Kernel hacking is taught at major universities, but under the title "Operating Systems".
We did no kernel programming in either the undergrad or graduate OS class at my undergraduate institution (I took both). It was all fairly low-level C systems programming. The undergrad OS class at my graduate institution (a top ten CS school) is the same way -- in fact, the OS class there is sometimes taught in Java. We did no kernel programming in any class I've taken so far, except for the graduate OS class at my graduate institution, and that was only because that was the project I happened to pick.
You learn OS theory, sure, and you almost need an understanding of a lot of that to know all of what's going on around the Linux kernel, but you're not going to learn about the Linux kernel in an OS class.
Then what does the "peer" mean? If anyone can comment, it's just "review".
This is just an argument about what "peer review" means, not about whether it's useful, how it compares to peer review, etc., but that term has a pretty specific meaning.
Wikipedia's entry starts with "Peer review (known as refereeing in some academic fields) is a process of subjecting an author's scholarly work or ideas to the scrutiny of others who are experts in the field."
Open Source doesn't mean that a company can't charge something. Look at Redhat. They sure as hell don't give everything away for free.
No, but CentOS does.
Open source means a change from a product-based market to a service-based market, at least to a FAR larger extent than is acknowledged by the typical "look at Red Hat" line. Remember, Red Hat makes most of its money because people want support. That's fine if you want to be in that business and can come up with some value to add to the product that only you can provide.
If you have something like TV schedules, there's no service there, unless you want to talk about providing the bandwidth to you to get it. If Bob generates the schedules and Fred buys them from Bob and starts giving them out, what's the incentive to go to Fred and pay for them? I certainly can't think of any added value Bob can give. So if Bob spends a good bit of money compiling the listings and the only people who actually buy them from him are those who want to actually support Bob while the people who just want the listings go to Fred, that could very easily be a losing proposition for Bob.
Anyway, pregnancy could somewhat easily be dealt with if they give out some sort of contraceptive. Ideally you'd want one of the long-term ones like the thing they put into your arm or whatever so there wouldn't be any chance for mistake. Though I don't know what they would want to do to get something like that declared safe for a long term space flight.
Then you could drag and drop the desired features onto the phone that is plugged into your computer via USB.
That would be the perfect phone.
Can you drag and drop different form factors, so that Bob can have a big, rich iPhone-like interface with a camera and touchscreen and whatnot, while Mary can have one of those really tiny flip phones that's not much bigger than your thumb and has physical buttons?
There's still a LOT of difference between what I consider a perfect phone and what you consider a perfect phone that is more than software. Heck, I'd go so far as to say that the hardware that's present and the form factor is a much bigger issue. Theoretically you could have a pluggable interface where you could buy just the hardware you need and such, but such a solution would probably give a product that works worse than just going out and getting something from the market today.
If points are subtracted for incorrect answers (say -1/4 pt to -1/2 for each one wrong), the effect of guesses can be taken out of the equation so that differences in scores actually reflect differences in knowledge.
How 'come you say that you can take out the effect of guessing if you subtract penalty points, but not if you don't:
First, because there is a large degree of scatter: a student choosing at random might do much better or much worse than this.
You can still do a lot better or worse than 0% if you're guessing.
I've never tried to work this out, but my intuition on this subject (and I have thought about it before) is that the two options are probably isomorphic; I suspect scores would follow the same distribution just scaled and shifted so that 25% in the "only count right answers" lands on 0% in the "penalize wrong answers" test. This may be right, or it may not, but it's definitely not clear (even after reading the "article") that this is the case.
There is one important caveat, and it illustrates what I see as the *real* benefit of this: it means that if you are only 2/3 of the way through the test and time is up, you don't have to run through and quickly fill in guesses for the remaining 1/3 of the test. My isomorphism probably only works (if it does at all) if you assume that any unanswered questions in the "penalize wrong answers" test are randomly filled in.
There's also Jonathan Coulton's I'm Your Moon, a love song from Charon to Pluto.
Watch this video for his introduction to the song, but you can download (for free -- CC license -- or buy) an MP3 here, since the video cuts out briefly every couple minutes.
Stop being dense. You don't patent a result -- you patent methods. (At least for utility patents. Design patents and biological patents are different, but neither of those apply here.)
So unless their system works by intercepting acoustic waves with an eardrum that vibrates tiny bones that move a liquid that triggers tiny hairs which send electrical sigals to a mass of neurons which somehow figures it out, no, the ear isn't prior art. Considering that we have not much better than "not a clue" how the brain actually associates the sound you hear to memory, I am skeptical that this is how their approach works.
Okay, I read Linus's whole post (the linked page was down, I had to use the google groups link someone else posted) and it's actually a lot more tempered and reasonable than that little quote reveals.
It was a big source of pride for them that they got the linux kernel to build in icc without patching. *eye roll*
Eye roll? Why?
