You would be foolish to think that they sit around at the Pentagon thinking about killing people for killing's sake, though. They think about how to most effectively change the disposition of their opponents.
No, maybe the US President & Cabinet consider ways to change disposition. By the time they hand a problem off to the Pentagon, the choice for a killing-based solution has already been made.
Nonviolent solutions are for the State Department.
The super gun was developed originally to take out satellite systems.
Maybe that was the intent, but it could never have worked. The cheapest, easiest way to get a projectile to go far is to include the propellant with the munition, like a rocket or missile, rather than using all energy at the first instant, like with a gun.
The largest gun ever used was constructed by Nazi Germany and had an effective range of less than 30 kilometers (and that's horizontal distance, it couldn't have got that far up). When they needed to go further, it was much easier and cheaper to invent V-2 rockets, than to try to make a yet larger gun.
This was rubbish because once the first incoming arrives the radar defence system is able to work out the launch location and it is destroyed.
It would've been discovered long before then, even before it was built. Forging a barrel that large would've required industrial capacity that Iraq has never come close to possessing (and that the US in fact no longer does, because huge steel objects are inefficient). Iraq can't even build their own 10cm tank guns, they'd never be able to build the 50+cm diameter weapon you'd need to threaten satellites.
they command undue attention in a conference room.
Height exaggerates social-skill tendencies. Tall people in a group are more visible than others, so both their successes and failures have a larger impact on percieved ranking in the community.
A tall person can either gravitate to the center-of-attention, flourishing in the light, or be pushed to the edges as a wallflower (where he won't obstruct the masses' view of the adored leader). There is less opportunity to be just a face in the crowd- you're a lead role, or you're nobody.
How many successful short-statured politicians can you name (at least since the age of mass media)?
Arguably, mass-media can help short people become greater politicians. You can't judge height from a talking head on TV. Stage management of televised appearances can conceal real stature.
Mass-media makes direct, personal communication less important to a candidate who's drumming up support. It can open doorways to a person who's not so good at taking charge in a crowd, but works great in front of the cameras.
(Look at US President GW Bush to see how mass-media has made his incoherent speech less of a handicap...)
height is a variable that can't concievably be manipulated by wage
I can concieve it. Wealth -> more+better food -> size -> height.
Sure, a person's height by the time she's earning serious wages is no longer changable, but that doesn't break the premise. You just have to look at it as an inter-generational cycle. Her children will be taller because she's wealthy, and they will also inherit some non-height factors which increase earnings.
Re:So Why Won't Doctor's Help Short People Grow?
on
Tall People Earn More
·
· Score: 1
So why won't doctor's help short people grow?
They do. The US FDA has just approved a big expansion of the range of heights considered appropriate for growth hormone prescriptions.
XML already exists. It has been defined already, and binary-encoding is not part of it.
There should always have been binary-encoding options as part of the XML specification, but now it's rather too late. XML parsers from many sources have already been deployed around the world.
Design shortcomings become expontentially harder to correct the longer a software project has gone on. By the time it's deployed for real systems in the wild, fundamental changes would be almost prohibitively disruptive.
Exactly. Bountyquest was only there to try invalidating patents for prior-art reasons.
But there is a whole separate criteria for patent validity: non-obviousness. It's that aspect which was missing in the "One Click" patent. Since BountyQuest didn't try to disprove non-obviousness (which is a much higher legal threshold to reach), it can't be blamed for failing to stop "One Click". (People don't tend to publish documents about obvious things)
If, in 1998, you had asked an experienced web developer "How can I let my returning customers order a initiate a product delivery with one button-press", a decent number of them would've been able to describe the exact system Bezos used.
If someone with ordinary skill in the field can guess the text of a patent by reading the title, it should never have been granted.
It's really, really expensive to get -96dB of quiet.
No, less than a million dollars could get that easily. That's not "really, really" expensive, on the scale the RIAA operates at.
There are 4 areas the recording industry spends money on: 1. Building & running recording studios. 2. Making & distributing CDs. 3. Promotion 4. Production (where I lump all the Human Resources tasks of joining together good writers, singers, musicians, and other "artists" to actually do the work of making music)
Of those, "Promotion" is by far where most of their investment goes. Tens of millions of dollars for even a mediorce album. But that's the one task which doesn't benefit customers at all- it's actually working against the buyer. All the promotion people really want can be accomplished for free by willing fans (on the internet, or elsewhere).
