I understand how group based healthcare works. My coverage is independent from my employer, so I'm paying 100% of the cost. I understand the $50/mo ($600/year) goes towards paying other people's coverage seeing as I haven't seen a doctor in years. I get that.
I was arguing the GGP (or GGGP at this point)'s assertion that nationalized healthcare like how it's done in the UK takes from some people to give to others. I also think that there's a moral hazard in not giving an economic incentive for people to remain as healthy as possible. We've apparently, as americans, lost any sense of taking care of ourselves, but sticking a price tag to it makes it so that individuals and employers alike have a vested interest in remaining more healthy.
Even in a group health plan, people who are using it less pay less.
I'm a republican. I've had my sig on the bottom since I joined/.
The reality is, we were promised lower spending and lower taxes 8 years ago and we were lied to. I would much rather have someone who up front says, "Hey, here's the programs we want to have" rather than, for instance, when the prescription drug benefit was pushed through the white house, the Medicare Administrator was threatened to keep quiet that the plan was going to cost $138 billion more than planned.
Thanks for the generalization, but if you look it up, Americans voluntarily donate 2x as much per capita than the closest European country.
It's not a selfishness thing. While there are the advantages, we see what happens to our neighbors up north and our brothers in the UK with their NHS, and it sucks. We just don't want that.
Because I am a healthy 23 year old male. When I'm here, I pay next to nothing for health insurance. I lived in the UK for a while. I was paying about 80 dollars a week for healthcare.
In one situation, I pay $50/month for healthcare. In the UK I was paying 160 pounds (about $320) a month for healthcare. I didn't go visit any health facilities while I was there, but I had a friend try to get an appointment to get some antibiotics and it took _three weeks_.
We could argue about whether it's worthwhile to society or not, but under your (assuming you're european) system, it is what the GP said. I was charged much much more for an (arguably) inferior service without regards to the probability of whether or not I would use it. That money was used so I could subsidize someone else's treatment.
Even if you hit the option to not have all the toolboxes show up in the alt-tab window (under ubuntu), then you have nothing to alt-tab to once all your documents are closed and you want to start a new one.
It's crufty and it shows. I use linux on all my machines, and despite trying my damndest to get the gimp to play nice, it simply won't.
"If you have a dual stack (ipv6 and ipv4) machine, you can communicate with ipv4 machines just as easily as an ipv6 machine."
Right, so that's exactly what the original poster said. IPv4 can't communicate with IPv6 - the IPv6 machine ALSO NEEDS TO BE DUAL STACK WITH IPv4. And use up one of those precious rare v4 addresses. So since your v6 server needs to have v4 on it to communicate with all the legacy v4 clients out there - you might as well save time and money and just leave it as straight v4 in the first place.
You're partially right. The dual-stack machine can be using nat (or whatever else) just like it is now. No change, that's the status quo. But, as each machine gets their own network to support IPV6 further and further out, the need for 6to4 translation diminishes until, at some point, the entire internet supports it.
Dual stack is a stopgap measure and a means to transition, not a permanent situation. The two situations I see are either permenantly using NAT (which works okay now, but what if we end up with 100 or 1000 or 1E6 as many devices as we have now) or moving to IPV6. It makes much more sense to go ahead and start transitioning now as opposed to when we really do exhaust the IPv4 address space.
Cache in your browser. Cache in your OS. Cache in your LAN router/proxy. Cache in your ISP. Cache in your peering hub. Cache wherever and whenever you can; design so you can trade off storage vs transmission speed, so you use whatever's locally cheaper.
Caching layer 7 data in layer 3 doesn't make sense because layer 3 just doesn't have enough information about the layers above it to be able to make decisions about what to keep and what to toss. All the examples you gave (ARP, DNS, RAM) etc... are great examples of good caching because the ARP resolver knows what's worthwhile and the DNS cache has different rules for what data to keep. You can cache things within your layer that you have knowledge about. At best, you can cache information that is BELOW your layer, but you can't cache upwards. That's why ARP, DNS, RAM, etc.. work so well. The IP layer, by design, doesn't have that knowledge, and can't cache efficiently.
Your idea of a distributed document fragment cache sounds like freenet. I think it's a great idea, but it belongs in layer 7, not 3.
Despite what your Fiancee says, that doesn't make sense.
If I were an ISP, I wouldn't care what kind of traffic was getting pumped over my network. Whether you're playing CS or SSHd into work or downloading movies, I wouldn't care (until the *IAA came caling).
