While the idea that our understanding of gravity is incorrect is very exciting to me... I don't think you can really draw the comparison between aether and dark matter.
It was easy to disprove the existence of aether with the Michelson-Morley experiment. Had that experiment not been possible it would have been very premature to jump to the conclusion that there is no aether. When it comes to dark matter, there is no easy experiment to disprove its existence and so it would be very rash to conclude that our understanding of gravity, which has worked extremely well for us for hundreds of years, is wrong.
Obviously the proper solution is for the emergency vehicles to drag a ten-mile strand of Cat5 behind them. Just imagine it -- an ambulance hurtling down the freeway with a blue network cable trailing behind, and a giant spool of Cat5 unreeling at 600 RPM... Tht tht tht tht tht tht tht
Ahh. I agree that PAE was a hideous piece of garbage. But I was referring specifically to the original segmentation in 8086 and above, which IMHO wasn't that bad and actually opened up a lot of options for the programmer.
I wouldn't have objected to having a segmented 16-bit addressing mode even if the chip had had a true 32-bit address mode. Maybe I'm crazy but I liked it.
Read the article. He was arrested for the crime of extortion, not spamming. As far as I am aware, there is no law stopping companies from paying somebody to send spam IM messages.
So why isn't the attorney general going after the advertisers? Because he can't enforce laws which don't exist, and I like things that way, thank you very much.
I didn't realize that Slashdot exists to interest you, personally. I think the discussion on this article is pretty much par for Slashdot. If you don't like it, read on to the next thing.
In order to preserve 16bit compatibility a dirty hack that was segmentation was required, that it was the best hack for the job makes it no less of a hack for a memory addressing scheme.
No. The idea of preserving compatibility with 16 bits apps was the hack, not the particular way they went about it. As you just said yourself, the elegant solution would be to ditch the idea of backward compatibility. Don't blame the solution when it's the problem that sucks.
Many processor architectures throughout computing history have used addressing modes. They are by no means "hacks."
What difference does it make to Linux developers of Mono applications whether Microsoft fails to implement their own spec?
That sort of defeats the point of.NET/Mono as a platform-independent development system, doesn't it? If you're coding in C#/Mono just because you like it, that's one thing, but touting it as the Linux version of.NET is dishonest because the idea of platform independence is one of the concepts of the.NET moniker.
I have no problem with Mono if they're just trying to implement the language/runtime, but they shouldn't bill it as.NET compatible if it's not.
I got a wild idea, why don't the "editors" of Slashdot do a 5 second google search before posting pointless Ask Slashdot questions like this and save us all a lot of time.
I never get comments like this.
First of all, and most important, if you think that Slashdot is a "waste of time" why do you bother with it? It's pretty lame to come in here commenting and then complain that you're wasting your time. "This sucks, now I have to do this guy's research for him." Uhhh.. No, you don't have to do anything. You could have just skipped the article if it sucked that much. Pointless bitching.
Second, if the question hadn't been posted, we'd have no excuse to banter on Slashdot about systems which support 16 GB of RAM. The fact that the guy needed an answer to a question, which he could have got somewhere else, is irrelevant. If he'd asked somewhere else, we'd have been deprived of this opportunity to talk about big-memory systems.
What exactly do you want Slashdot to be? Apparently, a mutual ego-stroking fest. I agree that the guy's question could have been answered with 5 minutes of Googling, but who gives a shit about some guy who doesn't know how to search? I come here to read and participate in the discussions.
There is no "reverse engineering" involved. These applications are written in C#, an open ECMA standard, and the open source Gtk+ toolkit.
The reverse engineering comes in when you try to emulate all the places where Microsoft has intentionally diverged from their own spec.
You can't even trust the MSDN documentation on simple Windows functions, how can you expect an entire application framework to actually adhere to the spec? Thinking you can just implement it straight out of the book is pretty naive. It IS reverse engineering.
Wow, you call that a hack? What other features-by-design do you consider "hacks?"