Of the other areas, #2 (making CDs) is totally obseleted by technology. And the barriers to entry for #1 (recording studio) have also come way down- the technology part is virtually free ($10,000), only the building is moderately expensive.
In the long run, they're left with #4 (Production) as the only thing they're still competing with. And that is where driven newcomers would manage to beat out the old guard.
IP addresses are network layer information. URLs are application level information.
You seem to be completely missing the point.
I just mentioned out Apache and IIS as big examples, but in reality, every Internet application is broken by NAT. I can't think of one major TCP/IP program that would work correctly if all computers in the world were NATted. Since universal NAT would destroy the internet, it follows that partial NAT is damaging to it (or so the reasoning goes)
You might respond, "But that's not what I was talking about". True. However, that's what Walker's article was all about. Your complain of "NAT bashing" was apparently a knee-jerk reaction, with no basis on having read the story.
If a commercial company is willing to publically sign code that is spyware, what exactly stops spyware?
Two things: 1. The risk of being stuck on the "spyware blacklist", similar to the "spammer's blacklist" he discusses. Once your company is listed there, many computers will automatically begin to reject any applications you've signed.
2. The "Trusted" environment within the PC will isolate applications from one another. A properly designed TC system will have a level of privilege separation, so that a block-puzzle you downloaded won't be able to read your email addressbook without raising big warning messages. (Social engineering can work around this barrier, if the installation instructions can trick users into installing programs with higher priviledge than they really need.)
Another tired complaint about how NAT is a terrible evil because it breaks badly designed applications.
NAT breaks apache. And IIS too. You call them "badly-designed applications"?
Re:Yes, and because you're not other people
on
Trusted Computing
·
· Score: 1
The theory of DRM is to protect copyright holders' rights.
The theory of DRM is to go beyond copyright holder's rights, by allowing them to technically restrict customers usage in ways that they legally can't. For example, copyright will expire in 95 years; DRM never will. Copyrighted products are resellable- DRM ones aren't. (Sure, a customer could negotiate a right to resell in exchange for a higher initial cost, but today you don't have to negotiate: they have no right to withhold resale) DRM is an end-run around the (already corporate-biased) copyright law.
The theory of TC is to help stop the swell of viruses and worms and spam and the like.
The theory of TC is to stop the spread of Napster, Kazaa, and the like. It's meant to protect DRM from tampering.
"Viruses and worms" are threats rolled out by proponents to bid for public support, but they are far from what TC was originally meant to do. The first conception of Trusted Computing was "I wish I could send someone some data, but not let him copy or print it". TC's biggest single feature is "remote attestation"; everything else is there to support it.
All of the vulnerabilities mentioned there are regarding stack-overflows (or UDP-bouncing), which are different things than encryption. The claim that it uses "military-grade encryption" is true.
This is a classic example of a logically fallacious ad hominem attack. Because someone working on a FREE project used traditional, error-prone C developement methods, you've decided to discount his opinions on the trends for large-scale networking applications.
There's no way to tell if the poker hand I describe in an email is real or if it's part of a encrypted message.
That could work a little, as might the general class of steganographic tricky.
But Walker already addressed that point just fine. Basically, they can make it so hard to evade the controls that the end result is indistinguishable from perfect control, even though 0.0001% of people can sneak around it for occasional small messages.
Re:Article is a bit off base in places
on
Trusted Computing
·
· Score: 1
but doesn't seem to realize these two facts: 2. Most users don't care about running their own website.
"In any case, the key lesson of the mass introduction of NAT is that it demonstrates, in a real world test, that the vast majority of Internet users do not notice and do not care that their access to the full range of Internet services and ability to act as a peer of any other Internet site has been restricted."
Anything not mainstream is underground, the only question is how far under.
That's already a rather pessimistic viewpoint. Acquiesed already, really. Saddam's Iraq was a still a dictatorship, even though handfulls of rebels could whisper together in the shadows.
But the creation of undergrounds is a human-resources problem; it depends on like-minded people finding each other and then going off to exchange their hidden knowledge. Until one of them can share crypto techniques with the other, they have no way of securely talking. The initial linkup MUST happen over normal channels.
In the past, those normal channels were safe. Everyone had long periods of isolation from any government monitoring, when associations could be struck and trust established. In the future, the pace of technology will make it more plausible for governments to monitor even everyday communication, letting them better quash underground cells before they ever form.
And they can take my general-purpose computer from me when they pry the keyboard from my cold dead hands.
How long has one computer ever been useful for you? Five years? Ten? Good luck keeping that thing functional once everything from Intel/AMD is DRM-approved.
