Well, by that logic, everything from Greece to India should be part of Macedonia. Alexander conquered it all ~2300 year ago, which is waaaay before China ever got to Tibet.
Also, we shouldn't give those territories back to Mexico, they should go back to Spain.
We can't use history as a guide for where borders should be created today. It seems like we should be able to, but it doesn't work. Borders have changed over the centuries, and entire countries and societies have come and gone, so there is no way to determine what belongs to whom. Sure it would be nice if everyone could have their "ancestral lands" back, but that brings up the questions: Which ancestors? When? What if they were sold and there is no record? What if they were just absorbed, without conflict, but without consent?
No, your historical idea just doesn't work - we live in today, not 700 years ago, or even 100 years ago.
Something to keep in mind is that each "generation" can be as short as the single division of a cell. In addition, don't underestimate the power of individual cells/organisms/creatures/animals to act in a parallel manner. There are insects (termites) that lay 7000 eggs a day, providing 7000 opportunities for some sort of evolutionary event. Looking at the aphid, in a single season, if all offspring lived to reproduce, there would be 1,560,000,000,000,000,000,000,000 (1.56x10^24) offspring. Reference here.
Multiply that by the total number of insects (including different species) and this is a lot of opportunities for mutations and evolutionary activity. Also, keep in mind things like viruses and bacteria that reproduce in even greater numbers.
I don't claim that this makes evolution any less amazing, but a billion+ years and all the parallelization possible from all types of life makes for an incredible amount of computation. Enough, perhaps to even determine the question to life, the universe and everything.
How about this one: A hospital does not have to be a single building or even a single cluster of buildings. There may be numerous branch clinics, rehab centers, affiliated offices, etc. that make up the hospital. If a hospital grows, but real estate isn't available next to the existing facilities, they may add the buildings a block or two over. In Chicago, the Northwestern University Hospital network has clinics, rehab centers and therapy centers all over the place - think different suburbs. So, if they can't connect to the Internet, they should have to build their own private network to connect their different facilities, and isn't economically feasible.
Oh, wait, you mentioned they could maintain separate systems - one on the Internet and one internal only, and they could transfer the data that way. Unfortunately, that isn't a feasible solution - all locations would be required to maintain their own separate copies of the records, so an Internet outage wouldn't prevent records from being obtained. This would be a logistical and administrative nightmare.
In addition, patient records also have to be accessed by insurance companies, ambulance companies and other health care providers. Without automated access this would require dozens of "CD transfers" from secure systems to another. A nurse administers medication and it needs to be sent to billing, the insurance company and added to their medical record. This needs to happen automatically, unless you feel there should be hundreds of employees burning CDs and moving them to a different machine.
Regarding the "known Internet threat", there is also a known threat with not being able to access a medical record in a timely fashion. This is a single high-profile case of the dangers of connecting to the Internet, but there are many cases of the dangers of not having timely access to medical records. These dangers are why there are medic alert bracelets.
Finally, I don't know who your insurance company is, but mine will cover me if I leave my door unlocked or if I have stuff that's out in my yard (picnic table, deck furniture, grill, etc.). In fact, where I used to live in Canada, there were cars stolen every week when people started them to warm them up and went back inside to keep warm. They were all covered by insurance, and that is unlocked, running and unattended. I think your final paragraph is just FUD designed to make it seem like it is the hospitals fault when it isn't.
The simple fact is, if hospitals don't connect to the Internet to share information, costs will go up since there will need to be a person transfering information manually, and they need to be paid. Also, there will need to be twice the number of systems maintained, which means more capital and maintenance costs. It will also mean more deaths, since some situations require medical information -NOW-, and not after someone has burned a CD and physically moved it.
This is why we invented things like encryption, firewalls and IDS systems. They might not be perfect, and can't protect against everything, but they can reduce the risk to an acceptable level. Perfect security doesn't exist.
We are using Xen in an educational environment - providing a network of VMs to each student so they can play around, break stuff, and get some practical experience.
For your iptables problem... I think you just need to add a rule to forward packets so they will be transfered through the software bridge. Make sure to use -I to insert the rule, appending the rule doesn't help since it will be dropped before it reaches the forward rule.
I agree with you that documentation isn't all that great right now, but our lab group and several other groups that I know of are working hard at fixing that issue. Virtualization is being used in many new and interesting ways, and the tools and documentation haven't caught up yet. Give it a few more months and things should be much improved. If nothing else, my lab group wants to have our tools and documentation released under the GPL by early May, and we are really focusing on having clear, detailed and correct documentation.
How much does it cost for a wired phone in your home/apartment/office? Call it about $20/month. (I know it is more in many places, but stay with me...)
Now, a broadband connection, depending on overall throughput, may cost $100/month.
Still with me?
OK - How many voice conversations can be supported over that broadband connection? More than 5? More than 10? More than 100? Anything over 5 means the provider is losing money. If someone cancels their wired phone and uses a broadband connection, the provider loses that revenue. Now, a single customer might not be too bad, the company may be providing the high-speed connection. However, it is still less revenue than there would have been.
