Speaking of reality, here's how prioritization is likely to go:
ISP sells higher tier of service to corporation A, so content to and from their sites gets to the endpoints more quickly.
As with any such contract, corporation A also include an SLA in the purchase, with performance measures. That SLA, like most, will have rebate clauses, just like they do now.
However, the ISP can't speak for all the other ISPs, in particular the ones at the edges. Since corporation A most likely is buying its internet connection from a backbone provider, the higher tier of service ends up making a little bit of difference, but not much due to the fact that QOS has to be coherently configured end to end to make a consistant difference.
Performance metrics for improved service are, predictably, not met. Corporation A realizes it has paid for nothing, and has been had by its ISP. Corporation A begins enforcing rebate clauses. ISP gets less for its 'tier' of service than it hoped, if anything at all.
If there's a dose of reality, its that residential users never had impact in the first place, and that prioritization will just set backbone providers at each other's throats more than they are now.
As for the Internet ocean analogy, it holds accurate until each and every ISP on the Internet sets up filtering and QOS the same - that's just the nature of QOS. Until then, it may as well be an ocean from the point of view of ISPs trying to sell traffic prioritization. Missing from the discussion is an understanding of how QOS works.
I think a 'tiered' internet is trouble from the start, but what about this scenerio: Your VOIP provider starts providing 911 service, and your 911 call gets squashed by your neighbor's video download. Under strict 'net neutrality', it is possible for this to happen, if unlikely.
Additionaly, the ability of backbone providers to influence the delivery of packets is quite limited in comparison to the 'last mile' provider. The ISP customers immediately connect to, if they choose to set QOS for some type of service from some content provider, will have a great deal more effect on download/upload speeds that backbone providers. That's just how QOS out at the edge works. Yes, backbone providers can influence packet delivery, but not nearly as much as edge providers.
The other problem with allowing provider to prioritize traffic is that once packets traverse provider boundries, all bets are off. Does anyone really think that Verizon/MCI/UUNet will treat AT&T's prioritized packets better or even on par with its own? After all, Verizon's own customers, like maybe giant-company-xyz, is paying to have their traffic prioritized, and all Verizon might have with AT&T is an aggreement that might not be worth as much as $$ from giant-company-xyz. If AT&T never sees all the router configs in Verizon's network, how can they claim that Verizon isn't honoring their QOS?
The internet is more like an ocean than it is a bunch of lakes and canals, and the telcos want to sell good weather and smooth sailing. AT&T will sell Disney, for example, a 'higher tier' of service for their streaming video on their backbone, but unless they can get each and every edge provider to go along, and each and every other entity that runs any kind of peering link at all on the Internet, it won't make as big a difference as they claim. My point is that even if telcos sell prioritization, its likely it won't stack up like they claim, due to the nature of the Internet itself. Then everybody will have to decide how to treat legitimate priority traffic, like 911 for example.
The entire debate looks to me as though it being framed in a misleading way.
Kodak's patent mentions previous research suggesting a correlation between age and the way pupils react to light. As a person gets older, their pupils have greater difficulty widening to cope with dim light, it says.
The company suggests that an age-verification system could take mug shots of a person from a set distance in controlled lighting, using a flash. Software would then measure the size of their red-eye dots to determine how wide their pupils are and make an estimate of their age.
I wonder if a picture of an older person with the red eyes in would fool such a sampling.
The point was not about the natural progess of things or whether or not we are the ones making choices, but rather if microsoft were to 'conquer the internet' (a highly deluded concept at its very best), information would move much less freely than it does now. And, if the church had somehow 'conquered' the printing press, it would certainly have used such a victory to supress information (which wouldn't have stopped change, true). It really wasen't my intention to compare the social setting now to europe then - you are quite correct in that the two situations are very different.
That and the link pointed to an article with a pic of steve ballsmer in which he looked nearly as psychotic as he talks.
"In the upcoming years we'll conquer the Internet."
If MS does manage to 'conquer the internet', that would be like the Catholic church successfully conquering that irritating 'printing press' when it first showed up. After, it was being used to print unauthorized material that was distributed by a network of individuals via unauthorized channels, worst of all information critical of the holy mother church. The horror.
The more they tighten their grasp, the more of the internet will slip through their fingers....
Actually, 'vista' will have been microsoft's first in-house ground-up build of an operating system since DOS. Dave Cutler and his team built NT/2K/XP for ms, and he would tell MS execs to drop certain feature requests when he though it was interfering the the delivery process for NT - there's nobody at MS to do that now.
