It seems to me it is perfectly reasonable that the students who did not sleep may have had their performance impaired because they were tired.
You're right, the experiment was not controlled for that (how would you control for that), but I find it extremely hard to believe that a sleeping brain is just sitting "idle" like you suggest.
Haven't you ever woken up in the morning and immediately seen the answer to some problem from the day before? Or the week before? It happens to me all the time.
Aren't you ever just sitting around and, out of the blue, comes a crystal clear idea that you swear you weren't even thinking about? These things aren't just random occurrences. I believe there is an entire world of thought in the brain that happens on a subconscious level, inaccessible to us except for the moments when those sub-faculties of the brain decide, for one reason or another, to express their thoughts to us consciously.
Have you ever sky dived, or climbed a precarious mountain, or flipped your bike over, or been in a car accident? Didn't you notice how your senses and thoughts suddenly became crystal clear? That's your brain shutting down all those "background processes" that normally run all the time. Realizing that you are in a dire situation, your brain essentially hands over the car keys to your conscious awareness, to give you the full abilities of your mind to extract yourself from the dangerous situation. This is why people do these "ridiculous" things like climb mountains. You take full command of your own mind in these situations.
I have absolutely no problem believing that complex, but subconscious, thought processes occur during sleep, and also during wakefulness.
Releasing under the GPL ensures you don't sell out, ever.
You're talking nonsense. At any moment, the copyright holder of a GPL licensed program can turn around and say "Okay, I'm closing the source, any further releases will now be under this new license." The authors can't prevent people from distirbuting/modifying the code that is already available, but they can continue their own closed source development and make releases under any terms they wish.
If you hold copyright on a work, then you always have total control over it, and I don't know of any way you could give up that control even if you wanted to, except for assigning the copyright to someone else (in theory, there's "Public Domain," but that's hard to achieve in reality).
moving mass from one object to the other therefore doesn't change anything
Mathematically incorrect.
The force of gravity is proportional to the PRODUCT of the two masses. Thus, if a mass p is transferred from a body of mass M to a body of mass m, the resulting gravitational force will be proportional to (M-p)*(m+p) which is decidedly not equal to M*m (unless the mass M was exactly p heavier than mass m to begin with).
Not true. Quicksort has less overhead than mergesort and is generally faster.
I think what he's referring to is efficiency on linked lists as opposed to arrays. Quicksort is slower than mergesort on lists because for each element of the list it has to decide whether to place that element into the left-hand sublist or the right-hand sublist, then fool around with a bunch of links. Mergesort has less overhead because you can just walk down to the middle of the list, and split the link there, without having to re-link a bunch of nodes onto different lists.
And quicksort can be made stable, if you want it to be. I believe the implementation given by somebody who replied to my original post, was a stable sort. Depending on the behavior of Python's list comprehensions (i.e., whether they traverse the list in-order), my implementation is also stable.
The WTC terrorists deliberately sought to kill innocent people. SPEWS just lists IP blocks, and if innocents are there, their IP blocks get listed
I see, kind of like how pro-life activists publish hit lists. They're just making a list, right? They're not responsible for what people do with that list, right?
You're a liar. RBL operators know damn well that there are innocent people who will be affected by their actions. It's part of their strategy. They can deny this, but their deception is pathetic. And you're just another fucking liar.
You're whining about a side-effect as if it's a motivation.
It's most certainly not a side-effect. What other reason is there to ban entire netblocks when only a single IP is the source of the spam? The only purpose I can see is to make those customers hostages to force them to leave their ISP. You can lie and say it's not intentional, but I see through it.
If the purpose of SPEWS was really to block spam from known bad hosts, then it would be a reactive system which places bans after specific evidence that a particular IP is misbehaving. By preemptively banning entire netblocks you are intentionally causing collateral damage among customers in an effort to damage the ISP.
In other words, you're using potential damage to innocent bystanders as a negotiation tactic.
I'm not trying to be inflammatory; these people may not be blowing up buildings or killing people, but they sure as hell operate in the same manner:
1. Find innocent users who are only weakly associated to the actual people you have a disagreement with -- in this case, customers on netblocks belonging to higher-level companies who, in one way or another, harbor spammers. (Analogy: people in WTC buildings whose only connection to American foreign policy is the fact that they are Americans.)
2. Hold these people hostage by making it impossible for them to use email, in an effort to extort them into placing pressure on the aforementioned higher-level company to stop harboring spammers (Analogy: blow up the building, threaten to do it again and again until demands are met.)
3. Endure a hostile backlash when your plan doesn't work out -- sysadmins, hackers, tech-savvy victims of this approach use massive DDoS attacks to take down multiple RBLs (Analogy: USA blows the shit out of Afghanistan, Iraq, redoubles efforts to decimate the terrorist networks.)