The segmentation mechanism was designed to allow the use of more memory while retaining compatibility with 16-bit, segment-naive code. It may not be beautiful but I can't think of a better way they could have implemented it.
The term "hack" is generally reserved for describing an inelegant implementation. In this case, it is the PROBLEM itself which is inelegant, not the solution to it. Segmentation was simple, easily understood, and it solved the problem.
This is due to the fact that you are working on 32-bit hardware that can only address 4GB directly, as far as I understand. Does Linux have this limit too? Or are there other 'tricks' that the Linux kernel applies to go above 4gb? Maybe other Slashdotters can elaborate on this.
On IA32, the limit exists on Linux as well as Windows. This is a hardware limitation of 32-bit addresses, as you pointed out. However, it's possible to "window" the higher memory into a fixed area under the 4 GB limit, similar to how XMM and EMM worked in the days of MS-DOS and extended memory managers.
As for memory being reserved for kernel space, that's a necessity in order for the kernel to be able to differentiate between user space and kernel space. It would be horrifically inefficient for the kernel to have to scan through a bunch of tables to figure out if a particular pointer is a kernel pointer or a user pointer. Setting an arbitrary boundary between kernel and user space lets the kernel do a simple test -- is this pointer greater than the kernel space limit? If so, it's a kernel pointer.
On Linux the boundary can be set at either 2 GB or 3 GB.
I'll grant you the right to hold your opinion about Apple fanboys, if you grant us the fact that not every Apple user falls under that definition.
I'm personally tired of being called a "fanboy" just because I decided to buy a Mini for the hell of it.
As for the topic at hand... I think Apple's true concern should be finding the individual who leaked the disc from inside the company. Perhaps they should offer to let the guy off the hook if he explains where the thing came from.
I'm glad the gamma-ray bursts are directed into space.
Ummm... We've had thunderstorms for billions of years, and presumably these gamma ray bursts as well. They are completely natural phenomena. If they could negatively affect us, we would have either evolved a method for coping millions of years ago, or we wouldn't even exist today.
OF COURSE the things are directed into space. Duh.
Maybe I'm missing something, but why not just get a second phone line and leave yourself dialed in to your ISP 24/7? Surely that'll cost you less that $40 per month.
Well, until mankind figures out a way around the pigeonhole problem -- which is NEVER -- this "problem" will always exist.
What we should be asking ourselves is, is there a way to construct a hashing algorithm for which the OPTIMAL method for finding collisions is a brute force search? So far it hasn't been done, and it hasn't been definitely proven to be possible or impossible, either.
I see a lot of people on these forums complaining that we should "just" make a hash algorithm that is unbreakable. It's a logical impossibility. Can you fit an infinite number of things into a finite number of holes and guarantee that each hole has at most one object in it? I hope people are capable of grasping that, at least.
Why don't you try it instead of being a fuck? It's trivial to install an xpi in your.mozilla directory in your home directory. I've done it many times.
As I mentioned in my post, it's not the best definition in the world, but the point remains that email is ridiculously easy to disrupt when compared to other methods of communicating over the internet.
Even with encryption I wouldn't describe email as "secure." People can still DoS you by swamping you with spam, or cause a mail server to drop your message without delivering it. As far as I'm concerned, a communication medium which lets attackers block messages from reaching their destinations is not really secure.
I realize that by that logic the entire internet isn't really secure, but email is significantly worse than other systems because (by default at least) it has no method for stopping unauthenticated parties from just swamping you with crap so you never see your messages.
No, you have to transmit MULTIPLE numbers.
You transmit the first Mersenne prime. Then you wait for a while. Then you transmit the next one. Wait again. Etc. It's much more efficient to send them in binary than unary (although this most recent prime would require over 2 million bits).
And just because aliens receive a signal with a bunch of strong, equally spaced pulses doesn't mean they'll automatically assume it's intelligent. There are plenty of natural cosmic phenomena which produce equally spaced pulses.
Can someone explain what the application/use these primes are for?