Now, we throw in a company that sets up in your neighborhood and offers you VoIP services over their broadband connection, so you don't need to keep your wired phone. If 10 people in the neighborhood do that, then the telecomm is 'losing' money. They have less revenue than before, and less profits, they have 1 new customer paying $100 for the broadband connection, but have lost 10 people paying $20, for a "loss" of $100.
Take it to the extreme, and have someone like Google provide phone service to anyone with an Internet connection. Imagine every person in the US cancels their regular phone service to use Google's service. The telecomms go bankrupt, or they have to increase the price of broadband by orders of magnitude. Yes, the company's may be bloated. Yes, tax dollars may have paid for the telecomm to run fiber. However, this was done so services could be fairly offered to everyone. Could my town of 800 have afforded to run fiber 30 miles to the nearest city? No. However, they can pay enough for service that it is profitable to maintain and manage that connection. It is the same with roads. Some roads use federal or state dollars to get paved. They may only connect 30 or 40 people, but that's the way things are. If we didn't do it this way, there would be extensive roads around cities, connected only by the Interstates. OK, I'm kind of off-topic now...
Anyway, the current market prices are because the revenue stream assumes that there will be wired home users paying more than their bandwidth is worth, as compared to a broadband Internet connection. If they lose those customers, it means that the cost for Internet bandwidth will rise - dramatically. So, they would rather have a company providing those services pay more, rather than having the cost pass on to all of their users.
Hmm, I'm curious, where does this technology actually exist (other than on paper and figments of engineer imaginations). The last time I checked 'we' dont have the technology to get humans reliably to/from low orbit, never mind anywhere near the moon.
True, this technology doesn't actually 'exist'. Just like the 2010 Fords and Chevys - they haven't been built yet. However, there is a lot of technology that -IS- down on paper. In fact, there are the original plans for the Saturn V rockets. It has come up before that "We can't build them anymore, we don't know how." or "We don't have the machines to build the parts anymore." OK, we don't. Ten years ago we couldn't build a 45nm chip even in a lab environment. Now, Intel will mass produce them in the next year. It isn't that the technology doesn't exist - it is that the technology hasn't been applied, and that is an entirely different matter. Techology isn't the barrier, the desire to apply that technology isn't there on a large enough scale to fund it.
So many folks seem to think that just because it was done in the 60's, it's easy, trivial, and a no brainer to go land on the moon. It's hard, expensive, and currently the technology to do it doesn't exist except on paper and in sci-fi literature. The closest thing the usa has to a manned lunar capabable piece of hardware is some rusty old Saturn V hulks sitting outside of some museums that relish 'the good old days' when america was actually a leader in the space race.
Hard? Yes, but we also have thousands of talented engineers - aerospace, mechanical and chemical, who are more than capable of doing this. (Unless you claim no one on Earth is as smart or talented as the engineers we had in the 60s.) Expensive? Yes. I'm not claiming it isn't. But it isn't more expensive than supporting a multi-year war/peace-keeping mission in Iraq. That is where the government has decided to spend a few hundred billion. In theory, (I said theory), we could leave Iraq and direct that money to NASA - the money -IS- there, it is just being spent on other things. As I mentioned above - the technology exists, we just aren't applying it to actually create the hardware.
In the 60's the landing was equatorial for a lot of technical reasons. Today, moon landings are sci-fi, for a lot more technical (combined with political and financial) reasons. Politicians may talk about going to the moon, but follow the money, it's not going to the moon, it's going to wars overseas. The talk of moon landings is nothing more than political rhetoric designed to gather up votes from folks that cant see the forest for all the trees, and actually believe that such things are in the plans of the administration. If it was actually in the administration plans, the mandate would be such as it was in the 60's, to get far enough into the program that it could NOT be cancelled at the expiry of the 8 year term, to much already invested. In reality, this administration is neatly talking the talk that allows for more talk, but not actually allocating funds to make it happen, then putting on schedules such that all the talk becomes a financial responsibility for a future administration. In laymans terms, that means, not gonna happen.
As I mentioned above - it isn't the technical reasons, it is the political reasons - and since the government has the power to set the budget, it is also a financial thing. I agree with you that in the current political structure it's not going to happen.
This article on/. is a perfect example of the propoganda working. So many folks seriously considering where a moon landing should occur, keeps the grassroots talk happening. Reality is, talk is cheap, and if the hardware is not being designed and built at this stage of the game, there is no program that needs to survive the change of administration coming in a relatively short timeframe (next election). That's when reality will start to h
While presented in a comedic way, it shows someone convincing a town that they have trouble because the billiard parlor in town has brought in a pool table. While billiards is OK, pool is a horrible game and will cause degradation in the children and cause them to stop doing their chores and become gamblers.
Just goes to show that new things are often looked upon as corruptive or causing some sort of lamentable behavior. I remember when Pac-Man came out and people objected since it was dangerous medically and that playing it was like running up 2-3 flights of stairs. Now it is considered harmless fun.
As mentioned previously, this sort of thing is being/has been done. One project I am familiar with is the Internet-scale Event and Attack Generation Environment (ISEAGE) project at Iowa State University.
Its webpage, has an overview of the project and documentation on its architecture and implementation. I think one of the key aspects of the project can be found in the overview: "Unlike computer-based simulations, real attacks will be played out against real equipment."
