Neutral Net Needs Twice the Bandwidth of Tiered

berberine writes with a link to Ars Technica, straight to an article discussing the differences between a net neutral internet and one that supports tiers of content. As you might imagine, our neutral internet is far more bandwidth-intensive; AT&T estimates it might require as much as twice the bandwidth of a tiered internet. From the article: "Corporate sponsorship of research doesn't automatically invalidate that research; what's needed is a close look at the actual results to determine if they were done correctly. According to David Isenberg, a long-time industry insider and proponent of 'dumb' (neutral) networks, the research itself is fine. In his view, it's simply obvious that a dumb network will require more peak capacity than a managed one. But extending that banal observation to make the claim that running a managed network is cheaper is, to Isenberg, not at all intuitive. For one thing, doubling the peak volume of a network does not mean spending twice as much money as it cost to build the original network."

Comments

And

[Umbral Blot]

And sometimes it is worth pursuing an outcome that is not maximally effecient for other reasons, a fact that people seem to overlook sometimes. So what if the internet is half as fast as it could be; that is an acceptable trade-off for a free and open internet.

Re:And

[ArcherB]

And sometimes it is worth pursuing an outcome that is not maximally effecient for other reasons, a fact that people seem to overlook sometimes.

Yeah, like the services provided by those that can't afford to bribe AT&T don't choked off! Even if they do bribe AT&T, if they don't bribe other line carriers, like say, Time Warner and Comcast (or whoever owns the wires), then AT&T's bandwidth is still going to be lower because of the choked traffic coming off the other lines. Traffic is only going to be as fast as the slowest segment.

Re:And

[morgan_greywolf]

Even then, let's suppose you were willing to accept a tiered Internet. How you tier makes a difference in whether it is maximally efficient for a given application. The reason we used managed networks in a corporate environment is because of corporate priorities -- financial transactions are more important than e-mail, so we segment off financial transactions and then give those transactions that must run over the same network as e-mail a higher priority over the e-mail, for instance.

The question is: How do we decide what traffic is more important on the Internet? Who pays? Who pays more? That's stupid. The benefits of a a free and open Internet far outweigh the inefficiencies of working with a basically unmanaged network. (Not that the Internet actually is completely unmanaged -- that's just not true. ISPs shape traffic on their own networks to improve customer connectivity to mail or webservers within the ISP's own network). The point of the Internet is to have a network where anything is possible. Tier it off and you'll make it about as useful as the television networks.

Re:And

[Hatta]

I just don't buy it. A neutral internet would transfer just as many bits as a prioritized one. The only way for a prioritized network to be better is if some of those bits are more important than others. If you assume that, then you're just begging the question.

Re:And

[UbuntuDupe]

Yeah, but you have to be sure you're not just pushing the unfreeness onto another aspect. Let's take the simplest kind of NN, where literally every packet has to be treated the same. (Some people want this. Not necessarily informed people, but whatever.) Then someone can just send packets non-stop, and since they're sending more packet transfer requests (whatever those are called) everyone else has to get in line behind those "dummy" requests. It then becomes a competition to see who can flood the network the most in order to get through. This is sort of similar to the price cap/shortage problem. Keep the price too low, and the good gets rationed by aggressiveness, like a PS3 or Wii at launch.

And take a guess who's best at flooding the internet with data they don't really need to send?

Re:And

[Saint Fnordius]

To put it differently, unmanaged traffic where the drivers get to decide which road they use themselves is less efficient than a traffic net where a central authority dictates to you which highway you're allowed to take. Of course an unmanaged net needs more throughput capacity overall, but in exchange the traffic doesn't require micromanaging. Part of why highways and trucks beat out rail service is because of that flexibility, of not being at the mercy of the switching stations and schedules.

Or consider an irrigation network with multiple sources and multiple outlets. You could either build all the pipes so that any of them could deliver maximum capacity, or have workers actively controlling the valves to distribute the water across the entire net so that one side doesn't overload. The latter solution doesn't require as robust a pipe, but requires a more complex valve system and somebody controlling it.

Re:And

[ghyd]

"So what if the internet is half as fast as it could be" Is it ? coming from a country where my ISP is very cheap, reliable, honest and has a lot of services (like tivo for free with the providen Adsl set top box, or your free very own TV channel to broadcast whatever you want, free phone calls for 50 countries, etc, etc), with extremely consistent 16mbps down 1mbps up in a town of 40.000, I have never noticed a problem with Internet speed. I don't now much, but isn't this debate just a side effect of the very bad ISP companies that US have? like, "oh, Internet is so bad, where's the problem from? neutrality!" rather than just stupid and unreliable ISPs, who cap everything, don't have a so great technical expertise, and don't even provide the (low) advertised speeds to begin with ?

Re: Re:And

[rnturn]

The point of the Internet is to have a network where anything is possible.

Heh, heh. I can remember when the phone companies wouldn't allow modems because (it seemed to those of us who used them, anyway) it allowed you to do things that the phone company hadn't thought of. "Sending bits across voice lines? NO! You'll have an expensive leased line installed if you want to do that. And you'll lease equipment from us, too. Or we'll cut off your service!"

Tier it off and you'll make it about as useful as the television networks.

You've hit the nail on the head. That's the model the phone companies are trying to emulate. It explains their ridiculous subscriber plans that include "Content by whoever".

I'm not at all surprised at the difficulty that the phone companies are having with the Internet. They had to be dragged -- kicking and screaming -- into accepting packet switched networks in the first place. My guess is that an entire generation of managers (or two) at these companies need to retire before we'll see anything like a basic understanding of the Internet in these companies' actions.

Re:And

[jack455]

Good post, except the implication you make here: "So what if the internet is half as fast as it could be; that is an acceptable trade-off for a free and open internet."
I even agree with the sentiment, but the study incorrectly implied or stated that doubling the peak capacity would double the costs. Even if that were true, not doubling the peak capacity would _not_ halve the "speed" of the internet. For what it's worth here's a selected quote from Ars quoting Isenberg, commenting on the study.

...
doubling the peak volume of a network does not mean spending twice as much money as it cost to build the original network. "The failure of the authors to extend the conclusions from capacity to raw costs of capacity is deliberately misleading," Isenberg says, "especially when the researchers invoked 'economic viability' and 'cost of capacity' in their introduction to the work." ...
According to Isenberg, the cheapest and best alternative is simply to build out dumb capacity: to "overprovision" by as much as 100 percent.

Re:And

[datapharmer]

That's a horrible analogy - It just doesn't hold up. The trains are actually far more efficient than highways as a system, but since the U.S. spends more on cleaning roadkill of the highways than total funding for Amtrak they don't have enough staff or working trains so things get off schedule which causes the entire system to break. Just because the U.S. can't manage a switched system doesn't mean it is bad... try telling someone in France that the highway is faster and more efficient than the train!

Unmanaged networks are inefficient and pointless. There is no damage in routing things to avoid network congestion, but tiered networks are bad too. A tiered network is like a toll road that has restricted carpool type lanes, but the number of passengers doesn't matter - how much you pay in tolls does.

Re:And

[dpilot]

> that is an acceptable trade-off for a free and open internet.

You make 2 mistakes:
First, that a free and open internet is desirable.
Second, assuming it is desirable, that it's more important than near-term profits.

I certainly wouldn't argue either point with you. In principle and in public, I don't think anyone would either, including government regulators. In practice and in private, I suspect that both arguments are toast.

Re: Re:And

[arivanov]

My guess is that an entire generation of managers (or two) at these companies need to retire before we'll see anything like a basic understanding of the Internet in these companies' actions.

I think you are somewhat mistaken here. They clearly see where the quick revenue opportunity lies in the current internet and this is all they are interested in - quick revenue without any further capex. They have all the understanding required for this one. The fact that it may kill long term revenue opportunities is not really relevant here. They are not interested in that and that has nothing to do with understanding of the internet. It has to do with actual way corporations function today.

Re:And

[Anonymous Coward]

I am confused by the purpose of this whole thing anyway.

Bandwidth is not simply "used" or some arbitrary quantity of data flying around aimlessly. It is a product of end users and computers transferring data that was requested by them. It was requested because the person needs it or wants it. How is tiering or managing the bandwidth going to change what people are doing and what they want to do and see online?

I view this like the distribution of gasoline in America. What ever method we use to distribute gas around the country has no direct effect on the quantity of gasoline we use on a daily basis. Managing the gas distribution process in a more efficient manner does not imply that people will use less gas.
What would have an effect is the price of the bandwidth and the gasoline. The price is not a direct result of the distribution management style either. Sounds like they are twisting the story to make it look like it is not the price they want to change, just the management of it. If you take away the supply, the price will go up. That is the goal here although it is disguised.

Re:And

[aaronl]

Trains and highway have different efficiencies, though. A train is an excellent way to move a lot of something a long distance. Highways are excellent for non-linear, lowest time transit, or local distribution. I couldn't take a train to my home, for example. I would need to take a taxi, bus, or personal car to get there from the train station. I couldn't take a train to work, since the time lost getting to the train, getting on the train, getting off the train, and getting from the station to work far exceeds the travel time to just drive.

Going between cities is where trains are the most useful. Moving about inside, or around, a city is where the highways are most needed. Rural areas, and there a lot of them in most every country, still need highways nearly all travel.

The unmanaged system of highways allows for all of the same things as trains, though less efficiently, but also allows for *substantially* more freedom of movement and independence of travel time. The right answer, as it always has been, is to use both.

BTW - it isn't just Amtrak that has problems in the US. Nearly all public transit systems are doing their best to approach complete uselessness. It is still faster and less expensive, for me to own, insure, and operate a car where I live than it is to use public transit. This is in metrowest Massachusetts, for reference. NYC is better, but the subway is still no picnic, and light rail can be hell there, too, but it's still a lot better than driving, usually.

For what it's worth, if rail was the better option in the US, business would use it more. As it turns out, you get more for your money by moving things around with trucks and planes. Transit times are much lower, and you can deal with changes in volume and the need to reroute things much more easily.

Re:And

[the eric conspiracy]

Some bits are more important. A UDP packet that is part of a RTP session when dropped may result in an unacceptable quality 911 phone call. A TCP packet part of a HTTP session will just get resent.

Re:And

[Retric]

Bandwidth is cheep, fiber is cheep, networks are cheep and they are only getting cheaper.

The silly thing is all this network neutrality talk / bribery has cost the AT&T / Cox / etc more than doubling the current available US network backbone would. Ahh well let the old system rot and soon enough new players are going to take over.

Re:And

[Sandbags]

According to studies, the most common traffic on the internet is torrent style P2P network traffic, and movie downloads. Soon we'll also be looking at IPTV as a massive bandwidth source. IPTV absolutely CANNOT be a second tier application or it will never succeed. IPTV being the centerpiece of most large ISP's future are going to have to make sure that traffic is uninterrupted, providing clear and perfect TV viewing (if there's download pauses, buffering, or any other quality issues it will be a billion dollar failure), is their goal. It will get one of the top tiers. Since the second 2 highest uses behind that will be illegal or at least questionable activity, how does ATT propose to limit that without impacting IPTV? Surely a smart hacker will VERY quickly figure out hot to make their traffic look like tier 1 packets, and recompose them client side.

By installing a tiered internet, all ATT will be doing will be to improve the speed of illegal services vs legal, and put everyone else doing business, atm, mail, sms, and other critical time sensitive in one of the worst speed catagories.

Because innovation rules the net, and any packet could be made to look like any content we want, the only way to tier the net is on an IP to IP bases, not packet to packet, and this would amount to a kind of internet segregation where those who want to/can afford to pay more, go faster. A network structured that way will either be ordered dismantled by the courts, or simply have few subscribers where there are other competitive offers.

