have one original idea of their own to contribute can do.
Do you honestly believe that every single one of Kerry's supporters is voracious news hound and super-informed political junkie, or do you think it's possible that their chosen ideology happens to jive with the facts this week? If we found WMD in Iraq tomorrow, I'm sure 50% of people polled, blinded by ideology, would still get it wrong. Then we could claim that Kerry supporters are vastly more ignorant than Bush. This proves NOTHING.
This is propaganda. Read what intelligent and informed people on both sides have to say, and then make a decision. Don't fall for this crap disguised as research.
Holding the machine up to a zero-defect standard is the logical fallacy.
Any voting system has potential for fraud, mistakes, etc. The issue should not be whether this system has that potential, but whether it's any better or worse than the current system, which isn't that great. You can't tell me that it's not within the realm of possibility for a purely mechanical machine to malfunction, or be rigged.
If something bad doesn't happen with these new machines, I'd be very surprised. That would worry me far more than hearing about a perfect election with no mistakes, as that's evidence that someone is truly incompetent or lying.
From the title, I thought the article was going to be about how you can easily rack up 250 rupies for the blue ring in Zelda by going to the money making game and hitting reset only when you lose.
Hey! Why don't they fund the new game that way?
Question: Blah.
Bush: I think it's important to blah. We've spent a lot of money to encourage blah. We need to restrain our spending except for blah.
Kerry: What Bush said, but not what Bush said. More money needs to be spent on blah than he says. I'm just the man to do it, because I fought in a war to defend blah.
Nader: Don't listen to them. They are in bed with EEEEEEVIL corporations who are trying to steal your children's college fund. They do this by fear-mongering. They will kill us all with unsafe cars if you don't elect me.
I'm a fiscal conservative. I'll give up everything except for national defense and highway construction, which the federal government is, and always has been, constitutionally mandated to provide for.
Now you've heard it.
If this were true, we'd have means testing of social security recipients.
As it is now, even the richest old people get their social security check each month.
What you're describing is welfare.