Probably better to remain silent than to speak and remove all doubt....
Perhaps just as interesting is that the more they do this kind of thing, the more dysfunctional windows becomes. Their XP dns client also caches failed lookups, which is not in keeping with either the letter or the spirit of the RFC. At first, this doesn't seem like much. However, if you VPN into a network that makes use of private IP addresses, some lookups will initially fail, and the failed results get cached. The failed lookups take precedence over entries in the hosts file. So, when we had people VPNing in and failing to get to internal web sites, we were confounded by this devient behaviour. In the end a script had to be pushed out to laptops to flush the dns cache, so normal lookups would work again. Had their DNS client implementation just stuck to the specs, everything would have worked normally. This nonsense is as likely to backfire as it is to actually help them.
The moral of the story is that the more they tighten their grasp, the more functionality slips through their fingers.
I think you're right, they probably could buy off some of the telcos, and arguably the largest ones. There are at first glance, however, two significant problems with this.
First, they will have to out-spend everyone else in order to do it. They might be able to pull this off at first, but it will get more and more costly in time, and it will cost them more than it does now to move the same amount of content. So, they are in fact guaranteeing that their costs will rise artifically. However, they may be able negotiate smaller increases with their own providers. The other half of that problem is that other telcos with whom Disney isn't a direct customer will most likely favor their own customers above a non-customer. Its that way now. Disney's 'prioritized' traffic will have to take a back seat on networks where a competitor is a direct customer and has an established busisness relationship with that telco. The cost to get their traffic practically prioritized every on the Internet will marginally increase as they buy their way into networks logically further away, and they'll end up in either a bidding war, or in court with competing content providers to see whose traffic gets the best treatment. Yet more $$ thrown at what is essentially a non-problem artifically created entirely by telcos, who then charge for a solution they can't really provide.
Secondly, prioritizing traffic Internet-wide, besides being impractical, will ultimatly carve up the Internet itself, leading to an Internet-wide state of diminished functionality. The reason is that telcos are trying to understand the Internet in terms of how they been doing business the past 50 years (can you blame them?).
The very concept of the Internet is too new for them. I've seen telecommunications people struggle with the notion on an internet of networks (before the internet was the size it is now), all freely allowing traffic to flow amonst each other to everyone's benefit, designed in fact to acheive that. Its confounding to them. I saw one teleco guy, a top telco engineer with a prominent fortune-100 global IT company, literally bury his head in his arms trying to understand what was meant by "allowing your system to make a connection anywhere on the 'internet', anywhere in the world". "What do you mean, some kind of global network that everyone is connected to and can just talk to each other? Just like that?". "yes". "How the hell does that work?" (pretty close to direct quotes. I don't think that company is on the fortune 100 list anymore).
What telcos are after is a return to the pre-internet state of things, not by deliberate intention, but by striving to preserve their business model through legislation. If they succeed, the content providers, even Disney, will eventually feel it. After all, no matter how big you are, you can't buy the Internet - just parts of it.
To put it another way, telcos are only familiar with lakes with canals between them. The internet is really more like the sea. Disney thinks they can buy good weather and smooth sailing.
It would seem Disney is confused about what it want, or who its friends are. Back in November of 2002, Disney sent a letter to the FCC asking "transmission network operators do not encumber relationships between their customers and destinations on the network."
Recently: "Walt Disney CEO Robert Iger weighed in on the network neutrality debate Monday with an opinion guaranteed to please his hosts here at the TelecomNext show -- in that he doesn't think any new legislation is needed."
Now, Disney wants to be a content provider, yet is siding with the telecoms in an endeavor that will ultimatly hurt content providers by trying in vain to prioritize selected traffic for selected content providers and consumers. Whoever it was at Disney that understood that trying to guarantee that a class of traffic gets prioritized fairly throughout all routers running the internet was virtually impossible must have left.
It would seem Disney want to feed them selves with one hand, and stab themselves with the other. What a Mickey Mouse operation.
So, will this mean that the Chinese government will officially let MS spy on them? This would certainly help. I can't imagine the chinese government being at all comfortable with that prospect. You'd think they'd develope their own distro with wine already set up to run those windows apps they'd think there to be a need for.
I just have a question. If someone comes into your house to service your furnace, for example, and they 'happen' to see pot growing next to the furnace, can they call the cops? Is that admissable evidence in a court of law? Because, MS is clearly planning on doing just that.