ISEAGE is approaching security from a real-world perspective, using real world devices. Sure, your software/hardware might be secure when the attacks are played against it; but is it secure when those attacks when there are dozens of attacks occuring simultaneously? What about when it is being hit by thousands of requests, or is under a DDoS attack? What happens when devices decide to start breaking the protocols, or the rules? What happens if a device physically fails? What is the effect of a device overheating during a DDoS attack? How do you simulate this/test for this other than hooking it up and hammering it with a DDoS attack?
This is the kind of information that is needed to prevent or mitigate an attack, but can't be found by reading code or running a scanner. How did the US figure out how to build rockets? We built some, they blew up, and better ones got built. The real world isn't the same as a lab.
Do you have any idea how many old viruses are still circulating?
I use the network at a major university and I get notified of newtear attacks at least once an hour. newtear targeted Windows 95 and NT. It has been around a LONG time, and still trying to infect new machines.
Take some time and Google for some of the big worms like Slammer and Code Red, and you will find people are still detecting them - thousands of times a day.
Regarding point 3, I have my AV set to scan every file I open, so clicking on a BrittenySpearsNaked.jpg.exe will cause it to be scanned and prevent the infection. You might want to look for this setting on your AV. (Scan on open or some such.)
Finally, AV is not the end-all, be-all of security; you should also have a firewall to prevent unwanted incoming and outgoing communications. AV is part of an overall security plan.
Bad behavior/design on the part of your anti-virus software doesn't mean you should throw up your hands and give up.
My concern would be that anything that could affect a portion of the disk would destroy more data. I know scratches that aren't noticed on a CD can make a DVD unreadable and, while a drive platter may not have the risk of scratches that optical storage does, the general idea is the same. A physical failure, such as a head alignment issue, that wouldn't be noticed with lower densities may be a factor with the higher densities.
Now, I don't have a solution to the problem, but I just want to point out that getting full performance out of something can raise new risks.
In the attacks, Paller said, the perpetrators "were in and out with no keystroke errors and left no fingerprints, and created a backdoor in less than 30 minutes. How can this be done by anyone other than a military organization?"
I don't know about anyone else, but when I type, I usually make a few mistakes. However, when I run a script it always runs the commands correctly. Even if it isn't a script, there is probably some attack case they are cutting and pasting the commands out of. It would be helpful to see how quickly individual characters are being typed, so it could be determined if they are typing the characters or not.
Also, having been in the military, there is no reason for them to train typists who never hit the wrong key.
Boss, I have a problem...
I got a call from the FBI that they wanted to meet with me regarding the use of the Kazaa program within our network. They think someone is downloading illegal music and kiddy porn. I was thinking we should block those programs to avoid liability, what do you think? Also, should we make it into a written policy just to be safe? I don't want to spend the next 2 weeks dealing with lawyers and law enforcement agents crawling through our networks.
Longer answer: No, I already have enough stuff that I need to focus on. If I needed to be distracted I would take off my headphones or check email every 5 minutes. If you could guarantee that the ads wouldn't be animated, flashy or colored in some way to try and draw my attention to it, then maybe. However, this would go against the whole point of advertising, where you want people to notice your ad.
I'm assuming you have never been in the military, and and have no real idea how orders are given and how they work.
First, there are certain orders called general orders and their variant, standing orders, which are passed down from someone in a command position. Someone below them in the chain-of-command can not change or violate those orders. For example, if a general has ordered that smoking is not allowed on a base, a major cannot tell me that it is OK to do so, even if they are above me in my chain-of-command. Also consider the don't ask, don't tell policy - even if someone asked and ordered you to tell them, you wouldn't have to tell.
A second aspect is an order has to be legal and the definition is sort of vague as a lawful order is an order that someone of reasonable intelligence would realize to not be illegal. Yes, it leaves some room for interpretation, but you can't have your military start a debate before deciding to do something.
Now, the use of deadly force is authorized only under a few conditions. I don't remember them all, but some of them are: To protect yourself from death or serious bodily harm, to protect others from death or serious bodily harm, to prevent the loss of property that is inherently dangerous to others (someone stealing explosives or something), and to prevent the escape of a known felon. So, if someone felt these didn't apply, then the order cannot be followed, since the order for the application of deadly force comes from the President, and only he can change it. Can you kill 500 people to save 1000? What about 501?
To further complicate things, is the Posse Comitatus Act of 1878 which states (among other things): The Secretary of Defense shall prescribe such regulations as may be necessary to ensure that any activity (including the provision of any equipment or facility or the assignment or detail of any personnel) under this chapter does not include or permit direct participation by a member of the Army, Navy, Air Force, or Marine Corps in a search, seizure, arrest, or other similar activity unless participation in such activity by such member is otherwise authorized by law. Info can be found here.
Now, at that time, terrorists were not considered a "special case" of attacker. If there were just criminal hijackers, then that would be a problem for the civil authorities, and not the military. Deciding to crash the plans into buldings just makes them mass murderers, (as we viewed these things at that point in time).
Now, it is also possible for certain positions within the military to be outside the normal chain-of-command. Ask yourself how a General or Admiral in a NATO post has to act. Can a US politician order them to give an order to NATO forces? Probably not, but that is a weird legal area.
