I can certainly see the majority of commerce moving to a non-anonymous net
It already has; it's just that this internet is layered on top of the public, anonymous network. Both users and servers are identified via trusted third parties: servers are generally vouched for by certificate authorities, and clients are generally vouched for by credit card companies. There's no technical reason clients cannot also use x509 certificates for authentication.
Technically speaking, Slashdot could move into this non-anonymous network tomorrow. It'd just require either a credit card or an SSL certificate for client authentication. It'd sure cut down on the trolling. Why hasn't Slashdot done this? Because the costs far outweigh the benefits.
Those who advocate a "new" internet are really talking about requiring solid authentication for everyone, all the time. They're asking everyone to shoulder the burden of universal authentication so the relatively few people who need it can reap the benefits. It's fundamentally unfair.
The same technique can be applied to the data network. If we separate the control information from the actual user data, we may achieve better security, as it would thwart any attempts to mess with the packet header, redirection attacks, prefix hijacks, or any of that other garbage
Your proposal is vague and meaningless. Control and user data are already separated on the Internet: it's not as if I can write a BGP message into a normal TCP stream and wreak havoc with routers. That would be the closest analogy to your phone example, and that problem was solved a long, long time ago.
Regular users can't run standard SMTP servers, since SMTP must listen on port 25. Port 25 is a privileged port that requires root privileges to bind.
In the future, when most users run under normal accounts, it might be helpful for mail servers to refuse connections with source-ports equal to or greater than 1024. (Being root is required to bind 1-1023).
Then again, ISP egress filtering for port 25 can have almost the same effect, today.
Unless the user receives a .desktop file, which works like a Windows shortcut or pif --- i.e., a.desktop file can be executed by double-clicking without it being marked executable. Congratulations, you lose.
Okay, so you have browsers do the OS X thing and mark downloaded.desktop files as being I'm-dangerous-I'm-from-the-Internet? Then you run into the Dancing Bunny Problem -- users will just click OK to see the cool thing they just downloaded.
Today, Linux (and to some extent, OS X) users are clueful enough to not descend to that level of stupidity. But as we see Windows decline in marketshare, more reckless users will end up in other operating systems and engage in the same kind of behavior that gives Windows such a bad reputation today.
We're dealing with a social problem here, not a technical one. Either we lock computers down so that users cannot run arbitrary programs on them, or we educate users to use their computers responsibly.
When people complain about SMTP being broken, they're not talking about SMTP. They're talking about the model that SMTP embodies: free-as-in-beer, unmetered email in which the receiver stores the message. SMTP is a simple, robust, and efficient way of implementing this model, and replacing it with a different wire protocol will change absolutely nothing about the spam problem.
Which aspect of the model would you change? Would you start charging for email? Would you implement a receiver-pulls-the-message system? Would you have email servers require certificates (which can be done within SMTP)?
Okay, so you know what you want to change? Good. Now advocate that instead of just whining that SMTP sucks.
Public key cryptography can be used to assert identity in an otherwise anonymous communication medium. Anonymity cannot, however, be layered on top of an attributed communication medium.
The Dancing Bunny Problem. There's an aphorism we seem to have forgotten: "never solve a social problem with a technical solution". The answer is more education for users; I don't see how any technical solution* can solve the dancing bunny problem.
* Well, S60-style platform security would go a long way, but I'd rather claw my eyes out with rusty 14.4k ISA modems than live in a world with locked-down computers for everyone
Custom work can rely on contract law to accomplish the goal of non-proliferation. After all, you're going to sign a contract for any custom work anyway: why not simply state that the customer is not allowed to redistribute the contracted-for work?
Piracy is a concern to those who are self employed programmers, making useful utilities that people pirate rather than pay for.
There are tried-and-true techniques for dealing with this problem, the most common of which is the shareware model: provide a free and freely-redistributable version with limited functionality, and a full-featured one that costs money. If the free version is sufficiently useful, people will purchase the for-pay one out of either a desire for the rest of the features (which you've proven exist by the quality of the free version), or our of a dense of moral obligation.
If it's good enough for iD software and DOOM, it's good enough for you.
Thoreau wrote, "When a man's conscience and the laws clash, it is his conscience that he must follow." It is not only our right, but our duty to disobey and unjust law. Making a law irrelevant and useless due to the sheer number of people disobeying it is one of the key factors in eliminating that law.
See the separate-but-equal laws, alcohol prohibition (juries nullified over 60% of prohibition cases toward the end), slavery (ever hear of the underground railroad?), and so on. Disobedience of the law has a long and dignified history, and so-called pirates are the latest in a long line of people working toward changing bad laws.