If it's cheaper to add bandwidth than it is to segregate packets anyway (and adding bandwidth is still a necessity either way), combined with the packet latency issues that would be inherent to a tiered structure (the additional time in ms necessary to analyze a packet before assigning priority), I'm guessing net neutrality will win even if the courts don't stand on our side.

Re:And

[morgan_greywolf]

According to studies funded by the Mafiaa, the most common traffic on the internet is torrent style P2P network traffic, and movie downloads.

There, fixed it for you.

Re:And

[enjerth]

I believe this is the whole point of the proposed tiered internet. Because providers have grossly oversold available bandwidth they're looking for ways to cover it up.

So of course "Net Neutrality" requires additional bandwidth.

Re:And

[Conspiracy_Of_Doves]

If the internet is to be tiered, the tiering should be based on the technology involved, not the organization that the packets are traveling to or from. For instance, VOIP should have precedence over web traffic, but all web traffic should be treated the same, and all VOIP traffic should be treated the same.

Re:And

[teh_chrizzle]

The benefits of a a free and open Internet far outweigh the inefficiencies of working with a basically unmanaged network.

you mean the benefits to the consumer. this in no way benefits AT&T. if the internet was just a big dumb pipe then anyone could get into the internet service business by simply connecting lines to backbones, which would be disastrous for AT&T.

a wide open, data-centric, unmanaged network might be perfect for network applications, but it would be a killing blow to the profit margins currently enjoyed by the industry. a complex and over managed network is far more profitable and far less likely to foster competition, and is therefore more far more desirable to the likes of AT&T.

he question is: How do we decide what traffic is more important on the Internet?

the answer is simple: the highest bidder has the most important traffic. if you want your traffic to be important, outbid the highest bidder.

Who pays? Who pays more?

again, very simple: everyone. everyone will pay. then they will pay more. after all, they own the net.

The point of the Internet is to have a network where anything is possible.

that may be, but the point of internet service is to make as much profit as possible while taking as little risk as possible, and investing as little capital as possible.

net neutrality is a symptom of competition. there is very little competition in the telco and cableco industries so there is very little net neutrality. if there were sufficient competition in the market, net neutrality would not be a debate, it would be a fact of life.

Re:And

[Maniac-X]

The net neutrality talk isn't making things cheaper, fiber is just becoming more proliferate and easy to manufacture. Overall bandwidth is increasing due to sheer inevitability. The internet will be ruined if it becomes tiered. It stands to put smaller ISPs out of business, and increase net cost for the end user, thus lining the pockets of the major ISP execs... sorry, but that's just not something most people want to see. Let's not turn the Internet into RIAA/MPAA v2.0.

Re:And

[pnotequalsnp]

Dear AT&T,

looks like you need 2x the bandwidth, "get on with it" (Monty Python)

Sincerely,

Net Neutrality

Re:And

[Remus Shepherd]

You can put it even simpler by using an analogy:

In many ways, Fascism is much more efficient than a Democracy. But that doesn't mean it's the best choice.

Re:And

[Anonymous Coward]

The silly thing is all this network neutrality talk / bribery has cost the AT&T / Cox / etc more than doubling the current available US network backbone would.

Re:And

[dodobh]

If road is faster for you, then you simply don't have enough rail infrastructure. If owning a private vehicle is better for you in an urban scenario, then your cities are misplanned. I suspect that paying for carbon credits would make our life interesting.

Mass transit works very well, as demonstrated in Europe and India (6 million commuters travel daily by local trains in Mumbai, that's the total public transit using population of NYC. Note that the non-train-feed bus using commuters are an equal number).

Re:And

[L0rdJedi]

Because innovation rules the net, and any packet could be made to look like any content we want, the only way to tier the net is on an IP to IP bases, not packet to packet, and this would amount to a kind of internet segregation where those who want to/can afford to pay more, go faster. A network structured that way will either be ordered dismantled by the courts, or simply have few subscribers where there are other competitive offers.

This is already the case though. People who can afford 10 or 100 mbit links to the net, which go faster, pay more. People who can't (or don't want to) get 8 meg cable all the way down to 768k DSL. This system hasn't been dismantled by courts or had anything really done to it. There's no reason to. If you want a faster connection, you pay for it, period.

Re:And

[Phisbut]

I just don't buy it. A neutral internet would transfer just as many bits as a prioritized one.

What I don't get is, from what I understand, a tiered internet is only screwing America. AT&T can charge all they want for the last mile, but with more and more of the Internet being installed elsewhere (and I bet more will come if it gets tiered in America), and more and more users not in America, Europeans, Asians and Australians will get excellent connections to servers in Europe, Asia and Australia, while Americans will have a crappy connection to just about anywhere (although one could argue that's already the case).

Pass all the laws you want in America, you'll soon have to accept that the Internet is no longer a US-only network.

P.S. What sucks though is that Canada will probably get sucked into whatever crapiness the US gets

Re:And

[GeckoX]

Yes, it really is Just That Simple.

Err, not even close actually.

In limited instances and certain cases, these points have some application. However, look at where mass public transit works well in the world. What is the common theme? Population density.

Now look at population density in North America...pockets of high density, not coincidentally these are the places where mass public transit is the best in NA. But for the most part, NA is not very dense population wise. It is simply prohibitively expensive to implement mass public transit as it has been in parts of europe and india.

Not saying it's impossible, just saying that it really is NOT Just That Easy.

Re:And

[perlchild]

And the consumer pays those costs...
AT&T sees it as a way to increase their profits, In other words, unless we boycott them, we sponsor their waste of money...
Perhaps now, any see why we could do better as an "Internet" without the largest players, at least until they're properly leashed?

Re:And

[hotdiggitydawg]

It is still faster and less expensive, for me to own, insure, and operate a car where I live than it is to use public transit. Shame - where I live, public transport is both cheaper and faster than owning your own car. Buses get priority lanes and can drive right through the city in a matter of minutes, even during rush hour. Plus parking is a real nightmare, petrol costs a fortune and . The fixed routes and timetables introduce a certain degree of inflexibility, but I can live with that - I save enough money by not owning a car that I could afford to hire one every second weekend if needed and still come out ahead.

Re:And

[aaronl]

Yes, many US cities are poorly laid out. There isn't anything that can be done about that now, so we must make due. Making due means buses and lots of personal vehicles for a lot of places.

Also, Europe doesn't have this crazy ever-present public transit system that many people seem to think. You still have to go by car if you are traveling to a smaller municipality, and you certainly need a car to get around in rural areas. Europe is just a lot smaller, and more built during times when there wasn't high speed transit available (ie: before cars).

As to the carbon credits, honestly, trains would be prohibitively expensive if I made up a new huge fee that you had to pay to run one. Carbon credits are the same thing: an arbitrary limitation that carries an arbitrary price tag. The *only* thing that would do is make things cost more. We'd still be driving cars, and we'd still have bad public transit, but the general public would just have a lot less money. All of the yahoos that say we should tax the hell out of gasoline ignore this problem, too. You don't make it fixed by making it more expensive, you just give the government more money and slow the economy.

It is fairly trivial to explain why carbon credits and the Kyoto Treaty do nothing except serve as a welfare system to third world and developing countries.

Re:And

[Carlos Laviola]

Why in the world would someone rely on a computer to dial emergency services? Think about how many things factor into placing a phone call over the internet versus a landline.

Re:And

[Fordiman]

I know what you mean. What's wrong with demanding that doubled bandwidth? As consumers and citizens, bandwidth is of high importance. We have jobs that require varying amounts of bandwidth, and a tiering system leaves out those who can't afford or aren't granted tiered status. I mean, be in the medical industry or have slow uploads? What's a developer to do? FTP takes throughput too, you know.

And debugging webpages isn't a fast process if I have to wait for someone's video surgery to pass through the tubes.

Yes, I went there.

The fact is that you can prioritize bandwidth all you want, but it won't help the impending data traffic jams we're all going to start getting stuck in. We need the infrastructure improved, not preferential treatment.

Give me 10M/s, or give me death!

Re:And

[lessermilton]

Tier it off and you'll make it about as useful as the television networks. Well, DUH!

That's exactly what they want - they don't want the end user to be an internet consumer, they want them to be a product that they can sell. Preferably if it means they make money off both ends of the bargin.

Re: Re:And

[Fordiman]

So, it's the fallacy of micromanagement; they're meddling in the rules to a degree that will harm future business - for quick profits. The quick profits let them keep their jobs and get tasty bonuses, while screwing their eventual successors.

Re:And

[Anonymous Coward]

Would you be able to walk one to four blocks to a personal rapid transit station, which would take you directly to your selected destination station?

Re:And

[Kadin2048]

Think about how many things factor into placing a phone call over the internet versus a landline.

You presume that there will always be a difference? I don't.

Re:And

[suggsjc]

I very much like the thought, but if any priority is given to any packet, then it will be exploited. Somebody will find a way to make their email packets look like VoIP packets so that they will have a higher priority.

So unless you can find a way to enforce morality to all of the end users, then there are going to be people who cheat the system (and ruin the good thoughts/concepts for everyone else). Perhaps morality isn't the best term, but you get the idea.

Re:And

[Anonymous Coward]

So you're saying the internet is a series of pipes... And here I thought it was a series of tubes.

Re:And

[Ana10g]

Hey! We have those! "Express Lanes". And guess what! NOBODY uses them! They'd rather sit in traffic than pay the extra tolls.

Re:And

[Ana10g]

I suspect that paying for carbon credits would make our life interesting.

Yup, Indulgences worked great for the Catholics! Carbon credits solve nothing, and create more bureaucracy and more waste of resources. What's needed is to fix the problem, not treat the symptoms.

Re:And

[arminw]

.....You presume that there will always be a difference?.....

There will be as long as the POTS remains basically a circuit switched system and the internet a packet system. A packet system makes more efficient use of the signal path, but a circuit switched setup delivers time critical information more reliably.

The Internet was never originally designed to deliver time critical data, such as live voice and video. It was conceived as a way to send digital data correctly even in the face of unreliable data transmission speeds and path interruptions. Data that did not make it on the first try can get resent until all the data EVENTUALLY gets to its destination. The Internet is being coerced into transmitting live voice and video even though this type of data was never part of the original basic design. To use the Internet technology for this kind of data, the "eventually" has to be turned in to a more or less "immediately". That means looking at the data and decide if it is of the "eventual" or "immediate" kind and doing the best effort to get the latter delivered, even at the expense of slowing down the former. Doing that has nothing whatever to do with "neutrality".

Re:And

[jafac]

I think another good model for this is how ethernet as a networking technology, beat out several other media-access technologies, like token-ring - specifically because of the overhead and complexity involved in keeping everything managed. Token-ring offered a nice set of diagnostic tools in the install package, along with the driver; because you NEEDED those tools on every network node. One bad node would break the ring, and then nobody could pass tokens.

Tim Berners-Lee made a good point about Network Neutrality. The open systems win out in the market place, because nobody wants to use a closed system. Take away Net Neutrality - and you will see the internet die.

Re:And

[arminw]

.......then you simply don't have enough rail infrastructure......

Before the automobile became popular in Germany, there was rail service to almost every town and even village. Over time most of these little local "Bimmelbahn" trains were either abandoned outright or replaced with buses. Every little village also had a grocery store where the normal daily things were within walking distance. Those too have largely been replaced by mega-chains that can sell the same things much cheaper and with a huge available variety. In Germany the number of reasonable neighborhood shopping opportunities still far exceeds those found in the Western parts of the US.

Compared to the American West, Germany is a very densely populated country, yet even there many people endure the hassles of using their car because it is still the lesser overall hassle in many daily routines. Especially in summer, the "Stau" (traffic jam) is the norm, rather than the exception. A moving metal carpet crawls along many sections of the Autobahn.

