Which in C, amounts to using setjmp() and longjmp(). How this is "better" than goto, I do not understand. Now you've got jumps jumping between totally different functions, magically causing whole chunks of call stack to vanish into thin air, potentially screwing up signal handling, etc. Hey, but at least you avoided "goto", right?
Obviously, in languages which support exceptions, it is better to use them. But we're talking about C here.
Is this how the OSS community at large operates? Instead of releasing patches, they release exploits?
There comes a point where releasing a patch would be ludicrous, because the problem and its solution are so obvious.
It would be like calling up Boeing to report that the wing has fallen off your airplane, explaining why that is a problem, and giving them detailed instructions how to fix it. They know what's wrong and how to fix it. The problem is motivating them to do it.
Time and again, MS has proven that the only motivation to fix problems is concrete exploits in the wild (and even then, they sometimes don't fix it).
the blades are adamantine (i think i got the metal right.... sure as hell isn't steel) and blades sharp enough down to an atomic level wouldn't need alot of force.
This isn't right. The sharp edge of a blade is used to initially cut the material so the rest of the blade, which is tapered into a wedge, can spread the cut open. The sharp edge just gets the process started.
You can split wood with an extremely dull ax, if there is a pre-existing crack in the wood, and you aim well. This demonstrates that it is the wedge shape of the blade which does the cutting, not the sharp edge.
It doesn't matter how sharp or hard the material of the blade is. It would require an unbelievable amount of force to spread metal apart with a blade.
My god... I thought this was one thing they taught us not to do in schoool. But here it is in Windows! My god, don't they screen for these things at the interview?
You're seeing an example of one of the very few instances where goto is considered "acceptable" to use. Sometimes you code a function which winds up a lot of complicated state, and a failure halfway through requires that you "unwind" the partially constructed state. This is most easily accomplished by having a "bailout ladder" which can be jumped into (via goto) from various points in the code above.
The only other solution involves lots of code duplication, or very bizarre function calls such as CleanupMyState(&context, 6) which just ends up use a Duff's Device in a switch() statement to simulate the use of goto in precisely such a manner, anyway.
When you find that the cleanest way to do something is goto, then the solution is goto. What is the point in cortorting your code just to follow a piece of dogma that was only meant as a guideline anyway? Remember, the point is clarity, not adherence to dogma.
pH is a logarithmic measure. Saying "A pH of near zero" is like saying "a near infinite number of people" i.e. nonsense.
How so? pH = -log10([H+]) -- negative base ten log of hydrogen ion concentration. A pH of 0 would imply:
pH = 0 = -log10([H+]) ==> [H+] = 1 mol/liter.
H+ solutions which are more concentrated than 1 mol/liter will have pH values below zero.
A "logarithmic scale" means that linear changes in the scale indicate exponential changes in some underlying quantity. For every change of 1 in pH number, the concentration of the solution changes by a factor of ten. Just because the graph of log(x) goes to minus infinity as x goes to zero doesn't mean a logarithmic scale has some kind of asymptote.
Learn before you post.
Re:New meaning for 'crashes'?
on
Hack Your Car
·
· Score: 1
If you mess around with cars it is infinitely more seriously. A car crash is no laughing matter, neither are serious breakdowns. Kill someone and rebooting is not an option.
If you killed someone, it would be your fault.
At any time, your engine could explode, seize, jump to 7000 RPM, whatever. With or without chip modifications. Something can always fuck up. As a driver you must be prepared to properly control the vehicle if the engine goes crazy. It is your responsibility.
I have seen an accelerator "floor" itself and take the car to 75 MPH before the driver decided to throw it into neutral. This, of course, immediately blew up the engine, but at least he wasn't hurtling uncontrollably at high speed. Obviously you should sacrifice the engine instead of people's lives.
You seem to be suggesting that if your engine were to go out, you'd swerve all over the road and kill people. This terrifies me. I don't think you should be driving.
What you're saying about copyright is correct; but that probably isn't what MS would come after you (and your open source project) for. It'd be patent and trade secret violations.
There's no such thing as a "trade secret violation" unless you are bound by an NDA. If the source is leaked and people not under NDA see it, the jig is up -- your trade "secret" is fucked.
That's the different between patents and trade secrets. With a patent, you must publish details of the invention publicly, but you have an exclusive right to license the use of that invention. With a trade secret, you have no legal protection against other people using it, but you don't disclose it publicly.