That's garbage. Parametrized queries have their place, sure. But if a quoting facility is letting "special" characters through, it has a bug and needs to be fixed. Guess what: one way of implementing parametrized queries is through automatic escaping!
1) We have hundreds of years of uranium even without reprocessing, and thousands if we do use reprocessing to recycle the fuel.
2) "Considerable" in the sense that you don't want to do it with your pinky finger, sure. But uranium mining is still *very* energy-positive. Even without petroleum, it could be done with biodiesel or electric equipment.
3) By the definition of half-life, the most dangerous substances have the shortest half-lives. After a decade cooling down in a tank, nuclear waste is safe enough to handle (briefly), and can be safely buried without worry. Radioactive materials either have a long half-life, or are dangerous, not both.
4) Of course a nuclear plant is expensive. But it's worth it. Each one produces a huge amount of power.
The CCR5-32 mutation, which this article discusses, is present naturally in 5-14% of Europeans. If HIV could easily develop a resistance to it, it would have already.
It's a shame that you typed such a long comment, because after a statement as factually incorrect as
The theories Obama supports...are based on those that have been shown by empirical evidence to be abject failures and have brought untold misery to the world. Soviet Union, North Korea and Zimbabwe for instance.
began with the idea that some kind of "action" is minimized* by what Nature (/God) does -- were in fact heavily motivated by mystical and religious ideas.
Fascinating. Do you have a source for all this? I find one of the best ways to understand a system is to get into the mindset of the people who created it and see the ideas that didn't work.
Nothing infuriated me more than listening to the globetrotting credit-card-spending children of upper-middle-class doctors rant angrily about white people and their money and privilege without a hint of irony
*sigh* Yes, I agree with you there. Perhaps I was mistaken in supposing sycodon was talking about policy, as I've heard others do. Unfortunately, being a bigoted asshole is bipartisan.
A logically wrong argument is wrong regardless of whether its result benefits the people we personally like. I wish people on both sides of the aisle would recognize this simple fact.
First, free speech and self-defense have nothing to do with economics. You'd be hard-pressed to find someone opposed to them in principle. I'm about as progressive as you can get, but I'm a near-absolutist on free speech. The answer to harmful speech is always more speech, and never censorship (unlike what the religious wing of the Republican party would claim.)
Second, low taxes are against your economic interest if the taxes are not progressive. The very rich will take their money and invest it in dubious assets, as we've seen in bubble after bubble after bubble over the last 30 years. On the other hand, that money, spend by the government on moderate-wage human resources and infrastructure, will generate a lot of real economic activity that increases everyone's prosperity. High, very progressive taxation is a good thing for society in general because it promotes real economic activity.
(My views here are essentially the same as those of Keynes, who coincidentally has been unfashionable for the past 30 years.)
Basing your political philosophy on reactionary opposition to a totalitarian regime that's been dead for 20 years is not a viable way of approaching the problems we face in today's world.
The moral of the story is that not all hypothesis are equally likely to be true, and so it makes no sense to give them equal weight. Doing so merely legitimizes those who try to further hypothesis for ulterior reasons.
So your defense of anti-intellectualism amounts to the observation that deep knowledge requires specialization? That's terribly weak.
Your Godwin mention is irrelevant at best, and a distraction at worst. Some bad people have had good educations, and some good people have had terrible educations. Education is not character. That does not invalidate the notion that those with an education are in general more competent in their chosen fields. You may not trust someone with a PhD to make all your decisions for you, but you'd be a fool not to trust him in the area of his expertise. (Unless, of course, you've done your own, equal research.)
Nobody is advocating putting rocket scientists in charge of mitigating climate change, or vica versa. That strawman is implied by your discussion of specialization.
What's really happening is conservatism rejects education as a qualification, and in some cases, views it as an antiqualification. This rejection is backwards and dangerous since it puts unqualified people in important positions.
I've always been fond of these lines of Oliver Goldsmith's verse:
Ye friends to truth, ye statesmen, who survey The rich man's joys increase, the poor's decay, 'Tis yours to judge how wide the limits stand Between a splendid and a happy land.
Claiming that progressives and conservatives both want to just screw over the middle class is exactly the kind of cynical, anti-rational mindset that's allowed the right to hold power over the last 30 years. It's poisoned the well of reason with hopelessness and despair.
The fact is that progressives do support a large, comfortable middle class, i.e., Goldsmith's "happy land", and imagining otherwise is conservative propaganda designed to keep people out of politics. Come on --- progressivism has always been about a quasiutilitarian idea of the greatest happiness for the greatest number.
