It's easy, just generate a random sequence and salt the plaintext with it before encrypting. The RNG would be easy to make because the card access is a manual process, with no scan being exactly like any other.
Not all RFID cards are read-only. EM4095 allows you to transmit into the token, and the token is free to do whatever it wants with the data.
There are many RFID readers on the market that are encrypted to some extent. See Indala.
The tag may only need to present its ID, then a link to an external database delivers the rest - your photo, your fingerprints, your other IDs, etc.
there is, however, the right to defend yourself from bears and murderous neighbors.
Most countries don't offer such a right to their citizens. If attacked, one is supposed to call the police and then wait patiently until they arrive and collect his (by then) cold body.
Size and weight. I don't want a large, flat, bulky device on my belt or ($deity forbid) in my pocket. My current phone is a flip phone, just a few inches long.
Cost of the hardware. A phone is a portable device, to be carried everywhere and to be scratched, dropped and otherwise abused. I want it to be reasonably cheap for that. I don't want to carry a jewel and protect it with my life.
Cost of the plan. Smartphone plans are more expensive, at no increase in my personal productivity or enjoyment of life. Right now cell companies are aggressively milking the market and the poor saps pay through the nose.
Quality of calls. That varies, but generally smartphones don't do better than a dumb phone.
Quality of the interface. My flip phone closes and protects the keypad from any accidental operation. Its firmware is simple enough to not contain bugs that would pocket-dial a number. Also I detest the idea of touching the screen.
Battery life. Dumb phones are certainly better in this aspect just because they have fewer hardware to run.
Screen size. The screen is too small for me to use, and for some reason I don't want to lose eyesight.
Data security. If I lose my phone there is nothing of value on it - no banking data, no personal emails.
I'm sure there are more, but this list is fairly representative already.
Yes, I had a few Panasonic phones, they are nice. Another benefit is that they offer better signal quality within their range, whereas a cell phone offers variable quality in a far greater range. When you are at home you probably don't care how well the phone might work in the next county.
Yet another benefit is that cordless phones use very low power (because the base is just a hundred feet away) and as result they don't cook your brains as much as a 1W cell phone does:-)
In other words, if you are a home worker you need to have a good phone. Polycom, of course, is one of the best, but Panasonic speakerphones are probably on the next notch.
With regard to the talk time, cordless phones are low-tech devices, and their batteries are cheap, sold at Fry's and replaceable under 30 seconds. Try to swap a battery in an iPhone...
Phones are out because I'm not really related to any specific phone. I may have my cell phone with me, or I may be at home, or I may be at the office, etc. And as I mentioned elsewhere, I don't trust Google with the price on last year's snow, let alone my phone number. I don't use SMS at all and it is disabled on my phone.
Authentication software on smartphones is also out because I don't have a smartphone and have no desire to ever get one. They are too bulky for my taste, and the mandatory (on AT&T) data plan adds nothing to my life - I'm near computers all the time already (except when I'm driving, and then I don't need a computer anyway.)
The Yubikey is a good idea, but it suffers from the need to have a USB port. On the other hand it does the input for you. Still I'd put a classical RSA token above Yubikey because it is a self-contained device that requires no hardware access to operate. This may be important if you are at a kiosk or at an Internet cafe where USB ports are disabled or inaccessible. If Gmail starts supporting RSA tokens then I will gladly buy one myself. If they support a mix of tokens then I'd be happy with Yubikey at home and an RSA token in my pocket.
The printed sheet of numbers is a hassle to carry, but in a pinch it will do.
There's a difference between saying, "Your performance is subpar. You really need to pick up the slack" and "You're a rat-like, miserable little shit".
That's the difference between two students.
When I was in school one of the guys knifed another one during the class, just for fun. Do you think you can say that his classroom performance was just subpar?
Because you can set the account settings to call you via a backup number (pehaps a landline; I hope you had landlines there in that little Nebraskan town) and have the code read out to you.
It's tough to do if you don't know your phone number ahead of time - and you need to. If you stay in a hotel they often don't have direct dial numbers, so no Gmail for you.
With regard to printed tables, that would be cumbersome but acceptable. Unfortunately that would drop Gmail's "ease of use" rating from 9+ to 0 because many people would be unable to log in. Some people don't even speak English, how are they going to listen to the codes - in what language? Will it be configurable? Will it support their language?