Re:Same false information about Trusted Computing
on
Trusted Computing
·
· Score: 1
False. That's just not how Trusted Computing works. You can copy all you want, it's decrypting that is made harder.
He's talking about something else. His whole point is that "Trusted Computing" (the concept, not the brand name) will go BEYOND what it's proponents are currently claiming.
"He's got a modem! Open fire, it must be a terrorist! Why else would he not use our beautiful Citizen's Internet, unless he has something to hide"
But seriously, in the long run (15+ years), they won't even have to ban modems. You won't have phone lines anymore, except things that run use VoIP. Sonic analysis and natural-language processors will be able to detect if those VoIP packets contain data inconsistent with verbal communication (even if computers can't understand speech, they'll soon be able to recognize it), and the police will come with a warrant.
The only workaround to that is a kind of "steganographic" concealment of your secret data ontop of legitimate traffic. The government would have to stringently punish anyone who researches, writes, or releases such code.
What sodding nazi ISP do you use for Cthulhu's sake? Or, more to the point, why the hell are you still using them?
I'm at work. My company makes custom TCP/IP applications, and over the past 20 years our customers have become increasingly inconvienced that we can no longer connect to them directly.
(It would be a fatal security risk for the Windows(tm) systems that may exist in the LAN)
any ISP that would stop me doesn't get a penny of my money
Which ISP is that, exactly? I've been through the websites of the top 4 "broadband" providers in the US, and they all require subscribers to agree that the applications they can use over TCP/IP is restricted (even if the provider has not yet implemented technical measures to block offenders, they have announced intent)
sucks that your connection is a piece of shit, guess YOU particularly (and the few people like you)
It's true that my selection of ports is more restrictive than average. However, by a big preponderance, the typical (US) internet user is not able to accept incoming connections.
If you add together all the AOL people, all the college students, all the corporate deskjockeys, and everyone on Comcast, AT&T, Verizon, Time-Warner, and RCN... well, that's much more than half of all people on the internet. Each of those services either currently blocks incoming connections, or has TermsOfService agreements that restrict what you're allowed to do with your TCP/IP pipe.
This is exactly the situation Walker describes in the first non-introductory section of his article.
Re:Lessig said it first
on
Trusted Computing
·
· Score: 2, Interesting
Now there are a few technical reasons why the internet CANNOT be retrofitted into pay-as-you-go content restricted affair.
I think that Walker's article does a good job at refuting those supposed technical reasons. If you can point out specifically how he's mistaken, please do so. The question of whether or not something is "techincally impossible" is always a difficult one, and the pattern throughout history is that something deemed "impossible" by one generation is achieved by the next.
The cost of metering the internet would be greater than the cost of providing it, free, to the world at large.
They don't have to do so much. Authoritarian control can be exercised without needing to meter each and every little packet. A government could simply declare that use of any encrypted protocol is illegal (the old "Clipper chip" proposal did something like this, good thing it died). Then randomly sniff packets, just 0.001% of them, until catching something that their automated systems cannot decode. The ISP points out the perpetrator, who is arrested while technicians comb over his PC for the offending code, and any other guilty evidence.
The Chilling Effect could be so strong that propagation of encryption could be effectively suppressed.
Finally restricting access to the internet would be like restricting access to our highway system.
That's using the "force of nature" argument. Restricting access to roads would be prohibitively difficult, because roadways are part of our physical world. The topology is determined by the 2-dimensional lay of the land. The internet, as an entirely artificial virtual world, obeys only rules invented by humans. What man built, man can unbuild.
I personally just mutter under my breath and pay it, but my wife actually routes around the toll roads useing back roads.
(Following is a pessimistic response. It might not happen, but it COULD) Within 5 years, to pay a toll, you'll need to have a radio-transponder installed in your car. Sure, you can still pay cash to an attendant, but it'll be $5 instead of $0.50. And eventually that attendant will be fired, and replaced with police cruisers to arrest nonpayers on the highway. 5 years after that, the toll points will be taken down and replaced with an automatic system that uses GPS to tell when you entered the "premium" roadway. One year later, speeding tickets will be automatically mailed if that GPS clocks you at going over 65 mph.
Two years after that, a politician will decide that the fairest way to allocate highway-maintenance taxes is based on actual road usage, and every mile you drive will be tolled.
Slashdot Mirror
You would be foolish to think that they sit around at the Pentagon thinking about killing people for killing's sake, though. They think about how to most effectively change the disposition of their opponents.