Of course, they don't have permission to rifle through every nook and cranny in my house just because they came in to fix the furnace...
The idea behind suggesting filtering was not to remove the traffic all at once, but as a way give dlink a reason to do NTP the right way. That's assuming that dlink catches on that ntp doesn't work for their products anymore, which they may not notice at all. I would guess that there's a good chance that dlink wouldn't care.
I'd think they could just firewall off just their ntp servers, and only allow certain networks in - their networks. Of course, it wouldn't be open anymore, but with PHBs trolling around like daleks, opening things up the general internet public is getting more and more difficult.
AT&T runs portions of the Internet backbone, and traffic from other countries can go through their network as well, like when computers in China go to microsoft's windows update site. Also, as a backbone provider, switching from one ISP to another may not keep your traffic from going through their network. Do a traceroute to various destinations, and its highly likely that no matter your ISP, you'll go through AT&T's network at some point. Even from another country.
The only viable way to keep traffic off of AT&T's network is for other backbone providers to refuse to route traffic through AT&T, and get alternative peering agreements up with other BB providers. This may not be a viable option, however, since AT&T carries enough traffic volume for the Internet that to effectively 'kick them off' the Internet may cause other BB providers to experience very heavy traffic loads.
If I was the government of a non-US country, I'd be canceling AT&T contracts today, given that AT&T did this on the sly.
"Although Bush and his top advisers have said that Earth is warming and human activity has contributed to this, they have questioned some predictions and caution that mandatory limits on carbon dioxide could damage the nation's economy."
Of course, the cost of doing nothing is much lower in the long run.
Yes, this is bad news for regular users, but its also bad for the big telcos. That's because if they start trying to sell traffic prioritization to people, they'll end up with egg on their face due to the very nature of the Internet, and everyone will lose. Regular customers will just lose first, but I think telcos will lose later.
The reason is that telcos think only in terms of their own networks, not in terms of the internet as a whole. For example, suppose I want to go to google video and so does Joe in Iowa. If Joe and I are both are customers AT&T, for example, and we both purchase some kind of fast streaming (steaming ?) video service from AT&T, and Google has direct uplink to AT&T, then we both will get faster video downloads. However, if Joe's traffic ever traverses another network like UUNet, then the fast steaming video service Joe paid for won't be so fast. Unless, that is, AT&T and Verizon/MCI (UUNet) have an agreement to honor each other's traffic prioritization.
Here's where it gets interesting. What if Verizon sells the same traffic prioritation to its customers? Are we to believe that Verizon will treat AT&T's 'prioritized' traffic with the same expediency as their own high-priority steaming video traffic? I think not. The interesting thing is that it doesn't matter if Joe is an AT&T customer or not - the chances of his traffic traversing non-AT&T link somewhere on the internet are pretty good, since there are steaming video providers all over the place, not just on AT&T's network.
The end result is that telcos may sell something to customers that they can't deliver, due to the nature of the Internet. What will happen in time, without 'net neutrality', is that telcos will try to re-engineer their networks to reduce the chances that their customers' traffic will ever traverse other provider's networks out on the internet.
Who will scream first will be business customers. They'll insist on SLAs when paying extra for 'prioritized' traffic, and SLAs nearly always include rebate clauses when things go wrong, and things will go wrong until the internet gets all partitioned up (and functionaly broken). My place of work hosts many hundreds of large commercial web sites, and I'll for sure enforce rebate clauses when the content we pay to have 'prioritized' doesn't move with the specified urgency. And, yes there are ways to determine how to measure whether or not traffic like steaming video is getting the performance promised in SLAs. I think what will happen is that big telcos will be at each other's throats for failure to honor each other's traffic prioritizations.
The Internet is an ocean, not a bunch of lakes. The telcos want to sell good weather and calm seas.
The only thing a 'tiered' internet will result in is poorer service to people who don't pay for 'prioritized' traffic - that you can bet on. Once that becomes apparent, of course people will start coughing up extra dough, and telcos will get a temporary boost to their bottom line. Of course, that is, until the internet starts to break down as telcos start to partition up the ocean into nice, managable lakes.
Slashdot Mirror
Speaking of reality, here's how prioritization is likely to go:
ISP sells higher tier of service to corporation A, so content to and from their sites gets to the endpoints more quickly.
As with any such contract, corporation A also include an SLA in the purchase, with performance measures. That SLA, like most, will have rebate clauses, just like they do now.