Google is getting this flak not because the need is not there, but because the implementation sucks. Besides, there is a ever-growing number of people who are annoyed with cell phonery, and when Google starts pushing those gadgets into every crevice those people are not happy. On top of that the summary ominously mentions that one day this optional feature may become mandatory. Loss of control is yet another thing that people hate. Looks like Google, with their billions of dollars, doesn't have a competent PR person (a psychologist) who can go through the harebrained output of engineers and prevent its publication before the damage is done.
Then engage your brain and consider the audience that uses Gmail - you don't think that virtually all of them have cell phones?
If you imply that most Gmail users are geeks, that's not so. Gmail is a simple email, and as result it is attractive to many people. Geeks, in fact, would be not so interested in Gmail because it is not very secure.
I do have a Gmail account, but I have a bunch of other email accounts, on real, paid-for servers, and I carefully choose which email to use for what.
I have a cell phone, but it doesn't receive SMS. I guess they could call the number, but I won't give it to Google anyway. If they make this system mandatory they'd better send me an RSA token, otherwise I'm done with them. The cost of their free service would be too high for me.
It's Google's decision, of course, to bet on teenagers who are married to their phones. Looking at vast history of stupid business decisions at Google, I'm sure they will do just that.
NAT on that level will cause so many customer complaints and support issues, that the problems and support issues introduced by phasing in IPv6 will seem like a piss in the Nile.
100% of residential customers are already behind one layer of NAT. Another layer won't change much. And if the customer wants a certain port forwarded through ISP's NAT into his own router (which then he can configure as he does today) then the ISP will set up an online service page, where for a measly sum of $1/mo you can buy such a service. If necessary they will move you into another NAT pool. This is trivial technically, and is a money maker financially.
On the other hand, IPv6 transition requires the ISP to replace most of their hardware and software, and if they have routers in people's homes then those also have to be scrapped and replaced. Then a tech has to be sent to each customer's home to analyze their network and give upgrade recommendations. I still have two XP boxes at home, and I'm typing this into an XP box at work - those beasts don't do IPv6 properly, and they are already out of maintenance by MS. That's not millions, that's billions of dollars. Compare to writing a custom NAT software... a work for one geek for a month.
I'm presenting this from the Devil's Advocate position because that's the position that the ISP's business people are in. They won't do IPv6 for the love of it. They will do whatever brings the most profit. And at this time it is most profitable to milk the IPv4 cow because scarcity of resources is where fortunes are made.
As websites migrate to IPv6 (without an IPv4 version), what IP address should the end user be directed to?
In the transition phase (which may take years) all reputable Web servers (businesses) will be running on hosted IPv4 [virtual] servers, pretty much as they are doing right now. Only cheap home users will be running IPv6 servers at home, and though they are welcome to do so, their audience will be very much limited. Many clients will never be upgraded to IPv6 just because they can't do it (I don't think even WinXP can do it properly; its IPv6 stack is half-baked.) There is a terrible amount of labor and skill and money that is required for a transition. I have 3 routers and 2 802.11a/n APs, none of them support IPv6. I personally can read enough HOWTOs to build an IPv6 system here (I have two boxes on IPv6 LAN connection already) but majority of non-geeks will be at complete loss, like a newbie looking at data binding examples in XAML.
then they are not being truthful in the reporting of their revenues to investors.
that means the revenues associated with the value of feature set B were reported to investors before they were actually produced and delivered.
I think they are in the clear as long as they are not contractually obligated to deliver the feature set B. This is very common in today's smartphones - just see how every OEM drags their feet with Android updates. In other words, they are good as long as they don't have to write in their books "Future liability - must spend money to design feature B."
hope none of their investors sues over it.
Well, if they sell "an infinitely upgradable router with this roadmap of future updates" in 2010, and expect to work non-stop on nothing but free, new features for it until year 2020 then I'm sure investors would have something to say here. But as long as the updates are done as part of a regular support cycle it won't be a problem. In most cases new features are just backported from other products, not written from scratch. There is also such a thing as "sunk cost" - if you have 10 firmware guys working support, you don't care too much *what* code they are producing, as long as the wolves are safe and the sheep aren't hungry.
The only catastrophe that is likely to strike is that all customers of a major ISP will find themselves one day on a 10.0.0.0/16 subnet, behind the NAT at the ISP office. And most of them will not even notice until their Skype connections start failing. Then it will be Skype's problem.
