Funny you claim "every government" is incapable of managing such a task, since both the countries I've lived (Norway and the UK) in has managed to do so just fine.
Just the thought of living in a country with privatized healthcare is abhorrent to me, amongst others because it creates little incentive for anyone to actually look at the big picture and put in place proper preventative programs.
In the UK, for example, not doing enough to prevent health problems directly costs the NHS money in more care. As a result they put in a huge amount of effort in programs to help people stop smoking, for example, because it comes out of THEIR budget when people later get cancer or other health problems as a result of smoking.
Because many of them have far higher re-offending rates. Murder is a crime few offenders do more than once even if they do get out of prison. On the other hand assaults, rape etc. are carried out by offenders that often repeat or escalate.
Except it is not a good deterrent. There is absolutely no proof that longer murder sentences leads to higher rates of murders.
Truth of the matter is that there are countries where murder carries long sentences with high murder rates (like the US) and with low murder rates (like the UK) and there are countries where murder carries short sentences with very low murder rates (like the Scandinavian countries) - there's no conclusive link between the length of sentences in these cases and frequency.
The point is that the majority of murders, premeditated or not, are done without any thought for the consequences. It is either done in affect or it is done in emotional states where you most certainly will not spend time worrying about whether you'll be locked up for life or "only" a handful of years.
Flash chips doesn't. Flash drives do. So what you are saying is really just that the the flash drive manufacturers is doing the same as the harddrive manufacturers. It has nothing to do with the capacity of the flash chips, which are typically reported in Mbits with M=1048576.
KB certainly "never ever" meant 1000 bytes either, since "K" isn't a SI prefix. It thoroughly amuse me when the "SI prefixes are holy" crowd keep writing KB instead of "kB" which is what you should be writing if you wish to actually use SI prefixes.
The C64 is a bad example, because it actually had 65K of RAM (notice, btw., that "K" is not a valid SI-prefix - the SI prefix is "k") - the 64K "normal" RAM, and 1K hardwired to be used as color data for the VIC chip.
But I can't remember ever seeing the C64 advertized as having 65K. Perhaps they did that in the US, but all of the magazines I have from the 80's say 64K.
Hey, you're posting on Slashdot, so obviously you want to be seen. Why aren't you posting your real name, your home address, your phone number and your e-mail address?
Why would you not want it to be seen you you are, since you're posting in public?
However a large number of domains ARE owned by private individuals, and the whois requirements means you either pony up for "protection" (third party services that put their name and address in whois and forward any requests to you) or leave your personal details available for anyone.
I can't see ANY reason why these details should be required to be public. It ought to be sufficient that the registrar has the details so they can be subpoeaned, and optionally request them to forward requests.
With the current system, nothing stops people with bad intentions from lying anyway - the current system only harms those who wants to stay honest.
It is highly relevant if you're trying to make money by buying all the tickets. Suppose you pay tickets for $100 million to win a $150 million jackpot, but that jackpot is paid out over 10 years, which is pretty normal for lotteries with large payouts.
First of all, inflation will reduce the value of that $150 million payout. But more importantly, if you invested $100 million in something else, with any sense at all you'd get a return on that money over those 10 years.
Assuming an average 5% return over the next 10 years on your $100m, which would mean a pretty conservative investment, you'd be left with more than $162 million at the end of the 10 years. Odds are you could make on average well above 5%, and could allow yourself to spend a portion of the returns and still end up with more than $150m at the end of it.
In that case, investing $100m to win $150m is a bad choice, whereas if the payout is immediate (or even just over a shorter period - how short depends on what return you'd expect to be able to make on investing the money) it would be a good investment.
If they were so amazingly smart as to know that humans would come along a few tens of million years after their time, yet stupid enough to worry about it.... Nah, I think I'll discount that idea.
The entire point of this thing is that it adds extra capabilities and performance to web apps. Presumably, unless the app developers choose to depend on features that are only available in Prism, you can still access those same apps on any browser.
Looking at one of his papers it looks like it's 90% just caching routing information in the routers.
Their routers calculate a hash from the IP headers, and then use that to parallelize out the actual routing, so that they can use cheaper DRAM instead of SRAM (performance of DRAM is a problem for high end routers), then they do a lookup in a hash table to see if they've come across this "flow" before. If so, a lot of information about QOS and about the route chosen initially is already stored, so that the router doesn't have to do much work on each packet. If the flow is "new" the normal routing steps are carried out.