So, to summarize... The SecDef has the ability to give orders and gives some every day - Google for SecDef orders or go to www.navy.mil and search for SecDef orders. However, there are laws and standing/general orders with Presidential authority that cannot be "ordered around". Also, there is the possibility that the military person has a position that puts them outside the normal chain-of-command for the military. So whoever reported the SecDef cannot give orders is wrong, but there are also people in certain positions who do not have the SecDef in their chain-of-command.
There are a lot of posts with incorrect or misleading information. Now IANAQCE, (I Am Not A Quantum Computing Expert) but I have taken a few graduate courses in Cryptography, and here is my understanding...
First, if there are problems that can be solved with a single algorithm from a standard computer, then it is practical for a quantum computer to do this. For example, factoring is a problem that can be handled this way. I also believe that the discrete log problem is solvable. However, if it is a problem that would require a "pre-solution" to be generated and then put into another algorithm as the seed it would not directly benefit from a quantum computer.
Second, if you have a 64 -qbit computer and a 128-bit key to crack, this isn't 2 cycles. You would like calculate the first 64 bits, then the second 64 bits but it doesn't work this way. If it did, then a 2-qbit computer would crack a 128-bit key in 64 cycles. It means that if you have an equal or greater number of qbits in your computer than in the key, you can crack it in a single operation. If you double the key length, you have to double the number of qbits in your computer to again crack it in 1 cycle. With regular computing if you add another bit, then it is twice as hard. With a quantum compter, it means you need to build a quantum computer with more qbits. So double the key bits means a quantum computer with double the qbits, not 2^n more qbits (where n is the original number of bits in the key).
Finally, quantum cryptography and quantum computing are not really related at all, other than using the word quantum. Quantum cryptography is about using the quantum properties of photons to exchange a key that can be used as a one-time pad. While it is possible to have intermediate systems to relay the key from source to destination, these must be trusted since they will view the key itself. Also, since the transmission requires the photon sender and receiver to be oriented in the same direction (+ or x), every intermediate system means you will lose half of your valid bits on average. So 4 intermediate systems means you only have 1/(2^4) or 1/16 of the original bits are valid at the receiver. Also, there is a small loss due to confirming those bits were not tampered with in transit. Plus there needs to be a separate channel so the receiver and sender can determine the index number of the original bits that were successfully transmitted.
An aside, cracking an algorithm doesn't mean it is suddenly useless. It may mean that instead of a complexity 2^128 it is now 2^123. It is "cracked" but is meaningless from a practical standpoint.
As an end note, what is scary isn't that we will have to change the encryption methods used in the future, but that all messages already sent and copied by someone can then be read easily. The government just needs to get copies of all the public keys used, save every message sent with those keys and wait for the quantum computer to catch up, then break the key and read the messages. Scary, isn't it? How many PGP or GPG messages could be stored on a 1TB disk? Let alone the storage that the NSA could put together.
Will future generations look back and wonder how we let it happen? Will we have to lie about how old we are so they don't blame us for letting it happen?
The formula for Pepsi is not a trade secret. The company went bankrupt early in the 20th century and their formula was revealed as a consequence. No idea where to find it though.
Why are you assuming 4 years of historical data? One year should be enough, and if you switched it over at the start of the tax year you wouldn't need any historical data. You would start from scratch. Assume 1k of data for each person to start since you need info on their current wage, deductions, etc.
Really, payroll is pretty simple for most of the cases, and can be done by hand in just a few minutes. (Not that this is preferred.) Since there is only 1 set of federal rules and 50 sets of state rules, even at 10MB each this is less than 1GB. Let's double it to 2GB, store it in RAM, and we could still have 6GB for processing.
Moving on from this, each company that has specific data for health care and other deductions would have to be loaded. Once this is done, let's assume it takes 1000 operations to calculate an individual's payroll, using the GPs assumption that the code has been optimized. We then have 200 million * 1000 = 200 billion. This ignores the fact that company data has to be loaded and data has to be properly outputed in some bank/printer readable format, but really it isn't too bad.
While having a single PC perform the work isn't practical, I believe the GP's point was that things are handled so inefficiently that a large company need several machines to calculate payroll because they are not efficiently using their CPU cycles.
Right, because there is no way you could build muscle mass without gravity...
I don't mean to be a troll, but don't overlook the obvious like the "power bands" that are widely used in aerobic classes. They are small and come in varying tensions. Also, if they want to arrive in excellent shape, they should get a BowFlex. It doesn't require gravity to produce tension and they could bolt it to the wall/floor and have it fold up. I know some will say this is a waste of mass/space in the capsule, but this is a serious medical issue for extended space travel.
Also, let's not forget that mass is still a constant, even if weight isn't. Moving a heavy weight quickly, and then slowing it down again before it sails out of your grasp and through the front window, would also give a workout.
Yes. But if there are multiple robots it might not matter if you take out one of them. Besides, people take a longer time to replace than equipment - especially if you only shoot higher ranking people. Plus morale takes a bigger hit when your friends die as compared to your equipment breaking.
I'm not saying you couldn't shoot the robot, just that it wouldn't usually be the best target.
Slashdot Mirror
Well, by that logic, everything from Greece to India should be part of Macedonia. Alexander conquered it all ~2300 year ago, which is waaaay before China ever got to Tibet.