Of course a/16 NAT would be a bit ambitious, considering the number of available ports at the NAT box, but a/24 is perfectly doable even if all 254 client IPs are in heavy use. That adds another 8 bits to the IPv4 space, and that ought to be good for about 254 more Earths at this moment.
And if for some reason you don't like the NAT then ISPs will gladly sell you, for an arm and a leg, a static IPv4 address. Businesses will be OK with that, and residential customers wouldn't care. ISP's NAT will be Skype-aware, and that covers pretty much all the needs of a typical consumer.
If you have signed an NDA and have the unfortunate set of circumstances put upon you, do you have any option that would not land up with you breaking at a minimum a contract, and at worst the law?
An NDA can't ask you to break the law. So if you are arrested it's because you took the NDA too far. NDA typically tells you to take "reasonable measures" to protect the information; it doesn't mean that you must defend it with your life or your freedom.
There's still an expectation that the software is fit for purpose.
But it's not you who defines the purpose, it's the manufacturer.
Imagine that someone took a MS flight simulator software, connected it to a real airplane somehow, switched the autopilot on, and the airplane fell to the ground. Can he sue Microsoft?
If the manufacturer tells you that the software is not fit for certain uses, you'd better believe it. Software for autopilots may come with formal proofs of correctness.
If I were the robot manufacturer I would say in the EULA that the control interface needs to be protected with an adequate firewall, and it's the responsibility of the user - we make robots, not firewalls. If a hacker went through the firewall then it's your fault - we didn't even sell you one.
At least radio has channels. IR links have only one channel, unless you want to use WDM:-) IR modems will be interfering with each other, so you still need some form of collision detection and avoidance, which further reduces the throughput. At least with microwave radios you have bandwidth to waste...
But of course if you want to do a good job then, as many people already said, you just lay some Ethernet cables and install a switch. The whole thread is about the case when some restless souls decided to play with old technologies that are suddenly new.
Install one or a couple of 802.11 access point; requires one power adapter and one (or zero, if WDS is used) Ethernet cable per AP. Everyone in the office can then access the network using 802.11 adapters built into most laptops and in many desktops. Get up to 54 Mbps link speed.
Rip up your ceiling and route hundreds of cables to hundreds of IR transceivers there. Buy one IR modem for each computer, connect with more wires. Test the configuration. Get a few Mbps link speed. If a wire gets damaged somewhere, rip up your ceiling again.
So which one makes more sense? Note that high speed links are necessary because you want fast access to servers that store your large files. With 802.11a/n you can get high data rates *and* constrain the covered area to your office. Each 5 GHz AP will cover radius of 10-20 meters just fine, especially if few walls are in between. WDS is also a good option at those bit rates.
Apple may be a lone knight in shining armor, but such a knight is still vulnerable to a mob attack done by dirty, illiterate barbarians with sticks. After all, Apple's technology was always pretty good, and that earned them what percentage of the market? IBM's PS/2 and OS/2 technologies were also good, but hordes of clones with Windows 3.1 casually stomped both into the ground.
One of the main advantages of IPv6 is we will be able to abolish NAT (for security, transparency, and reliability), yet you propose we adopt IPv6 and issue all Internet users with an... address translation device?
This is a popular request. What he really needs is an IPv6 firewall. Like NAT, it should auto-configure itself to allow TCP sessions that are initiated inside, and it should have manual controls to open ports for certain IPv6 hosts on the LAN. It also should have some form of a DNS server for the LAN because it will be hard to remember IPv6 addresses. DHCPv6 should be also present. In other words, it has to be a consumer box that I haven't seen anywhere yet.
Strictly speaking, there's no reason why anybody should need to run IPv4 on their home network
Except that pesky problem of about a billion IPv4-only devices - routers, printers, and all other stuff that you can't replace.
I bought a pair of 802.11a/n access points this week to do some bridging, and they work fine... in IPv4 mode. There is not a bit in them about IPv6, and these are super-new devices. There will be no hope of large scale IPv6 acceptance until all consumer grade hardware on store shelves supports it out of the box. Routers also need to do the right thing regardless of what IP version the ISP supports. The whole setup should be "hands off" - but today it requires a rocket scientist to just play on your LAN.
The USA can just promise that he won't be executed. It's not a politically viable option anyway (execution of a journalist.)
Instead he will be convicted for some lowly crime, given 10 years in prison, and sent to a place from which nobody ever comes back with intact body and mind. Ten years of torture is scarier than death.
for a huge majority, a car is absolutely necessary.
