While high-precision public servers are nice, most applications for NTP aren't sensitive to the amount of jitter introduced by consumer-grade endpoint (which I'd characterize as almost never exceeding 100ms, and often below 50ms). If you have an application where that much jitter in your NTP sync is an issue you need a local NTP server anyway, and quite possibly a local time source.
NTP requires long-term relationships among the hosts in the peer/server group. As implemented that means static IPs, but even if you changed the system to do repeated DNS lookups the NTP pool couldn't use hostnames -- the DNS-based pooling currently in use does not include any mechanism to distribute hostnames, nor do most NTP clients provide any method to easily consume such data even if it were available.
The NTP protocol doesn't support changing IPs -- there's a long-term relationship among hosts in an NTP group. Servers like yours that hop on and off the network are only useful for single-sync applications and therefore are not suitable for inclusion in an NTP pool.
First, that complaint is true of essentially all external busses, including SCSI, SAS, eSATA and virtually everything else except USB. They're setup that way for a reason -- DMA is much, much faster.
Second, memory access on modern busses is routed through an IOMMU. This provides both memory abstraction (which is vital on modern architectures) and allows the OS to control which devices, if any, can access a particular memory location.
No one is arguing that race doesn't exist. The argument is that race is a social construct, and that while genetically heritable traits may influence the way we define a race it's certainly not a reliable way to detect or define race.
Because they completed an engineering analysis and determined that the reduced operational costs and increased science opportunities were balanced by the increased risk. Heck, for all you know there is no significant increase in the risk, and the old landing area selection was based on unnecessarily conservative estimates of the landing precision, so landing further away would be purely detrimental.
I'm not sure what happens when you make your bed, but when I do the only impediment to spiders, mice, etc. would be "must enter along the perimeter", which doesn't actually seem like a big deal, as the perimeter is where such things are most likely to encounter the bed in the first place. It would probably keep larger animals out of my bed, but if the door isn't doing that already I've got bigger problems.
I agree. The form of an argument is the only reasonable way to determine if it's worthwhile. Leading with a counter-example is clear evidence that your argument is invalid and should be ignored.
If you're going to get all philosophical about what is "real" then you can't disprove YEC even with a time machine. You can't prove that anything you perceive is real -- or even that objective reality exists in the first place.
Usually they're designed to only allow SSL connections with valid certificate chains, but typically that's configurable. It's certainly not difficult to configure the HTTPS proxies I've used to ignore certificate chains, and there are valid reason you might want to do so. As you note it's a problem in both directions, as it breaks the end-to-end system and there's no side-channel in which to inform the end user of the situation, or allow the user to provide feedback about the desired action.
Since the devices are powered by light, and according to the article are essentially just an array of independent photodiodes, couldn't they just close their eyes to protect the implanted bits and submit the external bits for manual inspection like all other medical devices?
I hate the TSA as much as the next guy, but this seems like a pretty lame excuse.
FYI: I've recently executed an exclusive distribution contract for/. comments from users with IDs in the 592xxx series. All future distribution of such comments must be routed through me before they are submitted for display in the North American market. Please refrain from any further unauthorized commentary in this market via these channels.
Or maybe it's not legal to enforce distribution exclusivity agreements parties not subject to the agreement. But you seem to think it is, per your comments in sid 2790177, so I assume you'll be deferring to me for all future communication on this site.
I agree 100%. It's a risk analysis, and there are valid uses. I'm just saying the idea "it takes many x-ray exposures to cause cancer, so any single exposure is safe" is wrong.
But still requires MitM positioning, which means that some company in Russia can't do it without widespread support.
Their actual method of generating MD5 collisions with bad data is much more plausible, at least in the short term. In the long term it's easy to add a second hash to torrent files, so make collisions significantly more difficult to calculate, but in the short term it's a lot easier than getting everyone with the appropriate wire position to allow you to inject spoofed packets into their network, or buy/run your platform.
Because end-users are likely to download the source code, dig into the credits, and care about who wrote it originally instead of who sold it to them and provides support. That seems really likely.
And again, how is that different from the original BSD license, which requires accreditation?
The anti-BSD guys are an interesting crowd. Well they can continue to classify generational improvements in software as pilfering and lock out both closed-source and BSD developers from using any of their innovations -- after all, it's better to force some ideology than to actually work to improve the state of software.