Also, we shouldn't give those territories back to Mexico, they should go back to Spain.
We can't use history as a guide for where borders should be created today. It seems like we should be able to, but it doesn't work. Borders have changed over the centuries, and entire countries and societies have come and gone, so there is no way to determine what belongs to whom. Sure it would be nice if everyone could have their "ancestral lands" back, but that brings up the questions: Which ancestors? When? What if they were sold and there is no record? What if they were just absorbed, without conflict, but without consent?
No, your historical idea just doesn't work - we live in today, not 700 years ago, or even 100 years ago.
Something to keep in mind is that each "generation" can be as short as the single division of a cell. In addition, don't underestimate the power of individual cells/organisms/creatures/animals to act in a parallel manner. There are insects (termites) that lay 7000 eggs a day, providing 7000 opportunities for some sort of evolutionary event. Looking at the aphid, in a single season, if all offspring lived to reproduce, there would be 1,560,000,000,000,000,000,000,000 (1.56x10^24) offspring. Reference here.
Multiply that by the total number of insects (including different species) and this is a lot of opportunities for mutations and evolutionary activity. Also, keep in mind things like viruses and bacteria that reproduce in even greater numbers.
I don't claim that this makes evolution any less amazing, but a billion+ years and all the parallelization possible from all types of life makes for an incredible amount of computation. Enough, perhaps to even determine the question to life, the universe and everything.
How about this one: A hospital does not have to be a single building or even a single cluster of buildings. There may be numerous branch clinics, rehab centers, affiliated offices, etc. that make up the hospital. If a hospital grows, but real estate isn't available next to the existing facilities, they may add the buildings a block or two over. In Chicago, the Northwestern University Hospital network has clinics, rehab centers and therapy centers all over the place - think different suburbs. So, if they can't connect to the Internet, they should have to build their own private network to connect their different facilities, and isn't economically feasible.
Oh, wait, you mentioned they could maintain separate systems - one on the Internet and one internal only, and they could transfer the data that way. Unfortunately, that isn't a feasible solution - all locations would be required to maintain their own separate copies of the records, so an Internet outage wouldn't prevent records from being obtained. This would be a logistical and administrative nightmare.
In addition, patient records also have to be accessed by insurance companies, ambulance companies and other health care providers. Without automated access this would require dozens of "CD transfers" from secure systems to another. A nurse administers medication and it needs to be sent to billing, the insurance company and added to their medical record. This needs to happen automatically, unless you feel there should be hundreds of employees burning CDs and moving them to a different machine.
Regarding the "known Internet threat", there is also a known threat with not being able to access a medical record in a timely fashion. This is a single high-profile case of the dangers of connecting to the Internet, but there are many cases of the dangers of not having timely access to medical records. These dangers are why there are medic alert bracelets.
Finally, I don't know who your insurance company is, but mine will cover me if I leave my door unlocked or if I have stuff that's out in my yard (picnic table, deck furniture, grill, etc.). In fact, where I used to live in Canada, there were cars stolen every week when people started them to warm them up and went back inside to keep warm. They were all covered by insurance, and that is unlocked, running and unattended. I think your final paragraph is just FUD designed to make it seem like it is the hospitals fault when it isn't.
The simple fact is, if hospitals don't connect to the Internet to share information, costs will go up since there will need to be a person transfering information manually, and they need to be paid. Also, there will need to be twice the number of systems maintained, which means more capital and maintenance costs. It will also mean more deaths, since some situations require medical information -NOW-, and not after someone has burned a CD and physically moved it.
This is why we invented things like encryption, firewalls and IDS systems. They might not be perfect, and can't protect against everything, but they can reduce the risk to an acceptable level. Perfect security doesn't exist.
We are using Xen in an educational environment - providing a network of VMs to each student so they can play around, break stuff, and get some practical experience.
For your iptables problem... I think you just need to add a rule to forward packets so they will be transfered through the software bridge. Make sure to use -I to insert the rule, appending the rule doesn't help since it will be dropped before it reaches the forward rule.
I agree with you that documentation isn't all that great right now, but our lab group and several other groups that I know of are working hard at fixing that issue. Virtualization is being used in many new and interesting ways, and the tools and documentation haven't caught up yet. Give it a few more months and things should be much improved. If nothing else, my lab group wants to have our tools and documentation released under the GPL by early May, and we are really focusing on having clear, detailed and correct documentation.
How much does it cost for a wired phone in your home/apartment/office? Call it about $20/month. (I know it is more in many places, but stay with me...)
Now, a broadband connection, depending on overall throughput, may cost $100/month.
Still with me?
OK - How many voice conversations can be supported over that broadband connection? More than 5? More than 10? More than 100? Anything over 5 means the provider is losing money. If someone cancels their wired phone and uses a broadband connection, the provider loses that revenue. Now, a single customer might not be too bad, the company may be providing the high-speed connection. However, it is still less revenue than there would have been.
Now, we throw in a company that sets up in your neighborhood and offers you VoIP services over their broadband connection, so you don't need to keep your wired phone. If 10 people in the neighborhood do that, then the telecomm is 'losing' money. They have less revenue than before, and less profits, they have 1 new customer paying $100 for the broadband connection, but have lost 10 people paying $20, for a "loss" of $100.
