Ain't that the truth, and it's not just software bugs. My ADSL modem / NAT box overheats and reboots itself under any significant load unless you operate it upside down. In a example of true design genius the vent is on the bottom.
My home internet connections have previously suffered from enormous buffers in the DSLAM - setting off a big download could cause ping times to increase to about 2 seconds, rendering other interactive use of the connection impossible.
Still, either this has been fixed now or more modern versions of TCP are more sophisticated, because it doesn't seem to happen any more - at least not to the same degree.
Anyone who made a discovery of enough significance to overturn anthropogenic climate change would have made a reputation for himself enough to secure a funded career. Research is a competitive environment, and that would give you a huge advantage over those competing for funding - you don't make a name for yourself simply by supporting he orthodoxy. Astrologers, on the other hand, rely on the belief that astrology is true to make money.
If there was a consistent downwards trend in typical global temperature despite increasing CO2 concentration (if other factors, such as solar irradiance, stayed constant). There, that wasn't too hard. Alternatively, if atmospheric CO2 concentrations didn't increase despite our emissions (i.e. there were feedbacks). Or, rather less likely, if someone did a new measurement and determined that CO2 didn't absorb IR after all.
while ONE test that turns up FALSIFIED is usually fatal to a theory. (If it won't kill a theory it isn't a proper test.)
Meanwhile, outside an "introduction to Karl Popper" book, pieces of scientific evidence are often not 100% in favour of one theory or another, especially in a system with many different things interacting. At most, they merely have a "most likely interpretation". There was no one piece of evidence that singly led to AGW theory, and there's to unlikely to be one that singly disproves it.
The sceptics find flaws and outright fraud in the models and datasets and they are attacked and suppressed
No they didn't, and if you think they did you weren't looking closely. And suppressed? Last I checked Watts was still publishing his website, and sceptics still get disproportionately large coverage in the mainstream press.
I must admit, when I first heard about the idea to give/56 or/48 to everyone, it seemed ridiculous. I suspect most people don't have more than one subnet - but since there are about 8 million times as many/56s as there are people on Earth, maybe giving a/56 to everyone isn't so daft after all.
IIRC, the IPv6 policy is that unicast is only 2000::/3 for now - if we fill that, the allocation policy will be reassessed to be less generous. Hopefully they've been clear enough that other addresses are *not* invalid, so we don't have a repeat of the IPv4 class E debacle.
Is there a way for NAT64 to work without using DNS to fake AAAA records? If not, it's going to have trouble with things that handle IP addresses directly (P2P, Skype?). I suppose you could require applications to work it out themselves (after all, they had to handle the first round of NAT themselves), but that's not ideal.
Perhaps that was one thing that could have been done better, by having a way of advertising the address of a NAT64 gateway to IPv6 hosts. Oh well, too late now.
It is much better to rent a server at a data center. You get a much better connection there, and it's cheaper.
Well, that depends what you want. I hosted a telnet-based BBS for a while, for example, where it wasn't easier to do it your way. And an end-to-end network still has your option, of course.
No they won't. I've never needed to login to my home computer, and will never need to
So because you don't need something, nobody else does?
voip and video communications
Already go through a central server.
Which ones? That's inefficient and relies on someone being prepared to host a bandwidth-heavy service. How is that any better than sending the communications directly from one person to another? It also reduces innovation - with an end to end network, a hobbyist or small company can develop a new protocol and deploy it without extra costs. But not if they have to pay to set up a high bandwidth central server.
Games already require central servers.
I've played plenty of multiplayer games without using a central server, so no they don't. Or rather, one of the players was hosting it. Some games may do that now, but would they if NAT wasn't so prevelant? Your way imposes dependence on someone else setting up a service. Fine for big selling games while the owner can make money from the service, but not in general.
And as I said, an end-to-end internet does all of the things you want just as well. The key point is that a NAT-based internet has no advantage over an end-to-end one, even if all you do is connect to central servers. However it does reduce innovation and flexibility of the network.by limiting new services to big players. Also, NAT scales poorly because it's stateful.
You think that makes it harder for the government to control you?