No, maybe the US President & Cabinet consider ways to change disposition. By the time they hand a problem off to the Pentagon, the choice for a killing-based solution has already been made.
Nonviolent solutions are for the State Department.
The super gun was developed originally to take out satellite systems.
Maybe that was the intent, but it could never have worked. The cheapest, easiest way to get a projectile to go far is to include the propellant with the munition, like a rocket or missile, rather than using all energy at the first instant, like with a gun.
The largest gun ever used was constructed by Nazi Germany and had an effective range of less than 30 kilometers (and that's horizontal distance, it couldn't have got that far up). When they needed to go further, it was much easier and cheaper to invent V-2 rockets, than to try to make a yet larger gun.
This was rubbish because once the first incoming arrives the radar defence system is able to work out the launch location and it is destroyed.
It would've been discovered long before then, even before it was built. Forging a barrel that large would've required industrial capacity that Iraq has never come close to possessing (and that the US in fact no longer does, because huge steel objects are inefficient). Iraq can't even build their own 10cm tank guns, they'd never be able to build the 50+cm diameter weapon you'd need to threaten satellites.
they command undue attention in a conference room.
Height exaggerates social-skill tendencies. Tall people in a group are more visible than others, so both their successes and failures have a larger impact on percieved ranking in the community.
A tall person can either gravitate to the center-of-attention, flourishing in the light, or be pushed to the edges as a wallflower (where he won't obstruct the masses' view of the adored leader). There is less opportunity to be just a face in the crowd- you're a lead role, or you're nobody.
How many successful short-statured politicians can you name (at least since the age of mass media)?
Arguably, mass-media can help short people become greater politicians. You can't judge height from a talking head on TV. Stage management of televised appearances can conceal real stature.
Mass-media makes direct, personal communication less important to a candidate who's drumming up support. It can open doorways to a person who's not so good at taking charge in a crowd, but works great in front of the cameras.
(Look at US President GW Bush to see how mass-media has made his incoherent speech less of a handicap...)
height is a variable that can't concievably be manipulated by wage
I can concieve it. Wealth -> more+better food -> size -> height.
Sure, a person's height by the time she's earning serious wages is no longer changable, but that doesn't break the premise. You just have to look at it as an inter-generational cycle. Her children will be taller because she's wealthy, and they will also inherit some non-height factors which increase earnings.
So why won't doctor's help short people grow?
They do. The US FDA has just approved a big expansion of the range of heights considered appropriate for growth hormone prescriptions.
XML already exists. It has been defined already, and binary-encoding is not part of it.
There should always have been binary-encoding options as part of the XML specification, but now it's rather too late. XML parsers from many sources have already been deployed around the world.
Design shortcomings become expontentially harder to correct the longer a software project has gone on. By the time it's deployed for real systems in the wild, fundamental changes would be almost prohibitively disruptive.
That's not a valid use of the GPL. The license text requires that the entire license be included when you give out the program.
RMS recommends that authors of software shorter than the GPL simply use Public Domain release, to avoid textual overhead.
Exactly. Bountyquest was only there to try invalidating patents for prior-art reasons.
But there is a whole separate criteria for patent validity: non-obviousness. It's that aspect which was missing in the "One Click" patent. Since BountyQuest didn't try to disprove non-obviousness (which is a much higher legal threshold to reach), it can't be blamed for failing to stop "One Click". (People don't tend to publish documents about obvious things)
If, in 1998, you had asked an experienced web developer "How can I let my returning customers order a initiate a product delivery with one button-press", a decent number of them would've been able to describe the exact system Bezos used.
If someone with ordinary skill in the field can guess the text of a patent by reading the title, it should never have been granted.
have no power over how it's implemented.
They have POWER. They have $40 billion dollars of liquid power.
It's a free market. Microsoft(tm) should be able to either pay Akamai to use Windows(r) servers, or go to another company that does.
And if there's no company that does, it tells us a salient fact about the suitability of Windows for critical, high-capacity servers.
It's really, really expensive to get -96dB of quiet.
No, less than a million dollars could get that easily. That's not "really, really" expensive, on the scale the RIAA operates at.
There are 4 areas the recording industry spends money on:
1. Building & running recording studios.
2. Making & distributing CDs.
3. Promotion
4. Production (where I lump all the Human Resources tasks of joining together good writers, singers, musicians, and other "artists" to actually do the work of making music)
Of those, "Promotion" is by far where most of their investment goes. Tens of millions of dollars for even a mediorce album. But that's the one task which doesn't benefit customers at all- it's actually working against the buyer. All the promotion people really want can be accomplished for free by willing fans (on the internet, or elsewhere).