Take it to the extreme, and have someone like Google provide phone service to anyone with an Internet connection. Imagine every person in the US cancels their regular phone service to use Google's service. The telecomms go bankrupt, or they have to increase the price of broadband by orders of magnitude. Yes, the company's may be bloated. Yes, tax dollars may have paid for the telecomm to run fiber. However, this was done so services could be fairly offered to everyone. Could my town of 800 have afforded to run fiber 30 miles to the nearest city? No. However, they can pay enough for service that it is profitable to maintain and manage that connection. It is the same with roads. Some roads use federal or state dollars to get paved. They may only connect 30 or 40 people, but that's the way things are. If we didn't do it this way, there would be extensive roads around cities, connected only by the Interstates. OK, I'm kind of off-topic now...
Anyway, the current market prices are because the revenue stream assumes that there will be wired home users paying more than their bandwidth is worth, as compared to a broadband Internet connection. If they lose those customers, it means that the cost for Internet bandwidth will rise - dramatically. So, they would rather have a company providing those services pay more, rather than having the cost pass on to all of their users.
To respond to each of your points...
/. is a perfect example of the propoganda working. So many folks seriously considering where a moon landing should occur, keeps the grassroots talk happening. Reality is, talk is cheap, and if the hardware is not being designed and built at this stage of the game, there is no program that needs to survive the change of administration coming in a relatively short timeframe (next election). That's when reality will start to h
Hmm, I'm curious, where does this technology actually exist (other than on paper and figments of engineer imaginations). The last time I checked 'we' dont have the technology to get humans reliably to/from low orbit, never mind anywhere near the moon.
True, this technology doesn't actually 'exist'. Just like the 2010 Fords and Chevys - they haven't been built yet. However, there is a lot of technology that -IS- down on paper. In fact, there are the original plans for the Saturn V rockets. It has come up before that "We can't build them anymore, we don't know how." or "We don't have the machines to build the parts anymore." OK, we don't. Ten years ago we couldn't build a 45nm chip even in a lab environment. Now, Intel will mass produce them in the next year. It isn't that the technology doesn't exist - it is that the technology hasn't been applied, and that is an entirely different matter. Techology isn't the barrier, the desire to apply that technology isn't there on a large enough scale to fund it.
So many folks seem to think that just because it was done in the 60's, it's easy, trivial, and a no brainer to go land on the moon. It's hard, expensive, and currently the technology to do it doesn't exist except on paper and in sci-fi literature. The closest thing the usa has to a manned lunar capabable piece of hardware is some rusty old Saturn V hulks sitting outside of some museums that relish 'the good old days' when america was actually a leader in the space race.
Hard? Yes, but we also have thousands of talented engineers - aerospace, mechanical and chemical, who are more than capable of doing this. (Unless you claim no one on Earth is as smart or talented as the engineers we had in the 60s.) Expensive? Yes. I'm not claiming it isn't. But it isn't more expensive than supporting a multi-year war/peace-keeping mission in Iraq. That is where the government has decided to spend a few hundred billion. In theory, (I said theory), we could leave Iraq and direct that money to NASA - the money -IS- there, it is just being spent on other things. As I mentioned above - the technology exists, we just aren't applying it to actually create the hardware.
In the 60's the landing was equatorial for a lot of technical reasons. Today, moon landings are sci-fi, for a lot more technical (combined with political and financial) reasons. Politicians may talk about going to the moon, but follow the money, it's not going to the moon, it's going to wars overseas. The talk of moon landings is nothing more than political rhetoric designed to gather up votes from folks that cant see the forest for all the trees, and actually believe that such things are in the plans of the administration. If it was actually in the administration plans, the mandate would be such as it was in the 60's, to get far enough into the program that it could NOT be cancelled at the expiry of the 8 year term, to much already invested. In reality, this administration is neatly talking the talk that allows for more talk, but not actually allocating funds to make it happen, then putting on schedules such that all the talk becomes a financial responsibility for a future administration. In laymans terms, that means, not gonna happen.
As I mentioned above - it isn't the technical reasons, it is the political reasons - and since the government has the power to set the budget, it is also a financial thing. I agree with you that in the current political structure it's not going to happen.
This article on
While presented in a comedic way, it shows someone convincing a town that they have trouble because the billiard parlor in town has brought in a pool table. While billiards is OK, pool is a horrible game and will cause degradation in the children and cause them to stop doing their chores and become gamblers.
Just goes to show that new things are often looked upon as corruptive or causing some sort of lamentable behavior. I remember when Pac-Man came out and people objected since it was dangerous medically and that playing it was like running up 2-3 flights of stairs. Now it is considered harmless fun.
As mentioned previously, this sort of thing is being/has been done. One project I am familiar with is the Internet-scale Event and Attack Generation Environment (ISEAGE) project at Iowa State University.
Its webpage, has an overview of the project and documentation on its architecture and implementation. I think one of the key aspects of the project can be found in the overview: "Unlike computer-based simulations, real attacks will be played out against real equipment."
