> My favorite time scale is POSIX. They define a day as precisely 86400 "seconds". Obviously a POSIX second is not the same as an SI second.
That is not correct. POSIX speaks about SI seconds, it just specifies that 'unix time' is *approximate* number of (SI) seconds from the start of the epoch computed by a formula from date/time. When a leap second happens, the difference of this approximation from the exact number changes by one second and unix time experiences discontinuity (leap backwards).
Not really. It is true that before 1 CE (AD) there was 1 BCE (BC), but if you use integer numers for years, then 0 is equivalent to 1 BCE and -1 is equivalent to 2 BCE. See ISO 8601 for conventions for numeric representations of date and time.
Seems like many peering conflicts are related to excessive vertical integration of some providers. Verizon is currently both tier 1 provider and consumer ISP. Suppose it would be split to Verizon1, tier 1 provider, and Verizon 2, tier 2 consumer ISP. Verizon1 would be one of several upstream providers to Verizon2. What wold happen in such case if there is such kind of peering conflict between Verizon1 and Lever3? Verizon2 would immediately change its routing in a way that traffic between Level3 and Verizon2 would go through another upsream provider of Verizon2, which would threaten potential revenues of Verizon1 from Verizon2, therefore Verizon1 would have much stronger incentive to promptly solve the dispute.
Does labor laws apply for CEO position in California? In my country, positions like CEO and chair in board of directors are held outside of labor laws, based on specific mandate contract.
Socialism is an economic system where the state collects taxes to fund things like social programs, that allow people to do things like not work when they're sick without fear of, for example, starving to death.
Well, not really. Classical definition is that socialism is an economic system where means of production (i.e. capital) are publicly owned and managed (e.g. by the state or public cooperatives), while capitalism is an economic system where means of production are privately owned and managed. But press often mixes up socialists (who prefer socialism) and social-democratic parties (who prefer capitalism with socially sensitive regulations) and often uses 'socialists' as a shorthand for 'social-democratic parties'.
Rabbit starvation is unrelated by avitaminosis or missing essential acids (that would be more long term problem). It is caused by acquiring your energy mostly from proteins (i.e. you have to cut both fats and sugars to get it) and accumulation of excessive toxical byproducts of protein catabolism - proteins are not a good energy source for humans.
The inverse-square law only holds for something that radiates in a radial pattern.
More or less everything radiates in a radial pattern (has spherical wavefron) and is subject to the inverse-square law. Even lasers have some divergence. Better focus (by e.g. reflectors) would give you lower angle of divergence and therefore higher initial power density, but that is all.
x-rays are of a longer wavelength than visible light,
Definitely not. X-rays have significantly higher frequency and therefore shorter wavelength (380-740 nm for visible light and 0.01 - 10 nm for x-rays).
> Sad jokes aside - why aren't they implementing NAT64
NAT64 is generally more restrictive for IPv4 than common NAT, while does not have much advantages (if compared to IPv4NAT together with IPv6).
But there are other options like MAP-E, which solves both IPv4 exhaustion and IPv6 deployments with advantages (compared to CGNAT) for both users (better control over NAT) and ISPs (just stateless and easily scalable gateways).
Well, the referred wikipedia article says a different story:
In murder, U.S. police arrested 19,000 people for 26,000 murders, in which 75% were prosecuted and courts convicted 12,000 people. In Japan, 1,800 people were arrested for 1,300 murders, but prosecutors tried only 43%. Had the allegation that Japanese prosecutors use weak evidence mostly based on (forced) confessions to achieve convictions been true, the larger proportion of arrests would have resulted in prosecutions and eventual conviction. But the opposite is true. In fact, the data indicates that Japanese prosecutors bring charges only when the evidence is overwhelming and likelihood of conviction is near absolute, which gives a greater incentive for the accused to confess and aim for a lighter sentence, which, in turn, results in a high rate for confession.