Of the other areas, #2 (making CDs) is totally obseleted by technology. And the barriers to entry for #1 (recording studio) have also come way down- the technology part is virtually free ($10,000), only the building is moderately expensive.
In the long run, they're left with #4 (Production) as the only thing they're still competing with. And that is where driven newcomers would manage to beat out the old guard.
racial slurs about anyone else get modded to the stars.
What around here is a "racial slur", exactly?
I've seen one joke about food quality, which is not racial.
And I've seen one joke about the inability of Europeans to distinguish between Asians.
Which of those offends you, and why?
IP addresses are network layer information. URLs are application level information.
.
You seem to be completely missing the point
I just mentioned out Apache and IIS as big examples, but in reality, every Internet application is broken by NAT. I can't think of one major TCP/IP program that would work correctly if all computers in the world were NATted. Since universal NAT would destroy the internet, it follows that partial NAT is damaging to it (or so the reasoning goes)
You might respond, "But that's not what I was talking about". True. However, that's what Walker's article was all about. Your complain of "NAT bashing" was apparently a knee-jerk reaction, with no basis on having read the story.
If a commercial company is willing to publically sign code that is spyware, what exactly stops spyware?
Two things:
1. The risk of being stuck on the "spyware blacklist", similar to the "spammer's blacklist" he discusses. Once your company is listed there, many computers will automatically begin to reject any applications you've signed.
2. The "Trusted" environment within the PC will isolate applications from one another. A properly designed TC system will have a level of privilege separation, so that a block-puzzle you downloaded won't be able to read your email addressbook without raising big warning messages. (Social engineering can work around this barrier, if the installation instructions can trick users into installing programs with higher priviledge than they really need.)
Another tired complaint about how NAT is a terrible evil because it breaks badly designed applications.
NAT breaks apache. And IIS too. You call them "badly-designed applications"?
The theory of DRM is to protect copyright holders' rights.
The theory of DRM is to go beyond copyright holder's rights, by allowing them to technically restrict customers usage in ways that they legally can't. For example, copyright will expire in 95 years; DRM never will. Copyrighted products are resellable- DRM ones aren't. (Sure, a customer could negotiate a right to resell in exchange for a higher initial cost, but today you don't have to negotiate: they have no right to withhold resale) DRM is an end-run around the (already corporate-biased) copyright law.
The theory of TC is to help stop the swell of viruses and worms and spam and the like.
The theory of TC is to stop the spread of Napster, Kazaa, and the like. It's meant to protect DRM from tampering.
"Viruses and worms" are threats rolled out by proponents to bid for public support, but they are far from what TC was originally meant to do. The first conception of Trusted Computing was "I wish I could send someone some data, but not let him copy or print it". TC's biggest single feature is "remote attestation"; everything else is there to support it.
All of the vulnerabilities mentioned there are regarding stack-overflows (or UDP-bouncing), which are different things than encryption. The claim that it uses "military-grade encryption" is true.
This is a classic example of a logically fallacious ad hominem attack. Because someone working on a FREE project used traditional, error-prone C developement methods, you've decided to discount his opinions on the trends for large-scale networking applications.
There's no way to tell if the poker hand I describe in an email is real or if it's part of a encrypted message.
That could work a little, as might the general class of steganographic tricky.
But Walker already addressed that point just fine. Basically, they can make it so hard to evade the controls that the end result is indistinguishable from perfect control, even though 0.0001% of people can sneak around it for occasional small messages.
but doesn't seem to realize these two facts:
2. Most users don't care about running their own website.
He realized that fact, and stated it exactly:
"In any case, the key lesson of the mass introduction of NAT is that it demonstrates, in a real world test, that the vast majority of Internet users do not notice and do not care that their access to the full range of Internet services and ability to act as a peer of any other Internet site has been restricted."
Anything not mainstream is underground, the only question is how far under.
That's already a rather pessimistic viewpoint. Acquiesed already, really. Saddam's Iraq was a still a dictatorship, even though handfulls of rebels could whisper together in the shadows.
But the creation of undergrounds is a human-resources problem; it depends on like-minded people finding each other and then going off to exchange their hidden knowledge. Until one of them can share crypto techniques with the other, they have no way of securely talking. The initial linkup MUST happen over normal channels.