SPEWS and other "services" like it share the same moral motivation as terrorists, and I find their behavior repulsive, unethical, and deserving of punishment.
Hey, let's throw static/dynamic scoping into the mix just to further confuse people:
Python is a strongly, dynamically typed language with static variable scoping:-)
Not that there are a great many dynamically scoped languages around these days...
it most assuredly does not use n log n memory in the typical case
Are you sure? In the typical case, the list is split evenly in half, with n/2 items in the left side, and n/2 items in the right. Therefore at recursion frame k there are n/2+n/2=n items in memory. Since the list will split log n times, there are log n recursion frames, which means n log n memory in use at the deepest level of recursion.
These days, compilers on recent hardware are fast enough that link/compile times are simply not a bottleneck to development productivity.
Apparently you've never worked on a project with a "core" header file that gets #include'd by about 5000 source modules. Make one little diddly change to that header and you have to recompile 5000 files.
It can sure as hell be a massive waste of time. Now, whether or not it's good practice to structure your program in such a way that everything depends on a single header, is another issue entirely...
Your quicksort traverses the list three times in those spiffy list comprehensions.
Yep. Another "failing" of this implementation (and yours, too) is that it doesn't do the sort in-place, so it uses n log n memory. The in-place sort implementation is what makes most really fast implementations so hairy and difficult to understand.
do-while is rarely useful. The few times I actually find a use for it in C, I find myself thinking "Wow... I get to use do-while today!"
It's too easy to accidentally use do-while when you should have just used while. It actually makes the language less error-prone, because in those few cases where you do have an unconditional first pass, you are forced to structure the code differently and actually think about what you're doing.
You can always transform: do stmt while foo into stmt while foo stmt which isn't even longer (if stmt is actually many statements, it should be a function anyway). It's not worth introducing an abusable language construct just so you get to be lazy and not code a function when you should.
One of the things I initially judge a language by is how elegant it is to code the quicksort algorithm (yes, I'm aware Python has a built-in sort function -- that is not the point). Here's quicksort in Python:
def quicksort(list): if len(list) > 1: pivot = list[0] left = [x for x in list if x < pivot] right = [x for x in list if x > pivot] pivot = [x for x in list if x == pivot] return quicksort(left) + pivot + quicksort(right) return list
No, my point was if he did have some private IP in a logfile somewhere, there's no way to demonstrate that that IP came from Valve's internal LAN, as opposed to some other internal LAN.
"Oooh look, this guy made a little mistake, what an ignorant American!" Aren't you making a whole fucking boatload of assumptions yourself? Such as, for example, what my attitudes are toward the rest of the world, for one thing?
My point was that there are common private IP addresses, and common street names. s/28th street/Station Road/, okay?
I can think of far more important mistaken assumptions to make, than whether all cities have numbered streets in them. Such as Nixon giving the "V" sign to the citizens of Italy, for example.
A MAC would only show up in the ARP table (which is memory-only, not stored on disk), and only if he was physically connected to their LAN. Assuming he did connect to their network, it would have been over the Internet proper, and MAC addresses would never come into the equation.
I'd say simply bringing up the analogy I just mentioned would be sufficient. People might not be technically minded but they aren't idiots.
An even better analogy would be, "The crime was committed on 28th street. This guy's private records indicate that he's been to 28th street in the last several weeks. Hence, guilty as charged."
Except that's pretty much every city in the world has a "28th street." I'm quite sure a jury is capable of understanding this.
I'm an idiot because I know that Hungarians don't refer to their country as Hungary? And that I choose not to get all bunched up when someone misspells a word that isn't even the actual name of their country?
Yes, I'm really sure that every person in Hungary must want to kill me right now, for pointing out that they call their country by another name. How dare I take their side?
Smearing America at every chance like that is pretty lame.
Relax, I only brought it up because the OP did. He claimed it was stereotypically American to misspell the name of another country, I just countered that it's also stereotypically American to not even know what the hell the actual name of a country is.
And yes, I'm well aware it's a stereotype, I'm not saying all Americans are stupid or anything of that sort. Relax.
And if I was a Hungarian, I'd be pretty damned ticked off about it. The Germans refer to their country as "Deutschland," and their country code is appropriately ".de".
The Hungarians deserve the same, but were apparently marginalized by ISO along with probably a few dozen other countries.
The OP said it was ignorant to misspell the name Hungary. I'd say it's vastly more ignorant to assume that because we call a nation "Hungary" that Hungary must therefore be its actual name. That is the epitome of Ameri-centricity.
If I hadn't done anything wrong, I'd stick around to see what's being confiscated.
No point in doing that. They give you a receipt for anything they take, and they can't use and item as evidence in court without first proving that they gave you a receipt for that item. So everything they touched is going to be clearly itemized.