It's kind of like security through obscurity. With a patent you rely on force of law. With a trade secret you rely on people keeping their mouths shut. You might manage to keep it under wraps for years, but once it's out, you are fucked. Even if the person who leaked it was under NDA, the only recourse you have is against that particular individual. Your secret is still out, and suing the hell out of someone won't change that.
How do they get that much bandwidth over a satellite link?
At this point you need to be extremely careful with terminology. The question is not how they get bandwidth, but how they achieve the data rate. Bandwidth and data rate are proportional to each other only if the signal to noise ratio is a constant. People commonly use "bandwidth" when what they are actually referring to is "data rate."
Shannon's theorem states that the data rate of a channel is equal to the bandwidth times the logarithm of (1+SNR) where SNR is the signal to noise ratio.
Thus, you can increase the data rate two ways: expand the bandwidth, or increase signal strength (which increases the SNR). In reality it's done by a combination of both.
But then if I was searching for say "bill gates username password.xls" then clearly I had intent.
So what?
If the Yellow Pages published an entry saying "Phone number ringing directly to the red phone on President Bush's desk," do you really think it would be a crime to call that number?
They put it in the Yellow Pages for fuck's sake.
Are you trying to say it should be illegal for me to flip through the phone book looking for that number?
[ I wrote this script for my web site back when Verisign first tried this shit. Slashdot has screwed up the formatting somewhat. You need PHP to be installed on your web server, obviously. Please don't criticize my HTML, the purpose of this is not to demonstrate high-quality HTML. ]
<body> Th is page is computer-generated. It is not intended for any purpose you could possibly be interested in. The purpose of this page is to cause Google and other search engines to DDoS Verisign as they crawl the web for links. Click on the links if you want, they will all take you to Verisign's stupid "Wildcard" site. If you have any questions direct them to <a href="mailto:abuse@verisign.com">abuse@verisign.co m</a>.<p>
Could Verisign respond with a prompt that says "login" and then get your username and password, if they felt like it?
No. The username and password are authenticated by a cryptographic challenge. The password is never sent over the channel, in encrypted form or otherwise. It's a mathematical challenge protocol which only works if both sides already know what the correct password is.
It's impossible to set up a "fake" ssh server and steal people's passwords. This was one of the design points of ssh (and any other cryptographic service worth its salt).
Have you ever ordered pizza with a CC?
how much info do they take? riigghtt..
Every time I've ordered pizza on a CC, they've asked to see the card (which has the name on it) and I have to sign. What weird pizza place are you ordering from?
in the article, it says "and it is all legal", then it continues to talk about security breaches, so is the whole thing legal or not?
Just because "security" was breached doesn't mean a law was broken.
This is equivalent to a direct-line phone number that rings the Red Phone on the President's desk, then accidentally getting it listed in the Yellow Pages.
Can you really blame somebody for calling, when there's an entry in the phone book saying "Red Phone on President's Desk"? There are two levels of idiocy here: having the number listed in the phone book, and having that phone publicly accessible in the first place.
Question is.. do you trust the search engine(s) being used ?
Everyone keeps mentioning this. Can someone please explain how the hell the CC number is of any use whatsoever without knowing the name on the account? Or the expiration date?
That's like sending a password without the associated username. Which one is "password" and which one is "username" is pretty much interchangeable. Sending your CC number without your name seems about as risky as sending your name without your CC number -- i.e., not at all.
It's strongly typed, it has decent OO semantics, has freely available compilers, what's not to like:-)?
It has pointers, which you need in order to do pretty much anything useful. Pointers are a great brain strainer for many beginning programmers, and there are plenty of languages, Python for instance, where you don't have to worry about them.
In fact, now that I mention it, I think Python would be a beautiful first language to learn programming on.
Am I on another planet or have I just always been lucky in Linux? ctrl+x/c/v never fail
I usually resort to ctrl-c to copy something, and yes, this always seems to work. The reason people complain about this is that you're supposed to be able to simply highlight something and have it copy. This does not always work. Also, middle-click doesn't behave like you would expect in some applications, particularly Emacs.
I love Emacs, but I hate how it pastes. You would expect that middle-clicking would insert the text at the current pointer location, but this isn't what it does -- middle clicking moves the point to where you clicked, and then inserts the text there. As a result I usually paste stuff in the wrong place 99% of the time. It's a pain in the ass to to carefully click on the right spot, much easier to position the point with the arrow keys. Middle click should not move the point! (Yes, I know that you can use ctrl-y to paste out of the cut buffer at the point -- but the idea is to perform the whole operation with the mouse)
Until they can explain what magnetism and gravity is and how it's created.. and not just that "it's there.. we can measure it".. then there's no chance in hell that anyone can explain the rest..
