If you need to allocate 50 megs dynamically with C, the OS will look for the largest chunk of 50 megs; if it can't find it, it must set pointers on the fragmented ram.
Not on any modern OS I've ever seen. Operating systems that use paging, which is most of them, do not work this way. The memory they give to processes is not necessarily contiguous: it doesn't have to be, as it stores a mapping of virtual memory pages to real physical pages.
An array that spans, say, 50 pages could be scattered all over physical memory, but because its virtual addresses are still contiguous, that doesn't affect how it's accessed from within the process.
If the Java VM actually does allocate all the RAM at the beginning, that's horribly inefficient. It's much, much better to let the OS manage memory instead of letting a process take far more than it needs. Unless the java process is all you care about...
If that's how it happened, that's the single worst miscarriage of justice I've ever heard of in the States. No wonder crime is so rampant in some parts...
Finally, if you absolutely must not get caught... claim to be armed.
Why are you giving people advice on how to get away with shoplifting? Shoplifters should be caught. There's no justification for their taking things that don't belong to them. Or do you disagree with the concept of property?
You're quite good at pointing out logical fallacies in what other people are saying, I'll give you that. But in doing so, you're missing the point of what they're saying. I think I get it, so I'll tell you what I think.
For every thing that can be called objectively good, it's possible to find a way to show that it isn't. Fair enough. This means that nothing is objectively good. But then that in turn would suggest that the term "objectively good" is useless. So why concern ourselves with it? It can be useful to say that something is good if an overwhelming majority of people think it is, since we know that there must be something about it that convinces them of this. Maybe McDonald's isn't the best example, but maybe something like "having a drink of water when you're thirsty" is. There will always be some kooks who disagree, but who cares?
When people say that something is good, I think that this is what they mean. In this way we can say that the Beatles were a good band in a way that's meaningfully different from saying that those weird artists who make things out of poop are good artists. (hint: they are not.) It's true that it's not an objective truth, but nothing is.
In my few years on this Earth, I've found that a good way to live in reality is to ignore meaningless questions that can't be answered, such as: is it all really a dream? Are other people actually real, or am I imagining them? Am I in the Matrix? They're interesting to think about, but ultimately irrelevant to life on Earth, because in the end, we have to live our lives, real or not.
And yet which are true nonetheless. Serious mathematicians do not bother themselves with the question of whether or not certain basic axioms are true*: they accept them to be true and do useful things with them.
Pointing out that nothing in logic can be proven without relying on unprovable things is a useless statement: there is in fact nothing that we know that can be proven that conclusively. Only those who live in a fantasy world and reject reality waste time with these concerns.
* By this I am referring to the basic, uncontroversial axioms; not, say, the Axiom of Choice, which some people disagree about.
But it has one disadvantage. Suppose you're an ISP who has a few normal customers and one who'll use up as much of the available bandwidth as they possibly can. When the normal customers are largely idle, he'll get the lion's share of the full bandwidth, and when the normal customers are placing heavy demands on bandwidth, they ALL go slow. The heavy user sometimes sees amazing transfer speeds, but the normal users never get a chance to because of him.
An alternative, then, would be to have traffic priorities. A user who is constantly downloading and uploading gets as much bandwidth as he needs - when no one else is using it. But when someone who rarely makes heavy demands suddenly needs to, then he can get all the priority, and the heavy user can be throttled back beyond his 1/n fair share, so that others can have the chance to see good transfer speeds too.
Not effective? The Kamikaze pilots were, in 1945, the most effective anti-ship weapon the Japanese had. At one point they were sinking so many ships that if the sinkings had continued at that rate, they could sink every American ship in the Pacific with planes to spare. They were so deadly that the Americans had to change their tactics to defeat the Kamikazes (the "big blue blanket", heavier guns for AAA to blow planes apart instead of just making them come down, Corsairs on carriers, etc).
Kamikazes were VERY effective. They were essentially stand-off guided anti-ship missiles.
Unix's privilege separation wouldn't prevent something like, say, trashing all the user's files - files that are usually more important than the easily restored operating system. Don't be fooled into thinking that even Unix does security right.
doesn't it make sense to reduce the emission of greenhouse gases?
Not if the impact on the economy is as devastating as the Kyoto accord would be. There's no reason to take steps that'll reduce to poverty more people in developed nations just because it might have a positive effect on global warming.