What I _would_ care about would be things that affect my bottom line. If you're pumping tons of data, increasing my contention ratio and making me purchase more bandwidth, I'd care. If you're expecting 99.99999% uptime and the service goes down for 8 mins and you call to complain, I'd care.
Maybe it's current ISP policy, but if I were running an ISP, I would certainly care about people using an "obscene amount of bandwidth" if it caused congestion for my other customers.
He's not talking about syncing up a 15gig home directory. He's talking about producing 230gigs of data per month in deltas to whatever he's generating (I hope he's using rsync and not something naive).
Backing up 230 gigs/month is certainly business class usage. If "business" isn't a good adjective use "large" if you want. You don't have to be making money to need "business" features.
It isn't backwards compatible in any real sense with IPv4. You might as well switch to a different protocol entirely then switch to IPv6. IPv6 can talk back to IPv4 through crazy tunnels that nobody but people on slashdot understand. But nobody on IPv4 can talk with IPv6 easily (from my understanding, anyway)
With all due respect, it doesn't seem like you know what you're talking about. If you have a dual stack (ipv6 and ipv4) machine, you can communicate with ipv4 machines just as easily as an ipv6 machine. There is a special prefix that maps to each ipv4 address in the ipv6 space. That was designed that way so we could have a gradual transition.
Going the other way (ipv4 to ipv6) depends on tunneling solutions, but they were similarly developed to have gradual adoption.
You can't solve any of these problems by swapping out IPv4 and also not ripping out/redoing the entire network stack
1) You _can_ roam between different networks and keep your sessions alive, it just requires that the 'different networks' be run by the same company that can rewrite your packets. Mobile providers do that. Any other solution involves the public internet keeping track of where every 'roaming' user is at any time. Routing tables are already really full, adding in a potential billion or so new routes will make things worse.
2) You misunderstand what multicast means. Multicast means that all clients receive all the same data at the same time. The intention is for streaming media where you can pick up at any point without since the old data is uninteresting. Pushing that functionality down to the data layer doesn't make sense. Would you suggest that all network stacks would have to keep a cache of all possible documents people could request?
3) It's not sane because of routing issues. Again, routers have finite memory, you have to simplify the routing decisions. Making it so that any IP could be in any network and recieve packets is unrealistic.
4) You can still DDOS any protocol. If you flood enough data at it, you win. The most you can do is tell your upstream to block that traffic (and that's what's done today)
5) Authenticate against what? Cryptographically signing each packet you send? Who controls the keys, who verifies your identity? The good thing about keeping it how things are (at the application level) is that the APPLICATION gets to decide, not whoever wrote the network stack.
6) Nothing can solve that problem outside of a completely controlled, homogeneous network. If I can hack a computer and get control of it, then all the data that's sent from there looks like it's from there. If you made a network of unhackable machines (ha), then you could solove your problem.
a) You don't "own" the IP addresses, you are assigned an address by the IANA, and they don't let you "transfer" them from one person to another.
b) If you were to try to resell some of your own IP range to another, you would have to sell a ridiculously large number of addresses to be large enough for the global routers to pay attention to your route message.
All your responses are band-aids that would be totally unneeded without NAT. Does it not bother you to have developers spending extra time/effort getting the network to do what it wants (let you connect to another machine) instead of on other features they want?
and furthermore, unless you have an ungodly number of contigous addresses, parceling off and selling them won't work because the global routing tables ignore smaller route messages for space reasons. So, if I bought (say) a/24 from you, nobody would bother routing it (or at best, it would go a really non-optimal path)
Or similarly, if people realised oil is a finite resource and we need to start moving to a different one instead of the stopgap "LETS START DRILLING EVERYWHERE" idea we've had.
If your friendly government has a tool that it says will "only be used against the bad guys". What to keep you from being a "bad guy" when an unfriendly government comes to power (either by force or by coercion)
It is a US-hosted site with a majority of its users being americans. Of course he brought up the system we have in the US.
What did you expect? A dissertation on rights in every single country, state and municipality in the world? If you don't expect him to enumerate EVERY SINGLE system of rights on the planet, then you acknowledge that he has to choose what to talk about. If he has to choose what to talk about, wouldn't it make sense that he talks about what he is most knowledgable in?