Look, if you want to change the locked-away nature of software, change copyright law. Software gets 5 years of protection, but unobscured source must be placed in escrow, and after 5 years the source becomes public domain. Or you can keep code a trade secret indefinitely, but you have to forgo copyright protections. That would allow a balance of commercial and openness interests, without pitting the two against each other (IMHO unnecessarily -- society benefits both from widespread sharing and from commercial motivations), and would bring software in-line with other copyright-protected works, which are currently much easier to reproduce at the end of their protection period.
So if Microsoft starting publishing someone else's GPL'd software, and following all of the GPL license rules to the letter, how would any of that change what you're saying about credit going to the publisher rather than the original author? Is there some component of the GPL that requires publishers to advertise the original author? Wasn't that one of the original provision of the BSD license that people didn't like and eventually did away with?
It is hard, so long as the connections are using TCP. You need a MitM position to fake TCP, not just knowledge of the remote address.
And encryption has no bearing because torrents don't provide any method of authenticating each other. So if you can MitM to fake TCP streams you can MitM to fake encryption as well.
No, it's one x-ray. The chance from any single x-ray is very small, but ultimately it's exactly one x-ray energy photon that flips your cell from "normal" to "cancer", and that one photon will come from any single exposure. There are often contributing factors to that change; generally speaking a single DNA transcription error/etc. by itself is not enough to cause cancer, but even so there's a discrete change from "non-canerous" to "cancerous" and that change can easily be caused by a single ionization event.
Statistically X in 1 million kids will get cancer because of x-rays, not might get cancer. There's a well established link between ionizing radiation and cancer. The risk in any single exposure is statistically small, but if you're the unlucky one that gets cancer from the procedure the low population risk probably won't be much comfort.
And when the TSA starts preforming regular testing of their machines by an independent contractor, installing "x-ray on" indicators, training staff on the dangers of ionizing radiation, providing staff with personal safety equipment, and otherwise taking all the other basic precautions that every other industry using ionizing radiation already take, then we can talk about the relative danger, and compare it to the benefit.
Until then we can only assume that the TSA is operating death-rays, because the machines they're running are inherently dangerous and they can't be bothered to install basic safeguards.
Slashdot Mirror
I've got one better -- I actually had a pool user call my ISP and get me disconnected (temporarily) because I was "hacking" them on UDP port 123.
While high-precision public servers are nice, most applications for NTP aren't sensitive to the amount of jitter introduced by consumer-grade endpoint (which I'd characterize as almost never exceeding 100ms, and often below 50ms). If you have an application where that much jitter in your NTP sync is an issue you need a local NTP server anyway, and quite possibly a local time source.
NTP requires long-term relationships among the hosts in the peer/server group. As implemented that means static IPs, but even if you changed the system to do repeated DNS lookups the NTP pool couldn't use hostnames -- the DNS-based pooling currently in use does not include any mechanism to distribute hostnames, nor do most NTP clients provide any method to easily consume such data even if it were available.
The NTP protocol doesn't support changing IPs -- there's a long-term relationship among hosts in an NTP group. Servers like yours that hop on and off the network are only useful for single-sync applications and therefore are not suitable for inclusion in an NTP pool.
The explosives residue scan is for organic explosives. Gunpowder and similar substances are not detected. But fertilizer is.
First, that complaint is true of essentially all external busses, including SCSI, SAS, eSATA and virtually everything else except USB. They're setup that way for a reason -- DMA is much, much faster.
Second, memory access on modern busses is routed through an IOMMU. This provides both memory abstraction (which is vital on modern architectures) and allows the OS to control which devices, if any, can access a particular memory location.
No one is arguing that race doesn't exist. The argument is that race is a social construct, and that while genetically heritable traits may influence the way we define a race it's certainly not a reliable way to detect or define race.
Because they completed an engineering analysis and determined that the reduced operational costs and increased science opportunities were balanced by the increased risk. Heck, for all you know there is no significant increase in the risk, and the old landing area selection was based on unnecessarily conservative estimates of the landing precision, so landing further away would be purely detrimental.
I'm not sure what happens when you make your bed, but when I do the only impediment to spiders, mice, etc. would be "must enter along the perimeter", which doesn't actually seem like a big deal, as the perimeter is where such things are most likely to encounter the bed in the first place. It would probably keep larger animals out of my bed, but if the door isn't doing that already I've got bigger problems.
I agree. The form of an argument is the only reasonable way to determine if it's worthwhile. Leading with a counter-example is clear evidence that your argument is invalid and should be ignored.