If you're going to take out an insurance policy on our environment, don't take out one that will cost us more than the worst-case impact of the global warming itself... Do it economically or don't do it at all.
Don't you get it? They're not going to demand more money: at least not enough to make their wages competitive with ours. Because of their cost of living, they can live like kings on wages that would reduce American workers to starvation.
The disparity between the exchange rate and the cost of living in foreign countries is the whole root cause of outsourcing, and it isn't going to be solved soon, if ever.
It seems to me that it's just a matter of semantics. You can qualify the statement and say that an algorithm is O(n*log(n)) in the average case, and that statement is correct, even though it may be O(n^2) in the worst case. From what I recall of the algorithms course I took, they never said that O(f(n)) is defined specifically for the worst case of an algorithm. I think it's pretty much up to you how you mean it.
And after all, what matters is that people understand what you mean. If the person or people you're communicating with understand that you mean average case when you say O(f(n)) and don't specify average or worst case, then that would seem to me to be perfectly correct.
The OPs point is exactly that we should take the long view _before_ stuffing things up...
That isn't what he said, though. He said: "who gives a fuck about dosage," which is stupid simply because some dosages are okay. That was the statement I objected to. It's better to know what dosages carry what risks than to assume that any nonzero dosage is terribly wrong.
In essence, you seem to say "We don't know all the dangers, problems and threats in the underlying system, and we can't quantify all those we do know. So let's just Throw In Some More!"
No. My point was that in some cases, we DO know what the dangers are, and in those cases we shouldn't let stupid people make the decisions. Nuclear power is an excellent example of this. We can quantify the risks, and we know that it's a far better system than coal burning (many orders of magnitude better), but because idiots seem to rule the States, there hasn't been a new nuclear power plant in the States for 20 years or more.
And that is exactly why environmentalists are regarded by many as loudmouthed idiots. Not all are, but people with this attitude tarnish the reputation of the whole.
You clearly don't get it, so I'll fill you in. Dosage matters because there is no such thing as zero risk. You think mercury is so bad, and we need none of it in the atmosphere (or water or whatnot). But even if this could be achieved: then what about, say, radium? It's naturally occurring, 40 times as toxic as Plutonium, and kills you in the same way. How is it worse to die from mercury poisoning than from ordinary things like radium, or from anything else for that matter? There's always gonna be people dying from some cause or other, and people like you would focus on something that kills some small number of people while coal-burning power plants kill 10,000 people in the States per year.
Obviously we need to take the long view when it comes to managing the environment of our planet. That's just common sense, and granted, it is something that few people seem to understand as important. We need to understand the impact of things we do, and understand risks, and learn how to minimize them in the most efficient way we can. But what we don't need is to go off half-cocked about something retarded like tiny dosages of things that are much less dangerous, in those dosages, than things we encounter every day.
That most people are too stupid to comprehend this depresses me.
Ah, but it's not impossible. As far as I understand it, when someone requests a file with a given hash, the end result is not only that the source node and the end node have it, but several nodes in between. This is how things get cached and ultimately wind up having more than one source, thus providing anonymity.
A while back someone on this site explained how a successful attack could work. Basically, the RIAA (or whoever, really) would put lots and lots of machines with terabytes of nonsense data on Freenet. Each node would not only serve the data, but also continually request it from many other spam nodes. In this way it would get seen by Freenet as high-demand data and would tend to push real data out of cache. Then when Johnny Pirate looks for his favorite mp3s on Freenet... they just aren't there anymore.
There is no need to restrict their ability while browsing a web site...
Yes there goddamn is. If you had every possible action in a context menu, there'd be 20 or so items there and it'd be hard to find the ones that you DO use.
It's something I've always hated about Windows: there are always a whole ton of items in the context menu when I only ever use two or three items. Right-clicking in IE right now reveals 16 items, some of which are actually WORSE than useless: Set as Background, for instance, which kills my current background with whatever I'm right-clicking on. Pretty easy to accidentally hit it when I'm trying to save an image...
The point here is: don't place every possible action into a menu! KDE's menus are bloated! As for this:
However, in the screenshot, which contains 32 applications, only 7 are KDE applications! You can't claim the KDE menu is too blated because of all the other junk on the system...