Communicating with alien species, perhaps.
Mersenne primes have two interesting properties that might catch the attention of alien species: when written in binary, they are entirely composed of '1' bits; and, of course, they are prime.
A sure way to prove to another being that you are intelligent is to spew a bunch of numbers which all happen to be prime. The fact that they can be tranmitted using only '1' bits means the modulation is simple -- just send a series of pulses.
Once you're infected, in order to detect or clean, you have to cold boot from known clean media.
That still isn't good enough. I can imagine a virus that reflashes your BIOS to dynamically insert viral code into the boot sector as it boots from a floppy. Unless you jumper your BIOS so it can't be flashed, that is.
A lot of spyware gets on users' systems because they voluntarily install it. A box pops up and they mindlessly click "Okay."
This is perfectly possible with Firefox or any other browser that runs on Linux. Somebody writes some spyware, packages it in an XPI file and throws it up on a website. A user browses there, Firefox pops a box saying "Do you want to install this unsigned XPI?" and the user again mindlessly clicks "okay."
Installing Linux won't magically make people responsible.
Slashdot Mirror
It was easy to disprove the existence of aether with the Michelson-Morley experiment. Had that experiment not been possible it would have been very premature to jump to the conclusion that there is no aether. When it comes to dark matter, there is no easy experiment to disprove its existence and so it would be very rash to conclude that our understanding of gravity, which has worked extremely well for us for hundreds of years, is wrong.
Obviously the proper solution is for the emergency vehicles to drag a ten-mile strand of Cat5 behind them. Just imagine it -- an ambulance hurtling down the freeway with a blue network cable trailing behind, and a giant spool of Cat5 unreeling at 600 RPM... Tht tht tht tht tht tht tht
I wouldn't have objected to having a segmented 16-bit addressing mode even if the chip had had a true 32-bit address mode. Maybe I'm crazy but I liked it.
So why isn't the attorney general going after the advertisers? Because he can't enforce laws which don't exist, and I like things that way, thank you very much.
I didn't realize that Slashdot exists to interest you, personally. I think the discussion on this article is pretty much par for Slashdot. If you don't like it, read on to the next thing.
No. The idea of preserving compatibility with 16 bits apps was the hack, not the particular way they went about it. As you just said yourself, the elegant solution would be to ditch the idea of backward compatibility. Don't blame the solution when it's the problem that sucks.
Many processor architectures throughout computing history have used addressing modes. They are by no means "hacks."
That sort of defeats the point of .NET/Mono as a platform-independent development system, doesn't it? If you're coding in C#/Mono just because you like it, that's one thing, but touting it as the Linux version of .NET is dishonest because the idea of platform independence is one of the concepts of the .NET moniker.
I have no problem with Mono if they're just trying to implement the language/runtime, but they shouldn't bill it as .NET compatible if it's not.
I never get comments like this.
First of all, and most important, if you think that Slashdot is a "waste of time" why do you bother with it? It's pretty lame to come in here commenting and then complain that you're wasting your time. "This sucks, now I have to do this guy's research for him." Uhhh.. No, you don't have to do anything. You could have just skipped the article if it sucked that much. Pointless bitching.
Second, if the question hadn't been posted, we'd have no excuse to banter on Slashdot about systems which support 16 GB of RAM. The fact that the guy needed an answer to a question, which he could have got somewhere else, is irrelevant. If he'd asked somewhere else, we'd have been deprived of this opportunity to talk about big-memory systems.
What exactly do you want Slashdot to be? Apparently, a mutual ego-stroking fest. I agree that the guy's question could have been answered with 5 minutes of Googling, but who gives a shit about some guy who doesn't know how to search? I come here to read and participate in the discussions.
The reverse engineering comes in when you try to emulate all the places where Microsoft has intentionally diverged from their own spec.
You can't even trust the MSDN documentation on simple Windows functions, how can you expect an entire application framework to actually adhere to the spec? Thinking you can just implement it straight out of the book is pretty naive. It IS reverse engineering.