However, the ISP can't speak for all the other ISPs, in particular the ones at the edges. Since corporation A most likely is buying its internet connection from a backbone provider, the higher tier of service ends up making a little bit of difference, but not much due to the fact that QOS has to be coherently configured end to end to make a consistant difference.
Performance metrics for improved service are, predictably, not met. Corporation A realizes it has paid for nothing, and has been had by its ISP. Corporation A begins enforcing rebate clauses. ISP gets less for its 'tier' of service than it hoped, if anything at all.
If there's a dose of reality, its that residential users never had impact in the first place, and that prioritization will just set backbone providers at each other's throats more than they are now.
As for the Internet ocean analogy, it holds accurate until each and every ISP on the Internet sets up filtering and QOS the same - that's just the nature of QOS. Until then, it may as well be an ocean from the point of view of ISPs trying to sell traffic prioritization. Missing from the discussion is an understanding of how QOS works.
I think a 'tiered' internet is trouble from the start, but what about this scenerio: Your VOIP provider starts providing 911 service, and your 911 call gets squashed by your neighbor's video download. Under strict 'net neutrality', it is possible for this to happen, if unlikely.
Additionaly, the ability of backbone providers to influence the delivery of packets is quite limited in comparison to the 'last mile' provider. The ISP customers immediately connect to, if they choose to set QOS for some type of service from some content provider, will have a great deal more effect on download/upload speeds that backbone providers. That's just how QOS out at the edge works. Yes, backbone providers can influence packet delivery, but not nearly as much as edge providers.
The other problem with allowing provider to prioritize traffic is that once packets traverse provider boundries, all bets are off. Does anyone really think that Verizon/MCI/UUNet will treat AT&T's prioritized packets better or even on par with its own? After all, Verizon's own customers, like maybe giant-company-xyz, is paying to have their traffic prioritized, and all Verizon might have with AT&T is an aggreement that might not be worth as much as $$ from giant-company-xyz. If AT&T never sees all the router configs in Verizon's network, how can they claim that Verizon isn't honoring their QOS?
The internet is more like an ocean than it is a bunch of lakes and canals, and the telcos want to sell good weather and smooth sailing. AT&T will sell Disney, for example, a 'higher tier' of service for their streaming video on their backbone, but unless they can get each and every edge provider to go along, and each and every other entity that runs any kind of peering link at all on the Internet, it won't make as big a difference as they claim. My point is that even if telcos sell prioritization, its likely it won't stack up like they claim, due to the nature of the Internet itself. Then everybody will have to decide how to treat legitimate priority traffic, like 911 for example.
The entire debate looks to me as though it being framed in a misleading way.
"... I'm not sure that the markets are as worried about this as Slashdot readers are."
/. mood about msft's EU antitrust problems ...
I don't know if worried is the term that best characterizes the general
Kodak's patent mentions previous research suggesting a correlation between age and the way pupils react to light. As a person gets older, their pupils have greater difficulty widening to cope with dim light, it says. The company suggests that an age-verification system could take mug shots of a person from a set distance in controlled lighting, using a flash. Software would then measure the size of their red-eye dots to determine how wide their pupils are and make an estimate of their age.
I wonder if a picture of an older person with the red eyes in would fool such a sampling.
Would that include extinction of species with inadequate immune systems?
The point was not about the natural progess of things or whether or not we are the ones making choices, but rather if microsoft were to 'conquer the internet' (a highly deluded concept at its very best), information would move much less freely than it does now. And, if the church had somehow 'conquered' the printing press, it would certainly have used such a victory to supress information (which wouldn't have stopped change, true). It really wasen't my intention to compare the social setting now to europe then - you are quite correct in that the two situations are very different.
That and the link pointed to an article with a pic of steve ballsmer in which he looked nearly as psychotic as he talks.
"In the upcoming years we'll conquer the Internet."
If MS does manage to 'conquer the internet', that would be like the Catholic church successfully conquering that irritating 'printing press' when it first showed up. After, it was being used to print unauthorized material that was distributed by a network of individuals via unauthorized channels, worst of all information critical of the holy mother church. The horror.
The more they tighten their grasp, the more of the internet will slip through their fingers....
I wonder if it was developed in an underground lair? Using magma?
So the fox appoints a cat to protect the hens from the fox...
No doubt true. My point was that MS is building it in-house, something they haven't done since DOS. Well, I suppose WinCE might count.
Actually, 'vista' will have been microsoft's first in-house ground-up build of an operating system since DOS. Dave Cutler and his team built NT/2K/XP for ms, and he would tell MS execs to drop certain feature requests when he though it was interfering the the delivery process for NT - there's nobody at MS to do that now.