No, if the government want to do something then network design won't stop them. And I wouldn't care too much if piracy was impossible, since I don't do it as a rule. I just want the most flexible network possible.
Are you trolling? Do you really want a "walled garden" internet where we're limited to being passive consumers of big content?
Things "end users" may require end-to-end connectivity for:
p2p, voip and video communications, games, hosting servers, remote logins.
And that's just what I've done personally. Who knows what applications the future will hold? An "end to end" internet works just as well as NAT for connecting to central servers (better in fact, due to reduced complexity), but is much more flexible.
Average users don't see IP addresses at all, so I can't see any of those issues applying. Autoconfiguration is essential for those users anyway, and that shouldn't be any harder for V6.
Huh? Both theories of relativity have been quite well tested, special relativity especially so. Mass-energy equivalence, time dilation and mechanics (where different from Newtonian) have all had experimental tests. Similarly, for GR, differences in Mercury's orbit, gravitational time dilation (Pound-Rebka experment) and so forth.
There are a fair few unlimited, or at least "unlimited for practical purposes" ISPs available. Sky or Be, for example. I downloaded 200 GB one month, no problems.
The limited ones are generally the ones that use BT's backhaul from the exchange rather than doing their own (LLU), because BT charge a very high per-Mbps rate. Even then, it's enough for gaming.
The privacy extensions wouldn't provide any more privacy than you typically now get with IPv4. In IPv4, you typically get a/32 which identifies you - in IPv6, you'll get a/64,/56 or whatever. The privacy extensions only affect the last 64 bits - you can still be identified by the prefix that you were given by your ISP.
If you *do* have a landline, it makes sense to use it rather than a mobile where possible. I made about 2 hours of calls in the past month that cost me about 60p (plus line rental) but would have been £12 on my mobile (12p/minute PAYG). Consequently, I only use the mobile when I'm actually out of the house. And I'm known for being quiet, so I can't believe I'm a heavy user.
Of course, as you say, the internet connection is the main driving factor for having the landline.
There's really no excuse these days for a device not to be secure out of the box - i.e. you should be able to plug it straight into an unfirewalled network without problems. Security issues have been known about for years - even Microsoft's got on the ball now. I had a Vista box with a public IP and no separate firewall for months, and there were no problems.
100 is enough if it comes to that. My first car had 60 horsepower, allegedly, but I'm not convinced it was as high as that. Slow, certainly, but I got from one place to another safely.
Not sure how things work in the USA, but here in the UK new drivers are essentially prohibited from high power cars by the sheer cost of insurance.
It amazes me how much Brits are ok with being subjects of the Crown.
A rose by any other name...
I don't know where you get all your information, but the picture you paint doesn't resemble the Britain I live in at all. If I went by the media image, I'd think the US was a place where there was an even chance I'd be shot before the end of the week then bankrupted by the medical bills. Doesn't make it so.
I'm sure they won't like the comparison, but it's very reminiscent of Marxists and other far-left groups saying "The Soviet Union/Mao's China/Pol Pot's Cambodia weren't truly communist". This pattern is quite common amongst ideologues of various descriptions, both on the left and right.
I have no objection to first movers. As you rightly point out, someone has to start the ball rolling. And indeed the first mover usually doesn't have enough of an advantage to squash everyone else- but in some cases they do.
I was just trying to illustrate an example (with a purely rationalist, non-empirical argument) of how a natural monopoly can form in a free, unregulated market, because second movers will have no economically rational reason to try and compete. One of the things that the "race to the bottom" relies on is competition. If regulation is necessary to make that happen, so be it. If even that is impossible, a regulated monopoly is better than an unregulated one, as they won't be able to price-gouge so much.
What kind of entrepeneur has or could obtain the money required to build a large-scale electricity distribution system in competition with an established player?
The same kind which built the first one
The first one knew that he'd have the market to himself, so he'd be guaranteed a return. Doesn't apply to the second one.
If no one else can make a profit, then the price is already low by the market standard.