For those who don't understand the extent of the problem, CGN is also called NAT444
Well, CGN does not meany automatically NAT444. Because provider uses private range, users could get multiple IP addresses for all their devices, so they don't need to have a router with NAT. This is common in small wireless ISPs, where wifi devices connecting clients to an ISP network work as bridge (but could be also configured as router with NAT).
For accessing IPv4, there are some alternatives to central CGN - 4rd, NAT-E and NAT-T. They are based on idea of keeping complete NAT state in customer routers and assign port ranges to them (e.g. one customer gets 256 ports from one IP, so 256 customers could share one IP) and use IPv6 as an underlying transport protocol for that. Routers translating this to legacy IPv4 internet would be stateless and therefore much more robust, simpler and scalable. Not to mention that having NAT just in customer routers allows users for example configure some static mapping for local services.
It is just a question of power density and high power requirement of high-isp engines. If your engine has Isp 20000 s, it requires 20 times more power per 1 N of thrust than engine with Isp 1000 s.
You cannot just add bigger reactor (because that would significantly increase mass), you would need reactor and engine with higher specific power (power per mass of reactor+engine). Current combinations of (nuclear)reactor+(ion)engines has specific power ~ 200 W/kg max, for decent acceleration, you would need something like hunded times higher specific power.
On the other side, nuclear thermal engines have still small Isp (~ 1000 s), that leads to small deltaV and long times even for trips to Mars and similar near objects. There are some papers calculating VASIMR ion/plasma engines with current/not-too-distant nuclear reactors giving about half time for Mars trip compared to nuclear thermal rockets.
Slashdot Mirror
> My favorite time scale is POSIX. They define a day as precisely 86400 "seconds". Obviously a POSIX second is not the same as an SI second.
That is not correct. POSIX speaks about SI seconds, it just specifies that 'unix time' is *approximate* number of (SI) seconds from the start of the epoch computed by a formula from date/time. When a leap second happens, the difference of this approximation from the exact number changes by one second and unix time experiences discontinuity (leap backwards).
Well, i hope DST would be abandoned sooner.
There is a better way. Just wait several thousand years and then add one leap hour.
Not really. It is true that before 1 CE (AD) there was 1 BCE (BC), but if you use integer numers for years, then 0 is equivalent to 1 BCE and -1 is equivalent to 2 BCE. See ISO 8601 for conventions for numeric representations of date and time.
EU court (European Court of Justice) resides in fact in Luxembourg.
It is ECHR (European Court of Human Rights, unrelated to EU) that resides in Strasbourg.
Seems like many peering conflicts are related to excessive vertical integration of some providers. Verizon is currently both tier 1 provider and consumer ISP. Suppose it would be split to Verizon1, tier 1 provider, and Verizon 2, tier 2 consumer ISP. Verizon1 would be one of several upstream providers to Verizon2. What wold happen in such case if there is such kind of peering conflict between Verizon1 and Lever3? Verizon2 would immediately change its routing in a way that traffic between Level3 and Verizon2 would go through another upsream provider of Verizon2, which would threaten potential revenues of Verizon1 from Verizon2, therefore Verizon1 would have much stronger incentive to promptly solve the dispute.
Does labor laws apply for CEO position in California? In my country, positions like CEO and chair in board of directors are held outside of labor laws, based on specific mandate contract.
Socialism is an economic system where the state collects taxes to fund things like social programs, that allow people to do things like not work when they're sick without fear of, for example, starving to death.
Well, not really. Classical definition is that socialism is an economic system where means of production (i.e. capital) are publicly owned and managed (e.g. by the state or public cooperatives), while capitalism is an economic system where means of production are privately owned and managed. But press often mixes up socialists (who prefer socialism) and social-democratic parties (who prefer capitalism with socially sensitive regulations) and often uses 'socialists' as a shorthand for 'social-democratic parties'.
> The spread of the virus might have actually occurred
Well, it is generally accepted that the Black death was caused by bacteria (Yersinia pestis), not virus.