You'd make a poor scientist.
There are different levels of explanation of any phenomenon. At the highest level, magnetism is explained by saying "Magnetic fields emanate from magnetized objects." Of course, the question then arises as to what makes the magnetic objects magnetic. The answer is, "Magnetic objects are magnetic because they contain moving charges." Again, the question arises as to why moving charges produce a magnetic field. At this point, we turn to Einstein and relativity and say that the magnetic field is one facet of the electromagnetic field which distinguishes itself from the electric field by virtue of a particular reference frame.
Once again, the question arises as to where the electromagnetic field comes from. Again, the answer is, photons travel between charged particles which transfer momentum, and it is these particles which make up the electromagnetic field. Once again, the question arises as to why photons are able to transfer momentum between particles. This goes on ad infinitum.
At some point, you have to stop digging and admit that things occur simply because they occur. By definition there is no fundamental "cause" X which causes phenomenon Y, because you must then ask either, what causes X itself, or what causes X to cause Y (i.e., through what mechanism does the causality function).
The universe is composed of infinite layers of causality, and relationships between events. We have a hard enough time defining what exactly an event is, much less defining the nature of causality.
I challenge you to explain anything, even the simplest of things, to the most fundamental level of reality. I can always ask you "But why XYZ?" Just because this is the case, does not mean that we should not study things.
so let me get this straight... the economist is a reliable source for news about astrophysics?
If it had been in the New York Times, you would not have made this complaint -- and the New York Times doesn't specialize in anything, much less economics or astrophysics. At least The Economist can claim to be expert in at least one field.
This is not insightful. The veracity of a news story is not a function of where it is published, so long as the sources are reliable.
This just makes the team look like a bunch of unprofessional egotists. They've already had to change the name twice because they were stomping on someone else's toes. I'm sure they still haven't learned, and "FireFox" is going to turn out to be taken by someone else.
I guess they figure, it's a Mozilla product, therefore they are more 31337 than anyone else and can just pull names out of a hat.
Stupid, stupid stupid. Why not call it "Mozilla Streamline," since that's the point of the fucking thing anyway, and just get the stupid naming problems over and done with?
Oh, and not to mention, referring my friends to use a browser which keeps changing its name every couple of months makes me look like a retard. I'm going to stop referring people to it.
It doesn't say "Batch File Virus," it says "Batch Trojan Generator." You are aware that the term "batch" is far, far older than its usage in the Microsoft world, right?
In general, the term "batch" refers to a scriptable process which runs over a large data set without operator intervention. On DOS, it refers to a "batch" of commands to be run automatically. In this case, it clearly refers to the mass-production of viruses with different characteristics "in batches." It doesn't mean the output is a.BAT file, for crissake!
Slashdot Mirror
I hate it when folks
Try to type inside the box
Please, don't strike "Enter."
Which in C, amounts to using setjmp() and longjmp(). How this is "better" than goto, I do not understand. Now you've got jumps jumping between totally different functions, magically causing whole chunks of call stack to vanish into thin air, potentially screwing up signal handling, etc. Hey, but at least you avoided "goto", right?
Obviously, in languages which support exceptions, it is better to use them. But we're talking about C here.
There comes a point where releasing a patch would be ludicrous, because the problem and its solution are so obvious.
It would be like calling up Boeing to report that the wing has fallen off your airplane, explaining why that is a problem, and giving them detailed instructions how to fix it. They know what's wrong and how to fix it. The problem is motivating them to do it.
Time and again, MS has proven that the only motivation to fix problems is concrete exploits in the wild (and even then, they sometimes don't fix it).
This isn't right. The sharp edge of a blade is used to initially cut the material so the rest of the blade, which is tapered into a wedge, can spread the cut open. The sharp edge just gets the process started.
You can split wood with an extremely dull ax, if there is a pre-existing crack in the wood, and you aim well. This demonstrates that it is the wedge shape of the blade which does the cutting, not the sharp edge.
It doesn't matter how sharp or hard the material of the blade is. It would require an unbelievable amount of force to spread metal apart with a blade.
You're seeing an example of one of the very few instances where goto is considered "acceptable" to use. Sometimes you code a function which winds up a lot of complicated state, and a failure halfway through requires that you "unwind" the partially constructed state. This is most easily accomplished by having a "bailout ladder" which can be jumped into (via goto) from various points in the code above.
The only other solution involves lots of code duplication, or very bizarre function calls such as CleanupMyState(&context, 6) which just ends up use a Duff's Device in a switch() statement to simulate the use of goto in precisely such a manner, anyway.