Do you think she put those applications into the menu HERSELF? If not, how did they get there? She said she just installed the Fedora RPMs. Regardless of whether or not the included apps are strictly a part of KDE or not, they were there in what she had to work with, and a stupidly bloated menu was the result. Nothing unfair in pointing out that it's bad thing in a review of an desktop environment.
Besides, even the K menu is itself too bloated, and there's no way you can say that it's not a part of KDE. My Windows Start Menu has 9 items, 7 of which are useful to me. They all have big target areas and it's easy to get the one I want right away. KDE's K menu has 24 - all of them put there by KDE. It's such a pain that it turned me off KDE when I tried it out. Gnome was a significantly better desktop as far as I was concerned (with the exception of their whole "take effect immediately" guideline - I absolutely loathe it).
Namely this is one of the poorer reviews I have read on OSNews, and that is saying ALOT since they are normally quite bad.
I find it funny how she says "KDE is great" right in the conclusion of her review, and still you find faults with her most sensible objections. I think you'd find fault with ANYTHING bad that anyone said about KDE.
I can think of such a situation. Actually, someone else thought it up for me:
...(distribution of) a binary module that contains snippets of object code directly compiled from GPLed methods in the kernel (because they were inlined)...
from Hobbex's post.
Hobbex was of the opinion that this is copyright infringement. I think that the use of inlined functions could fall under fair use. The rest of the code isn't being distributed at all, so the GPL wouldn't have to be agreed to, and normal copyright law should let you do this.
The licenses for most software are designed blah blah communist propaganda...
Now why does it say most licences are designed to take away your freedom? If copyrights are truely free market, that shouldn't be the case!
Sir, that statement refers to END USER LICENSE AGREEMENTS. You know, the things that people click through that nobody really agrees to? They aren't real contracts, they're just licenses that the software publishers hope you agree to. But copyright is a different animal altogether.
I don't see how anyone can be against copyright. If people like you and RMS and Linus want to release their software to everyone, under the GPL, that's great. It's a good thing. But forcing EVERYONE to do the same? That roughly is what the end effect of no copyrights would be. What is wrong with you, man?
(By the way, for you mods on crack, the top quote about communist propaganda was a JOKE.)
Those are good points. And I know you already have a few responses, but I must get my two cents in... again.
If indeed they don't have any AVGAS around, or they do but need all that they have, then they are acting perfectly reasonably. However, the article said:
(they) refuse to give him the fuel, saying they do not want to encourage tourism in the Antarctic.
and this implies not that they don't have any, but that they do and they don't want to sell it to him. It doesn't say it for sure, but it is implied. So if this is indeed the case, then they are being complete dicks.
And thus there was no reason for the crackhead mods to mod me down as "flamebait". (I know this wasn't your fault, of course...)
Your post was interesting and informative. Very much so, in fact. But I think you're missing the point.
Basically, fuck what the rules say. Rarely is one legally compelled to help another in need. That doesn't mean you shouldn't do it. They could sell him some fuel, and make a profit doing so, and send him on his way. It would be very easy for them to do this. The fact that they don't have to do this, according to the rules, doesn't mitigate the fact that they're a bunch of assholes.
But if we never land there, how will we ever discover life on Europa? From orbit somehow? If life on Europa only exists deep in its oceans, how will that work?
We can't accomplish anything without risking something.
If the wrapper is under the GPL and the nVidia driver itself is linked with this wrapper code..what licence is the core part of the driver under?
Interesting thought... I think it works in this case because nVidia owns the copyright to both the closed-source part of the driver AND the GPL part. They can distribute the GPLed part because they are complying with the terms of the GPL, and they can distribute the closed-source part because they own it. It may well violate the GPL that covers the other part of their code, but the beauty of this is that no one can sue them for it except for... themselves.
Slashdot Mirror
An array that spans, say, 50 pages could be scattered all over physical memory, but because its virtual addresses are still contiguous, that doesn't affect how it's accessed from within the process.
If the Java VM actually does allocate all the RAM at the beginning, that's horribly inefficient. It's much, much better to let the OS manage memory instead of letting a process take far more than it needs. Unless the java process is all you care about...
You're quite good at pointing out logical fallacies in what other people are saying, I'll give you that. But in doing so, you're missing the point of what they're saying. I think I get it, so I'll tell you what I think.