If you're going to get all philosophical about what is "real" then you can't disprove YEC even with a time machine. You can't prove that anything you perceive is real -- or even that objective reality exists in the first place.
If you define god to be unobservable then it doesn't really matter if god exists or not.
Usually they're designed to only allow SSL connections with valid certificate chains, but typically that's configurable. It's certainly not difficult to configure the HTTPS proxies I've used to ignore certificate chains, and there are valid reason you might want to do so. As you note it's a problem in both directions, as it breaks the end-to-end system and there's no side-channel in which to inform the end user of the situation, or allow the user to provide feedback about the desired action.
Since the devices are powered by light, and according to the article are essentially just an array of independent photodiodes, couldn't they just close their eyes to protect the implanted bits and submit the external bits for manual inspection like all other medical devices?
I hate the TSA as much as the next guy, but this seems like a pretty lame excuse.
FYI: I've recently executed an exclusive distribution contract for /. comments from users with IDs in the 592xxx series. All future distribution of such comments must be routed through me before they are submitted for display in the North American market. Please refrain from any further unauthorized commentary in this market via these channels.
Or maybe it's not legal to enforce distribution exclusivity agreements parties not subject to the agreement. But you seem to think it is, per your comments in sid 2790177, so I assume you'll be deferring to me for all future communication on this site.
I agree 100%. It's a risk analysis, and there are valid uses. I'm just saying the idea "it takes many x-ray exposures to cause cancer, so any single exposure is safe" is wrong.
Which is pretty easy, given a peer-to-peer client. By definition if one end does it the other end will as well.
But still requires MitM positioning, which means that some company in Russia can't do it without widespread support.
Their actual method of generating MD5 collisions with bad data is much more plausible, at least in the short term. In the long term it's easy to add a second hash to torrent files, so make collisions significantly more difficult to calculate, but in the short term it's a lot easier than getting everyone with the appropriate wire position to allow you to inject spoofed packets into their network, or buy/run your platform.
Because end-users are likely to download the source code, dig into the credits, and care about who wrote it originally instead of who sold it to them and provides support. That seems really likely.
And again, how is that different from the original BSD license, which requires accreditation?
The anti-BSD guys are an interesting crowd. Well they can continue to classify generational improvements in software as pilfering and lock out both closed-source and BSD developers from using any of their innovations -- after all, it's better to force some ideology than to actually work to improve the state of software.
Look, if you want to change the locked-away nature of software, change copyright law. Software gets 5 years of protection, but unobscured source must be placed in escrow, and after 5 years the source becomes public domain. Or you can keep code a trade secret indefinitely, but you have to forgo copyright protections. That would allow a balance of commercial and openness interests, without pitting the two against each other (IMHO unnecessarily -- society benefits both from widespread sharing and from commercial motivations), and would bring software in-line with other copyright-protected works, which are currently much easier to reproduce at the end of their protection period.
So if Microsoft starting publishing someone else's GPL'd software, and following all of the GPL license rules to the letter, how would any of that change what you're saying about credit going to the publisher rather than the original author? Is there some component of the GPL that requires publishers to advertise the original author? Wasn't that one of the original provision of the BSD license that people didn't like and eventually did away with?
It is hard, so long as the connections are using TCP. You need a MitM position to fake TCP, not just knowledge of the remote address.
And encryption has no bearing because torrents don't provide any method of authenticating each other. So if you can MitM to fake TCP streams you can MitM to fake encryption as well.
No, it's one x-ray. The chance from any single x-ray is very small, but ultimately it's exactly one x-ray energy photon that flips your cell from "normal" to "cancer", and that one photon will come from any single exposure. There are often contributing factors to that change; generally speaking a single DNA transcription error/etc. by itself is not enough to cause cancer, but even so there's a discrete change from "non-canerous" to "cancerous" and that change can easily be caused by a single ionization event.
Statistically X in 1 million kids will get cancer because of x-rays, not might get cancer. There's a well established link between ionizing radiation and cancer. The risk in any single exposure is statistically small, but if you're the unlucky one that gets cancer from the procedure the low population risk probably won't be much comfort.
And when the TSA starts preforming regular testing of their machines by an independent contractor, installing "x-ray on" indicators, training staff on the dangers of ionizing radiation, providing staff with personal safety equipment, and otherwise taking all the other basic precautions that every other industry using ionizing radiation already take, then we can talk about the relative danger, and compare it to the benefit.
Until then we can only assume that the TSA is operating death-rays, because the machines they're running are inherently dangerous and they can't be bothered to install basic safeguards.