The original company spent a huge amount building the network, and is paying off that upfront cost from what it charges its customer - plus a profit. Now, suppose someone else comes along and takes the customers away by undercutting him - what does he do? Well, he'll have to cut his prices. Even if he cuts them to the point where he's covering his recurring costs but not taking enough money to pay back his loans for the infrastructure, that's still better than having no income at all. Both companies will rationally compete themselves down [*] to the point where they're *both* loss-making in that situation. Since the new company knows that's what would happen in a true competition, he doesn't enter. Hence, first-mover advantage.
That competing entrepreneurs are able to come in if he dare make to much of a profit - there doesn't even need to be any, actually competing, at all. The threat is stronger than the execution.
What kind of entrepeneur has or could obtain the money required to build a large-scale electricity distribution system in competition with an established player? Especially an established player can use its status to guarantee that the competitor can't make a profit - by undercutting as long as necessary. Given the particular economics of electricity supply (big fixed costs, relatively small variable ones), it makes sense to sell at a lower price if necessary rather than leave capacity idle as a result of people going to his competitor. Who's going to build a competing grid when he knows he won't make a profit from it?
Your link was notable for its lack of detail in its examples, making it impossible to assess whether what it says is correct or not. But since it's from an Austrian economist I guess I shouldn't be surprised - they were never fans of empiricism. However there are enough recent examples of dominant players crowding out competitors (e.g. my previous example, British Telecom), or cartels price-fixing to show the flaw with a pure free market. The fact that government mandated monopolies also existed doesn't change that, or mean they are the only possible form of monopoly (e.g. Standard Oil).
Nobody said those things should be free. But developed countries have measures to ensure that people can obtain them even if they couldn't afford them in a pure free market - precisely because they are essential. But that's unrelated.
Infrastructure is a prime candidate for regulation, because of its "natural monopoly" character - in the absence of regulation, it's very easy for a dominant company to squeeze everyone else out and then exploit their monopoly. Telephone and internet connections got a lot cheaper and faster here (UK) when the dominant phone company were required to let other providers use their network.
Slashdot Mirror
Ain't that the truth, and it's not just software bugs. My ADSL modem / NAT box overheats and reboots itself under any significant load unless you operate it upside down. In a example of true design genius the vent is on the bottom.
My home internet connections have previously suffered from enormous buffers in the DSLAM - setting off a big download could cause ping times to increase to about 2 seconds, rendering other interactive use of the connection impossible.
Still, either this has been fixed now or more modern versions of TCP are more sophisticated, because it doesn't seem to happen any more - at least not to the same degree.
Global warming was predicted before there was the ability to measure it, as far back as the 19th century.
There was even a short film made about it in 1958.
Anyone who made a discovery of enough significance to overturn anthropogenic climate change would have made a reputation for himself enough to secure a funded career. Research is a competitive environment, and that would give you a huge advantage over those competing for funding - you don't make a name for yourself simply by supporting he orthodoxy. Astrologers, on the other hand, rely on the belief that astrology is true to make money.
Tell me of a test that would falsify AGW theory?
If there was a consistent downwards trend in typical global temperature despite increasing CO2 concentration (if other factors, such as solar irradiance, stayed constant). There, that wasn't too hard. Alternatively, if atmospheric CO2 concentrations didn't increase despite our emissions (i.e. there were feedbacks). Or, rather less likely, if someone did a new measurement and determined that CO2 didn't absorb IR after all.
while ONE test that turns up FALSIFIED is usually fatal to a theory. (If it won't kill a theory it isn't a proper test.)
Meanwhile, outside an "introduction to Karl Popper" book, pieces of scientific evidence are often not 100% in favour of one theory or another, especially in a system with many different things interacting. At most, they merely have a "most likely interpretation". There was no one piece of evidence that singly led to AGW theory, and there's to unlikely to be one that singly disproves it.
The sceptics find flaws and outright fraud in the models and datasets and they are attacked and suppressed
No they didn't, and if you think they did you weren't looking closely. And suppressed? Last I checked Watts was still publishing his website, and sceptics still get disproportionately large coverage in the mainstream press.
I must admit, when I first heard about the idea to give /56 or /48 to everyone, it seemed ridiculous. I suspect most people don't have more than one subnet - but since there are about 8 million times as many /56s as there are people on Earth, maybe giving a /56 to everyone isn't so daft after all.