In the past, those normal channels were safe. Everyone had long periods of isolation from any government monitoring, when associations could be struck and trust established. In the future, the pace of technology will make it more plausible for governments to monitor even everyday communication, letting them better quash underground cells before they ever form.
And they can take my general-purpose computer from me when they pry the keyboard from my cold dead hands.
How long has one computer ever been useful for you? Five years? Ten? Good luck keeping that thing functional once everything from Intel/AMD is DRM-approved.
False. That's just not how Trusted Computing works. You can copy all you want, it's decrypting that is made harder.
He's talking about something else. His whole point is that "Trusted Computing" (the concept, not the brand name) will go BEYOND what it's proponents are currently claiming.
Let's set up some dedicated modem links.
"He's got a modem! Open fire, it must be a terrorist! Why else would he not use our beautiful Citizen's Internet, unless he has something to hide"
But seriously, in the long run (15+ years), they won't even have to ban modems. You won't have phone lines anymore, except things that run use VoIP. Sonic analysis and natural-language processors will be able to detect if those VoIP packets contain data inconsistent with verbal communication (even if computers can't understand speech, they'll soon be able to recognize it), and the police will come with a warrant.
The only workaround to that is a kind of "steganographic" concealment of your secret data ontop of legitimate traffic. The government would have to stringently punish anyone who researches, writes, or releases such code.
What sodding nazi ISP do you use for Cthulhu's sake? Or, more to the point, why the hell are you still using them?
I'm at work. My company makes custom TCP/IP applications, and over the past 20 years our customers have become increasingly inconvienced that we can no longer connect to them directly.
(It would be a fatal security risk for the Windows(tm) systems that may exist in the LAN)
any ISP that would stop me doesn't get a penny of my money
Which ISP is that, exactly? I've been through the websites of the top 4 "broadband" providers in the US, and they all require subscribers to agree that the applications they can use over TCP/IP is restricted (even if the provider has not yet implemented technical measures to block offenders, they have announced intent)
sucks that your connection is a piece of shit, guess YOU particularly (and the few people like you)
It's true that my selection of ports is more restrictive than average. However, by a big preponderance, the typical (US) internet user is not able to accept incoming connections.
If you add together all the AOL people, all the college students, all the corporate deskjockeys, and everyone on Comcast, AT&T, Verizon, Time-Warner, and RCN... well, that's much more than half of all people on the internet. Each of those services either currently blocks incoming connections, or has TermsOfService agreements that restrict what you're allowed to do with your TCP/IP pipe.
This is exactly the situation Walker describes in the first non-introductory section of his article.
Now there are a few technical reasons why the internet CANNOT be retrofitted into pay-as-you-go content restricted affair.
I think that Walker's article does a good job at refuting those supposed technical reasons. If you can point out specifically how he's mistaken, please do so. The question of whether or not something is "techincally impossible" is always a difficult one, and the pattern throughout history is that something deemed "impossible" by one generation is achieved by the next.
The cost of metering the internet would be greater than the cost of providing it, free, to the world at large.
They don't have to do so much. Authoritarian control can be exercised without needing to meter each and every little packet. A government could simply declare that use of any encrypted protocol is illegal (the old "Clipper chip" proposal did something like this, good thing it died). Then randomly sniff packets, just 0.001% of them, until catching something that their automated systems cannot decode. The ISP points out the perpetrator, who is arrested while technicians comb over his PC for the offending code, and any other guilty evidence.
The Chilling Effect could be so strong that propagation of encryption could be effectively suppressed.
Finally restricting access to the internet would be like restricting access to our highway system.
That's using the "force of nature" argument. Restricting access to roads would be prohibitively difficult, because roadways are part of our physical world. The topology is determined by the 2-dimensional lay of the land. The internet, as an entirely artificial virtual world, obeys only rules invented by humans. What man built, man can unbuild.
I personally just mutter under my breath and pay it, but my wife actually routes around the toll roads useing back roads.
(Following is a pessimistic response. It might not happen, but it COULD)
Within 5 years, to pay a toll, you'll need to have a radio-transponder installed in your car. Sure, you can still pay cash to an attendant, but it'll be $5 instead of $0.50. And eventually that attendant will be fired, and replaced with police cruisers to arrest nonpayers on the highway. 5 years after that, the toll points will be taken down and replaced with an automatic system that uses GPS to tell when you entered the "premium" roadway. One year later, speeding tickets will be automatically mailed if that GPS clocks you at going over 65 mph.
Two years after that, a politician will decide that the fairest way to allocate highway-maintenance taxes is based on actual road usage, and every mile you drive will be tolled.