Now,
2. your own government doesn't stick to the rules anyway I'll be the first to bash our government on our technology policy, but your quote, while factually accurate, is misleading. Yes, there is a big controversy over the government wiretapping without a warrant, but that doesn't change (what the article is talking about) the ability to be anonymous. We still have free internet cafes and other points we can get to the internet anonymously and post dissident material, which is a bedrock of our society. The court even struck down a state anti-spam law because it removed the right to anonymity.
for me i consider privacy a right, but anonymity is purely dependant on the situation. should scammers have the right to post shit anonymously? of course they don't, hence it's not a "right".
I don't know where you're from, but in a number of jurstictions (including, I would assume all democracies), the right to privacy _is_ a right. It is in the US, and it is in the UK/EU.
In fact, I think that the right to anonymity (in terms of speech) is a fundamental right in a free and open society.
I'm not sure if that was intended to be a joke, but that's been debunked. Pencils don't work in space because the graphite floats around and shorts out your electronics.
I don't see why you keep crapping on skype all over this thread. The alternative clients you suggest don't fulfill the problems you have with skype.
If you're worried about the security of your data, using AIM doesn't even use encryption for either the text or the data. If both of your clients are behind NAT and can't connect directly, AIM will proxy your data to an AOL-hosted server. Skype does the same thing, except the data is encrypted by the different endpoints.
The skype protocol has been reversed engineered, and it is the same process that hackers have been going through for years: figure out the protocol, they change it, repeat. There's a nice writeup (of all places) on wikipedia.
You can get idealogical about skype being closed source, but very honestly, I don't even bother to read the source of open source software.
There was a rocket that NASA launched which detonated in the boost phase. They pulled the RTG out of the ocean, fixed the dings on the casing and reused it in another satelite.
Slashdot Mirror
Hopefully the other responses will see this post.
I understand how group based healthcare works. My coverage is independent from my employer, so I'm paying 100% of the cost. I understand the $50/mo ($600/year) goes towards paying other people's coverage seeing as I haven't seen a doctor in years. I get that.
I was arguing the GGP (or GGGP at this point)'s assertion that nationalized healthcare like how it's done in the UK takes from some people to give to others. I also think that there's a moral hazard in not giving an economic incentive for people to remain as healthy as possible. We've apparently, as americans, lost any sense of taking care of ourselves, but sticking a price tag to it makes it so that individuals and employers alike have a vested interest in remaining more healthy.
Even in a group health plan, people who are using it less pay less.
I'm a republican. I've had my sig on the bottom since I joined /.
The reality is, we were promised lower spending and lower taxes 8 years ago and we were lied to. I would much rather have someone who up front says, "Hey, here's the programs we want to have" rather than, for instance, when the prescription drug benefit was pushed through the white house, the Medicare Administrator was threatened to keep quiet that the plan was going to cost $138 billion more than planned.
Thanks for the generalization, but if you look it up, Americans voluntarily donate 2x as much per capita than the closest European country.
It's not a selfishness thing. While there are the advantages, we see what happens to our neighbors up north and our brothers in the UK with their NHS, and it sucks. We just don't want that.
Because I am a healthy 23 year old male. When I'm here, I pay next to nothing for health insurance. I lived in the UK for a while. I was paying about 80 dollars a week for healthcare.
In one situation, I pay $50/month for healthcare. In the UK I was paying 160 pounds (about $320) a month for healthcare. I didn't go visit any health facilities while I was there, but I had a friend try to get an appointment to get some antibiotics and it took _three weeks_.
We could argue about whether it's worthwhile to society or not, but under your (assuming you're european) system, it is what the GP said. I was charged much much more for an (arguably) inferior service without regards to the probability of whether or not I would use it. That money was used so I could subsidize someone else's treatment.
Even if you hit the option to not have all the toolboxes show up in the alt-tab window (under ubuntu), then you have nothing to alt-tab to once all your documents are closed and you want to start a new one.
It's crufty and it shows. I use linux on all my machines, and despite trying my damndest to get the gimp to play nice, it simply won't.
I'm going to invoke the slippery slope argument on giving the state of kentucky the ability to block its citizens access to internet sites.
I will then cut and paste the "The hangman came for x but I wasn't x, so I didn't care" poem.
This was posted already, come on admins. At least try to act like you know what's going on.
Fair enough that that's how her company is run. It doesn't make sense though business wise. And, like you said, her company is untypical.