When you find that the cleanest way to do something is goto, then the solution is goto. What is the point in cortorting your code just to follow a piece of dogma that was only meant as a guideline anyway? Remember, the point is clarity, not adherence to dogma.
What is the pH of hogwash? Are hogs acidic or basic?
Anybody know?
How so? pH = -log10([H+]) -- negative base ten log of hydrogen ion concentration. A pH of 0 would imply:
pH = 0 = -log10([H+]) ==> [H+] = 1 mol/liter.
H+ solutions which are more concentrated than 1 mol/liter will have pH values below zero.
A "logarithmic scale" means that linear changes in the scale indicate exponential changes in some underlying quantity. For every change of 1 in pH number, the concentration of the solution changes by a factor of ten. Just because the graph of log(x) goes to minus infinity as x goes to zero doesn't mean a logarithmic scale has some kind of asymptote.
Learn before you post.
If you killed someone, it would be your fault.
At any time, your engine could explode, seize, jump to 7000 RPM, whatever. With or without chip modifications. Something can always fuck up. As a driver you must be prepared to properly control the vehicle if the engine goes crazy. It is your responsibility.
I have seen an accelerator "floor" itself and take the car to 75 MPH before the driver decided to throw it into neutral. This, of course, immediately blew up the engine, but at least he wasn't hurtling uncontrollably at high speed. Obviously you should sacrifice the engine instead of people's lives.
You seem to be suggesting that if your engine were to go out, you'd swerve all over the road and kill people. This terrifies me. I don't think you should be driving.
There's no such thing as a "trade secret violation" unless you are bound by an NDA. If the source is leaked and people not under NDA see it, the jig is up -- your trade "secret" is fucked.
That's the different between patents and trade secrets. With a patent, you must publish details of the invention publicly, but you have an exclusive right to license the use of that invention. With a trade secret, you have no legal protection against other people using it, but you don't disclose it publicly.
It's kind of like security through obscurity. With a patent you rely on force of law. With a trade secret you rely on people keeping their mouths shut. You might manage to keep it under wraps for years, but once it's out, you are fucked. Even if the person who leaked it was under NDA, the only recourse you have is against that particular individual. Your secret is still out, and suing the hell out of someone won't change that.
At this point you need to be extremely careful with terminology. The question is not how they get bandwidth, but how they achieve the data rate. Bandwidth and data rate are proportional to each other only if the signal to noise ratio is a constant. People commonly use "bandwidth" when what they are actually referring to is "data rate."
Shannon's theorem states that the data rate of a channel is equal to the bandwidth times the logarithm of (1+SNR) where SNR is the signal to noise ratio.
Thus, you can increase the data rate two ways: expand the bandwidth, or increase signal strength (which increases the SNR). In reality it's done by a combination of both.
So what?
If the Yellow Pages published an entry saying "Phone number ringing directly to the red phone on President Bush's desk," do you really think it would be a crime to call that number?
They put it in the Yellow Pages for fuck's sake.
Are you trying to say it should be illegal for me to flip through the phone book looking for that number?
[ I wrote this script for my web site back when Verisign first tried this shit. Slashdot has screwed up the formatting somewhat. You need PHP to be installed on your web server, obviously. Please don't criticize my HTML, the purpose of this is not to demonstrate high-quality HTML. ]
o m</a>.<p>
.= chr(rand() % 26 + 97);
<html>
<head>
<title>Verisign slammer</title>
</head>
<body>
Th is page is computer-generated. It is not intended for any purpose you could
possibly be interested in. The purpose of this page is to cause Google and other search engines to DDoS Verisign as they crawl the web for links. Click on the links if you want, they will all
take you to Verisign's stupid "Wildcard" site. If you have any questions
direct them to <a href="mailto:abuse@verisign.com">abuse@verisign.c
<?php
for($i = 0; $i < 50; $i++)
{
for($j = 0; $j < 4; $j++)
{
$host = "";
for($k = 0; $k < 16; $k++)
{
$host
}
$host = "http://www.$host.net";
echo "<a href=\"$host\">$host</a>\n";
}
echo "<br>";
}
?>
</body>
</html>
Controversy implies the existence of opposing viewpoints. This isn't controversial at all -- we all know this idea is fucked.
No. The username and password are authenticated by a cryptographic challenge. The password is never sent over the channel, in encrypted form or otherwise. It's a mathematical challenge protocol which only works if both sides already know what the correct password is.
It's impossible to set up a "fake" ssh server and steal people's passwords. This was one of the design points of ssh (and any other cryptographic service worth its salt).