For every thing that can be called objectively good, it's possible to find a way to show that it isn't. Fair enough. This means that nothing is objectively good. But then that in turn would suggest that the term "objectively good" is useless. So why concern ourselves with it? It can be useful to say that something is good if an overwhelming majority of people think it is, since we know that there must be something about it that convinces them of this. Maybe McDonald's isn't the best example, but maybe something like "having a drink of water when you're thirsty" is. There will always be some kooks who disagree, but who cares?
When people say that something is good, I think that this is what they mean. In this way we can say that the Beatles were a good band in a way that's meaningfully different from saying that those weird artists who make things out of poop are good artists. (hint: they are not.) It's true that it's not an objective truth, but nothing is.
In my few years on this Earth, I've found that a good way to live in reality is to ignore meaningless questions that can't be answered, such as: is it all really a dream? Are other people actually real, or am I imagining them? Am I in the Matrix? They're interesting to think about, but ultimately irrelevant to life on Earth, because in the end, we have to live our lives, real or not.
That's just my philosophy.
Pointing out that nothing in logic can be proven without relying on unprovable things is a useless statement: there is in fact nothing that we know that can be proven that conclusively. Only those who live in a fantasy world and reject reality waste time with these concerns.
* By this I am referring to the basic, uncontroversial axioms; not, say, the Axiom of Choice, which some people disagree about.
That's a good way of doing it.
But it has one disadvantage. Suppose you're an ISP who has a few normal customers and one who'll use up as much of the available bandwidth as they possibly can. When the normal customers are largely idle, he'll get the lion's share of the full bandwidth, and when the normal customers are placing heavy demands on bandwidth, they ALL go slow. The heavy user sometimes sees amazing transfer speeds, but the normal users never get a chance to because of him.
An alternative, then, would be to have traffic priorities. A user who is constantly downloading and uploading gets as much bandwidth as he needs - when no one else is using it. But when someone who rarely makes heavy demands suddenly needs to, then he can get all the priority, and the heavy user can be throttled back beyond his 1/n fair share, so that others can have the chance to see good transfer speeds too.
Seems more fair to me.
Not effective? The Kamikaze pilots were, in 1945, the most effective anti-ship weapon the Japanese had. At one point they were sinking so many ships that if the sinkings had continued at that rate, they could sink every American ship in the Pacific with planes to spare. They were so deadly that the Americans had to change their tactics to defeat the Kamikazes (the "big blue blanket", heavier guns for AAA to blow planes apart instead of just making them come down, Corsairs on carriers, etc).
Kamikazes were VERY effective. They were essentially stand-off guided anti-ship missiles.
Unix's privilege separation wouldn't prevent something like, say, trashing all the user's files - files that are usually more important than the easily restored operating system. Don't be fooled into thinking that even Unix does security right.
If you're going to take out an insurance policy on our environment, don't take out one that will cost us more than the worst-case impact of the global warming itself... Do it economically or don't do it at all.
Don't you get it? They're not going to demand more money: at least not enough to make their wages competitive with ours. Because of their cost of living, they can live like kings on wages that would reduce American workers to starvation.
The disparity between the exchange rate and the cost of living in foreign countries is the whole root cause of outsourcing, and it isn't going to be solved soon, if ever.
It seems to me that it's just a matter of semantics. You can qualify the statement and say that an algorithm is O(n*log(n)) in the average case, and that statement is correct, even though it may be O(n^2) in the worst case. From what I recall of the algorithms course I took, they never said that O(f(n)) is defined specifically for the worst case of an algorithm. I think it's pretty much up to you how you mean it.
And after all, what matters is that people understand what you mean. If the person or people you're communicating with understand that you mean average case when you say O(f(n)) and don't specify average or worst case, then that would seem to me to be perfectly correct.
You clearly don't get it, so I'll fill you in. Dosage matters because there is no such thing as zero risk. You think mercury is so bad, and we need none of it in the atmosphere (or water or whatnot). But even if this could be achieved: then what about, say, radium? It's naturally occurring, 40 times as toxic as Plutonium, and kills you in the same way. How is it worse to die from mercury poisoning than from ordinary things like radium, or from anything else for that matter? There's always gonna be people dying from some cause or other, and people like you would focus on something that kills some small number of people while coal-burning power plants kill 10,000 people in the States per year.