In the competition between rail and road a big factor often ignored by public transit boosters is that of time and convenience. Rail is definitely the most efficient way to move a given mass from point A to B but seldom the most convenient. Riding public transit while toting a weeks worth of groceries just isn't something most people enjoy. Stopping at a grocery after work is seldom an option for anyone who commutes on public transit. Therefore many forget about transit, because the need (want) to do their shopping at lunch break or after work.

In India of course, automobiles are largely unaffordable to most and roads are poor or marginal. Therefore trains are a preferred mode of transport.

Re:And

[Conspiracy_Of_Doves]

I could see someone making his email packets look like VOIP packets, but how would you get the receiving email server to recognize them, unless you are only emailing one or a few other servers that way where you have a deal with the admins of those servers.

Re:And

[doom]

Now look at population density in North America...pockets of high density, not coincidentally these are the places where mass public transit is the best in NA. But for the most part, NA is not very dense population wise. It is simply prohibitively expensive to implement mass public transit as it has been in parts of europe and india.

Not saying it's impossible, just saying that it really is NOT Just That Easy.

And what I think you're missing is that the low-density areas have essentially been created by the US government prioritizing highway funding. Suburbia only seems to make sense if you hide a lot of the costs that make it possible. If you stop hiding the costs, suburbia will at the very least stop sprawling quite as fast as it is.

Re:And

[doom]

aaronl wrote:

Yes, many US cities are poorly laid out. There isn't anything that can be done about that now, so we must make due. Making due means buses and lots of personal vehicles for a lot of places.

There's nothing that can be done about it? Absolutely nothing? How about we try revising the zoning laws so that it's not illegal to build real cities?

Re:And

[doom]

Carbon credits are the same thing: an arbitrary limitation that carries an arbitrary price tag. The *only* thing that would do is make things cost more. We'd still be driving cars, and we'd still have bad public transit, but the general public would just have a lot less money. All of the yahoos that say we should tax the hell out of gasoline ignore this problem, too. You don't make it fixed by making it more expensive, you just give the government more money and slow the economy.

So your opinion is that the law of supply and demand has been repealed? How else can you explain the notion that if it cost more to pollute people won't find ways to pollute less (e.g. by using mass transit in preference to private cars)?

What I think is going on here is that you've got it stuck in your head that suburbia is The Natural Order of Things, and no one in America will ever be induced to abandon their suburban ways no matter how much they have to pay... "you'll pry that gas spigot out of our mouths from our cold dead fingers!"

(The other problem you have, is you've decided a priori that nothing that comes out of the mouth of those left wing whackos can possibly have an ounce of truth to it...)

Re:And

[TrilateralRegression]

But that's tiering by hardware, the case in point is tiering by software.

Re:And

[Anonymous Coward]

Too many people are throwing out the baby with the bathwater on this topic.

There's nothing wrong with implementing QoS in a network. Different types of traffic have different service requirements, and the network can more efficiently and effectively meet your needs if you tell the net what those requirements are.

"Network neutrality" in this sense is a political issue. The problem appears when the QoS is applied in a biased manner, in an attempt to disadvantage particular customers. It's generally considered wrong if, say, Comcast shafts all VoIP traffic except its own. "Neutrality" is neutral to the customer, not neutral to the traffic type. VoIP traffic could receive better service with QoS -- you just want to be sure that any VoIP traffic gets the same treatment, regardless of the user, so as to avoid self-serving manipulation of the traffic.

A carrier simply must be required to provide a given service at the same rate to any customer, and disallowed from charging differential rates for the same QoS. Certainly, one degenerate way to achieve this goal is to eliminate all QoS from the network, so that it impossible to offer other services. But that's an overly extreme approach to the problem.

Re:And

[dcam]

Some bits are more important. A UDP packet that is part of a RTP session when dropped may result in an unacceptable quality 911 phone call.

In which case they should not have been using UDP. What you meant to say was a number of UDP packets dropped resulting in unacceptable quality.

Re:And

[mikael]

That's already happened with "Virgin Media". Virgin announced their new Video-On-Demand services, then at the same time announced that they would be "shaping traffic" and "capping excessive bandwidth usage hogs". If bandwidth was limited, where did all the capacity for VOD come from?

Re:And

[timeOday]

Yeah, my first response was, only twice the bandwidth? You could run a whole ISP with a lousy 1 mbit T1 10 years ago, how many times has it doubled since then? Clearly neutrality is a bargain!

Re:And

[An Onerous Coward]

I thought most VOIP systems used UDP. Do you know something the people over at Skype don't? Or am I confused about what you're trying to say?

Re: Re:And

[complete loony]

If you want a true neutral network, force ISP's to split their business into 3 separate business units.

• Network access
• Server hosting
• Services

Each layer of their business should be profitable, and be forced to pay the same wholesale rate as external competitors. These days I don't care about my ISP's email account, user web space, newsgroups servers and a number of the other services they provide. And just because they have a government granted monopoly on the last mile of network access, should not allow them to gouge me on their other services while stifling competition.

Re:And

[dcam]

No, the use UDP and it is the appropriate protocol. I guess that was a nitpick with the comment.

A brief description:
TCP - connection oriented and fault tolerant. A TCP packet is sent over an established connection. When a TCP packet is sent, the reciever confirms that the packet has been recieved.
UDP - connectionless protocol. No connection is established for packets to be sent. No confirmation is given that a packet is recieved.

Implications of this are:
- UDP can send data faster, but you are likely to lose some packets.
- TCP can send data slower, and you are unlikely to lose any packets.

All that said, there are some cases where TCP is better than UDP. If you don't want to lose a single packet (eg serving pages of HTTP, where the loss of a packet means a corrupted page) TCP is the better protocol. If the loss of a single packet does not matter than UDP is better (eg a control system which is sending constant adjustments like move left, move right, where the loss of a single command does not cause issues because there are constant adjustments).

Applying all this to VOIP, the loss of a single packet (or more) means you lose a bit of sound, which isn't much of a problem. On the other hand latency is a big issue in VOIP. UDP is better for latency (no overhead for the connection, no confirmation etc). Hence VOIP generally uses UDP (although they may have a control connection over TCP).

So my nitpick was that the loss of a single packet is to be expected over UDP, given the nature of the protocol. However the loss of a large number of packets would be an issue.

Does that all make sense?

Re:And

[aaronl]

Well, I suppose you could knock down substantial portions of US cities to build light rail. Of course, besides being kind of dumb, you also wouldn't have anyone left to ride them. If you go to the mid-west, you have "real cities", I suppose, it's just that there still isn't good public transit.

Re:And

[aaronl]

Again, just because you've decided that the price of gas should be artificially inflated does not make public transit magically work for the majority of situations. Now, if you forcibly move everyone closer together, build new cities, and provide excellent public transit, this could work. Of course, you'd also probably also have a civil war, and your plan would collapse.

What is more likely to happen is that the cost of fuel will rise. Correspondingly, the price of goods and services will increase. People will have little choice but to pay the increased costs. Eventually, you may well have more carpooling and public transit. However, you may also have an increase in base pay, as people will demand more money so that they can afford to pay for the ramifications of the artificially increased fuel prices. The trend will be for businesses to leave the US for countries that don't do your plan, the US economy will shrink, and we'll use less energy.

You see, your situation does not reduce the supply of fuel, but only increases the price. If fuel wasn't available, then you have a change. Of course, this will likely return you to warfare, as the public will still demand fuel. Destroying First World economies because you think gas being expensive will fix things is pretty short sighted. Just because Europe does it, doesn't make it right. Politics aside, Europe is settled in a wildly different manner than the US.

And actually, I've decided that very little truth comes out of the mouths of the right *or* the left. I know full well that we need to find other energy sources, but I also know that most of the highly publicly touted reasons for the need are complete bullshit.

Re:And

[Saint Fnordius]

Yes, the analogy breaks down because I used two different means of transportation. A better version would have been to compare two different switching systems for a common transportation mode: a centrally managed switching network versus a network where every junction is free to make its own decisions. With a centrally managed network, one central station monitors traffic all over the rail net, sending trains on different routes to balance the load. In an independent net, the junction stations need larger switching yard capacity for peak times, and cannot rely upon headquarters to route traffic around them.

The point is that a savings in capacity through centralised load balancing (something the tiered internet promises) is offset by the cost of managing the net. It also presents an achilles' heel, a single weak point that would severely affect performance should it be damaged.

Mind you, I have avoided the subject of preferential treatment, where the transportation company charges different rates based on which passengers are sitting in the train. This would also raise costs as each train would be inspected as soon as it reaches the first junction, taking the time to cross-check if the passengers paid for preferential treatment or not. The head switcher has to field the report from the first junction, then consult his list of preferred customers, then finally calculate the route with the preference weighing in instead of merely by load.

Now that I've considered it more, the call for a tiered internet has nothing to do with managing the network and load balancing, and depends only upon whether the network allows for preferential treatment or not. In other words, they are like conductors who stop a train to check the passengers' tickets and then assign the train a priority based on whether the passengers paid more for preferential treatment or not. In the internet, they inspect every packet upon receipt, passing it along only after checking to see how big a bribe the originator paid. In the end, all passengers/packets are inconvenienced as traffic is held up, and their prices rise due to the wages of the inspector.

Re:And

[Saint Fnordius]

The point is fooling the routers in between the two end nodes, spoofing until the packet is opened. This makes more sense if, say, a certain VoIP protocol is given priority over a P2P protocol*. The P2P client could be redesigned to spoof the appearance of the VoIP packet, thus sneaking into the higher priority treatment, and only the end computers know the difference. Or if a small ISP had two networks but had to travel across AT&T's network, they could spoof their traffic as high-priority VoIP, removing the fake wrapping once it re-enters their network and then routing it to the proper server.

*for this argument assume it's a legitimate P2P service

Re:And

[dodobh]

Riding public transit while toting a weeks worth of groceries just isn't something most people enjoy. Stopping at a grocery after work is seldom an option for anyone who commutes on public transit.

Why would I want to buy a weeks worth of groceries, when I can have quality stuff home delivered on call (I mean that literally, I can call the vendor and ask for stuff to be home delivered, and I'll get in within 15 minutes, max, at no extra cost. Competition is nice.)?

Re:And

[dodobh]

No, the point is that carbon credits are an economic penalty on the pollution caused by excessive private transport. Pollution is currently an Externality and essentially free to the polluters. I consider it a fair price for the polluters to pay more for their luxuries.

Re:And

[xappax]

Then someone can just send packets non-stop

First, you can already send packets non-stop, but only up to the bandwidth cap you've been given. If you have a 3mb/s connection, you can send 3 megabits worth of packets each second and no more - which is exactly as it should be. You paid for that capacity, and should be able to saturate it with whatever traffic you like and have it routed.

Second, even if someone could do that, why? The purpose of sending a packet onto the internet isn't to send the packet anywhere, it's to send information somewhere. I guess if you just wanted to be an asshole, you could send a whole bunch of packets all the time and try to slow the internet down for others, but how does that help you move information better? I guess you could take your data stream and send each packet in it 5 times to ensure that it gets priority, but then you're slowing down your own traffic too, since packet #16 has to get processed 5 times before the router is ready for packet #17.

Maybe I'm just being snobby, but it seems like you don't have a clear enough understanding of the way routed networks operate to be making economic metaphors about them, let alone implying that network neutrality advocates are uninformed. If I'm wrong, please educate me - I would be glad to learn more if I'm missing something.

Re:And

[UbuntuDupe]

Maybe I wasn't clear. When I flood the network with packets like that, it has to devote a greater share of the resources to me. When I want to send "serious" data, I just swap out the dummy packets with serious ones. I could even charge other users for the "service" of not flooding the network like that, so that some of their packets can get that "equal treatment". This screws over smaller users.

And I don't understand why you feel snobby. I *already agreed* that this is not what competent NN advocates have in mind. (Though the Wikipedia article mentions some floozy who wants exactly this.) I was just trying to show how crude NN approaches can simply cause the unfairness to manifest differently rather than eliminate it.