It seems like this guy's first priority was to sound an alarm...
Makes sense to me.
"Hey guys. I thought I'd let you know, there's a bunch of cops searching my apartment right now, so if I end up missing by the end of the day, you'll know what happened. In that case, could you please call a lawyer for me?"
It's the kind of thing friends tell each other, don't you think?
Slashdot Mirror
You're right, the experiment was not controlled for that (how would you control for that), but I find it extremely hard to believe that a sleeping brain is just sitting "idle" like you suggest.
Haven't you ever woken up in the morning and immediately seen the answer to some problem from the day before? Or the week before? It happens to me all the time.
Aren't you ever just sitting around and, out of the blue, comes a crystal clear idea that you swear you weren't even thinking about? These things aren't just random occurrences. I believe there is an entire world of thought in the brain that happens on a subconscious level, inaccessible to us except for the moments when those sub-faculties of the brain decide, for one reason or another, to express their thoughts to us consciously.
Have you ever sky dived, or climbed a precarious mountain, or flipped your bike over, or been in a car accident? Didn't you notice how your senses and thoughts suddenly became crystal clear? That's your brain shutting down all those "background processes" that normally run all the time. Realizing that you are in a dire situation, your brain essentially hands over the car keys to your conscious awareness, to give you the full abilities of your mind to extract yourself from the dangerous situation. This is why people do these "ridiculous" things like climb mountains. You take full command of your own mind in these situations.
I have absolutely no problem believing that complex, but subconscious, thought processes occur during sleep, and also during wakefulness.
You're talking nonsense. At any moment, the copyright holder of a GPL licensed program can turn around and say "Okay, I'm closing the source, any further releases will now be under this new license." The authors can't prevent people from distirbuting/modifying the code that is already available, but they can continue their own closed source development and make releases under any terms they wish.
If you hold copyright on a work, then you always have total control over it, and I don't know of any way you could give up that control even if you wanted to, except for assigning the copyright to someone else (in theory, there's "Public Domain," but that's hard to achieve in reality).
Mathematically incorrect.
The force of gravity is proportional to the PRODUCT of the two masses. Thus, if a mass p is transferred from a body of mass M to a body of mass m, the resulting gravitational force will be proportional to (M-p)*(m+p) which is decidedly not equal to M*m (unless the mass M was exactly p heavier than mass m to begin with).
Yeah! They're stealing from the rightful owners of the moon! Oh wait!
From the article (you DID read it right): they are estimating there is a total of about 1,100,000 metric tons of He3 on the moon.
Now, the moon weighs 7.4e22 kilograms. Even if we remove all 1.1e6 metric tons of He3, the mass of the moon will only change by 1 part in 67 trillion.
And that's assuming we were somehow capable of mining every last gram of He3 -- A complete impossibility.
I think what he's referring to is efficiency on linked lists as opposed to arrays. Quicksort is slower than mergesort on lists because for each element of the list it has to decide whether to place that element into the left-hand sublist or the right-hand sublist, then fool around with a bunch of links. Mergesort has less overhead because you can just walk down to the middle of the list, and split the link there, without having to re-link a bunch of nodes onto different lists.
And quicksort can be made stable, if you want it to be. I believe the implementation given by somebody who replied to my original post, was a stable sort. Depending on the behavior of Python's list comprehensions (i.e., whether they traverse the list in-order), my implementation is also stable.
I see, kind of like how pro-life activists publish hit lists. They're just making a list, right? They're not responsible for what people do with that list, right?
You're a liar. RBL operators know damn well that there are innocent people who will be affected by their actions. It's part of their strategy. They can deny this, but their deception is pathetic. And you're just another fucking liar.
It's most certainly not a side-effect. What other reason is there to ban entire netblocks when only a single IP is the source of the spam? The only purpose I can see is to make those customers hostages to force them to leave their ISP. You can lie and say it's not intentional, but I see through it.
If the purpose of SPEWS was really to block spam from known bad hosts, then it would be a reactive system which places bans after specific evidence that a particular IP is misbehaving. By preemptively banning entire netblocks you are intentionally causing collateral damage among customers in an effort to damage the ISP.
In other words, you're using potential damage to innocent bystanders as a negotiation tactic.
That makes you a terrorist, in my book.
1. Find innocent users who are only weakly associated to the actual people you have a disagreement with -- in this case, customers on netblocks belonging to higher-level companies who, in one way or another, harbor spammers. (Analogy: people in WTC buildings whose only connection to American foreign policy is the fact that they are Americans.)
2. Hold these people hostage by making it impossible for them to use email, in an effort to extort them into placing pressure on the aforementioned higher-level company to stop harboring spammers (Analogy: blow up the building, threaten to do it again and again until demands are met.)