IIRC, the IPv6 policy is that unicast is only 2000::/3 for now - if we fill that, the allocation policy will be reassessed to be less generous. Hopefully they've been clear enough that other addresses are *not* invalid, so we don't have a repeat of the IPv4 class E debacle.
Is there a way for NAT64 to work without using DNS to fake AAAA records? If not, it's going to have trouble with things that handle IP addresses directly (P2P, Skype?). I suppose you could require applications to work it out themselves (after all, they had to handle the first round of NAT themselves), but that's not ideal.
Perhaps that was one thing that could have been done better, by having a way of advertising the address of a NAT64 gateway to IPv6 hosts. Oh well, too late now.
It is much better to rent a server at a data center. You get a much better connection there, and it's cheaper.
Well, that depends what you want. I hosted a telnet-based BBS for a while, for example, where it wasn't easier to do it your way. And an end-to-end network still has your option, of course.
No they won't. I've never needed to login to my home computer, and will never need to
So because you don't need something, nobody else does?
voip and video communications
Already go through a central server.
Which ones? That's inefficient and relies on someone being prepared to host a bandwidth-heavy service. How is that any better than sending the communications directly from one person to another? It also reduces innovation - with an end to end network, a hobbyist or small company can develop a new protocol and deploy it without extra costs. But not if they have to pay to set up a high bandwidth central server.
Games already require central servers.
I've played plenty of multiplayer games without using a central server, so no they don't. Or rather, one of the players was hosting it. Some games may do that now, but would they if NAT wasn't so prevelant? Your way imposes dependence on someone else setting up a service. Fine for big selling games while the owner can make money from the service, but not in general.
And as I said, an end-to-end internet does all of the things you want just as well. The key point is that a NAT-based internet has no advantage over an end-to-end one, even if all you do is connect to central servers. However it does reduce innovation and flexibility of the network.by limiting new services to big players. Also, NAT scales poorly because it's stateful.
You think that makes it harder for the government to control you?
No, if the government want to do something then network design won't stop them. And I wouldn't care too much if piracy was impossible, since I don't do it as a rule. I just want the most flexible network possible.
Are you trolling? Do you really want a "walled garden" internet where we're limited to being passive consumers of big content?
Things "end users" may require end-to-end connectivity for:
p2p, voip and video communications, games, hosting servers, remote logins.
And that's just what I've done personally. Who knows what applications the future will hold? An "end to end" internet works just as well as NAT for connecting to central servers (better in fact, due to reduced complexity), but is much more flexible.
Average users don't see IP addresses at all, so I can't see any of those issues applying. Autoconfiguration is essential for those users anyway, and that shouldn't be any harder for V6.
Huh? Both theories of relativity have been quite well tested, special relativity especially so. Mass-energy equivalence, time dilation and mechanics (where different from Newtonian) have all had experimental tests. Similarly, for GR, differences in Mercury's orbit, gravitational time dilation (Pound-Rebka experment) and so forth.
There are a fair few unlimited, or at least "unlimited for practical purposes" ISPs available. Sky or Be, for example. I downloaded 200 GB one month, no problems.
The limited ones are generally the ones that use BT's backhaul from the exchange rather than doing their own (LLU), because BT charge a very high per-Mbps rate. Even then, it's enough for gaming.
The privacy extensions wouldn't provide any more privacy than you typically now get with IPv4. In IPv4, you typically get a /32 which identifies you - in IPv6, you'll get a /64, /56 or whatever. The privacy extensions only affect the last 64 bits - you can still be identified by the prefix that you were given by your ISP.
The -ize ending is valid in non-American English too, it isn't an Americanism.
I think that's called NAT-PT, which was discussed but ultimately rejected.
If you *do* have a landline, it makes sense to use it rather than a mobile where possible. I made about 2 hours of calls in the past month that cost me about 60p (plus line rental) but would have been £12 on my mobile (12p/minute PAYG). Consequently, I only use the mobile when I'm actually out of the house. And I'm known for being quiet, so I can't believe I'm a heavy user.
Of course, as you say, the internet connection is the main driving factor for having the landline.