Re:And

[AK Marc]

Mass transit works very well,

No, it does not. Growing up in Dallas, I have tried many times to use public transportation. It failed me miserably. The last time I tried was calling the transportation center and getting them to run the "optimal" transportation from my home to my work. It would start with a half mile walk (part of which was in the wrong direction), followed by a bus ride. Then a transfer. Then another transfer. Then to a stop not far from work. However, the 2 hour trip could have been walked in about 3 hours. When the mass transit is about 50% faster than walking, it is a dismal failure.

If owning a private vehicle is better for you in an urban scenario, then your cities are misplanned.

Hmmm, and I'm thinking that if a private vehicle is so problematic, that your cities are misplanned. Mass transit works in European cities because they weren't planned. Mass transit doesn't work in the US cities because they were planned. If you think that's backward, tell me the age of Paris (decent mass transit) and, say, Dallas. Which one was around long before zoning and such existed? Which one has tiny roads that wind around? The lack of planning made private vehicle use more trouble, and made mass transit more attractive.

Oh, and the 2 hour mass-transit solution was to "fix" a 10-15 minute drive (10 minutes clear traffic, 15 minutes rush-hour traffic). And the cost of the fare was about twice the fuel price in a private car.

Re:And

[doom]

What is more likely to happen is that the cost of fuel will rise. Correspondingly, the price of goods and services will increase. People will have little choice but to pay the increased costs. Eventually, you may well have more carpooling and public transit.

And what I think would happen is you would nearly instantly get an increase in things like car-pooling (and an avoidance of more-or-less unnecessary trips, of which there are many because of the government-provided illusory "free" nature of the resources in use -- look up the "traffic evaporation" phenomena some time, and contemplate what that means exactly).

Re:And

[doom]

Well, I suppose you could knock down substantial portions of US cities to build light rail.

Light rail is in fact, fairly easy to drop in any place where roads exist, and only requires bulldozing cities in your over-heated rhetoric.

Of course, besides being kind of dumb,

Your mama.

you also wouldn't have anyone left to ride them.

Unlimited gasoline being a prerequisite for life-as-we-know-it.

If you go to the mid-west, you have "real cities", I suppose,

You have real cities any place that was built before the post-WWII zoning regulations kicked in, provided they weren't destroyed in an ill-advised attempt at making room for parking lots and freeways.

it's just that there still isn't good public transit.

I dunno where you've been looking. I lived for a few years in Pocatello, Idaho, and even there I could get around on their bus-lines (and get to work on the commuter busses provided by the INEL site). Getting in and out of town by rail or plane wasn't exactly difficult either. Admittedly, if you didn't live somewhere like Salt Lake City, it was difficult to go skiing on weekends by riding a bus, and for that sort of thing I resorted to car-pooling.

Re:And

[xappax]

When I flood the network with packets like that, it has to devote a greater share of the resources to me.

When you flood the network with 3mbps of dummy packets, it forces the network to devote about as much resources as if you were sending 3mbps of real data. Since you've paid for a 3mbps pipe, you have the right to consume that much resources, no questions asked, and the ISP has (hopefully) factored that into how much capacity they sell and build. Since you can't go over 3mbps no matter how legitimate or illegitimate your traffic is, you don't really have the ability to disrupt anyone else's service - both the ISP and its customers expect that you may at any and all times be using the full 3mbps.

When I want to send "serious" data, I just swap out the dummy packets with serious ones.

Sure, but then when you want to send a "serious" packet, it'll have to get in line behind all the dummy packets you've been firing non-stop. Dummy packets can't "save a spot" for a serious packet in the routing queue, so having dummy packets in the routing queue won't give your serious packets better priority. In fact, it's possible it could give them slightly lower priority since the router has to first deal with the dummy packets that preceded the serious one.

I could even charge other users for the "service" of not flooding the network like that, so that some of their packets can get that "equal treatment".

No, you couldn't. It's like claiming you can charge people at the same hotel as you for the "service" of not shitting on every bed you see. You've paid for access to your room, they've paid for access to theirs, so they don't care if you're shitting on beds because the only one you're able to mess up is your own. The only way you could realistically mess up all the beds is by paying to book them all, but then there's nobody to shake down, and you're out a lot of money.

I was just trying to show how crude NN approaches can simply cause the unfairness to manifest differently rather than eliminate it.

It's good that you're trying to come up with exploits and weaknesses in possible net neutrality scenarios, and I don't think you should stop. But you should understand that the ideas you've suggested so far are wrong - giving all packets equal treatment does not open up a network to any of the attacks you described.

Re:And

[Sandbags]

I'm not talking about connection speed, I'm talking about data type speed throttling. I may pay for a 10Mbit connection, but my ISP may still decide video and P2P traffic get capped at 512Kbps. Were I to download from "approved" partner sites, only then could I get 10Mbit. This means that to get real speed I may have to pay MORE to get my data from approved (partner) services, or services owned by the ISP. That's not "who wants more pays more" that's "if the ISP gets paid more, I can go faster" Services that pay the ISP will get prefered traffic, and smaller companies that can't compete with major network providers get little or no bandwidth. This creates artificial monopolies as well as an ISP may be contracted to only allow 1 popular MP3 site, and if I want music from someone other than Apple.com, it will download slower.

only twice?

[TheSHAD0W]

Pshaw. That's easy. The major cost of bandwidth is running all the fiber. Doubling the bandwidth adds maybe 10%?

Re:only twice?

[JordanL]

I believe they mean doubling *existing* bandwidth, in which case it would indeed cost a lot... that would mean doubling the entire infrastructure of our internet backbone.

But again, that doesn't necessarilly mean that its less cost-effective.

Re:only twice?

[Anonymous Coward]

This keeps happening anyway, the infrastructure is getting upgraded all the time, more cables are laid etc etc. I think it's more the case that if the internet was tiered the traffic would drop a bit before carrying on with its typical inexorable rise... meaning they could lay off the upgrades for a few years - at the same time as raking in the cash by charging more! Doesn't seem like good value from my end.

Re:only twice?

[gbjbaanb]

not necessarily - I understand there is a lot of spare capacity in the internet backbones already, dark fibre and the like, so switching it all on would not be as costly as digging up the roads all over again.

However. I don't know about you but I buy my bandwidth from a reseller of a retail arm of the telco who did dig up the roads in the first place. This means that I will never get twice the bandwidth for less money because the reseller I get it from purchases bandwidth directly. ie, they buy X mbps for £Y. If they wanted to double the bandwidth for all their users, they'd have to buy another X mbps which would still cost £Y. So, for all us end-users the cost literally would be double.

Re:only twice?

[amorsen]

I believe they mean doubling *existing* bandwidth, in which case it would indeed cost a lot... that would mean doubling the entire infrastructure of our internet backbone.

In most cases it would just mean using more expensive line cards or perhaps replacing routers. It's very rare that fibers are used to capacity.

Question

[BlueLightSpecial]

Would there be any way to change our neutral internet to a tiered internet without loads of downtime? Or wouldn't it take many people down to do this switch?

Re:Question

[aadvancedGIR]

if(pPacket->port == CST_EMULE_DEFAULT_PORT)
{
        delete(pPacket);
}

Where can I get my check?

Re:Question

[oyenstikker]

If your ISP goes down to implement priority queuing, your ISP sucks. There should be no downtime. There probably won't be for major links. Residential customers of ISPs like Time Warner will probably see downtime, but how to tell that it is tiered changeover downtime and not the usual we-spent-all-our-money-on-yachts-instead-of-upgrad ing-our-aging-infrastructure downtime?

Show me the bandwidth

[Anonymous Coward]

But we're already paying eight times the cost of neutral net bandwidth, so in what way is this study relevant to the consumer?

Re:Show me the bandwidth

[enrevanche]

It is relevant in that this study can be used by the congressman that have been paid for by AT&T to oppose net neutrality.

It is relevant because it will allow AT&T to make a system for which they can charge vastly more than they do now.

It is relevant because it will allow AT&T to reduce your choice more and more over time and to take bigger and bigger pieces of the internet pie.

It is relevant because it will allow AT&T to force more and more companies to deal directly with them for connectivity if their customers want any access to the AT&Ts customers (or shall we call them victims.)

Well one thing to consider

[JimboFBX]

Did they factor in the speed hit you get when you have to make the extra effort to throttle individual packets based on a priority table? What about latency- does a neutral net have lower latency than a tiered one?

Neutral Net is The Devil!

[Anonymous Coward]

AT&T estimates it might require as much as twice the bandwidth of a tiered internet. AT&T also informed its customers that it would be unable to protect the children from evil monster internet pedophiles and that neutral net was found to be causing cancer in third world countries where protection against the silent killer is non-existent.

Between sobs, AT&T's CEO released a statement where he told everyone of a harrowing story where neutral net gained access to his home and forcibly raped him in front of his family.

Wait a second, doesn't AT&T make mad bank if net neutrality is abolished? Like, 'conflict of interest,' n' shit?

Re:Neutral Net is The Devil!

[morgan_greywolf]

AT&T also informed its customers that it would be unable to protect the children from evil monster internet pedophiles and that neutral net was found to be causing cancer in third world countries where protection against the silent killer is non-existent.

In related news, AT&T and Verizon have joined forces to form the Fight Neutral Net Coalition. Pwn Joornet, president of the Coalition said today that he feels this fight is important. "The Neutral Net causes AIDS, it causes global warming, it contributes to heart disease, and worst of all, it eats small children for breakfast. Will someone please think of the children?", said Joornet in a press conference earlier this week.

President George W. Bush, who was last seen laughing while holding up checks from AT&T and Verizon at the National Bank of Texas, was unable to be reached for comment.

government regulation vs free market

[Attila Dimedici]

I tend to support limiting government regulation. This is an issue that I find myself very conflicted on. I have seen studies that make a good case that insisting on net neutrality is the scenario that favors expanding bandwidth to the highest degree. The problem is that ISP's are generally government created semi-monopolies, so unless we force the government to change the rules eliminating this monopoly status, government regulation is necessary to maintain the public interest. On the other hand, I don't trust politicians to pass a "net neutrality" law that doesn't contain some additional onerous clauses. To sum up, I think net neutrality is in the best interest of everyone (even the ISP's in the long run), but I am afraid to support "net neutrality" laws because I suspect they will be something other than advertised.

Re:government regulation vs free market

[Anonymous Coward]

The problem is that ISP's are generally government created semi-monopolies Uh, no, the problem is that you basically can't create a small-scale version of AT&T or whoever and grow it over time into something that can compete with them. There's no "local market" for the internet, there's nobody that will choose an internet that only talks to a 10-mile radius no matter how cheap it is. Unless you can connect to the rest of the internet on a level, neutral playing field there is no way for you to compete. You can't go to the bank and say 'hey give me this 100 billion dollar loan so I can set up a national infrastructure and start a price war with the comm monopolies' and expect them to ever lend you the money. Anytime you as a middle class person have no prayer of ever setting up a competing company and, over time, displacing a company there needs to be regulation. Fundamentally, net neutrality means that any ISP can compete with any other ISP at any scale. You can start an ISP for your block if you want and think you can provide a better service than somebody else.

Re:government regulation vs free market

[vertinox]

I support Ron Paul, but in general I will argue that where there is no free market, it should be regulated to have one. Otherwise, it is just a controlled economy by a non-governmental entity.

Seeing that telecommunications is by default a monopoly due to barriers of entry, it should be regulated to ensure competition and free market can be perused.

Re:government regulation vs free market

[Attila Dimedici]

Telecommunications exists as a semi-monopoly because of government regulation. Most of the barriers to entry are government regulations in one form or another.

What AT&T actually means to say is....

[Anonymous Coward]

..."We really want to double the price of what you're paying us for bandwidth".