For an even larger majority money is absolutely necessary. But should we allow them to rob banks?
The social security provides small money disbursements to those who are in need. Similarly, buses exist for people who can't drive for one reason or another. You aren't defending people without licenses, who were convicted of DUI, for example? You aren't suggesting that blind people should drive? If not then why it is OK for otherwise incapable people to drive? If a guy can't find a couple hundred dollars per year for the insurance then it's guaranteed that in case of an accident you will be left holding the bag.
Let's apply this to other sectors of the economy and see how far our nation continues to survive.
It was always applied to other sectors of the economy. Only recently few people started saying that they can pick and choose which laws to obey.
Slashdot Mirror
Frankly how do you encrypt one way traffic?
there is, however, the right to defend yourself from bears and murderous neighbors.
Most countries don't offer such a right to their citizens. If attacked, one is supposed to call the police and then wait patiently until they arrive and collect his (by then) cold body.
My personal reasons for not having a smartphone:
I'm sure there are more, but this list is fairly representative already.
Yes, I had a few Panasonic phones, they are nice. Another benefit is that they offer better signal quality within their range, whereas a cell phone offers variable quality in a far greater range. When you are at home you probably don't care how well the phone might work in the next county.
Yet another benefit is that cordless phones use very low power (because the base is just a hundred feet away) and as result they don't cook your brains as much as a 1W cell phone does :-)
In other words, if you are a home worker you need to have a good phone. Polycom, of course, is one of the best, but Panasonic speakerphones are probably on the next notch.
With regard to the talk time, cordless phones are low-tech devices, and their batteries are cheap, sold at Fry's and replaceable under 30 seconds. Try to swap a battery in an iPhone...
What the heck would your "preferred" system look like?
Like this.
Phones are out because I'm not really related to any specific phone. I may have my cell phone with me, or I may be at home, or I may be at the office, etc. And as I mentioned elsewhere, I don't trust Google with the price on last year's snow, let alone my phone number. I don't use SMS at all and it is disabled on my phone.
Authentication software on smartphones is also out because I don't have a smartphone and have no desire to ever get one. They are too bulky for my taste, and the mandatory (on AT&T) data plan adds nothing to my life - I'm near computers all the time already (except when I'm driving, and then I don't need a computer anyway.)
The Yubikey is a good idea, but it suffers from the need to have a USB port. On the other hand it does the input for you. Still I'd put a classical RSA token above Yubikey because it is a self-contained device that requires no hardware access to operate. This may be important if you are at a kiosk or at an Internet cafe where USB ports are disabled or inaccessible. If Gmail starts supporting RSA tokens then I will gladly buy one myself. If they support a mix of tokens then I'd be happy with Yubikey at home and an RSA token in my pocket.
The printed sheet of numbers is a hassle to carry, but in a pinch it will do.
There's a difference between saying, "Your performance is subpar. You really need to pick up the slack" and "You're a rat-like, miserable little shit".
That's the difference between two students.
When I was in school one of the guys knifed another one during the class, just for fun. Do you think you can say that his classroom performance was just subpar?
So you'll take advice from a doctor but not the crack-whore handcuffed in the police station?
It highly depends on the nature of the advice.
Because you can set the account settings to call you via a backup number (pehaps a landline; I hope you had landlines there in that little Nebraskan town) and have the code read out to you.
It's tough to do if you don't know your phone number ahead of time - and you need to. If you stay in a hotel they often don't have direct dial numbers, so no Gmail for you.
With regard to printed tables, that would be cumbersome but acceptable. Unfortunately that would drop Gmail's "ease of use" rating from 9+ to 0 because many people would be unable to log in. Some people don't even speak English, how are they going to listen to the codes - in what language? Will it be configurable? Will it support their language?
Google is getting this flak not because the need is not there, but because the implementation sucks. Besides, there is a ever-growing number of people who are annoyed with cell phonery, and when Google starts pushing those gadgets into every crevice those people are not happy. On top of that the summary ominously mentions that one day this optional feature may become mandatory. Loss of control is yet another thing that people hate. Looks like Google, with their billions of dollars, doesn't have a competent PR person (a psychologist) who can go through the harebrained output of engineers and prevent its publication before the damage is done.
Then engage your brain and consider the audience that uses Gmail - you don't think that virtually all of them have cell phones?
If you imply that most Gmail users are geeks, that's not so. Gmail is a simple email, and as result it is attractive to many people. Geeks, in fact, would be not so interested in Gmail because it is not very secure.