So it's a method for allowing techniques that are typically considered too expensive to carry out in large scale backbone routers to be used to optimize bandwidth usage.
There are no explicit circuits - if a router fails, preceding routers in the chain will just reroute elsewhere, and if the "new" routers in the chain are flow routers, they'd create a new flow and proceed as if nothing had happened.
Of course two systems layered can be more secure than either system separately, but they can also be more insecure.
If either system can allow uncontrolled access to hardware or disk if there's a bug, for example, having two systems instead of one would be less secure.
Similarly if the hosting system has a bug or configuration problem that allow an attack against the hosting system to grant access to the hosted system, then obviously security will be weakened. Some vm systems have a way to get password-free access to the vm's if you have root access to the host, for example (though that's debatably not a huge problem - if you have root access you could install a modified kernel and reboot unless the system has been hardened a lot more than what people usually do) so any root compromise of the host system would compromise any and all vm's on such systems.
It's hard to get away from the thought that ANY additional network service potentially increases the possible methods of attacks, and so hypervisors that are network accessible does have a potential risk.
On the other hand, virtualization also allows a lot of additional isolation that people might not be able to afford otherwise - not everyone can afford to run a single service per server to prevent security holes in one from affecting others. In cases like that, virtualization may mean your applications may face significantly reduced privileges. Personally I think that benefit does have significant potential to reduce the attack vectors far more than adding a hypervisor will increase it. Particularly since the hypervisor is likely to receive far more scrutiny than an app put together for a small business with a handful of developers.
If you use virtualization to concentrate more services on the same hardware, you almost inevitably increase the risk of weaker security: For each additional service you add additional potential for misconfigured vm's and broken applications, and if there are then bugs in the hypervisor that allow you to break out of a vm and gain access to the hypervisor or other vm's you suddenly have a far greater problem than before you concentrated things.
I think Theo is making a point that's worth keeping in mind, but in his usual obnoxious way, and I think it is a far smaller problem than he makes it.
When England and Australia passed very restrictive firearms regulations a few years ago, property crime and especially violent crime in both countries took a significant jump upward right afterward, and have stayed up.
Regardless whether you are right about this or not, England has one of the lowest rates of guncrime in the world, and certainly compared to the US the overall rate of violent crimes is miniscule too. Don't know about Australia.
You'll also easily find plenty of countries with restrictive firearms laws that have very low crime rates, so at the very least the correlation is not so straightforward. It might very well be right that restrictive firearms laws cause crime rates to increase in places where large number of weapons are in circulation, for example, or where police are routinely armed, or where poverty is high.
None of the available e-paper solutions are anywhere near the thickness or flexibility needed to make it practical to produce devices consisting of more than one or two "sheets". I doubt we'll see devices anything remotely like what you're suggesting for decades vs. at most two-"sheet" devices.
There's several different technologies already doing similar things to this.
The oldest being "single system image" clusters. DEC/Compaq/HP had an SSI project for Linux a while back, see openssi.org.
There's also a number of platforms such as 3Tera's AppLogic that abstracts away the invididual servers using virtualization. AppLogic looks fantastic in many respects - you get a nice GUI to connect together "components" to applications that can span hundreds of CPU's, and you can then deploy applications on a "grid" of host nodes and have their system provision everything, or you can snapshot an application or create a scaled down copy for development/testing etc.
It's great technology, and actually not too badly priced (prices for hosted AppLogic grids are roughly the same as for managed hosting of similarly spec'ed servers from some of the providers I've looked at).
You'll see a lot of these spring up over the next few years - deploying large scale apps "manually" is a real pain, and most successful companies that have to scale end up reinventing a lot of operational infrastructure. Systems such as AppLogic finally start making it possible to outsource or purchase a lot of that stuff and spend your inhouse resources on your core competencies instead.
Where you'll really see this start to shine is when people start cycling hardware. Cycling hardware to keep using the most cost effective platform is damn resource intensive (time/manpower) if your operational infrastructure isn't automated. Virtualization is a godsend for that. If I can constantly pick the most cost effective hardware and abstract it away completely from our apps, and take down or add hardware nodes transparently, then that'll save me a huge amount of time and money.
I absolutely agree. If I'm traveling for a long vacation I may check in a suitcase (mostly because I'm then going with my wife, and she couldn't travel with carry-on only if her life depended on it), otherwise I'll always do carry-on only.