Stop the presses!

[Orange Crush]

They mean to say that a network with arbitrary caps and rate limiting consumes less bandwidth than an unrestricted one? Say it ain't so!

Next up: Conserve water by tying a knot in your garden hose.

Re:Stop the presses!

[dw]

>They mean to say that a network with arbitrary caps and rate limiting consumes less bandwidth than an unrestricted one? Say it ain't so!

To look at it another way. A provider desiring to guarantee QOS... latency, jitter and minimum bandwidth for services such as VoIP, without having the benifit of having control over that bandwidth, would need to have a lot more bandwidth to meet those expectations.

This is just restating the idea that QOS enforcement becomes irrelevant with enough bandwidth.

Re:Stop the presses!

[enrevanche]

This is not about QOS, ii is about control.

Re:Stop the presses!

[dw]

> This is not about QOS, ii is about control.

This is about selling services, and being able to provide better services than the competition.

I believe the mistake being made here is in trying to argue that providers don't have the right to build tiered infrastructure. They own their own infrastructure, and have the right to build and offer any kind of tiered services they wish... and the pricing of those offerings are drastically dropping thanks to technologies like MPLS, as apposed to dedicated circuits such as T3s...

The debate should be focused on whether a provider can give QOS to a customer who's data competes with other public internet traffic... along a shared path.

Tiered *networking* is going to continue to develop. We just need to to make sure it doesn't develop into a tiered internet.

Wait a second...

[vigmeister]

When I am looking at leasing an internet connection at home, I equate bandwidth with speed and this is a reasonably rational assumption (today).

Analyzing the situation and pluggin in numbers,
Assume that the bandwidth available is fixed. What they're essentially saying is that either all of us can get 50BjBps (Bajillion Bps) regardless of the importance of our packets, or using a pareto distribution, 20% of us will get 80BjBps and 80% will get 20BjBps effectively?

I know these are rough numbers, But damn if I know which one I'd prefer... I think at the end of the day, a clearly defined set of standards for prioritization needs to first be developed by an independent body (ICANN/ISO/IEEE?). Once that is done, we can debate net neutrality. Right now, none of us actually know what is going to be prioritized. If streaming video for doctors performing live surgery is prioritized, I'm OK with that. If companies can buy priority for commercial, then I am kind of opposed to it unless I am guaranteed that these priority purchases will subsidize my connection.

Maybe they can have two levels of internet access: Neutral internet access (~$50 p.m) and Tiered access (~$10 p.m). Then let these levels fight it out. Of course, the implementation is unclear to me as I am not network engineer. To think about it, isn't this tiered in itself?

Cheers!

Re:Wait a second...

[Xeth]

If what you want is a reduction in latency, how about just making that one more thing that you purchase when you're buying a connection? If I want my doctors to transmit videos, well then I'll buy the 10 mbit/100msec package. If I'm just messing around with email, then I'll buy a consumer package. By simply making latency another part of the purchasing decision, the market will (in this case) work. Why bother with nasty things like tiers and payola?

Yes, I realize that a single ISP doesn't control all the wires my packets run over, but they can make the appropriate purchases. Why should we treat latency as any different from bandwidth? Both are determined by all the intervening lines. We just need to sell on both.

Re:Wait a second...

[An Onerous Coward]

You're missing a critical point. If you choose not to prioritize bandwidth, the routing process is dead simple:

1) Read the destination IP from the packet header.
2) Look up the destination in the routing table.
3) Send the packet out the appropriate pipe.

When you decide to prioritize, you add a lot more steps.

1) Read the destination IP from the packet header.
2) Read more of the packet to determine the type of communication and prioritize it accordingly.
3) Check the queue to see if anything of a higher priority needs to be sent first.
4) Send those.
5) Look up the destination in the routing table.
6) Send the packet out the appropriate pipe.

Okay, three more steps. But the point is, the amount of processing needed for each packet is greatly increased, so the same hardware handles far fewer packets.

As Cory Doctorow likes to repeat over and over and over, the Internet2 people have tried every prioritization scheme imaginable, and have never found anything that relieves congestion quite as well as simply adding bandwidth to a stupid network. A stupid network can be blazing fast, but as you add more brains to it, it has to slow down and think about what it's doing. Intelligence should live on the edge of the network.

The only, only, only reason to make the network "priority aware" (given that we can increase capacity easily) is to allow the Telcos to use QOS guarantees to charge certain customers up the wazoo.

Almost on Topic: Globaltics WORSE than Goatse!

[Anonymous Coward]

Didn't know it was possible but you just passed up goatse on my annoying-o-meter. And you're rapidly approaching the GNAA ...

Re:Almost on Topic: Globaltics WORSE than Goatse!

[Anonymous Coward]

goatse!!!

Then look at ALL the regulation here

[Anonymous Coward]

So why get rid of "common carrier" and its' requirements when you still have the easements? The public donated infrastructure? The licensing monopoly? The legal protection of neutral network operator when you are no longer a neutral operator?

Are you torqued by government interference in a free market when it
protects artists (copyrights)
protects investors (patents)
protects corporations (trademarks and plc registration)
protects markets (corn/oil/wood subsidies)
removes external competitors (import tarrifs)

?

Re:Then look at ALL the regulation here

[sid0]

Are you torqued by government interference in a free market when it protects artists (copyrights) protects investors (patents) protects corporations (trademarks and plc registration)

No, no, no. Government is to protect property, and without property there can be no free market. (I do recognise intellectual property as property, unlike some others: I just think the current system needs reform in prior art and such, and the time limits are stupid. I'm in favour of a copyright system in which automatic copyright is for, say, 30 years, and you have to register to extend it up to, say, 60 years. Patents should be valid for a lot less, of course.)

IP is a "government granted monopoly" in the same way as a piece of land you own is a "government granted monopoly". IP is NECESSARY in a free market.

protects markets (corn/oil/wood subsidies) removes external competitors (import tarrifs)

Yes and yes. I'm not a fan of subsidies and protectionism, especially in the developed world. I dislike lobbying as well -- it is clearly misuse of government power. Protectionism and lobbying are not legitimate parts of a free market.

Re:Then look at ALL the regulation here

[zotz]

"I'm in favour of a copyright system in which automatic copyright is for, say, 30 years, and you have to register to extend it up to, say, 60 years. Patents should be valid for a lot less, of course."

"IP is a "government granted monopoly" in the same way as a piece of land you own is a "government granted monopoly". IP is NECESSARY in a free market."

I seriously don't understand your thinking here. Should your rights to your real property also expire after 30 years?

all the best,

drew

Re:Then look at ALL the regulation here

[sid0]

Ah, I didn't mean that. I should rephrase my statement somewhat: I do recognise intellectual property to be like property. I didn't mean both were identical.

The "government granted monopoly" thinking is something that I'm just peeved at. Real, tangible property is as much a "government granted monopoly" as IP is -- which is to say, not at all. Just to clarify, A implying X and B implying X does not mean that A and B are the same.

Re:Then look at ALL the regulation here

[Lockejaw]

protects investors (patents)

This is deliciously ironic.

Re:Then look at ALL the regulation here

[zotz]

"The "government granted monopoly" thinking is something that I'm just peeved at."

But copyrights and patents are government granted monopolies.

You can protect your stuff yourself without the government, just keep all your ideas secret. Now sure, the government should protect your person so that someone does not torture you into giving up your secrets to them...

all the best,

drew

Re:Then look at ALL the regulation here

[sid0]

Is the ownership of a piece of land a government granted monopoly? I mean, you can do almost anything you like with this land. You can forbid any other citizen from trespassing. You have a monopoly over your land, don't you?

Re:Then look at ALL the regulation here

[Anonymous Coward]

except that property is defined as mutually exclusive. property is not an a priori; human invented the institution of private property as a social construction to solve the problem of conflict over mutually exclusive artifacts in the world (i.e. physical things). ideas and data are not mutually exclusive. if I use your car, you are precluded from using that same car. but if I use your idea, you can still use it too.

Re:Then look at ALL the regulation here

[BoberFett]

I'm not the OP, but yes. All of those instances of government meddling really piss me off. I can understand the initial desire for government to interfere in markets, acting to create stability or incentive. Unfortunately it never takes long for those best intentions to lead straight to hell.

Copyright is a good idea. But it's run amok and now nothing ever enters the public domain.

Patents are a good idea. But it's run amok and now the little guy can't invent anything without being sued by a patent troll.

Corn subsidies stabilized the price of corn for farmers. But now it's considered an entitlement and between that and sugar tariffs, everything we eat is loaded with corn syrup. How much has this country spent fighting diabetes, heart disease, and everything else that comes with obesity caused in no small part by a diet of corn syrup? Was it worth it?

All that said, I do agree with net neutrality, for the very reasons you stated. The telcoms got filthy rich using their monopoly status and by stringing lines across everybody elses land without having to pay rent. They now want to change the rules of the game, in which case I think every American who has a phone company line running through their yard should send them a bill for rent. If they get to change the game, so do we.

Re:Then look at ALL the regulation here

[BunnyClaws]

Yeah, I have a government granted monopoly of my backyard. I am preventing competition in the market by owning my backyard. No, owning a piece of land is not a government granted monopoly. This is Slashdot where did all the Libertarians go?

Re:Then look at ALL the regulation here

[sid0]

I hope you realise I was being sarcastic. :)

Re:Then look at ALL the regulation here

[BunnyClaws]

I do now. :) I've heard crazier arguments against capitalism so sometimes it is hard to tell if someone is being serious or not.

... And so what?

[Anonymous Coward]

A toll road needs fewer lanes than a main highway. Heck, if you make the toll high enough, you won't need any lanes at all. So? The ISP's interest doesn't align with net users: ISPs want to maximize profits, which will require restricting traffic. Their view of the Internet is a toll road, not a superhighway.

The new way to spin "net neutrality is bad"

[grev]

Majority of people don't know what net neutrality is, they don't care, and they never will. Now, whenever the issue is brought up in the mainstream news or whatever, big business can talk about how it's half as efficient, in addition to being communist and un-American. I can only imagine how this will turn out.

Re:The new way to spin "net neutrality is bad"

[Magada]

OH NOES! Twice the bandwidth needed! ZOMG! WTFBBQ! Pwnt!
The argument is easily defused, imho, by the simple observation that building and manning the infrastructure needed will actually create jobs, as well as provide new growth opportunity for all sorts of businesses - the new, neutral, high-bandwidth Internet could even become something like the highway and hydro projects undertaken as part of the New Deal - a way to energize the whole of the economy by targeted investments in infrastructure.

Re:The new way to spin "net neutrality is bad"

[sjames]

The counter argument goes: net neutrality is share and share alike. Those who oppose it want the wealthiest corporations to have a feast and let the rest of us fight over scraps they occasionally drop on the floor.

Re:The new way to spin "net neutrality is bad"

[slazzy]

I bet a non-neutral internet would actually only need about %1 of the bandwidth we use now, because nobody will be using it.

KISS

[smittyoneeach]

Your existing newtorks are built upon relatively simple, freely available protocols.
Any hypothetical or actual throughput you think you'll gain from sexing up the infrastructure will come at the cost of lots of pain. Buggy code, code with bugs inserted for nefarious purposes...
I hope that there will always be "plain old networks" available. If a company wants to come up with some slick product and sell it to the sheep, fine. That's capitalism. I just wont have much compassion during the winter, when the sheep are asking WTF their wool went.

Re:KISS

[Anonymous Coward]

I think the key dimension on this is time.

At an instant, the 'optimal' solution is managed. But this changes. And a managed network will require network wide upgrades to 'correctly' support new 'services'. This will introduce huge costs and essentially prevent service introduction - killing the goose that laid the golden egg.

Interesting AT&T (etc) have lots of experience with this. IN (Intelligent Networks) failed for exactly this reason.