Probably better to remain silent than to speak and remove all doubt....
Just remember what the acronym stands for: End User Loses Always.
Perhaps just as interesting is that the more they do this kind of thing, the more dysfunctional windows becomes. Their XP dns client also caches failed lookups, which is not in keeping with either the letter or the spirit of the RFC. At first, this doesn't seem like much. However, if you VPN into a network that makes use of private IP addresses, some lookups will initially fail, and the failed results get cached. The failed lookups take precedence over entries in the hosts file. So, when we had people VPNing in and failing to get to internal web sites, we were confounded by this devient behaviour. In the end a script had to be pushed out to laptops to flush the dns cache, so normal lookups would work again. Had their DNS client implementation just stuck to the specs, everything would have worked normally. This nonsense is as likely to backfire as it is to actually help them.
The moral of the story is that the more they tighten their grasp, the more functionality slips through their fingers.
I think you're right, they probably could buy off some of the telcos, and arguably the largest ones. There are at first glance, however, two significant problems with this.
First, they will have to out-spend everyone else in order to do it. They might be able to pull this off at first, but it will get more and more costly in time, and it will cost them more than it does now to move the same amount of content. So, they are in fact guaranteeing that their costs will rise artifically. However, they may be able negotiate smaller increases with their own providers. The other half of that problem is that other telcos with whom Disney isn't a direct customer will most likely favor their own customers above a non-customer. Its that way now. Disney's 'prioritized' traffic will have to take a back seat on networks where a competitor is a direct customer and has an established busisness relationship with that telco. The cost to get their traffic practically prioritized every on the Internet will marginally increase as they buy their way into networks logically further away, and they'll end up in either a bidding war, or in court with competing content providers to see whose traffic gets the best treatment. Yet more $$ thrown at what is essentially a non-problem artifically created entirely by telcos, who then charge for a solution they can't really provide.
Secondly, prioritizing traffic Internet-wide, besides being impractical, will ultimatly carve up the Internet itself, leading to an Internet-wide state of diminished functionality. The reason is that telcos are trying to understand the Internet in terms of how they been doing business the past 50 years (can you blame them?).
The very concept of the Internet is too new for them. I've seen telecommunications people struggle with the notion on an internet of networks (before the internet was the size it is now), all freely allowing traffic to flow amonst each other to everyone's benefit, designed in fact to acheive that. Its confounding to them. I saw one teleco guy, a top telco engineer with a prominent fortune-100 global IT company, literally bury his head in his arms trying to understand what was meant by "allowing your system to make a connection anywhere on the 'internet', anywhere in the world". "What do you mean, some kind of global network that everyone is connected to and can just talk to each other? Just like that?". "yes". "How the hell does that work?" (pretty close to direct quotes. I don't think that company is on the fortune 100 list anymore).
What telcos are after is a return to the pre-internet state of things, not by deliberate intention, but by striving to preserve their business model through legislation. If they succeed, the content providers, even Disney, will eventually feel it. After all, no matter how big you are, you can't buy the Internet - just parts of it.
To put it another way, telcos are only familiar with lakes with canals between them. The internet is really more like the sea. Disney thinks they can buy good weather and smooth sailing.
It would seem Disney is confused about what it want, or who its friends are. Back in November of 2002, Disney sent a letter to the FCC asking "transmission network operators do not encumber relationships between their customers and destinations on the network."
Recently: "Walt Disney CEO Robert Iger weighed in on the network neutrality debate Monday with an opinion guaranteed to please his hosts here at the TelecomNext show -- in that he doesn't think any new legislation is needed."
Now, Disney wants to be a content provider, yet is siding with the telecoms in an endeavor that will ultimatly hurt content providers by trying in vain to prioritize selected traffic for selected content providers and consumers. Whoever it was at Disney that understood that trying to guarantee that a class of traffic gets prioritized fairly throughout all routers running the internet was virtually impossible must have left.
It would seem Disney want to feed them selves with one hand, and stab themselves with the other. What a Mickey Mouse operation.
So what it boils down to is that priates will get a faster version of vista (if it ever ships) than people suckered into paying for it.
The customer is always right after all - right where we want them.
So, will this mean that the Chinese government will officially let MS spy on them? This would certainly help. I can't imagine the chinese government being at all comfortable with that prospect. You'd think they'd develope their own distro with wine already set up to run those windows apps they'd think there to be a need for.