"If you have a dual stack (ipv6 and ipv4) machine, you can communicate with ipv4 machines just as easily as an ipv6 machine."
Right, so that's exactly what the original poster said. IPv4 can't communicate with IPv6 - the IPv6 machine ALSO NEEDS TO BE DUAL STACK WITH IPv4. And use up one of those precious rare v4 addresses. So since your v6 server needs to have v4 on it to communicate with all the legacy v4 clients out there - you might as well save time and money and just leave it as straight v4 in the first place.
You're partially right. The dual-stack machine can be using nat (or whatever else) just like it is now. No change, that's the status quo. But, as each machine gets their own network to support IPV6 further and further out, the need for 6to4 translation diminishes until, at some point, the entire internet supports it.
Dual stack is a stopgap measure and a means to transition, not a permanent situation. The two situations I see are either permenantly using NAT (which works okay now, but what if we end up with 100 or 1000 or 1E6 as many devices as we have now) or moving to IPV6. It makes much more sense to go ahead and start transitioning now as opposed to when we really do exhaust the IPv4 address space.
Cache in your browser. Cache in your OS. Cache in your LAN router/proxy. Cache in your ISP. Cache in your peering hub. Cache wherever and whenever you can; design so you can trade off storage vs transmission speed, so you use whatever's locally cheaper.
Caching layer 7 data in layer 3 doesn't make sense because layer 3 just doesn't have enough information about the layers above it to be able to make decisions about what to keep and what to toss. All the examples you gave (ARP, DNS, RAM) etc... are great examples of good caching because the ARP resolver knows what's worthwhile and the DNS cache has different rules for what data to keep. You can cache things within your layer that you have knowledge about. At best, you can cache information that is BELOW your layer, but you can't cache upwards. That's why ARP, DNS, RAM, etc.. work so well. The IP layer, by design, doesn't have that knowledge, and can't cache efficiently.
Your idea of a distributed document fragment cache sounds like freenet. I think it's a great idea, but it belongs in layer 7, not 3.
Despite what your Fiancee says, that doesn't make sense.
If I were an ISP, I wouldn't care what kind of traffic was getting pumped over my network. Whether you're playing CS or SSHd into work or downloading movies, I wouldn't care (until the *IAA came caling).
What I _would_ care about would be things that affect my bottom line. If you're pumping tons of data, increasing my contention ratio and making me purchase more bandwidth, I'd care. If you're expecting 99.99999% uptime and the service goes down for 8 mins and you call to complain, I'd care.
Maybe it's current ISP policy, but if I were running an ISP, I would certainly care about people using an "obscene amount of bandwidth" if it caused congestion for my other customers.
He's not talking about syncing up a 15gig home directory. He's talking about producing 230gigs of data per month in deltas to whatever he's generating (I hope he's using rsync and not something naive).
Backing up 230 gigs/month is certainly business class usage. If "business" isn't a good adjective use "large" if you want. You don't have to be making money to need "business" features.
It isn't backwards compatible in any real sense with IPv4. You might as well switch to a different protocol entirely then switch to IPv6. IPv6 can talk back to IPv4 through crazy tunnels that nobody but people on slashdot understand. But nobody on IPv4 can talk with IPv6 easily (from my understanding, anyway)
With all due respect, it doesn't seem like you know what you're talking about. If you have a dual stack (ipv6 and ipv4) machine, you can communicate with ipv4 machines just as easily as an ipv6 machine. There is a special prefix that maps to each ipv4 address in the ipv6 space. That was designed that way so we could have a gradual transition.
Going the other way (ipv4 to ipv6) depends on tunneling solutions, but they were similarly developed to have gradual adoption.
You can't solve any of these problems by swapping out IPv4 and also not ripping out/redoing the entire network stack
1) You _can_ roam between different networks and keep your sessions alive, it just requires that the 'different networks' be run by the same company that can rewrite your packets. Mobile providers do that. Any other solution involves the public internet keeping track of where every 'roaming' user is at any time. Routing tables are already really full, adding in a potential billion or so new routes will make things worse.
2) You misunderstand what multicast means. Multicast means that all clients receive all the same data at the same time. The intention is for streaming media where you can pick up at any point without since the old data is uninteresting. Pushing that functionality down to the data layer doesn't make sense. Would you suggest that all network stacks would have to keep a cache of all possible documents people could request?
3) It's not sane because of routing issues. Again, routers have finite memory, you have to simplify the routing decisions. Making it so that any IP could be in any network and recieve packets is unrealistic.