Re:KISS

[smittyoneeach]

Interesting AT&T (etc) have lots of experience with this.

Interesting, your tacit assertion that corporate knowledge persists.

Re:KISS

[Aladrin]

Interesting, your tacit assertion that corporate knowledge persists.

Interesting, your use of English is.

Re:KISS

[smittyoneeach]

Meine frau is von Deutschland. Learning to speak German has given me a taste for retro syntax.

Net Neutrality

[Tuoqui]

Is not about a dumb internet. It is about an internet that does not discriminate based on entry or exit points and/or the protocol being used except where such discrimination will benefit the overall network performance.

Net Neutrality Positive
VOIP Packets receiving priority (because lag and bandwidth throttling reduce performance of VOIP technologies)
Prioritizing Gaming traffic of popular/well used games (IE. MMOs, FPS over internet, etc...)

Net Neutrality Negative
Throttling Bandwidth on P2P applications (This is the big concern on most ISPs, they admittedly do suck up a lot of bandwidth)
Extorting Money from websites who have not paid large sums of money for faster service (YouTube-wannabes)
Delaying or Denying packets coming from X-Network (because they didn't pay extortion money)

Ways to fix things... Run more Fiber. It should not be as hard as it was before since many of the tunnels and such have been made already.

Re:Net Neutrality

[daskro]

How does Net Neut prioritize VOIP and gaming traffic? If anything it does the opposite since it treats all the traffic the same.

Re:Net Neutrality

[TubeSteak]

Ways to fix things... Run more Fiber. It should not be as hard as it was before since many of the tunnels and such have been made already. That is not the answer.
There is already plenty of fiber is lying around.

The real issue is hardware to light up the fiber and then to switch the packets.
That is where the ISPs are trying to cheap out.

Re:Net Neutrality

[Tuoqui]

No I was just saying that if it improves the performance of the network functions it is justified.

VOIP is a very time sensitive protocol. If your ping times become laggy and/or your bandwidth fails to meet the required 64kbps that a DS0 channel requires then the audio becomes choppy and/or delayed and laggy. This is perfectly acceptable in video communications but audio with lagging response is to say the least an annoyance (Notice on News reporters on the scene (or astronauts) they ask a question and sometimes theres a delay of up to 2+ seconds before they respond? Thats the sorta thing we're talking about here. Cant do much about the astronaut scenario since thats a limitation on the laws of physics). FPS games fall under similar things since delays getting too long renders a game unplayable.

So yes if you want voice communications over the internet then you want them to prioritize that traffic. What you DONT want them doing is intentionally throttling and/or delaying packets (IE. Torrents) that have no particular performance reason for being delayed other than they've not bothered to lay enough fiber for the bandwidth to their customers.

So in short yes you want protocols and programs which require more 'real-time' response to be prioritized over torrents but they should not be throttling connections and/or engaging in active filtering (Censorship) of websites because they refuse to pay for their bandwidth twice (once to get it out, once to get it to customers in a quick and expedient manner).

Re:Net Neutrality

[Anonymous Coward]

except where such discrimination will benefit the overall network performance.

You killed your own point in the very first sentence.

Re:Net Neutrality

[norminator]

No I was just saying that if it improves the performance of the network functions it is justified.
Then unfortunately, you're saying that a non-neutral network is justified, since the whole gist of TFA was that neutrality hurts the performance of the network. They can justify whatever performance claims they want. The whole thing is kind of a red herring, I think, since net neutrality isn't saying that packet types couldn't be discriminated against, but rather that packets shouldn't be given lower priority based on their source/destination. I think there could be a tiering solution that would give VOIP a higher priority, and certain other applications which are important and time sensitive, without discriminating against who is sending and who is receiving, and that would help bandwidth issues.

My questions:
1) Sure, a tiered internet would have some effect on peak bandwidth, but wouldn't the average bandwidth requirements be the same? (Unless, of course, internet usage declines dramatically, because it takes too long to load web pages and download files from non-ISP-approved sources...), and

2) I know that the network hardware companies are pushing for a tiered internet alongside the telcos... How would the cost of implementing all of their new hardware and software tiering solutions compare to the cost of increasing the available bandwidth?

As a side note, I think it's funny how the network hardware companies want a tiered internet so they have an excuse to roll out new products, while the telcos want the tiered internet so they have an excuse to not have to create more of what they create. I guess that's because the hardware companies don't get paid for their products until they can ship them, whereas, the telcos can get (and have been getting) paid for products years ago that they're still stalling on rolling out.

Re:Net Neutrality

[someone300]

There are three network neutrality positions:

1. Dumb neutral networks: Packets aren't prioritised, charged, changed in any way based on their data or their endpoints. Proponents of this system assure it's possible provided you have lots of bandwidth available.
2. QoS neutral network: Packets are prioritised based on the nature of the data involved. VoIP and gaming is prioritised, followed by burst transfers like page requests, followed by bulk transfers like torrents and downloads which eat the spare bandwidth. Higher priority service can't be assigned based on endpoints or based - You can't charge to improve a particular companies traffic.
3. Tiered network: You can pay for higher quality internet traffic or to be prioritised over other types of traffic.

The first two positions generally seem to be considered in support of network neutrality, whereas the latter against network neutrality. The second one's idea being that latency involved with sending a file is unimportant compared to the latency of a VoIP packet. Essentially, this means VoIP gets prioritised, but it doesn't necessarily mean it impacts on the speed of the file transfer. It's supposed to be a cooperative rather than a capitalist idea, where the reason one type of packet is prioritised over another is for everyone's benefit without. I subscribe to this view, since I believe that no matter how much you increase the bandwidth available, torrent data and file transfers are probably going to increase to fill the available bandwidth, so certain types should be prioritised.

I would, however, be satisfied with either 1 or 2 being implemented. I am completely against 3, though.

Re:Net Neutrality

[fwr]

Yay! Someone (sic) who actually has a clue! I hear a lot of rants on Slashdot about net neutrality, most of which are clueless. Even people who profess to be network engineers, but who are not necessarily experienced in VoIP or QoS, don't really have a clue. Just because you know EIGRP, OSPF, IS-IS, BGP, doesn't mean you know jack about QoS. For a primer there are eight things that would add to the delay of a packet. In order they are serialization, propagation, queuing, forwarding, shaping, network, codec, and compression. By far the biggest concern, and the largest delay, is serialization. However, serialization is a fixed value based on the bandwidth of the link (and each hop between the destinations would add a serialization delay). On speeds over T3 the serialization delay is so small that QoS engineers usually don't take it into consideration, so the major serialization delay would be on the first hop out to the ISP and the last hop to the destination user. You would hope that all of the hops in-between would be at least T3 speeds.

Propagation is just what it says, how long it takes the signal to go down the line. Most people use 70% c for this value, and it's based on distance, so it's also a fixed value. Skip queuing for now, as that's where the main problem comes in. Forwarding is how much it takes the network gear to process and forward the packet from the input interface/queue to the output queue. This should be negligible give current day hardware switched packet gear. Shaping is self-imposed delay of the sending of traffic so that you don't go over the CIR on links such as Frame Relay, and shouldn't be applicable to what we are discussing. "Network" is some unknown delay within a service provider's network, such as a Frame Relay network. Since we are talking about service providers' networks network delay is not applicable. Codec and compress, these are VoIP specific delays in coding the voice patterns by the DSP in the VoIP device, compressing it if applicable, and putting it in the VoIP phone's output queue. They are not applicable to this discussion.

So we are left with going back to queuing. Queuing delay is the amount of delay sitting in an output queue because there are other packets being transmitted, or waiting to be transmitted before yours. Why would they be waiting, because the bandwidth is exceeded, the utilization for the link is at 100%. The service provider over-proscribed their links, etc. How much is the delay? Well, it's the serialization delay for all of the packets waiting before you to be transmitted. Since serialization delay is usually ignored for links over T3 speeds you wouldn't think that they would add up to much, unless the link were really, really, over-proscribed. Here's where the major QoS tools come into play, where you can maybe have two queues serviced by the outbound interface. Say you have one priority queue for VoIP traffic and one other queue for all other traffic. If there is ever any traffic in the VoIP queue you would always transmit that next instead of a packet in the regular queue. Maybe you designate a specific bandwidth available for the higher priority traffic for a particular interface. Maybe you have more than just one class of packets, more than just high priority and normal, and maybe you do weighted early random dropping of packets in the normal priority queue, base on the classification, so that bulk transfer protocols such as FTP would back off sooner, reduce the TCP window size, and not effect the other traffic.

Keep in mind what we are talking about here though. If the pipe isn't full there is generally no reason for QoS or tiering of network traffic on high speed links. Are the links full now? I don't know, you'd have to ask the service providers. Will they become full soon? Possibly, this could just be a preemptive step by the service providers to avoid any issue as the links fill up. However, I suspect this is a major push by service providers to be able to increase revenue and charge for what

Re:Net Neutrality

[sjames]

The problem is if any shaping whatsoever is permitted, it will be used strategically to increase revenue while reducing the overall performance of the network. If tiered service is permitted, they'll do it openly. If shaping is permitted for performance enhancement only, you can bet they'll "inadvertantly misconfigure" their QoS to implement tiered service without admitting it. It's hard to prove such things to be intentional and strategic.

OTOH if it's simply forbidden, there can be no excuses.

Re:Net Neutrality

[whit3]

According to ATT's study, network neutrality is positive. The establishment
of network neutrality doubles the demand for ... ATT's product: gigabytes of
transferred data.

If ATT believes this study, they will be lobbying for net neutrality.
If they don't lobby for net neutrality, we can draw at least one conclusion.

Re:Net Neutrality

[pab4250]

Running more fiber/adding capacity is absolutely the answer. With enough bandwith, companies investing in the pipe can have a tiered system and still have plenty capacity for everyone. It would be an incentive for the companies to pay for a build out.

TFA

[Robber Tom]

It's worth noting that AT&T funded the study. They MAY have a vested interest.

Bandwidth

[Renraku]

Bandwidth is a funny resource.

Imagine if you had a tree that bore fruit once or twice a day. But if you did not eat the fruit within an hour, it spoiled. There's no point in trying to conserve the fruit unless your demand is higher than the output of the tree.

Its always good to have say, 10% free. Out of ten fruit, leave one so that any surprise visitors might have a quick snack as well.

Of course, the other reason you might try to conserve it is to create artificial demand. Now, half of your crop goes to waste. You sell the other half for very high prices saying that your supply just can't keep up with demand and that you must sell them at a higher price due to the whole free market thing.

Point is, every fruit you don't sell will be useless in an hour. But its better to let a fruit rot than to sell it for a decent price, after all.

Re:Bandwidth

[TooMuchToDo]

Excellent non-car analogy. And hats off to reminding me of "The Grapes of Wrath."

Re:Bandwidth

[mpeg4codec]

You totally lost me with that analogy. Could you rephrase it in terms of cars?

Re:Bandwidth

[soapthgr8]

To make this work, you have to assume that the location is perpetually cold and rainy and is near a large body of salt water. Imagine if you had an auto factory that produced one car (unpainted and no clear coat) every 6-minutes. If you did not paint the car or have it moved within an hour, body cancer would set in and the car would rot. There's no point in trying to conserve the cars unless your demand is higher than the output of the auto factory. Its always good to have say, 10% free. Out of 10 cars, leave one so that any surprise visitors might have a quick getaway as well. Of course, the other reason you might try to conserve them is to create artificial demand. Now, half of your hourly production goes to waste. You sell the other half for very high prices saying that your supply just can't keep up with demand and that you must sell them at a higher price due to the whole free market thing. Point is, every car you don't sell will be useless in an hour. But its better to let a car rot than to sell it for a decent price, after all.