Obviously we need to take the long view when it comes to managing the environment of our planet. That's just common sense, and granted, it is something that few people seem to understand as important. We need to understand the impact of things we do, and understand risks, and learn how to minimize them in the most efficient way we can. But what we don't need is to go off half-cocked about something retarded like tiny dosages of things that are much less dangerous, in those dosages, than things we encounter every day.
That most people are too stupid to comprehend this depresses me.
Ah, but it's not impossible. As far as I understand it, when someone requests a file with a given hash, the end result is not only that the source node and the end node have it, but several nodes in between. This is how things get cached and ultimately wind up having more than one source, thus providing anonymity.
A while back someone on this site explained how a successful attack could work. Basically, the RIAA (or whoever, really) would put lots and lots of machines with terabytes of nonsense data on Freenet. Each node would not only serve the data, but also continually request it from many other spam nodes. In this way it would get seen by Freenet as high-demand data and would tend to push real data out of cache. Then when Johnny Pirate looks for his favorite mp3s on Freenet... they just aren't there anymore.
This would ruin the utility of the network.
Exactly how can Intel make Microsoft do anything they don't want to do?
No, if you're dumb you'll short it. For that to work effectively, you not only have to know that it will go down, but when. And you don't.
And the amount of money you could lose is limited only by your imagination.
It's something I've always hated about Windows: there are always a whole ton of items in the context menu when I only ever use two or three items. Right-clicking in IE right now reveals 16 items, some of which are actually WORSE than useless: Set as Background, for instance, which kills my current background with whatever I'm right-clicking on. Pretty easy to accidentally hit it when I'm trying to save an image...
The point here is: don't place every possible action into a menu! KDE's menus are bloated! As for this:
Do you think she put those applications into the menu HERSELF? If not, how did they get there? She said she just installed the Fedora RPMs. Regardless of whether or not the included apps are strictly a part of KDE or not, they were there in what she had to work with, and a stupidly bloated menu was the result. Nothing unfair in pointing out that it's bad thing in a review of an desktop environment.
Besides, even the K menu is itself too bloated, and there's no way you can say that it's not a part of KDE. My Windows Start Menu has 9 items, 7 of which are useful to me. They all have big target areas and it's easy to get the one I want right away. KDE's K menu has 24 - all of them put there by KDE. It's such a pain that it turned me off KDE when I tried it out. Gnome was a significantly better desktop as far as I was concerned (with the exception of their whole "take effect immediately" guideline - I absolutely loathe it).
I find it funny how she says "KDE is great" right in the conclusion of her review, and still you find faults with her most sensible objections. I think you'd find fault with ANYTHING bad that anyone said about KDE.
from Hobbex's post. Hobbex was of the opinion that this is copyright infringement. I think that the use of inlined functions could fall under fair use. The rest of the code isn't being distributed at all, so the GPL wouldn't have to be agreed to, and normal copyright law should let you do this.
That's my non-lawyer opinion, anyway.
Yes... what a "good" idea that is. Something goes wrong with the plane, and you wind up in space.
Nope, can't see any problem with that idea...
I don't see how anyone can be against copyright. If people like you and RMS and Linus want to release their software to everyone, under the GPL, that's great. It's a good thing. But forcing EVERYONE to do the same? That roughly is what the end effect of no copyrights would be. What is wrong with you, man?
(By the way, for you mods on crack, the top quote about communist propaganda was a JOKE.)
If indeed they don't have any AVGAS around, or they do but need all that they have, then they are acting perfectly reasonably. However, the article said:
and this implies not that they don't have any, but that they do and they don't want to sell it to him. It doesn't say it for sure, but it is implied. So if this is indeed the case, then they are being complete dicks.
And thus there was no reason for the crackhead mods to mod me down as "flamebait". (I know this wasn't your fault, of course...)
Your post was interesting and informative. Very much so, in fact. But I think you're missing the point.
Basically, fuck what the rules say. Rarely is one legally compelled to help another in need. That doesn't mean you shouldn't do it. They could sell him some fuel, and make a profit doing so, and send him on his way. It would be very easy for them to do this. The fact that they don't have to do this, according to the rules, doesn't mitigate the fact that they're a bunch of assholes.
But if we never land there, how will we ever discover life on Europa? From orbit somehow? If life on Europa only exists deep in its oceans, how will that work?
We can't accomplish anything without risking something.
Beauty, ain't it?