Every time I've ordered pizza on a CC, they've asked to see the card (which has the name on it) and I have to sign. What weird pizza place are you ordering from?
Just because "security" was breached doesn't mean a law was broken.
This is equivalent to a direct-line phone number that rings the Red Phone on the President's desk, then accidentally getting it listed in the Yellow Pages.
Can you really blame somebody for calling, when there's an entry in the phone book saying "Red Phone on President's Desk"? There are two levels of idiocy here: having the number listed in the phone book, and having that phone publicly accessible in the first place.
Everyone keeps mentioning this. Can someone please explain how the hell the CC number is of any use whatsoever without knowing the name on the account? Or the expiration date?
That's like sending a password without the associated username. Which one is "password" and which one is "username" is pretty much interchangeable. Sending your CC number without your name seems about as risky as sending your name without your CC number -- i.e., not at all.
What am I missing here?
It has pointers, which you need in order to do pretty much anything useful. Pointers are a great brain strainer for many beginning programmers, and there are plenty of languages, Python for instance, where you don't have to worry about them.
In fact, now that I mention it, I think Python would be a beautiful first language to learn programming on.
I usually resort to ctrl-c to copy something, and yes, this always seems to work. The reason people complain about this is that you're supposed to be able to simply highlight something and have it copy. This does not always work. Also, middle-click doesn't behave like you would expect in some applications, particularly Emacs.
I love Emacs, but I hate how it pastes. You would expect that middle-clicking would insert the text at the current pointer location, but this isn't what it does -- middle clicking moves the point to where you clicked, and then inserts the text there. As a result I usually paste stuff in the wrong place 99% of the time. It's a pain in the ass to to carefully click on the right spot, much easier to position the point with the arrow keys. Middle click should not move the point! (Yes, I know that you can use ctrl-y to paste out of the cut buffer at the point -- but the idea is to perform the whole operation with the mouse)
You'd make a poor scientist.
There are different levels of explanation of any phenomenon. At the highest level, magnetism is explained by saying "Magnetic fields emanate from magnetized objects." Of course, the question then arises as to what makes the magnetic objects magnetic. The answer is, "Magnetic objects are magnetic because they contain moving charges." Again, the question arises as to why moving charges produce a magnetic field. At this point, we turn to Einstein and relativity and say that the magnetic field is one facet of the electromagnetic field which distinguishes itself from the electric field by virtue of a particular reference frame.
Once again, the question arises as to where the electromagnetic field comes from. Again, the answer is, photons travel between charged particles which transfer momentum, and it is these particles which make up the electromagnetic field. Once again, the question arises as to why photons are able to transfer momentum between particles. This goes on ad infinitum.
At some point, you have to stop digging and admit that things occur simply because they occur. By definition there is no fundamental "cause" X which causes phenomenon Y, because you must then ask either, what causes X itself, or what causes X to cause Y (i.e., through what mechanism does the causality function).
The universe is composed of infinite layers of causality, and relationships between events. We have a hard enough time defining what exactly an event is, much less defining the nature of causality.
I challenge you to explain anything, even the simplest of things, to the most fundamental level of reality. I can always ask you "But why XYZ?" Just because this is the case, does not mean that we should not study things.
If it had been in the New York Times, you would not have made this complaint -- and the New York Times doesn't specialize in anything, much less economics or astrophysics. At least The Economist can claim to be expert in at least one field.
This is not insightful. The veracity of a news story is not a function of where it is published, so long as the sources are reliable.
I guess they figure, it's a Mozilla product, therefore they are more 31337 than anyone else and can just pull names out of a hat.
Stupid, stupid stupid. Why not call it "Mozilla Streamline," since that's the point of the fucking thing anyway, and just get the stupid naming problems over and done with?
Oh, and not to mention, referring my friends to use a browser which keeps changing its name every couple of months makes me look like a retard. I'm going to stop referring people to it.
It doesn't say "Batch File Virus," it says "Batch Trojan Generator." You are aware that the term "batch" is far, far older than its usage in the Microsoft world, right?
In general, the term "batch" refers to a scriptable process which runs over a large data set without operator intervention. On DOS, it refers to a "batch" of commands to be run automatically. In this case, it clearly refers to the mass-production of viruses with different characteristics "in batches." It doesn't mean the output is a .BAT file, for crissake!
Insurance companies don't operate that way. What they will do is jack up the premiums of people who do not use the device.
I hate to drop the conversation down a notch, but how can an alley, an empty space between two buildings, be manufactured? ;-)