I do have a Gmail account, but I have a bunch of other email accounts, on real, paid-for servers, and I carefully choose which email to use for what.
I have a cell phone, but it doesn't receive SMS. I guess they could call the number, but I won't give it to Google anyway. If they make this system mandatory they'd better send me an RSA token, otherwise I'm done with them. The cost of their free service would be too high for me.
It's Google's decision, of course, to bet on teenagers who are married to their phones. Looking at vast history of stupid business decisions at Google, I'm sure they will do just that.
NAT on that level will cause so many customer complaints and support issues, that the problems and support issues introduced by phasing in IPv6 will seem like a piss in the Nile.
100% of residential customers are already behind one layer of NAT. Another layer won't change much. And if the customer wants a certain port forwarded through ISP's NAT into his own router (which then he can configure as he does today) then the ISP will set up an online service page, where for a measly sum of $1/mo you can buy such a service. If necessary they will move you into another NAT pool. This is trivial technically, and is a money maker financially.
On the other hand, IPv6 transition requires the ISP to replace most of their hardware and software, and if they have routers in people's homes then those also have to be scrapped and replaced. Then a tech has to be sent to each customer's home to analyze their network and give upgrade recommendations. I still have two XP boxes at home, and I'm typing this into an XP box at work - those beasts don't do IPv6 properly, and they are already out of maintenance by MS. That's not millions, that's billions of dollars. Compare to writing a custom NAT software... a work for one geek for a month.
I'm presenting this from the Devil's Advocate position because that's the position that the ISP's business people are in. They won't do IPv6 for the love of it. They will do whatever brings the most profit. And at this time it is most profitable to milk the IPv4 cow because scarcity of resources is where fortunes are made.
As websites migrate to IPv6 (without an IPv4 version), what IP address should the end user be directed to?
In the transition phase (which may take years) all reputable Web servers (businesses) will be running on hosted IPv4 [virtual] servers, pretty much as they are doing right now. Only cheap home users will be running IPv6 servers at home, and though they are welcome to do so, their audience will be very much limited. Many clients will never be upgraded to IPv6 just because they can't do it (I don't think even WinXP can do it properly; its IPv6 stack is half-baked.) There is a terrible amount of labor and skill and money that is required for a transition. I have 3 routers and 2 802.11a/n APs, none of them support IPv6. I personally can read enough HOWTOs to build an IPv6 system here (I have two boxes on IPv6 LAN connection already) but majority of non-geeks will be at complete loss, like a newbie looking at data binding examples in XAML.
then they are not being truthful in the reporting of their revenues to investors.
that means the revenues associated with the value of feature set B were reported to investors before they were actually produced and delivered.
I think they are in the clear as long as they are not contractually obligated to deliver the feature set B. This is very common in today's smartphones - just see how every OEM drags their feet with Android updates. In other words, they are good as long as they don't have to write in their books "Future liability - must spend money to design feature B."
hope none of their investors sues over it.
Well, if they sell "an infinitely upgradable router with this roadmap of future updates" in 2010, and expect to work non-stop on nothing but free, new features for it until year 2020 then I'm sure investors would have something to say here. But as long as the updates are done as part of a regular support cycle it won't be a problem. In most cases new features are just backported from other products, not written from scratch. There is also such a thing as "sunk cost" - if you have 10 firmware guys working support, you don't care too much *what* code they are producing, as long as the wolves are safe and the sheep aren't hungry.
Wait until the catastrophe strikes.
The only catastrophe that is likely to strike is that all customers of a major ISP will find themselves one day on a 10.0.0.0/16 subnet, behind the NAT at the ISP office. And most of them will not even notice until their Skype connections start failing. Then it will be Skype's problem.
Of course a /16 NAT would be a bit ambitious, considering the number of available ports at the NAT box, but a /24 is perfectly doable even if all 254 client IPs are in heavy use. That adds another 8 bits to the IPv4 space, and that ought to be good for about 254 more Earths at this moment.
And if for some reason you don't like the NAT then ISPs will gladly sell you, for an arm and a leg, a static IPv4 address. Businesses will be OK with that, and residential customers wouldn't care. ISP's NAT will be Skype-aware, and that covers pretty much all the needs of a typical consumer.
If you have signed an NDA and have the unfortunate set of circumstances put upon you, do you have any option that would not land up with you breaking at a minimum a contract, and at worst the law?