The Linux kernel has disclaimers saying that it's made for GCC and they don't care if it works on other compilers, and it makes pretty heavy use of GCC extensions. The kernel isn't portable code compiler-wise almost at all.
Or better yet rout their call back to their number (if its local or an 800) It would be great to have them call their own receptionest..
The only drawback of that approach is that it may not be legal to record the conversation and post it online since you are not a party to the conversation.
I wonder what you would need to do to make it kosher? Maybe include you in on the call but still route back, then just you be really quiet? (This assumes you're in a one-party state of course.)
Same reason we have most airport security: it provides the illusion of security.
(Okay, so it probably helps a little bit. Personally, I suspect that door locks would very possibly either prevented or more likely substantially lessened the severity of 9/11. I also think that the GP is right, and we would have just seen a different attack.)
If the door was locked, how long would it take for the pilot to open the door if the terrorist started executing one passanger a minute until the door was opened?
On the flip side, how many of those executions would it take before the passengers turned *all* the flights into flight 93?
Then it's just a race to see whether the passengers react or the pilot caves first. My suggestion: with the terrorists with knives instead of guns, the passengers.
The point isn't that you know the difference, the point is that if you have to think about the difference your code could probably be improved by splitting it.
Why is
*p++ = *q++;
better than
*p = *q; p++; q++;
?
Answer: unless for some reason you are coding on a screen that only shows you 3 lines of text at once, it isn't.
Why make the programmer (i.e. you in a couple weeks) think harder about what the code does then you need to?
loqi said it best: "Expressions are like sex: they're better without side-effects."
Slashdot Mirror
I know that #2 sounds remarkably similar to what's going on in Mac OS X's new "Time Machine" feature as well as ZFS.
And, be fair, Vista's "previous versions" feature. They've got it too, and before Apple.
I know it's not an innovation on their part, but why mention OS X and not it?
Mainframes have been running VMs for years.
Years? More like decades. IBM more or less invented virtualization back in the 60s for the System/390, and it lives on today as z/VM.
Not anymore, apparently
And then it'll want a straw, and who knows what will happen after that.
Kernel hacking is taught at major universities, but under the title "Operating Systems".
We did no kernel programming in either the undergrad or graduate OS class at my undergraduate institution (I took both). It was all fairly low-level C systems programming. The undergrad OS class at my graduate institution (a top ten CS school) is the same way -- in fact, the OS class there is sometimes taught in Java. We did no kernel programming in any class I've taken so far, except for the graduate OS class at my graduate institution, and that was only because that was the project I happened to pick.
You learn OS theory, sure, and you almost need an understanding of a lot of that to know all of what's going on around the Linux kernel, but you're not going to learn about the Linux kernel in an OS class.
Then what does the "peer" mean? If anyone can comment, it's just "review".
This is just an argument about what "peer review" means, not about whether it's useful, how it compares to peer review, etc., but that term has a pretty specific meaning.
Wikipedia's entry starts with "Peer review (known as refereeing in some academic fields) is a process of subjecting an author's scholarly work or ideas to the scrutiny of others who are experts in the field."
Open Source doesn't mean that a company can't charge something. Look at Redhat. They sure as hell don't give everything away for free.
No, but CentOS does.
Open source means a change from a product-based market to a service-based market, at least to a FAR larger extent than is acknowledged by the typical "look at Red Hat" line. Remember, Red Hat makes most of its money because people want support. That's fine if you want to be in that business and can come up with some value to add to the product that only you can provide.
If you have something like TV schedules, there's no service there, unless you want to talk about providing the bandwidth to you to get it. If Bob generates the schedules and Fred buys them from Bob and starts giving them out, what's the incentive to go to Fred and pay for them? I certainly can't think of any added value Bob can give. So if Bob spends a good bit of money compiling the listings and the only people who actually buy them from him are those who want to actually support Bob while the people who just want the listings go to Fred, that could very easily be a losing proposition for Bob.
STDs should be a factor...
Did you forget a "not"?
Anyway, pregnancy could somewhat easily be dealt with if they give out some sort of contraceptive. Ideally you'd want one of the long-term ones like the thing they put into your arm or whatever so there wouldn't be any chance for mistake. Though I don't know what they would want to do to get something like that declared safe for a long term space flight.
Then you could drag and drop the desired features onto the phone that is plugged into your computer via USB.
That would be the perfect phone.
Can you drag and drop different form factors, so that Bob can have a big, rich iPhone-like interface with a camera and touchscreen and whatnot, while Mary can have one of those really tiny flip phones that's not much bigger than your thumb and has physical buttons?
There's still a LOT of difference between what I consider a perfect phone and what you consider a perfect phone that is more than software. Heck, I'd go so far as to say that the hardware that's present and the form factor is a much bigger issue. Theoretically you could have a pluggable interface where you could buy just the hardware you need and such, but such a solution would probably give a product that works worse than just going out and getting something from the market today.