Rabbit starvation is unrelated by avitaminosis or missing essential acids (that would be more long term problem). It is caused by acquiring your energy mostly from proteins (i.e. you have to cut both fats and sugars to get it) and accumulation of excessive toxical byproducts of protein catabolism - proteins are not a good energy source for humans.
More or less everything radiates in a radial pattern (has spherical wavefron) and is subject to the inverse-square law. Even lasers have some divergence. Better focus (by e.g. reflectors) would give you lower angle of divergence and therefore higher initial power density, but that is all.
Definitely not. X-rays have significantly higher frequency and therefore shorter wavelength (380-740 nm for visible light and 0.01 - 10 nm for x-rays).
It is probably the same. Small wireless ISPs use Linux/PC-based ISP-wide 1:N NATs for years, they just don't call that CGNAT.
Main difference between NAT and CGNAT is that you buy CGNAT from Cisco ;-) .
> Sad jokes aside - why aren't they implementing NAT64
NAT64 is generally more restrictive for IPv4 than common NAT, while does not have much advantages (if compared to IPv4NAT together with IPv6).
But there are other options like MAP-E, which solves both IPv4 exhaustion and IPv6 deployments with advantages (compared to CGNAT) for both users (better control over NAT) and ISPs (just stateless and easily scalable gateways).
Well, 'crowdsourced justice' is essentially an euphemism for lynch.
Well, the referred wikipedia article says a different story:
In murder, U.S. police arrested 19,000 people for 26,000 murders, in which 75% were prosecuted and courts convicted 12,000 people. In Japan, 1,800 people were arrested for 1,300 murders, but prosecutors tried only 43%. Had the allegation that Japanese prosecutors use weak evidence mostly based on (forced) confessions to achieve convictions been true, the larger proportion of arrests would have resulted in prosecutions and eventual conviction. But the opposite is true. In fact, the data indicates that Japanese prosecutors bring charges only when the evidence is overwhelming and likelihood of conviction is near absolute, which gives a greater incentive for the accused to confess and aim for a lighter sentence, which, in turn, results in a high rate for confession.
Well, it was the jury who sent the person to prison, not DEA.
Sorry, names of alternatives are not 'NAT-E' and 'NAT-T', but 'MAP-E' and 'MAP-T'.
For those who don't understand the extent of the problem, CGN is also called NAT444
Well, CGN does not meany automatically NAT444. Because provider uses private range, users could get multiple IP addresses for all their devices, so they don't need to have a router with NAT. This is common in small wireless ISPs, where wifi devices connecting clients to an ISP network work as bridge (but could be also configured as router with NAT).
For accessing IPv4, there are some alternatives to central CGN - 4rd, NAT-E and NAT-T. They are based on idea of keeping complete NAT state in customer routers and assign port ranges to them (e.g. one customer gets 256 ports from one IP, so 256 customers could share one IP) and use IPv6 as an underlying transport protocol for that. Routers translating this to legacy IPv4 internet would be stateless and therefore much more robust, simpler and scalable. Not to mention that having NAT just in customer routers allows users for example configure some static mapping for local services.
Just wondering whether the employee was fired, or promoted to the management.
It is just a question of power density and high power requirement of high-isp engines. If your engine has Isp 20000 s, it requires 20 times more power per 1 N of thrust than engine with Isp 1000 s.
You cannot just add bigger reactor (because that would significantly increase mass), you would need reactor and engine with higher specific power (power per mass of reactor+engine). Current combinations of (nuclear)reactor+(ion)engines has specific power ~ 200 W/kg max, for decent acceleration, you would need something like hunded times higher specific power.
On the other side, nuclear thermal engines have still small Isp (~ 1000 s), that leads to small deltaV and long times even for trips to Mars and similar near objects. There are some papers calculating VASIMR ion/plasma engines with current/not-too-distant nuclear reactors giving about half time for Mars trip compared to nuclear thermal rockets.