I can pack a full weeks worth of clothes in carry-on easily, plus my laptop, charger, phone charger, assorted cables etc. and something to read. For longer trips I'll use the laundry facilities at the hotel (tip: a lot of hotels will have a cheap self service laundry room hidden away somewhere if you don't want to use the overpriced laundry service - ask for it).
I usually can't even be bothered bringing liquids through security anymore. I buy water for the trip and minimal toiletries after security (but the hotel will have at least shampoo and body lotion etc. anyway, and most will supply toothpaste etc. if you ask for it - most hotels apart from the real budget ones _will_ provide you what you need, and for most basic stuff you won't pay extra).
It makes travel a ahell of a lot more pleasant.
And pick your carry-on bag with care. My newest one is a really nice (but not particularly expensive) samsonite bag that has a lot of useful space, but more importantly it has four solid (and replacable - the wheels on my last bag got ruined far too quickly) wheels that will roll sideways... Sounds like it doesn't make much difference, but try running through a busy airport, or pushing a carry-on that is probably well over the weight-limit (I've never had that be a problem, thankfully) down the aisle of the plane rather than dragging or carrying it, and having a bag that's easy to maneuver and move through narrow spaces makes a big difference.
My wife thinks all the small things I care about are crazy, but when you spend 16 hours door to door each way every 6-8 weeks, making it pleasant becomes an obsession;)
I've thought the same thing. Added screening just forces potential terrorists to target something else - witness the failed attack at Glasgow airport for example (burning car droven into the terminal building - unfortunately for the idiots that did it, it didn't blow up), or the attacks on the London underground and buses that amazingly only killed 51 or so people.
Security theater: having to take off my shoes and other gear, can't have anything metal in carry-on luggage, severely limited amounts of liquids (I get thirsty easily and it can take a while for the stews to pass around refreshments). I don't feel any safer.
WTF?!?
I fly from Heathrow, London to SFO regularly. I ALWAYS take 2-4 litres of water with me unless I've gotten a business upgrade. The limitation is on bringing liquids through security - anything you buy after security is fine to take on board everywhere I've flown from over the last few years. I also always have metal stuff in my carry-on - I usually don't check any bags as my trips are generally for a week, and I can just barely stuff everything I need for a week in my carry-on. That includes my laptop and lots of gadgets and cables. You just can't have anything sharp
I agree with you I don't feel any safer though, but then I didn't feel unsafe to start with.
Most full price airlines offer that these days in business class, more or less. It's not really like a sleeper compartment - you get a chair that goes almost or completely flat, with legrests that extend and go almost flat, so you get close to a full length bed. You also get privacy screens to the side, so you can close yourself relatively well off.
Don't know what US airlines offers something similar - United (the only US airline I've personal experience with) has announced they'll start rolling it out this year, at least on their 777's, but I haven't seen it so far on the flights I take.
Not really. Economy Plus usually give you about 5-8" extra legroom depending on airline. Hardly sleeper compartments. I've flown Economy Plus on BA a bit, and it's worth the extra money, but you're paying for the extra space only - no extra service (service is the same as for regular economy). If you fly between the US and the UK, though, United is often a lot better value. Especially if you do more than 2-4 trips (depending on length), as once you get to Premier in Mileage Plus you get the premium economy seats at no extra cost.
Slashdot Mirror
Just the thought of living in a country with privatized healthcare is abhorrent to me, amongst others because it creates little incentive for anyone to actually look at the big picture and put in place proper preventative programs.
In the UK, for example, not doing enough to prevent health problems directly costs the NHS money in more care. As a result they put in a huge amount of effort in programs to help people stop smoking, for example, because it comes out of THEIR budget when people later get cancer or other health problems as a result of smoking.
I've heard they have this newfangled thing called TV now.
Because many of them have far higher re-offending rates. Murder is a crime few offenders do more than once even if they do get out of prison. On the other hand assaults, rape etc. are carried out by offenders that often repeat or escalate.
Truth of the matter is that there are countries where murder carries long sentences with high murder rates (like the US) and with low murder rates (like the UK) and there are countries where murder carries short sentences with very low murder rates (like the Scandinavian countries) - there's no conclusive link between the length of sentences in these cases and frequency.
The point is that the majority of murders, premeditated or not, are done without any thought for the consequences. It is either done in affect or it is done in emotional states where you most certainly will not spend time worrying about whether you'll be locked up for life or "only" a handful of years.
If that precision isn't needed, because the meaning is understood, then what the fuck is the problem with using K=1024?