ISEAGE is approaching security from a real-world perspective, using real world devices. Sure, your software/hardware might be secure when the attacks are played against it; but is it secure when those attacks when there are dozens of attacks occuring simultaneously? What about when it is being hit by thousands of requests, or is under a DDoS attack? What happens when devices decide to start breaking the protocols, or the rules? What happens if a device physically fails? What is the effect of a device overheating during a DDoS attack? How do you simulate this/test for this other than hooking it up and hammering it with a DDoS attack?
This is the kind of information that is needed to prevent or mitigate an attack, but can't be found by reading code or running a scanner. How did the US figure out how to build rockets? We built some, they blew up, and better ones got built. The real world isn't the same as a lab.
Do you have any idea how many old viruses are still circulating?
I use the network at a major university and I get notified of newtear attacks at least once an hour. newtear targeted Windows 95 and NT. It has been around a LONG time, and still trying to infect new machines.
Take some time and Google for some of the big worms like Slammer and Code Red, and you will find people are still detecting them - thousands of times a day.
Regarding point 3, I have my AV set to scan every file I open, so clicking on a BrittenySpearsNaked.jpg.exe will cause it to be scanned and prevent the infection. You might want to look for this setting on your AV. (Scan on open or some such.)
Finally, AV is not the end-all, be-all of security; you should also have a firewall to prevent unwanted incoming and outgoing communications. AV is part of an overall security plan.
Bad behavior/design on the part of your anti-virus software doesn't mean you should throw up your hands and give up.
My concern would be that anything that could affect a portion of the disk would destroy more data. I know scratches that aren't noticed on a CD can make a DVD unreadable and, while a drive platter may not have the risk of scratches that optical storage does, the general idea is the same. A physical failure, such as a head alignment issue, that wouldn't be noticed with lower densities may be a factor with the higher densities.
Now, I don't have a solution to the problem, but I just want to point out that getting full performance out of something can raise new risks.
I don't know about anyone else, but when I type, I usually make a few mistakes. However, when I run a script it always runs the commands correctly. Even if it isn't a script, there is probably some attack case they are cutting and pasting the commands out of. It would be helpful to see how quickly individual characters are being typed, so it could be determined if they are typing the characters or not.
Also, having been in the military, there is no reason for them to train typists who never hit the wrong key.
That day is going to come in the second half of next year with the release of Windows Vista.
See here for details.
You could try this:
Boss, I have a problem...
I got a call from the FBI that they wanted to meet with me regarding the use of the Kazaa program within our network. They think someone is downloading illegal music and kiddy porn. I was thinking we should block those programs to avoid liability, what do you think? Also, should we make it into a written policy just to be safe? I don't want to spend the next 2 weeks dealing with lawyers and law enforcement agents crawling through our networks.
You don't fight the fights you can win; you fight the fights that need fighting.
Thank you EFF for standing up for my rights.
Short answer: No.
Longer answer: No, I already have enough stuff that I need to focus on. If I needed to be distracted I would take off my headphones or check email every 5 minutes. If you could guarantee that the ads wouldn't be animated, flashy or colored in some way to try and draw my attention to it, then maybe. However, this would go against the whole point of advertising, where you want people to notice your ad.
Just my $0.02.
'nuff said.
I'm assuming you have never been in the military, and and have no real idea how orders are given and how they work.
First, there are certain orders called general orders and their variant, standing orders, which are passed down from someone in a command position. Someone below them in the chain-of-command can not change or violate those orders. For example, if a general has ordered that smoking is not allowed on a base, a major cannot tell me that it is OK to do so, even if they are above me in my chain-of-command. Also consider the don't ask, don't tell policy - even if someone asked and ordered you to tell them, you wouldn't have to tell.
A second aspect is an order has to be legal and the definition is sort of vague as a lawful order is an order that someone of reasonable intelligence would realize to not be illegal. Yes, it leaves some room for interpretation, but you can't have your military start a debate before deciding to do something.
Now, the use of deadly force is authorized only under a few conditions. I don't remember them all, but some of them are: To protect yourself from death or serious bodily harm, to protect others from death or serious bodily harm, to prevent the loss of property that is inherently dangerous to others (someone stealing explosives or something), and to prevent the escape of a known felon. So, if someone felt these didn't apply, then the order cannot be followed, since the order for the application of deadly force comes from the President, and only he can change it. Can you kill 500 people to save 1000? What about 501?
To further complicate things, is the Posse Comitatus Act of 1878 which states (among other things): The Secretary of Defense shall prescribe such regulations as may be necessary to ensure that any activity (including the provision of any equipment or facility or the assignment or detail of any personnel) under this chapter does not include or permit direct participation by a member of the Army, Navy, Air Force, or Marine Corps in a search, seizure, arrest, or other similar activity unless participation in such activity by such member is otherwise authorized by law. Info can be found here.
Now, at that time, terrorists were not considered a "special case" of attacker. If there were just criminal hijackers, then that would be a problem for the civil authorities, and not the military. Deciding to crash the plans into buldings just makes them mass murderers, (as we viewed these things at that point in time).
Now, it is also possible for certain positions within the military to be outside the normal chain-of-command. Ask yourself how a General or Admiral in a NATO post has to act. Can a US politician order them to give an order to NATO forces? Probably not, but that is a weird legal area.