Slashdot Mirror
It already has; it's just that this internet is layered on top of the public, anonymous network. Both users and servers are identified via trusted third parties: servers are generally vouched for by certificate authorities, and clients are generally vouched for by credit card companies. There's no technical reason clients cannot also use x509 certificates for authentication.
Technically speaking, Slashdot could move into this non-anonymous network tomorrow. It'd just require either a credit card or an SSL certificate for client authentication. It'd sure cut down on the trolling. Why hasn't Slashdot done this? Because the costs far outweigh the benefits.
Those who advocate a "new" internet are really talking about requiring solid authentication for everyone, all the time. They're asking everyone to shoulder the burden of universal authentication so the relatively few people who need it can reap the benefits. It's fundamentally unfair.
Tor, The MixMaster anonymous remailer, Freenet, and public proxy systems, among other places. We do have fairly good anonymity.
Your proposal is vague and meaningless. Control and user data are already separated on the Internet: it's not as if I can write a BGP message into a normal TCP stream and wreak havoc with routers. That would be the closest analogy to your phone example, and that problem was solved a long, long time ago.
Regular users can't run standard SMTP servers, since SMTP must listen on port 25. Port 25 is a privileged port that requires root privileges to bind.
In the future, when most users run under normal accounts, it might be helpful for mail servers to refuse connections with source-ports equal to or greater than 1024. (Being root is required to bind 1-1023).
Then again, ISP egress filtering for port 25 can have almost the same effect, today.
Unless the user receives a .desktop file, which works like a Windows shortcut or pif --- i.e., a .desktop file can be executed by double-clicking without it being marked executable. Congratulations, you lose.
Okay, so you have browsers do the OS X thing and mark downloaded .desktop files as being I'm-dangerous-I'm-from-the-Internet? Then you run into the Dancing Bunny Problem -- users will just click OK to see the cool thing they just downloaded.
Today, Linux (and to some extent, OS X) users are clueful enough to not descend to that level of stupidity. But as we see Windows decline in marketshare, more reckless users will end up in other operating systems and engage in the same kind of behavior that gives Windows such a bad reputation today.
We're dealing with a social problem here, not a technical one. Either we lock computers down so that users cannot run arbitrary programs on them, or we educate users to use their computers responsibly.
When people complain about SMTP being broken, they're not talking about SMTP. They're talking about the model that SMTP embodies: free-as-in-beer, unmetered email in which the receiver stores the message. SMTP is a simple, robust, and efficient way of implementing this model, and replacing it with a different wire protocol will change absolutely nothing about the spam problem.
Which aspect of the model would you change? Would you start charging for email? Would you implement a receiver-pulls-the-message system? Would you have email servers require certificates (which can be done within SMTP)?
Okay, so you know what you want to change? Good. Now advocate that instead of just whining that SMTP sucks.
Public key cryptography can be used to assert identity in an otherwise anonymous communication medium. Anonymity cannot, however, be layered on top of an attributed communication medium.
The Dancing Bunny Problem. There's an aphorism we seem to have forgotten: "never solve a social problem with a technical solution". The answer is more education for users; I don't see how any technical solution* can solve the dancing bunny problem.
* Well, S60-style platform security would go a long way, but I'd rather claw my eyes out with rusty 14.4k ISA modems than live in a world with locked-down computers for everyone
Custom work can rely on contract law to accomplish the goal of non-proliferation. After all, you're going to sign a contract for any custom work anyway: why not simply state that the customer is not allowed to redistribute the contracted-for work?
There are tried-and-true techniques for dealing with this problem, the most common of which is the shareware model: provide a free and freely-redistributable version with limited functionality, and a full-featured one that costs money. If the free version is sufficiently useful, people will purchase the for-pay one out of either a desire for the rest of the features (which you've proven exist by the quality of the free version), or our of a dense of moral obligation.
If it's good enough for iD software and DOOM, it's good enough for you.
Thoreau wrote, "When a man's conscience and the laws clash, it is his conscience that he must follow." It is not only our right, but our duty to disobey and unjust law. Making a law irrelevant and useless due to the sheer number of people disobeying it is one of the key factors in eliminating that law.
See the separate-but-equal laws, alcohol prohibition (juries nullified over 60% of prohibition cases toward the end), slavery (ever hear of the underground railroad?), and so on. Disobedience of the law has a long and dignified history, and so-called pirates are the latest in a long line of people working toward changing bad laws.
