That's true -- the accelerator in Trieste apparently emits harmful radiation to the surrounding countryside.
Relative to the possible dangers, they're fairly safe.:-)
At CHESS we had a fun "no ladders" rule, since the experimental areas were separated from the beamline by a 10' concrete wall and no ceiling. Climbing a ladder would get you a direct view of the beamline and an unhealthy shot of X-rays.
It is impossible. They can accelerate to quite close to the speed of light, though. Perhaps the article authors were inconveniently rounding. Not very helpful, considering that most particle accelerators produce particles traveling faster than.99c, and there's a vast difference between.99c and c.
If only they'd use the proper units for near-speed-of-light travel.
The summary authors don't seem to understand. Nobody at CERN reported the malfunction? I assume they mean "reported to the press" -- otherwise, how did they fix a 30-ton transistor without telling anyone.
Anyway, things malfunction and break on particle accelerators constantly. They're devilishly difficult to maintain properly. (They operate in extreme fail-safe modes, so failures harmless but common.)
That's not correct. Shannon's information theory is closely tied to thermodynamics. In it, acquiring information is a reduction in entropy. (It's a local reduction in entropy, meaning that more entropy must be created elsewhere. It makes for an interesting argument why entropy must always increase. It's easy to show that entropy must always go one direction or the other. Our world-view, in which information is always being acquired, can only exist in an a universe where entropy always increases.)
The 9/11 connection is just a PR plug, though -- those properties of steel have been known for a long time. (I can't guarantee they were known when the building were built, but certainly they were known prior to the event.)
Re:It gives you something just as bad...
on
Review: Spore
·
· Score: 1
SecuROM run by a non-admin user runs a non-admin userland service. It doesn't run anything at the admin, kernel, or driver level. It seems like this means it doesn't qualify as a rootkit.
I realize calling things you don't like "rootkits" is popular, but that hardly makes it correct.
If you're talking about simple bot malware that doesn't employ a good rootkit, then there a number of different programs that will output all of your network activity. (Disk activity's a little trickier, I think.) A rough familiarity with what you would normally be connecting to and what programs are knowingly running on your machine (e.g., automatic update software), and something like this will be sufficient.
Note that in the above case, something that searches for discrepancies between what is reported by the OS and what isn't won't do you any good, since the malware is hiding nothing. It can be difficult to automatically differentiate intentional from unintentional usage.
Now, if the bot is using a rootkit, this is a subject of much interest. If you want the tool to run on your machine, then it's a race to see if the malware author or forensic tool author is more clever. Of course, if connectivity to the Internet goes through a router you have access to, you can engineer some way to compare actual network communication with what's reported on the machine. None of these are simple or easy to use.
The same technology couldn't be used to stop DoS attacks -- the connections are broken by forging TCP RST packets. Most DoS attacks don't use TCP. They're also doing a particular sort of network pattern detection that catches BitTorrent but won't necessarily catch a bot.
Granted, there are methods for detecting bots and methods for silencing their traffic.
With botnets, you can get a pretty good idea by comparing external network logs to user-initiated communication. If they're not talking to their C&C, they're not doing much.
It's "not doing anything" on fairly short time-scales. Neither allocating a thread nor paging a small amount of data takes significant resources. If the thread gives you zero benefit, sure, it's stupid to do. If the design has a use, though, then inactive threads are nearly free. (They were cheap, but not free, to create, and maintaining them takes only a tiny amount of resources.)
Your friend isn't quite right. The computation is for an ideal black body. The incident energy flux is modified by the Earth's albedo or absorptivity, which is a function of wavelength. The emitted energy flux is modified by its emissivity, also a function of wavelength.
Generally absorptivity and emissivity are roughly the same. However, within the Earth, objects absorb Solar radiation at wavelengths with high absorptivity and reradiate them at wavelengths with low emissivity. This significantly increases the temperature of the Earth. (There are some other effects, too -- like gradually losing part of our atmosphere to space.) This is why "greenhouse gases" are interesting -- they pass certain wavelengths of radiation and reflect others that result in Earth temperature increase.
There's a fairly good description of this on Wikipedia's Stefan-Boltzmann Law page.
Slashdot Mirror
Not that any reporter would use it (nobody would understand it), but the convenient measure is rapidity, since it's additive.
That's true -- the accelerator in Trieste apparently emits harmful radiation to the surrounding countryside.
Relative to the possible dangers, they're fairly safe. :-)
At CHESS we had a fun "no ladders" rule, since the experimental areas were separated from the beamline by a 10' concrete wall and no ceiling. Climbing a ladder would get you a direct view of the beamline and an unhealthy shot of X-rays.
It is impossible. They can accelerate to quite close to the speed of light, though. Perhaps the article authors were inconveniently rounding. Not very helpful, considering that most particle accelerators produce particles traveling faster than .99c, and there's a vast difference between .99c and c.