The segmentation mechanism was designed to allow the use of more memory while retaining compatibility with 16-bit, segment-naive code. It may not be beautiful but I can't think of a better way they could have implemented it.
The term "hack" is generally reserved for describing an inelegant implementation. In this case, it is the PROBLEM itself which is inelegant, not the solution to it. Segmentation was simple, easily understood, and it solved the problem.
On IA32, the limit exists on Linux as well as Windows. This is a hardware limitation of 32-bit addresses, as you pointed out. However, it's possible to "window" the higher memory into a fixed area under the 4 GB limit, similar to how XMM and EMM worked in the days of MS-DOS and extended memory managers.
As for memory being reserved for kernel space, that's a necessity in order for the kernel to be able to differentiate between user space and kernel space. It would be horrifically inefficient for the kernel to have to scan through a bunch of tables to figure out if a particular pointer is a kernel pointer or a user pointer. Setting an arbitrary boundary between kernel and user space lets the kernel do a simple test -- is this pointer greater than the kernel space limit? If so, it's a kernel pointer.
On Linux the boundary can be set at either 2 GB or 3 GB.
I'm personally tired of being called a "fanboy" just because I decided to buy a Mini for the hell of it.
As for the topic at hand... I think Apple's true concern should be finding the individual who leaked the disc from inside the company. Perhaps they should offer to let the guy off the hook if he explains where the thing came from.
Ummm... We've had thunderstorms for billions of years, and presumably these gamma ray bursts as well. They are completely natural phenomena. If they could negatively affect us, we would have either evolved a method for coping millions of years ago, or we wouldn't even exist today.
OF COURSE the things are directed into space. Duh.
Maybe I'm missing something, but why not just get a second phone line and leave yourself dialed in to your ISP 24/7? Surely that'll cost you less that $40 per month.
What we should be asking ourselves is, is there a way to construct a hashing algorithm for which the OPTIMAL method for finding collisions is a brute force search? So far it hasn't been done, and it hasn't been definitely proven to be possible or impossible, either.
I see a lot of people on these forums complaining that we should "just" make a hash algorithm that is unbreakable. It's a logical impossibility. Can you fit an infinite number of things into a finite number of holes and guarantee that each hole has at most one object in it? I hope people are capable of grasping that, at least.
The frickin' AC did better than that.
You're a troll.
As I mentioned in my post, it's not the best definition in the world, but the point remains that email is ridiculously easy to disrupt when compared to other methods of communicating over the internet.
I realize that by that logic the entire internet isn't really secure, but email is significantly worse than other systems because (by default at least) it has no method for stopping unauthenticated parties from just swamping you with crap so you never see your messages.
And just because aliens receive a signal with a bunch of strong, equally spaced pulses doesn't mean they'll automatically assume it's intelligent. There are plenty of natural cosmic phenomena which produce equally spaced pulses.
So why don't YOU learn before you speak?
Yeah, except that the Mersenne primes are well known and thus useless for cryptography -- at least, if you want any security.
Communicating with alien species, perhaps.
Mersenne primes have two interesting properties that might catch the attention of alien species: when written in binary, they are entirely composed of '1' bits; and, of course, they are prime.
A sure way to prove to another being that you are intelligent is to spew a bunch of numbers which all happen to be prime. The fact that they can be tranmitted using only '1' bits means the modulation is simple -- just send a series of pulses.
That still isn't good enough. I can imagine a virus that reflashes your BIOS to dynamically insert viral code into the boot sector as it boots from a floppy. Unless you jumper your BIOS so it can't be flashed, that is.
A lot of spyware gets on users' systems because they voluntarily install it. A box pops up and they mindlessly click "Okay."
This is perfectly possible with Firefox or any other browser that runs on Linux. Somebody writes some spyware, packages it in an XPI file and throws it up on a website. A user browses there, Firefox pops a box saying "Do you want to install this unsigned XPI?" and the user again mindlessly clicks "okay."
Installing Linux won't magically make people responsible.