Flash chips doesn't. Flash drives do. So what you are saying is really just that the the flash drive manufacturers is doing the same as the harddrive manufacturers. It has nothing to do with the capacity of the flash chips, which are typically reported in Mbits with M=1048576.
KB certainly "never ever" meant 1000 bytes either, since "K" isn't a SI prefix. It thoroughly amuse me when the "SI prefixes are holy" crowd keep writing KB instead of "kB" which is what you should be writing if you wish to actually use SI prefixes.
But I can't remember ever seeing the C64 advertized as having 65K. Perhaps they did that in the US, but all of the magazines I have from the 80's say 64K.
Bytes isn't a SI unit, so what the SI-prefix has always been has no relevance.
Why would you not want it to be seen you you are, since you're posting in public?
I can't see ANY reason why these details should be required to be public. It ought to be sufficient that the registrar has the details so they can be subpoeaned, and optionally request them to forward requests.
With the current system, nothing stops people with bad intentions from lying anyway - the current system only harms those who wants to stay honest.
First of all, inflation will reduce the value of that $150 million payout. But more importantly, if you invested $100 million in something else, with any sense at all you'd get a return on that money over those 10 years.
Assuming an average 5% return over the next 10 years on your $100m, which would mean a pretty conservative investment, you'd be left with more than $162 million at the end of the 10 years. Odds are you could make on average well above 5%, and could allow yourself to spend a portion of the returns and still end up with more than $150m at the end of it.
In that case, investing $100m to win $150m is a bad choice, whereas if the payout is immediate (or even just over a shorter period - how short depends on what return you'd expect to be able to make on investing the money) it would be a good investment.
If they were so amazingly smart as to know that humans would come along a few tens of million years after their time, yet stupid enough to worry about it.... Nah, I think I'll discount that idea.
It lets you run multiple processes, but you need to pass an extra command line option to achieve it. Can't remember the name.
Their routers calculate a hash from the IP headers, and then use that to parallelize out the actual routing, so that they can use cheaper DRAM instead of SRAM (performance of DRAM is a problem for high end routers), then they do a lookup in a hash table to see if they've come across this "flow" before. If so, a lot of information about QOS and about the route chosen initially is already stored, so that the router doesn't have to do much work on each packet. If the flow is "new" the normal routing steps are carried out.
So it's a method for allowing techniques that are typically considered too expensive to carry out in large scale backbone routers to be used to optimize bandwidth usage.
There are no explicit circuits - if a router fails, preceding routers in the chain will just reroute elsewhere, and if the "new" routers in the chain are flow routers, they'd create a new flow and proceed as if nothing had happened.
If either system can allow uncontrolled access to hardware or disk if there's a bug, for example, having two systems instead of one would be less secure.
Similarly if the hosting system has a bug or configuration problem that allow an attack against the hosting system to grant access to the hosted system, then obviously security will be weakened. Some vm systems have a way to get password-free access to the vm's if you have root access to the host, for example (though that's debatably not a huge problem - if you have root access you could install a modified kernel and reboot unless the system has been hardened a lot more than what people usually do) so any root compromise of the host system would compromise any and all vm's on such systems.
It's hard to get away from the thought that ANY additional network service potentially increases the possible methods of attacks, and so hypervisors that are network accessible does have a potential risk.
On the other hand, virtualization also allows a lot of additional isolation that people might not be able to afford otherwise - not everyone can afford to run a single service per server to prevent security holes in one from affecting others. In cases like that, virtualization may mean your applications may face significantly reduced privileges. Personally I think that benefit does have significant potential to reduce the attack vectors far more than adding a hypervisor will increase it. Particularly since the hypervisor is likely to receive far more scrutiny than an app put together for a small business with a handful of developers.
If you use virtualization to concentrate more services on the same hardware, you almost inevitably increase the risk of weaker security: For each additional service you add additional potential for misconfigured vm's and broken applications, and if there are then bugs in the hypervisor that allow you to break out of a vm and gain access to the hypervisor or other vm's you suddenly have a far greater problem than before you concentrated things.
I think Theo is making a point that's worth keeping in mind, but in his usual obnoxious way, and I think it is a far smaller problem than he makes it.
Regardless whether you are right about this or not, England has one of the lowest rates of guncrime in the world, and certainly compared to the US the overall rate of violent crimes is miniscule too. Don't know about Australia.
You'll also easily find plenty of countries with restrictive firearms laws that have very low crime rates, so at the very least the correlation is not so straightforward. It might very well be right that restrictive firearms laws cause crime rates to increase in places where large number of weapons are in circulation, for example, or where police are routinely armed, or where poverty is high.