3. Endure a hostile backlash when your plan doesn't work out -- sysadmins, hackers, tech-savvy victims of this approach use massive DDoS attacks to take down multiple RBLs (Analogy: USA blows the shit out of Afghanistan, Iraq, redoubles efforts to decimate the terrorist networks.)
SPEWS and other "services" like it share the same moral motivation as terrorists, and I find their behavior repulsive, unethical, and deserving of punishment.
Not that there are a great many dynamically scoped languages around these days...
I guess that shows I do much more time analysis than space analysis...
Are you sure? In the typical case, the list is split evenly in half, with n/2 items in the left side, and n/2 items in the right. Therefore at recursion frame k there are n/2+n/2=n items in memory. Since the list will split log n times, there are log n recursion frames, which means n log n memory in use at the deepest level of recursion.
Have I missed something subtle here?
Apparently you've never worked on a project with a "core" header file that gets #include'd by about 5000 source modules. Make one little diddly change to that header and you have to recompile 5000 files.
It can sure as hell be a massive waste of time. Now, whether or not it's good practice to structure your program in such a way that everything depends on a single header, is another issue entirely...
Yep. Another "failing" of this implementation (and yours, too) is that it doesn't do the sort in-place, so it uses n log n memory. The in-place sort implementation is what makes most really fast implementations so hairy and difficult to understand.
It's too easy to accidentally use do-while when you should have just used while. It actually makes the language less error-prone, because in those few cases where you do have an unconditional first pass, you are forced to structure the code differently and actually think about what you're doing.
You can always transform: do stmt while foo into stmt while foo stmt which isn't even longer (if stmt is actually many statements, it should be a function anyway). It's not worth introducing an abusable language construct just so you get to be lazy and not code a function when you should.
One of the things I initially judge a language by is how elegant it is to code the quicksort algorithm (yes, I'm aware Python has a built-in sort function -- that is not the point). Here's quicksort in Python:
def quicksort(list):
if len(list) > 1:
pivot = list[0]
left = [x for x in list if x < pivot]
right = [x for x in list if x > pivot]
pivot = [x for x in list if x == pivot]
return quicksort(left) + pivot + quicksort(right)
return list
I'd say this speaks for itself. Enjoy.
In five years, that statement will sound a lot like, "Sounds like a very good reason for me to stay off this 'Internet' thing."
Guess who'll be working at Burger King, flipping burger orders that people punched in on their cell phones...
No, my point was if he did have some private IP in a logfile somewhere, there's no way to demonstrate that that IP came from Valve's internal LAN, as opposed to some other internal LAN.
My point was that there are common private IP addresses, and common street names. s/28th street/Station Road/, okay?
I can think of far more important mistaken assumptions to make, than whether all cities have numbered streets in them. Such as Nixon giving the "V" sign to the citizens of Italy, for example.
A MAC would only show up in the ARP table (which is memory-only, not stored on disk), and only if he was physically connected to their LAN. Assuming he did connect to their network, it would have been over the Internet proper, and MAC addresses would never come into the equation.
An even better analogy would be, "The crime was committed on 28th street. This guy's private records indicate that he's been to 28th street in the last several weeks. Hence, guilty as charged."
Except that's pretty much every city in the world has a "28th street." I'm quite sure a jury is capable of understanding this.
Yes, I'm really sure that every person in Hungary must want to kill me right now, for pointing out that they call their country by another name. How dare I take their side?
Christ buddy, sit down.
Relax, I only brought it up because the OP did. He claimed it was stereotypically American to misspell the name of another country, I just countered that it's also stereotypically American to not even know what the hell the actual name of a country is.
And yes, I'm well aware it's a stereotype, I'm not saying all Americans are stupid or anything of that sort. Relax.
And if I was a Hungarian, I'd be pretty damned ticked off about it. The Germans refer to their country as "Deutschland," and their country code is appropriately ".de".
The Hungarians deserve the same, but were apparently marginalized by ISO along with probably a few dozen other countries.
The OP said it was ignorant to misspell the name Hungary. I'd say it's vastly more ignorant to assume that because we call a nation "Hungary" that Hungary must therefore be its actual name. That is the epitome of Ameri-centricity.
No point in doing that. They give you a receipt for anything they take, and they can't use and item as evidence in court without first proving that they gave you a receipt for that item. So everything they touched is going to be clearly itemized.
It seems like this guy's first priority was to sound an alarm...
Makes sense to me.
"Hey guys. I thought I'd let you know, there's a bunch of cops searching my apartment right now, so if I end up missing by the end of the day, you'll know what happened. In that case, could you please call a lawyer for me?"
It's the kind of thing friends tell each other, don't you think?