An NDA can't ask you to break the law. So if you are arrested it's because you took the NDA too far. NDA typically tells you to take "reasonable measures" to protect the information; it doesn't mean that you must defend it with your life or your freedom.
There's still an expectation that the software is fit for purpose.
But it's not you who defines the purpose, it's the manufacturer.
Imagine that someone took a MS flight simulator software, connected it to a real airplane somehow, switched the autopilot on, and the airplane fell to the ground. Can he sue Microsoft?
If the manufacturer tells you that the software is not fit for certain uses, you'd better believe it. Software for autopilots may come with formal proofs of correctness.
If I were the robot manufacturer I would say in the EULA that the control interface needs to be protected with an adequate firewall, and it's the responsibility of the user - we make robots, not firewalls. If a hacker went through the firewall then it's your fault - we didn't even sell you one.
At least radio has channels. IR links have only one channel, unless you want to use WDM :-) IR modems will be interfering with each other, so you still need some form of collision detection and avoidance, which further reduces the throughput. At least with microwave radios you have bandwidth to waste...
But of course if you want to do a good job then, as many people already said, you just lay some Ethernet cables and install a switch. The whole thread is about the case when some restless souls decided to play with old technologies that are suddenly new.
Why is this a bad idea again?
So which one makes more sense? Note that high speed links are necessary because you want fast access to servers that store your large files. With 802.11a/n you can get high data rates *and* constrain the covered area to your office. Each 5 GHz AP will cover radius of 10-20 meters just fine, especially if few walls are in between. WDS is also a good option at those bit rates.
Apple may be a lone knight in shining armor, but such a knight is still vulnerable to a mob attack done by dirty, illiterate barbarians with sticks. After all, Apple's technology was always pretty good, and that earned them what percentage of the market? IBM's PS/2 and OS/2 technologies were also good, but hordes of clones with Windows 3.1 casually stomped both into the ground.
One of the main advantages of IPv6 is we will be able to abolish NAT (for security, transparency, and reliability), yet you propose we adopt IPv6 and issue all Internet users with an... address translation device?
This is a popular request. What he really needs is an IPv6 firewall. Like NAT, it should auto-configure itself to allow TCP sessions that are initiated inside, and it should have manual controls to open ports for certain IPv6 hosts on the LAN. It also should have some form of a DNS server for the LAN because it will be hard to remember IPv6 addresses. DHCPv6 should be also present. In other words, it has to be a consumer box that I haven't seen anywhere yet.
Strictly speaking, there's no reason why anybody should need to run IPv4 on their home network
Except that pesky problem of about a billion IPv4-only devices - routers, printers, and all other stuff that you can't replace.
I bought a pair of 802.11a/n access points this week to do some bridging, and they work fine ... in IPv4 mode. There is not a bit in them about IPv6, and these are super-new devices. There will be no hope of large scale IPv6 acceptance until all consumer grade hardware on store shelves supports it out of the box. Routers also need to do the right thing regardless of what IP version the ISP supports. The whole setup should be "hands off" - but today it requires a rocket scientist to just play on your LAN.
Right. And this has happened outside of fiction when?
The parent post already answered this question. Specifically, the year was 1963, the year of patsies.
The USA can just promise that he won't be executed. It's not a politically viable option anyway (execution of a journalist.)
Instead he will be convicted for some lowly crime, given 10 years in prison, and sent to a place from which nobody ever comes back with intact body and mind. Ten years of torture is scarier than death.
What about those 5%? Their wages would undoubtedly fall, because the demand for unskilled labor will always be outstripped by the supply.
You can't hold the progress back. The best you can do is prepare the society for changes before they occur.
for a huge majority, a car is absolutely necessary.
For an even larger majority money is absolutely necessary. But should we allow them to rob banks?
The social security provides small money disbursements to those who are in need. Similarly, buses exist for people who can't drive for one reason or another. You aren't defending people without licenses, who were convicted of DUI, for example? You aren't suggesting that blind people should drive? If not then why it is OK for otherwise incapable people to drive? If a guy can't find a couple hundred dollars per year for the insurance then it's guaranteed that in case of an accident you will be left holding the bag.
Let's apply this to other sectors of the economy and see how far our nation continues to survive.
It was always applied to other sectors of the economy. Only recently few people started saying that they can pick and choose which laws to obey.
Some good reading on the subject: http://www.econlib.org/library/Enc/MinimumWages.html