I just have a question. If someone comes into your house to service your furnace, for example, and they 'happen' to see pot growing next to the furnace, can they call the cops? Is that admissable evidence in a court of law? Because, MS is clearly planning on doing just that.
Of course, they don't have permission to rifle through every nook and cranny in my house just because they came in to fix the furnace...
The idea behind suggesting filtering was not to remove the traffic all at once, but as a way give dlink a reason to do NTP the right way. That's assuming that dlink catches on that ntp doesn't work for their products anymore, which they may not notice at all. I would guess that there's a good chance that dlink wouldn't care.
I'd think they could just firewall off just their ntp servers, and only allow certain networks in - their networks. Of course, it wouldn't be open anymore, but with PHBs trolling around like daleks, opening things up the general internet public is getting more and more difficult.
AT&T runs portions of the Internet backbone, and traffic from other countries can go through their network as well, like when computers in China go to microsoft's windows update site. Also, as a backbone provider, switching from one ISP to another may not keep your traffic from going through their network. Do a traceroute to various destinations, and its highly likely that no matter your ISP, you'll go through AT&T's network at some point. Even from another country.
The only viable way to keep traffic off of AT&T's network is for other backbone providers to refuse to route traffic through AT&T, and get alternative peering agreements up with other BB providers. This may not be a viable option, however, since AT&T carries enough traffic volume for the Internet that to effectively 'kick them off' the Internet may cause other BB providers to experience very heavy traffic loads.
If I was the government of a non-US country, I'd be canceling AT&T contracts today, given that AT&T did this on the sly.
"Although Bush and his top advisers have said that Earth is warming and human activity has contributed to this, they have questioned some predictions and caution that mandatory limits on carbon dioxide could damage the nation's economy."
Of course, the cost of doing nothing is much lower in the long run.
"You can give your heart to Jesus, but your ass belongs to the Core!"
Yes, this is bad news for regular users, but its also bad for the big telcos. That's because if they start trying to sell traffic prioritization to people, they'll end up with egg on their face due to the very nature of the Internet, and everyone will lose. Regular customers will just lose first, but I think telcos will lose later.
The reason is that telcos think only in terms of their own networks, not in terms of the internet as a whole. For example, suppose I want to go to google video and so does Joe in Iowa. If Joe and I are both are customers AT&T, for example, and we both purchase some kind of fast streaming (steaming ?) video service from AT&T, and Google has direct uplink to AT&T, then we both will get faster video downloads. However, if Joe's traffic ever traverses another network like UUNet, then the fast steaming video service Joe paid for won't be so fast. Unless, that is, AT&T and Verizon/MCI (UUNet) have an agreement to honor each other's traffic prioritization.
Here's where it gets interesting. What if Verizon sells the same traffic prioritation to its customers? Are we to believe that Verizon will treat AT&T's 'prioritized' traffic with the same expediency as their own high-priority steaming video traffic? I think not. The interesting thing is that it doesn't matter if Joe is an AT&T customer or not - the chances of his traffic traversing non-AT&T link somewhere on the internet are pretty good, since there are steaming video providers all over the place, not just on AT&T's network.
The end result is that telcos may sell something to customers that they can't deliver, due to the nature of the Internet. What will happen in time, without 'net neutrality', is that telcos will try to re-engineer their networks to reduce the chances that their customers' traffic will ever traverse other provider's networks out on the internet.
Who will scream first will be business customers. They'll insist on SLAs when paying extra for 'prioritized' traffic, and SLAs nearly always include rebate clauses when things go wrong, and things will go wrong until the internet gets all partitioned up (and functionaly broken). My place of work hosts many hundreds of large commercial web sites, and I'll for sure enforce rebate clauses when the content we pay to have 'prioritized' doesn't move with the specified urgency. And, yes there are ways to determine how to measure whether or not traffic like steaming video is getting the performance promised in SLAs. I think what will happen is that big telcos will be at each other's throats for failure to honor each other's traffic prioritizations.
The Internet is an ocean, not a bunch of lakes. The telcos want to sell good weather and calm seas.
The only thing a 'tiered' internet will result in is poorer service to people who don't pay for 'prioritized' traffic - that you can bet on. Once that becomes apparent, of course people will start coughing up extra dough, and telcos will get a temporary boost to their bottom line. Of course, that is, until the internet starts to break down as telcos start to partition up the ocean into nice, managable lakes.
Well, it was interesting while it lasted.
He gets others to work for him. That's why he's rich.