If only they'd use the proper units for near-speed-of-light travel.
The summary authors don't seem to understand. Nobody at CERN reported the malfunction? I assume they mean "reported to the press" -- otherwise, how did they fix a 30-ton transistor without telling anyone.
Anyway, things malfunction and break on particle accelerators constantly. They're devilishly difficult to maintain properly. (They operate in extreme fail-safe modes, so failures harmless but common.)
Come on. You didn't even read the summary. It's the vendors of the traffic cameras that want this -- obviously so they can sell more cameras.
So, you don't have an encrypted filesystem, then? Because that's really the only way they won't find one.
Laws rarely address implementation specifics.
That's not correct. Shannon's information theory is closely tied to thermodynamics. In it, acquiring information is a reduction in entropy. (It's a local reduction in entropy, meaning that more entropy must be created elsewhere. It makes for an interesting argument why entropy must always increase. It's easy to show that entropy must always go one direction or the other. Our world-view, in which information is always being acquired, can only exist in an a universe where entropy always increases.)
The 9/11 connection is just a PR plug, though -- those properties of steel have been known for a long time. (I can't guarantee they were known when the building were built, but certainly they were known prior to the event.)
SecuROM run by a non-admin user runs a non-admin userland service. It doesn't run anything at the admin, kernel, or driver level. It seems like this means it doesn't qualify as a rootkit.
I realize calling things you don't like "rootkits" is popular, but that hardly makes it correct.
So ask someone who went to DEFCON for a copy.
I think the explanation is quite simple. "We don't know what we're doing."
The other squid -- a Superconducting Quantum Interference Device.
You have to use something like squid, but it's because of magnetic hysteresis. (I could explain, but Wikipedia is pretty acccurate.)
It's possible in theory, but in practice, it's technology that law enforcement doesn't have access to.
If you're talking about simple bot malware that doesn't employ a good rootkit, then there a number of different programs that will output all of your network activity. (Disk activity's a little trickier, I think.) A rough familiarity with what you would normally be connecting to and what programs are knowingly running on your machine (e.g., automatic update software), and something like this will be sufficient.
Note that in the above case, something that searches for discrepancies between what is reported by the OS and what isn't won't do you any good, since the malware is hiding nothing. It can be difficult to automatically differentiate intentional from unintentional usage.
Now, if the bot is using a rootkit, this is a subject of much interest. If you want the tool to run on your machine, then it's a race to see if the malware author or forensic tool author is more clever. Of course, if connectivity to the Internet goes through a router you have access to, you can engineer some way to compare actual network communication with what's reported on the machine. None of these are simple or easy to use.
Botnets account for less traffic than P2P file sharing (a few percent).
The same technology couldn't be used to stop DoS attacks -- the connections are broken by forging TCP RST packets. Most DoS attacks don't use TCP. They're also doing a particular sort of network pattern detection that catches BitTorrent but won't necessarily catch a bot.
Granted, there are methods for detecting bots and methods for silencing their traffic.
With botnets, you can get a pretty good idea by comparing external network logs to user-initiated communication. If they're not talking to their C&C, they're not doing much.
Seems reasonable; I'm not really familiar with data for the Moon or Mars.
The overall theory for the Stefan-Boltzmann argument is right, but as usual, there's some important subtleties. :-)
It's "not doing anything" on fairly short time-scales. Neither allocating a thread nor paging a small amount of data takes significant resources. If the thread gives you zero benefit, sure, it's stupid to do. If the design has a use, though, then inactive threads are nearly free. (They were cheap, but not free, to create, and maintaining them takes only a tiny amount of resources.)
Why would a man not screw around as much as possible?
In short, because our young are vulnerable after birth, require a fairly large energy investment, and are few in number.
Monogamy actually appears in a number of different animal species.
Your friend isn't quite right. The computation is for an ideal black body. The incident energy flux is modified by the Earth's albedo or absorptivity, which is a function of wavelength. The emitted energy flux is modified by its emissivity, also a function of wavelength.
Generally absorptivity and emissivity are roughly the same. However, within the Earth, objects absorb Solar radiation at wavelengths with high absorptivity and reradiate them at wavelengths with low emissivity. This significantly increases the temperature of the Earth. (There are some other effects, too -- like gradually losing part of our atmosphere to space.) This is why "greenhouse gases" are interesting -- they pass certain wavelengths of radiation and reflect others that result in Earth temperature increase.
There's a fairly good description of this on Wikipedia's Stefan-Boltzmann Law page.
If they're not doing anything, the userspace data will be paged out.
Appropriately, 250 GB / month has the same dimensions as 1 Mbit / sec.
It's just a rather coarser measure than you usually use for bandwidth.
Everyone makes this kind of argument. It's called a slippery slope argument. The problem is that it's not really valid.