4) You can still DDOS any protocol. If you flood enough data at it, you win. The most you can do is tell your upstream to block that traffic (and that's what's done today)
5) Authenticate against what? Cryptographically signing each packet you send? Who controls the keys, who verifies your identity? The good thing about keeping it how things are (at the application level) is that the APPLICATION gets to decide, not whoever wrote the network stack.
6) Nothing can solve that problem outside of a completely controlled, homogeneous network. If I can hack a computer and get control of it, then all the data that's sent from there looks like it's from there. If you made a network of unhackable machines (ha), then you could solove your problem.
6)
Because:
a) You don't "own" the IP addresses, you are assigned an address by the IANA, and they don't let you "transfer" them from one person to another.
b) If you were to try to resell some of your own IP range to another, you would have to sell a ridiculously large number of addresses to be large enough for the global routers to pay attention to your route message.
All your responses are band-aids that would be totally unneeded without NAT. Does it not bother you to have developers spending extra time/effort getting the network to do what it wants (let you connect to another machine) instead of on other features they want?
Take into the effect that most Computers are not servers, and don't need a Real IP Address.
No no no no no. You are breaking a fundamental property of the internet with that. uPNP et al are just bandaids to cover up that fact.
Machines _should_ have publically routable IP addresses.
and furthermore, unless you have an ungodly number of contigous addresses, parceling off and selling them won't work because the global routing tables ignore smaller route messages for space reasons. So, if I bought (say) a /24 from you, nobody would bother routing it (or at best, it would go a really non-optimal path)
Or similarly, if people realised oil is a finite resource and we need to start moving to a different one instead of the stopgap "LETS START DRILLING EVERYWHERE" idea we've had.
I would think accelerometers aren't used. Another point of failure and all.
Cool, a soon as there's an "open" device that supports all the features, I'm in.
Where did I put my DIY chip foundry.....
If your friendly government has a tool that it says will "only be used against the bad guys". What to keep you from being a "bad guy" when an unfriendly government comes to power (either by force or by coercion)
It is a US-hosted site with a majority of its users being americans. Of course he brought up the system we have in the US.
What did you expect? A dissertation on rights in every single country, state and municipality in the world? If you don't expect him to enumerate EVERY SINGLE system of rights on the planet, then you acknowledge that he has to choose what to talk about. If he has to choose what to talk about, wouldn't it make sense that he talks about what he is most knowledgable in?
Now,
2. your own government doesn't stick to the rules anyway
I'll be the first to bash our government on our technology policy, but your quote, while factually accurate, is misleading. Yes, there is a big controversy over the government wiretapping without a warrant, but that doesn't change (what the article is talking about) the ability to be anonymous. We still have free internet cafes and other points we can get to the internet anonymously and post dissident material, which is a bedrock of our society. The court even struck down a state anti-spam law because it removed the right to anonymity.
for me i consider privacy a right, but anonymity is purely dependant on the situation. should scammers have the right to post shit anonymously? of course they don't, hence it's not a "right".
I don't know where you're from, but in a number of jurstictions (including, I would assume all democracies), the right to privacy _is_ a right. It is in the US, and it is in the UK/EU.
In fact, I think that the right to anonymity (in terms of speech) is a fundamental right in a free and open society.
damnit.
Even with viewing all the comments D2 often will show a post and it's GP, but not the middle post. I thought you were replying to ivan256...
I'm not sure if that was intended to be a joke, but that's been debunked. Pencils don't work in space because the graphite floats around and shorts out your electronics.
I don't see why you keep crapping on skype all over this thread. The alternative clients you suggest don't fulfill the problems you have with skype.
If you're worried about the security of your data, using AIM doesn't even use encryption for either the text or the data. If both of your clients are behind NAT and can't connect directly, AIM will proxy your data to an AOL-hosted server. Skype does the same thing, except the data is encrypted by the different endpoints.
The skype protocol has been reversed engineered, and it is the same process that hackers have been going through for years: figure out the protocol, they change it, repeat. There's a nice writeup (of all places) on wikipedia.
You can get idealogical about skype being closed source, but very honestly, I don't even bother to read the source of open source software.
Also, it just works (tm).
There was a rocket that NASA launched which detonated in the boost phase. They pulled the RTG out of the ocean, fixed the dings on the casing and reused it in another satelite.