There's really no excuse these days for a device not to be secure out of the box - i.e. you should be able to plug it straight into an unfirewalled network without problems. Security issues have been known about for years - even Microsoft's got on the ball now. I had a Vista box with a public IP and no separate firewall for months, and there were no problems.
100 is enough if it comes to that. My first car had 60 horsepower, allegedly, but I'm not convinced it was as high as that. Slow, certainly, but I got from one place to another safely.
Not sure how things work in the USA, but here in the UK new drivers are essentially prohibited from high power cars by the sheer cost of insurance.
It amazes me how much Brits are ok with being subjects of the Crown.
A rose by any other name...
I don't know where you get all your information, but the picture you paint doesn't resemble the Britain I live in at all. If I went by the media image, I'd think the US was a place where there was an even chance I'd be shot before the end of the week then bankrupted by the medical bills. Doesn't make it so.
I'm sure they won't like the comparison, but it's very reminiscent of Marxists and other far-left groups saying "The Soviet Union/Mao's China/Pol Pot's Cambodia weren't truly communist". This pattern is quite common amongst ideologues of various descriptions, both on the left and right.
I have no objection to first movers. As you rightly point out, someone has to start the ball rolling. And indeed the first mover usually doesn't have enough of an advantage to squash everyone else- but in some cases they do.
I was just trying to illustrate an example (with a purely rationalist, non-empirical argument) of how a natural monopoly can form in a free, unregulated market, because second movers will have no economically rational reason to try and compete. One of the things that the "race to the bottom" relies on is competition. If regulation is necessary to make that happen, so be it. If even that is impossible, a regulated monopoly is better than an unregulated one, as they won't be able to price-gouge so much.
What kind of entrepeneur has or could obtain the money required to build a large-scale electricity distribution system in competition with an established player?
The same kind which built the first one
The first one knew that he'd have the market to himself, so he'd be guaranteed a return. Doesn't apply to the second one.
If no one else can make a profit, then the price is already low by the market standard.
The original company spent a huge amount building the network, and is paying off that upfront cost from what it charges its customer - plus a profit. Now, suppose someone else comes along and takes the customers away by undercutting him - what does he do? Well, he'll have to cut his prices. Even if he cuts them to the point where he's covering his recurring costs but not taking enough money to pay back his loans for the infrastructure, that's still better than having no income at all. Both companies will rationally compete themselves down [*] to the point where they're *both* loss-making in that situation. Since the new company knows that's what would happen in a true competition, he doesn't enter. Hence, first-mover advantage.
[*] Or form a cartel, if not legally prohibited.
That competing entrepreneurs are able to come in if he dare make to much of a profit - there doesn't even need to be any, actually competing, at all. The threat is stronger than the execution.
What kind of entrepeneur has or could obtain the money required to build a large-scale electricity distribution system in competition with an established player? Especially an established player can use its status to guarantee that the competitor can't make a profit - by undercutting as long as necessary. Given the particular economics of electricity supply (big fixed costs, relatively small variable ones), it makes sense to sell at a lower price if necessary rather than leave capacity idle as a result of people going to his competitor. Who's going to build a competing grid when he knows he won't make a profit from it?
Your link was notable for its lack of detail in its examples, making it impossible to assess whether what it says is correct or not. But since it's from an Austrian economist I guess I shouldn't be surprised - they were never fans of empiricism. However there are enough recent examples of dominant players crowding out competitors (e.g. my previous example, British Telecom), or cartels price-fixing to show the flaw with a pure free market. The fact that government mandated monopolies also existed doesn't change that, or mean they are the only possible form of monopoly (e.g. Standard Oil).
And I didn't go to a government school.
Nobody said those things should be free. But developed countries have measures to ensure that people can obtain them even if they couldn't afford them in a pure free market - precisely because they are essential. But that's unrelated.
Infrastructure is a prime candidate for regulation, because of its "natural monopoly" character - in the absence of regulation, it's very easy for a dominant company to squeeze everyone else out and then exploit their monopoly. Telephone and internet connections got a lot cheaper and faster here (UK) when the dominant phone company were required to let other providers use their network.