That's garbage. Parametrized queries have their place, sure. But if a quoting facility is letting "special" characters through, it has a bug and needs to be fixed. Guess what: one way of implementing parametrized queries is through automatic escaping!
Nuclear power is our best bet.
1) We have hundreds of years of uranium even without reprocessing, and thousands if we do use reprocessing to recycle the fuel.
2) "Considerable" in the sense that you don't want to do it with your pinky finger, sure. But uranium mining is still *very* energy-positive. Even without petroleum, it could be done with biodiesel or electric equipment.
3) By the definition of half-life, the most dangerous substances have the shortest half-lives. After a decade cooling down in a tank, nuclear waste is safe enough to handle (briefly), and can be safely buried without worry. Radioactive materials either have a long half-life, or are dangerous, not both.
4) Of course a nuclear plant is expensive. But it's worth it. Each one produces a huge amount of power.
The CCR5-32 mutation, which this article discusses, is present naturally in 5-14% of Europeans. If HIV could easily develop a resistance to it, it would have already.
CNAMEs are good. Using multiple A records is a bad idea that will screw up Kerberos, confuse ssh, and eat your children.
Also, beware when you first start a job and all the production servers are running never-updated Gentoo behind a m0n0wall gateway.
It's a shame that you typed such a long comment, because after a statement as factually incorrect as
I just can't bring myself to read any further.
Fascinating. Do you have a source for all this? I find one of the best ways to understand a system is to get into the mindset of the people who created it and see the ideas that didn't work.
*sigh* Yes, I agree with you there. Perhaps I was mistaken in supposing sycodon was talking about policy, as I've heard others do. Unfortunately, being a bigoted asshole is bipartisan.
A logically wrong argument is wrong regardless of whether its result benefits the people we personally like. I wish people on both sides of the aisle would recognize this simple fact.
First, free speech and self-defense have nothing to do with economics. You'd be hard-pressed to find someone opposed to them in principle. I'm about as progressive as you can get, but I'm a near-absolutist on free speech. The answer to harmful speech is always more speech, and never censorship (unlike what the religious wing of the Republican party would claim.)
Second, low taxes are against your economic interest if the taxes are not progressive. The very rich will take their money and invest it in dubious assets, as we've seen in bubble after bubble after bubble over the last 30 years. On the other hand, that money, spend by the government on moderate-wage human resources and infrastructure, will generate a lot of real economic activity that increases everyone's prosperity. High, very progressive taxation is a good thing for society in general because it promotes real economic activity.
(My views here are essentially the same as those of Keynes, who coincidentally has been unfashionable for the past 30 years.)
And your statement invalidates my convervatism-is-anti-rationalism thesis how?
Thank you!
Basing your political philosophy on reactionary opposition to a totalitarian regime that's been dead for 20 years is not a viable way of approaching the problems we face in today's world.
I was about to discuss this topic when I realized that Carl Sagan has already done so, and far more eloquently than I can manage.
The Dragon in my Garage
The moral of the story is that not all hypothesis are equally likely to be true, and so it makes no sense to give them equal weight. Doing so merely legitimizes those who try to further hypothesis for ulterior reasons.
So your defense of anti-intellectualism amounts to the observation that deep knowledge requires specialization? That's terribly weak.
Your Godwin mention is irrelevant at best, and a distraction at worst. Some bad people have had good educations, and some good people have had terrible educations. Education is not character. That does not invalidate the notion that those with an education are in general more competent in their chosen fields. You may not trust someone with a PhD to make all your decisions for you, but you'd be a fool not to trust him in the area of his expertise. (Unless, of course, you've done your own, equal research.)
Nobody is advocating putting rocket scientists in charge of mitigating climate change, or vica versa. That strawman is implied by your discussion of specialization.
What's really happening is conservatism rejects education as a qualification, and in some cases, views it as an antiqualification. This rejection is backwards and dangerous since it puts unqualified people in important positions.
I've always been fond of these lines of Oliver Goldsmith's verse:
Ye friends to truth, ye statesmen, who survey
The rich man's joys increase, the poor's decay,
'Tis yours to judge how wide the limits stand
Between a splendid and a happy land.
Claiming that progressives and conservatives both want to just screw over the middle class is exactly the kind of cynical, anti-rational mindset that's allowed the right to hold power over the last 30 years. It's poisoned the well of reason with hopelessness and despair.
The fact is that progressives do support a large, comfortable middle class, i.e., Goldsmith's "happy land", and imagining otherwise is conservative propaganda designed to keep people out of politics. Come on --- progressivism has always been about a quasiutilitarian idea of the greatest happiness for the greatest number.