This looks like double-speak to me

[Fross]

couldn't this be re-interpreted as saying that if they were to run a tiered network, they would have no problem throttling its bandwidth to 50%, in order to ensure the content *they* prioritise gets through unhindered?

fud fud fud fud fud fud fud ....

[tomstdenis]

fud fud fud ...

When in doubt, spread fud. Just like the myth of "the evils of socialized medicine." Tell the same lie enough and people start spouting it themselves. Now, for example, you have uneducated ignorant folk yelling as loud as can be that "commie-loving socialized medicine is no good," despite the fact that in many countries it works sufficient enough to increase the average lifespan of their citizens. [and for the record, I think Michael Moore is full of shit, so don't lump me in with that sensationalizing lying sack of shit].

Same thing here. The telcos will tell us over and over that "this it the way things must be in a god fearing red blooded free america" and people will eat it up. In 10 years you'll hear the same ignorant uneducated folk spouting on about the evils of a "neutral network" as being commie and evil.

People really need to learn to research context. Then they'll see through the BS of modern lobbyists/advertisers and be straight.

Tom

Re:fud fud fud fud fud fud fud ....

[UbuntuDupe]

Wait -- for all your complaining about health-related fud, aren't you the one who thinks a valid way to calculate fraction of health care costs due to the legal system, would be to divide visible awards (or visible awards + lawyer costs) by total health care expenditures? And then to say, "oops, that's .5%, guess it's not a big deal, even if doctors spend half their salaries on liability insurance" ?

Re:fud fud fud fud fud fud fud ....

[tomstdenis]

I think that health care should be universal. But you have to present both sides of the issue. Yes, you benefit with increased health coverage, less red tape [everything is covered, no hassling about which treatment is available], etc. But you DO pay more in taxes for it. It's not free, it's universal. And it works because most people don't need it on a regular basis but pay for it anyways. For example, for every 1 person in a hospital [or currently going through a treatment] there is probably a 100 who aren't, if not more. The problem is when you're only goal in life is to maximize profit. You just can't reason why a business should be non-profit (e.g. break even on the costs). Then those same people spread the fud about "evil socialized services."

Similarly, I think the net should be universal as well. I pay my ISP $X dollars/mo for my connection and I should be able to hit any other IP that is routable through the net. It just makes sense. People have to realize there is more to life than being richer than your neighbour [who ironically most people ignore anyways].

And that's as simple as it gets. Most people in France, the UK and Cuba [three spots that Asshat Moore went to] do with a lot less than their American counterparts. Despite what they showed in the film, not everyone in france lives in a 4 bedroom apartment with two cars, no debt, a plasma tv, etc... I know a bunch of peeps from the UK, Ireland, France, etc and while they're not bad off, they're not living the same type of life [e.g. material wise].

Middle class Americans for the most part couldn't fathom living in a semi-detached house, with no car, taking public transit, etc. That's what poor people do afterall!

The irony of course is the more they cling to the HMO bullying the less coverage they'll actually receive.

Tom

Re:fud fud fud fud fud fud fud ....

[An Onerous Coward]

Moore wasn't trying to show his audience the lifestyle of the average Frenchman. He was trying to show the lifestyle of an average French doctor, to counteract the FUD that says we need to preserve our star-spangled HMO-based system, because that's the only system where doctors are allowed to make as much as they can.

Re:fud fud fud fud fud fud fud ....

[tomstdenis]

I didn't get that from the film, maybe I missed the introduction bit. Either way, I think most doctors in the USA are more worried about malpractice insurance fees or settling their debt than being rich. Maybe if our [canada/usa] governments helped doctors pay down their debt, and remove malpractice suits from law they wouldn't need to get paid so much and health coverage would be affordable.

My point though was that while the UK and France have what looks like good medical coverage that is state provided, it isn't free. You pay for it through your taxes, which last I checked were higher than what I pay here in Ontario. That they're willing to put up with paying for it shows a combination of two things. That many people are likely for it and wish to pay the taxes, and that those who aren't are too lazy to fight it.

But eitherway, I'm sure if you told a doctor that they would be out of school debt instantly [or say cut it in half], and free from meritless lawsuits they would gladly work for a lower pay.

Tom

Re:fud fud fud fud fud fud fud ....

[An Onerous Coward]

I think most doctors get into the field out of a desire to help and heal people. But when they leave their training with mammoth debt, it forces them to pick a high-paying specialty and seek out a private practice to join. That way, they can make enough money to get rid of their debts. I've heard that the U.S. has way too many specialists compared to the number of general practitioners, and private practices often cater to elective medical procedures.

Malpractice fees are high (though they're not what's driving the recent explosion in health costs), and it makes sense that they'd put pressure on doctors to find more lucrative options.

I think there are a couple of government programs that pay medical school bills for doctors who are willing to spend a few years working in rural areas or free clinics. Those programs could probably be greatly expanded.

Twice the bandwidth

[vsavkin]

Is this really a big deal?
In much cases, fast and dumb solution is cheaper than smart, but slow.
If you have not enough busses - just upgrade to faster transceivers as optical fiber can carry practically unlimited bandwidth.
And you can look at computers for another example - upgrade processor, busses is faster and cheaper then optimizing applications.

Re:Twice the bandwidth

[enrevanche]

This is absolutely true in the abstract sense and if AT&T were charged with making an efficient system it is what they would be trying to do. What they are really try to is make a revenue maximization system. If this costs three times as much to the consumer (in infrastructure, bureaucracy and equipment) so that AT&T can double it's profits then the hell with the consumer.

From "bandwidth crisis" to deep profit inspection?

[AHuxley]

In Capitalist West at&t profits by you sending packets.
In Soviet Russia at&t profits by spy sending packets to you.

So, basically....

[plazman30]

AT&T is saying that their existing customer base is getting crappy service because they don't have enough bandwidth to support their customer base. So, instead of, I don't know....dealing with the problem by making infrastructure changes they should have made 5-10 years ago, instead they want to charge content providers for access to their inadequate pipes.

With all the patching Automatic Update does, I'm surprised that Microsoft isn't all over a neutral net. They may have to pay a fortune to ISPs. Then again, they would just probably create a "Windows Update Pro" and charge users for "faster access" to updates.

Andy

Re:So, basically....

[DragonWriter]

With all the patching Automatic Update does, I'm surprised that Microsoft isn't all over a neutral net. They may have to pay a fortune to ISPs.

They can afford to pay a fortune to ISPs, especially if it means competitors (like every Linux distro that is gratis as well as libre) that can't instantly suffers a major disadvantage in pushing updates.

Re:So, basically....

[plazman30]

I hope the whole f***ing planet blocks ALL US IPs from getting to their sites. Maybe that will teach AT&T a lesson. Most first world nations offer 10 Mbit bidirectional pipes, some offer 100.

The net is neutral now, and I don't see anyone complaining, except for AT&T. Put on your big girl panties AT&T, and FIX YOUR BACKEND!

Andy

Google Translate:

[fishthegeek]

I ran the text through google translate and this is what happened:

Researchers at AT&T were very concerned that bandwidth would be further commoditized if the government does not act to prevent it. If At&t is required to treat everyone the same, then the consumer is free to choose the services that they want based on something called "quality of service" rather than a more practical method of choosing.... say... oh I don't know... uhm... a method of choosing based on how profitable it is for At&t. Having the consumer choose services based on what benefits At&t is a much more practical and convenient way for the consumer to purchase services over the Internet.

At&t is very concerned about the bewildering number of options that the American consumer has available, and with the best interest of our customers at heart, At&t should assist the consumer by limiting the number of choices immediately.

Spokesmen for At&t quickly said that "We do not want to the consumer to get the full unfettered benefit of the Internet because then we would have to actually add infrastructure to meet demand.

Re:Google Translate:

[Anonymous Coward]

Of course, you got this biased translation only because the money hoarders at Google have rigged their translator, Google being the ultimate freeloader on the hard work of the ISPs ;)

Re:Google Translate:

[badfish99]

Google being the ultimate freeloader on the hard work of the ISPs
Wow: Google get to use the network for free!!?? I never knew that </sarcasm>

Re:Google Translate:

[Anonymous Coward]

so AT&T wants to F****n kill google now? Watch the chairs fly!

Re:Google Translate:

[Nom du Keyboard]

I ran the text through google translate and this is what happened:

And now you know why Google will be the first Internet service relegated to the slowest service possible. Can't have this getting out to everyone. Heck, they might even put it on their main page -- something to fill up all that unused white space.

Re:Google Translate:

[Namlak]

I ran the text through google translate and this is what happened

Odd, when I ran it through, I got this...

"Google buys AT&T, changes name to AT&T."

Re:Google Translate:

[GWBasic]

Translation: A non-neutral net makes AT&T more money.

A Little Confused

[BackwardEngineer]

So, we're using twice as much bandwidth right now since we don't have the whole tiered structure in place?

In other words...

[Em Ellel]

Allowing traffic through requires more bandwidth than blocking traffic.

Whomever got paid to "research" this - I admire your ability to get paid for stating the obvious.

-Em

Re:In other words...

[bhmit1]

Allowing traffic through requires more bandwidth than blocking traffic.

More importantly, they basically said that when they are allowed to have a tiered internet, they intend on blocking half of the traffic. If you're a generator of traffic to an isp, and you're not paying the tariff/extortion, guess what half you're in?

Re: Bandwidth is cheap

[Anonymous Coward]

That study may make a nice headline, but bandwidth is fairly easy and cheap to provide, the backbones are DWDM with 10Gb/s per Lambda with each fiber strand carrying 40-100 individual lambdas... The bandwidth problem for most US homes is in the last mile, which a tiered architecture isn't going to help.

The tiered aspect should be on the client side

[MikeRT]

One would think it would be about economies of scale, stupid AT&T, et al. Seriously, if you offer a base broadband package for $10/month with 2GB of download bandwidth included, and $0.25/GB after that, I bet that would reliably generate a lot more revenue, in a more efficient way, than mucking around with websites, contracts, etc. Anyone remember the telecoms trying to make companies like Google out to be robber barons, foisting all of the costs onto the public? That's how ridiculous it's gotten. Unlimited bandwidth may be sensible someday, but not right now. The rest of the network just isn't up to handling many users maxing out their pipe every month. Metered bandwidth would solve that in a market-friendly way.

Re:The tiered aspect should be on the client side

[enrevanche]

NO, no no no. This will provide the most efficient and cost effective system for the consumer. Not only would the system be easy to maintain, it would allow audits to determine whether a provider is reasonably charging for the semi-monopolized service.

But

By building a huge regulated infrastructure, the comprehension of the system is too difficult. Plus, AT&T knows that consumer will become more and more dependent on the internet. They know that they can get consumers to eventually pay substantially more to get the services they need. The consumer will never know what hit him and he will be "thankful" that AT&T has sorted it all out for him.

Re:The tiered aspect should be on the client side

[AusIV]

I agree. I've never understood why broadband works the way it does. You're given a limit to how much data you can transfer per second. It doesn't matter whether you're using your full bandwidth every second of the month, or use ten percent of your full bandwidth a few times a month you pay the same price.

Now, your broadband provider pays the same cost to transfer 1 kb whether you're downloading a web page or a movie - you just get a lot more bits with a movie. It's never made since to me that someone who transfers 20 GB using p2p a month pays the same as someone who transfers 20 MB a month checking their e-mail just because they're capable of getting their bits at the same speed.

It also doesn't make sense to me that the broadband providers think it would be more profitable to make a distinction between http traffic, p2p traffic, legitimate video downloading traffic, gaming traffic, etc. Aside from the fact that this would require some infrastructure changes to filter the different packets, a bit is a bit, and while some activities transfer more bits, each bit costs the same to transfer. If they charged for the bandwidth you actually use instead of the bandwidth you're capable of using, it's fairer to the consumer - they pay for what they get, rather than what they could get - and it's more profitable for them - they can adjust bandwidth costs according to what a bit costs to transfer.