If points are subtracted for incorrect answers (say -1/4 pt to -1/2 for each one wrong), the effect of guesses can be taken out of the equation so that differences in scores actually reflect differences in knowledge.
How 'come you say that you can take out the effect of guessing if you subtract penalty points, but not if you don't:
First, because there is a large degree of scatter: a student choosing at random might do much better or much worse than this.
You can still do a lot better or worse than 0% if you're guessing.
I've never tried to work this out, but my intuition on this subject (and I have thought about it before) is that the two options are probably isomorphic; I suspect scores would follow the same distribution just scaled and shifted so that 25% in the "only count right answers" lands on 0% in the "penalize wrong answers" test. This may be right, or it may not, but it's definitely not clear (even after reading the "article") that this is the case.
There is one important caveat, and it illustrates what I see as the *real* benefit of this: it means that if you are only 2/3 of the way through the test and time is up, you don't have to run through and quickly fill in guesses for the remaining 1/3 of the test. My isomorphism probably only works (if it does at all) if you assume that any unanswered questions in the "penalize wrong answers" test are randomly filled in.
It's not that they can't do exams, but the high-pressure scenario really freaks them out.
OTOH, if you can't deal with high-pressure scenarios you should be trying to become most kinds of doctors or some kinds of lawyers.
There's also Jonathan Coulton's I'm Your Moon, a love song from Charon to Pluto.
Watch this video for his introduction to the song, but you can download (for free -- CC license -- or buy) an MP3 here, since the video cuts out briefly every couple minutes.
Stop being dense. You don't patent a result -- you patent methods. (At least for utility patents. Design patents and biological patents are different, but neither of those apply here.)
So unless their system works by intercepting acoustic waves with an eardrum that vibrates tiny bones that move a liquid that triggers tiny hairs which send electrical sigals to a mass of neurons which somehow figures it out, no, the ear isn't prior art. Considering that we have not much better than "not a clue" how the brain actually associates the sound you hear to memory, I am skeptical that this is how their approach works.
Okay, I read Linus's whole post (the linked page was down, I had to use the google groups link someone else posted) and it's actually a lot more tempered and reasonable than that little quote reveals.
Wow, that's certainly a weirder version of what "go to hell" means than what I'm used to.
Yes, the article is inconsistent and stupid. See the table at the bottom of the last page of the article.
What are YOU smoking? It's the *article* that (apparently, according to a couple other posts too) says that Pidgin isn't available for Windows.
This is America. We've been surrendering our rights for the past 50 years with a dumb smile on our faces.
I'm sorry, I believe you misspelled "about 225".
I wonder if Microsoft compiles Windows using the Intel compilers?
Microsoft uses its own cl.exe to build Windows.
It was a big source of pride for them that they got the linux kernel to build in icc without patching. *eye roll*
Eye roll? Why?
The Linux kernel has disclaimers saying that it's made for GCC and they don't care if it works on other compilers, and it makes pretty heavy use of GCC extensions. The kernel isn't portable code compiler-wise almost at all.
Or better yet rout their call back to their number (if its local or an 800) It would be great to have them call their own receptionest..
The only drawback of that approach is that it may not be legal to record the conversation and post it online since you are not a party to the conversation.
I wonder what you would need to do to make it kosher? Maybe include you in on the call but still route back, then just you be really quiet? (This assumes you're in a one-party state of course.)
Then why do we now have locks on the doors?
Same reason we have most airport security: it provides the illusion of security.
(Okay, so it probably helps a little bit. Personally, I suspect that door locks would very possibly either prevented or more likely substantially lessened the severity of 9/11. I also think that the GP is right, and we would have just seen a different attack.)
If the door was locked, how long would it take for the pilot to open the door if the terrorist started executing one passanger a minute until the door was opened?
On the flip side, how many of those executions would it take before the passengers turned *all* the flights into flight 93?
Then it's just a race to see whether the passengers react or the pilot caves first. My suggestion: with the terrorists with knives instead of guns, the passengers.
Why is better than ?
Answer: unless for some reason you are coding on a screen that only shows you 3 lines of text at once, it isn't.
Why make the programmer (i.e. you in a couple weeks) think harder about what the code does then you need to?
loqi said it best: "Expressions are like sex: they're better without side-effects."
Yeah I know that most cases it wouldn't matter, but I can think of quite a few worst-case scenarios with a lot of lookups.
You're not going to trigger any cache effects that matter because you use a const instead of #define.
In any case, I know that you can modify const values by getting a pointer, casting to non-const, and then modifying it.
If you want to provoke undefined behavior... "can" is too strong a word. You meant to say "may be able to in a non-portable and unreliable way".