None of the available e-paper solutions are anywhere near the thickness or flexibility needed to make it practical to produce devices consisting of more than one or two "sheets". I doubt we'll see devices anything remotely like what you're suggesting for decades vs. at most two-"sheet" devices.
The oldest being "single system image" clusters. DEC/Compaq/HP had an SSI project for Linux a while back, see openssi.org.
There's also a number of platforms such as 3Tera's AppLogic that abstracts away the invididual servers using virtualization. AppLogic looks fantastic in many respects - you get a nice GUI to connect together "components" to applications that can span hundreds of CPU's, and you can then deploy applications on a "grid" of host nodes and have their system provision everything, or you can snapshot an application or create a scaled down copy for development/testing etc.
It's great technology, and actually not too badly priced (prices for hosted AppLogic grids are roughly the same as for managed hosting of similarly spec'ed servers from some of the providers I've looked at).
You'll see a lot of these spring up over the next few years - deploying large scale apps "manually" is a real pain, and most successful companies that have to scale end up reinventing a lot of operational infrastructure. Systems such as AppLogic finally start making it possible to outsource or purchase a lot of that stuff and spend your inhouse resources on your core competencies instead.
Where you'll really see this start to shine is when people start cycling hardware. Cycling hardware to keep using the most cost effective platform is damn resource intensive (time/manpower) if your operational infrastructure isn't automated. Virtualization is a godsend for that. If I can constantly pick the most cost effective hardware and abstract it away completely from our apps, and take down or add hardware nodes transparently, then that'll save me a huge amount of time and money.
I can pack a full weeks worth of clothes in carry-on easily, plus my laptop, charger, phone charger, assorted cables etc. and something to read. For longer trips I'll use the laundry facilities at the hotel (tip: a lot of hotels will have a cheap self service laundry room hidden away somewhere if you don't want to use the overpriced laundry service - ask for it).
I usually can't even be bothered bringing liquids through security anymore. I buy water for the trip and minimal toiletries after security (but the hotel will have at least shampoo and body lotion etc. anyway, and most will supply toothpaste etc. if you ask for it - most hotels apart from the real budget ones _will_ provide you what you need, and for most basic stuff you won't pay extra).
It makes travel a ahell of a lot more pleasant.
And pick your carry-on bag with care. My newest one is a really nice (but not particularly expensive) samsonite bag that has a lot of useful space, but more importantly it has four solid (and replacable - the wheels on my last bag got ruined far too quickly) wheels that will roll sideways... Sounds like it doesn't make much difference, but try running through a busy airport, or pushing a carry-on that is probably well over the weight-limit (I've never had that be a problem, thankfully) down the aisle of the plane rather than dragging or carrying it, and having a bag that's easy to maneuver and move through narrow spaces makes a big difference.
My wife thinks all the small things I care about are crazy, but when you spend 16 hours door to door each way every 6-8 weeks, making it pleasant becomes an obsession ;)
I've thought the same thing. Added screening just forces potential terrorists to target something else - witness the failed attack at Glasgow airport for example (burning car droven into the terminal building - unfortunately for the idiots that did it, it didn't blow up), or the attacks on the London underground and buses that amazingly only killed 51 or so people.
WTF?!?
I fly from Heathrow, London to SFO regularly. I ALWAYS take 2-4 litres of water with me unless I've gotten a business upgrade. The limitation is on bringing liquids through security - anything you buy after security is fine to take on board everywhere I've flown from over the last few years. I also always have metal stuff in my carry-on - I usually don't check any bags as my trips are generally for a week, and I can just barely stuff everything I need for a week in my carry-on. That includes my laptop and lots of gadgets and cables. You just can't have anything sharp
I agree with you I don't feel any safer though, but then I didn't feel unsafe to start with.
Don't know what US airlines offers something similar - United (the only US airline I've personal experience with) has announced they'll start rolling it out this year, at least on their 777's, but I haven't seen it so far on the flights I take.
Not really. Economy Plus usually give you about 5-8" extra legroom depending on airline. Hardly sleeper compartments. I've flown Economy Plus on BA a bit, and it's worth the extra money, but you're paying for the extra space only - no extra service (service is the same as for regular economy). If you fly between the US and the UK, though, United is often a lot better value. Especially if you do more than 2-4 trips (depending on length), as once you get to Premier in Mileage Plus you get the premium economy seats at no extra cost.