So, to summarize... The SecDef has the ability to give orders and gives some every day - Google for SecDef orders or go to www.navy.mil and search for SecDef orders. However, there are laws and standing/general orders with Presidential authority that cannot be "ordered around". Also, there is the possibility that the military person has a position that puts them outside the normal chain-of-command for the military. So whoever reported the SecDef cannot give orders is wrong, but there are also people in certain positions who do not have the SecDef in their chain-of-command.
There are a lot of posts with incorrect or misleading information. Now IANAQCE, (I Am Not A Quantum Computing Expert) but I have taken a few graduate courses in Cryptography, and here is my understanding...
/(2^4) or 1/16 of the original bits are valid at the receiver. Also, there is a small loss due to confirming those bits were not tampered with in transit. Plus there needs to be a separate channel so the receiver and sender can determine the index number of the original bits that were successfully transmitted.
First, if there are problems that can be solved with a single algorithm from a standard computer, then it is practical for a quantum computer to do this. For example, factoring is a problem that can be handled this way. I also believe that the discrete log problem is solvable. However, if it is a problem that would require a "pre-solution" to be generated and then put into another algorithm as the seed it would not directly benefit from a quantum computer.
Second, if you have a 64 -qbit computer and a 128-bit key to crack, this isn't 2 cycles. You would like calculate the first 64 bits, then the second 64 bits but it doesn't work this way. If it did, then a 2-qbit computer would crack a 128-bit key in 64 cycles. It means that if you have an equal or greater number of qbits in your computer than in the key, you can crack it in a single operation. If you double the key length, you have to double the number of qbits in your computer to again crack it in 1 cycle. With regular computing if you add another bit, then it is twice as hard. With a quantum compter, it means you need to build a quantum computer with more qbits. So double the key bits means a quantum computer with double the qbits, not 2^n more qbits (where n is the original number of bits in the key).
Finally, quantum cryptography and quantum computing are not really related at all, other than using the word quantum. Quantum cryptography is about using the quantum properties of photons to exchange a key that can be used as a one-time pad. While it is possible to have intermediate systems to relay the key from source to destination, these must be trusted since they will view the key itself. Also, since the transmission requires the photon sender and receiver to be oriented in the same direction (+ or x), every intermediate system means you will lose half of your valid bits on average. So 4 intermediate systems means you only have 1
An aside, cracking an algorithm doesn't mean it is suddenly useless. It may mean that instead of a complexity 2^128 it is now 2^123. It is "cracked" but is meaningless from a practical standpoint.
As an end note, what is scary isn't that we will have to change the encryption methods used in the future, but that all messages already sent and copied by someone can then be read easily. The government just needs to get copies of all the public keys used, save every message sent with those keys and wait for the quantum computer to catch up, then break the key and read the messages. Scary, isn't it? How many PGP or GPG messages could be stored on a 1TB disk? Let alone the storage that the NSA could put together.
Hope that was helpful.
Unless it is a duck.
Or does that mean all ducks are witches? *Looks suspiciously at ducks on nearby pond.*
To handle all scenarios my foil headwear are two-ply. One has the shiny side out, the other has the shiny side in.
Will future generations look back and wonder how we let it happen? Will we have to lie about how old we are so they don't blame us for letting it happen?
The formula for Pepsi is not a trade secret. The company went bankrupt early in the 20th century and their formula was revealed as a consequence. No idea where to find it though.
Why are you assuming 4 years of historical data? One year should be enough, and if you switched it over at the start of the tax year you wouldn't need any historical data. You would start from scratch. Assume 1k of data for each person to start since you need info on their current wage, deductions, etc.
Really, payroll is pretty simple for most of the cases, and can be done by hand in just a few minutes. (Not that this is preferred.) Since there is only 1 set of federal rules and 50 sets of state rules, even at 10MB each this is less than 1GB. Let's double it to 2GB, store it in RAM, and we could still have 6GB for processing.
Moving on from this, each company that has specific data for health care and other deductions would have to be loaded. Once this is done, let's assume it takes 1000 operations to calculate an individual's payroll, using the GPs assumption that the code has been optimized. We then have 200 million * 1000 = 200 billion. This ignores the fact that company data has to be loaded and data has to be properly outputed in some bank/printer readable format, but really it isn't too bad.
While having a single PC perform the work isn't practical, I believe the GP's point was that things are handled so inefficiently that a large company need several machines to calculate payroll because they are not efficiently using their CPU cycles.
Right, because there is no way you could build muscle mass without gravity...
I don't mean to be a troll, but don't overlook the obvious like the "power bands" that are widely used in aerobic classes. They are small and come in varying tensions. Also, if they want to arrive in excellent shape, they should get a BowFlex. It doesn't require gravity to produce tension and they could bolt it to the wall/floor and have it fold up. I know some will say this is a waste of mass/space in the capsule, but this is a serious medical issue for extended space travel.
Also, let's not forget that mass is still a constant, even if weight isn't. Moving a heavy weight quickly, and then slowing it down again before it sails out of your grasp and through the front window, would also give a workout.
Yes. But if there are multiple robots it might not matter if you take out one of them. Besides, people take a longer time to replace than equipment - especially if you only shoot higher ranking people. Plus morale takes a bigger hit when your friends die as compared to your equipment breaking.
I'm not saying you couldn't shoot the robot, just that it wouldn't usually be the best target.