I also don't understand this desire to charge people to connect to their network, even if they're already paying t connect to other networks. I pay my cable company to be connected to the entire internet - not just other sites hosted on their network. If it's expensive for them to connect me to other networks, that price should be reflected in the price I pay as a customer, not the price paid by the people I'm already paying to connect to.

Strawman to Confuse and Distract from the Issue

[ClarkEvans]

Having a Neutral Network doesn't mean that it can't have different levels of service. The U.S. Postal system is neutral, it doesn't [1] favor big companies over small ones, or some content over others. It has *classes* of mail, and charges by class. A tiered network can work the same way, packets do have a place to mark their priority. There is no reason why we can't develop ways to change content priority. Real-time stuff can mark their packets higher, and bulk-mail can mark the packets for overnight delivery ;)

What they mean to do, however, is different. It is to charge differently based on who you are, who your communicating to, and what sort of content your sending. This is nothing more than corporate censorship. It is wrong.

[1] Well, it just now there are proposals to start pricing USPS mail based on the amount of mail you send. This is clearly wrong since it favors larger corporations over smaller ones, and hence serves to limit competition.

Re:Strawman to Confuse and Distract from the Issue

[sid0]

Why? Think of it as a bulk order. Bulk orders are great business. Does anyone complain when a large corporation gets software for less per piece than a smaller corporation, or an individual, would?

Twice as useful

[Experiment 626]

The purpose of a network is to transmit data. It receives usage when a customer sends packets over it. By AT&T's own admission, a neutral network is twice as useful to customers as a tiered one, but they want the tiered one anyway since it increases profit margins and allows them to blackmail Web sites.

AT&T estimates!!! PLEASE

[enrevanche]

This is just another industry sponsored "proof" for use by lobbyists.

Corporate sponsorship of research doesn't automatically invalidate that research Right, I'm sure Phillip Morris would agree. Industry doesn't sponsor research that it doesn't already know what the conclusion will be.

Why?

[Actually, I do RTFA]

I read the Ars article, and tried to get through the AT&T study (going to try again after more coffee). As I read this research: If companies are allowed to drop "unimportant" packets to the sidelines, while only guaranteeing 1/3 of the packets as fast delivery as otherwise necessary, they only need 50% of the bandwidth.

Assuming that my analysis is correct:

1. No shit. Airlines recently announced that if customers are willing to extend travel times by flying around the world on empty seats to get where they are going, they can cut the number of planes in the sky by XX%.
2. Who cares. Over the backbone of the internet, there should be enough bandwidth.
3. The want to fuck me. Who cares if someone else's packets get dropped. My packets are obviously higher priority to me. If John Doe down the street really needs tons of packets, let him pay more. I don't think anyone opposes tiering on the consumer side.

Re:Why?

[Chirs]

"I don't think anyone opposes tiering on the consumer side."

Um...yah. I do. Your example is not tiering, but rather simply paying for bandwidth.

Tiering would allow your ISP to define their traffic shaping, so that maybe Vonage VoIP packets get dropped while AT&T's VoIP packets get through because AT&T owns the network. Or they decide that newsgroups are unimportant and so now they take forever to access. Or maybe your neighbor fires up his TV-over-IP and all the sudden your dropped packet rate spikes because he's taking up all the bandwidth.

Re:Why?

[Actually, I do RTFA]

No, what you are describing is content based tiering (which is evil). I meant tiering, just on the consumer side. For instance, I would love to have a guaranteed connection of X, with an up to Y ( Y > X ) connection speed when bandwidth was available, assuming it was cheaper than just Y bandwidth. That is, have a tier 1 connection of X, and a tier 2 connection of Y-X. As long I can use my VOIP and some small webbrowsing on the side.

Article title misleading

[ao_coder]

Neutral net COULD need AS MUCH AS twice the bandwidth. Let's not spread worst-case scenario memes just because it looks good as a headline.

are they not selling bandwidth?

[fermion]

I do not know if they are saying that the need for bandwidth is a good or bad thing. What I do know is that most people who sell stuff are not looking to sell less of it. They may package it so they sell less at a higher profit each time, but they are trying to sell more. For instance ATT is now trying to sell integrated packages to pay for all the cable they are laying out. These packages coincidentally are in some cases asking you to pay for stuff that you can get free on the internet. Additionally bandwidth is not so limited as to need be rationed during peak time, such as, for instance, is energy.

If ATT considers bandwidth to be a problem, here might be why. The baby bells as the 90's were ending felt the need to halt the progress that was being made on opening up the bandwidth market. They slashed prices on DSL and created real impediments to competitors. For instance, it is not always possible to have multiple third party services on one line. In the process they created a situation in which bandwidth was very cheap, and service was very bad. If the study is correct, I think we will find taht ATT is undersupplying the market in bandwidth, assuming that everyone will buy their new services. IN fact I would say it is, after only 7 years, nearly impossible to find a DSL plan that actually provide the level of service we were used to in the late 90's.

So ? as long as i PAY for that bandwidth,

[unity100]

who are they to tell me how to use it for god's sakes. i bought it, i use it, and you cant tell me anything about it. dont sell it if you dont want to.

title should be

[Anonymous Coward]

neutral net provides twice the bandwidth for the same cost

Dark Fiber

[Ranger]

So what happened to all that optical fiber that was laid but unused during the dot com era? Did it finally get used? Every company wants to emulate DeBeers and create artificial scarcity, so they can jack up the price. If spam were ever gotten under control I would imagine that there would be no need to increase bandwidth.

Smells like FUD to me.

Re:Dark Fiber

[the eric conspiracy]

There is a lot of dark fiber in the ground along major backbones, but that doesn't mean the switching and last mile distribution to use it is in place. And a lot of that fiber is not up to specs needed for 100Gbps DWDM.

Bill Microsoft for malware traffic then

[gig]

For years a significant portion of Internet bandwidth is faulty Windows computers distributing malware to each other because Microsoft deviated from standard industry practice with regards to network security.

If you're going to start being stingy about bandwidth I suggest network providers bill Microsoft until their tire fires are put out.

More like "So?"

[Anonymous Coward]

AT&T want more money but they don't want customers to see they're being charged more. So they want to remove Net Neutrality. So they have to call it something else (who would buy "network hegemony"?) so they call it "Tiered service" (forget that selling low/med/hi speed broadband is "tiered"). But people aren't really buying it, so you have to show that this is a good thing.

So show that bandwidth is going to cost 2x as much for NN and you can either justify doubling the cost to the users or justift the Tiered service as "cheaper".

So what would you expect AT&T to do?

"Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.

It's been 43 minutes since you last successfully posted a comment"

Just Because

[pentalive]

Just because it takes 2 x the bandwidth for a neutral network does not mean more infrastructure needs to be built and more cost needs to be expended. Who says we are not currently operating at 1/2 or less of the available bandwidth. What about all that "dark fiber" I've heard so much about??

Our neutral internet?

[DCheesi]

As you might imagine, our neutral internet is far more bandwidth-intensive;

"Our neutral internet"? Obviously the OP doesn't realize that ISPs are already managing network flow, and have been since such management technologies first became available to them. Colleges and other ISPs already try to identify and downgrade bandwidth-intensive torrents and such (if not block them outright), and cable ISPs already give special priority to their own network services (cable-co. VoIP plans, etc.).

This then brings up another point: There are two different issues being lumped together under the "net neutrality" name. The first is simply traffic-shaping and bandwidth management in general; the second is allowing ISPs to charge individual content providers for higher priority within their networks. One is already in place, and is often (but not always) a very good thing. The other is objectionable on several fronts.

I think most people who advocate "net neutrality" are objecting to the second issue, whereas most pro-telco arguments (including this AT&T study) are focused on the first issue.

What Happened to All the Overcapacity?

[rbegga]

To the Bandwidth Providers:

We keep hearing these arguments from the Telco's and Cable COs about how much more difficult it will be to build and maintain an open Internet because of the bandwidth requirements that imposes. Enlighten us as to why this is now a problem considering the major Telecom bust that occurred a few years back was due to the overcapacity you had built into your networks? Google is going around buying up dark fiber from you guys while you're complaining about lack of infrastructure? Nonsense. I don't believe you guys can't figure out a model to make this work for you and us without getting the government involved.

Re:What AT&T actually means to say is....

[Inglix the Mad]

No kidding.

"Corporate sponsorship of research doesn't automatically invalidate that research..."

No, but when it's AT&T you look 3 times because this is the same company that lied about the impact Net users had on the phone circuit system so they could try and get per minute charges levied. For those not in the know:

In the 90's the now AT&T phone company claimed that Internet users whom stayed online were a danger to the system as they took up so many resources. In Wisconsin they tried to get per minute charges levied against people whom used the net. Why? Well you have to travel back in time a bit and realize the Bell business model was "by the call" for local calls. If you made 5 calls a month instead of 50, you were costing them money. Fortunately some college kids, UC Berkeley I believe, bought some used phone company hardware and showed that creating the dial-tone was the drain on the system, thus net users whom stayed connect for long periods of time actually used less resources than your typical phone caller. Why? Net users might make 3 or 4 calls a day (Internet had started going unlimited) while the typical home user might make twice that (unless they had teenagers, then all bets were off).

I actually trust the cable company more than a phone company, and I don't trust the cable company much.

Duh

[the eric conspiracy]

It should be obvious that if you don't prioritize traffic based on QoS requirements that you will need more bandwidth. This has been a basic given for many years now. The question is what will it cost to prioritize the traffic to meet a given QoS level vs. just adding bandwidth.

There are a lot lot of people who think the various prioritization schemes that have been proposed just won't work because they are not scalable - while a fast dumb core is.

To me the problem with prioritization is that it is just harder to implement, and once it is in place it makes management harder. Also it tends to place limits as to what you can do on the IP network. Fast-dumb doesn't have these problems.

The problem with free market and the internet.

[Repossessed]

AT&T is the primary company pushing to be allowed to do this. I am a Comcast subscriber. This is my traceroute to google.com.

3 ge-5-4-ur01.saltlakecity.ut.utah.comcast.net (68.87.170.161) 9.116 ms 9.247 ms *
4 te-9-4-ar01.saltlakecity.ut.utah.comcast.net (68.87.170.9) 9.021 ms * 9.210 ms
5 12.116.47.117 (12.116.47.117) 19.295 ms 20.255 ms 19.232 ms
6 tbr1.dvmco.ip.att.net (12.122.86.250) 46.279 ms 46.672 ms 45.820 ms
7 tbr2.sffca.ip.att.net (12.122.12.133) 45.180 ms 45.821 ms 45.441 ms
8 ggr3.sffca.ip.att.net (12.122.82.149) 47.504 ms 47.508 ms 47.932 ms
9 att-gw.sanfran.level3.net (192.205.33.82) 167.304 ms 48.359 ms 45.286 ms
10 vlan69.csw1.SanJose1.Level3.net (4.68.18.62) 57.119 ms 49.613 ms 52.738 ms
I also point out that we already have a tiered network. so many MB/s costs so many dollars. Both for the provider *and* the consumer. AT&T is trying to make companies pay *again*. This shouldn't need more laws. This should be classified as extortion.

That said, I'm wary of net neutrality laws, Because from my understanding, the network is already managed. One of the local ISPs CEO did an interview in Wired, where he talked about how his company was already giving priority, based on what customers demanded and what needed the priority most. (VOIP service for example, gets high priority because disruption there matters more than elsewhere.)

This doesn't mean we shouldn't have net neutrality laws, Just that we need to be very careful about writing them, so that legitimate (non extortion) methods can still be used.*

*Though while we're at it, I wouldn't mind seeing it made illegal for college campuses to restrict how dorm students use their internet. There's really no excuse for cutting off somebodies access to communication (IRC is usually the first thing to get hit with idiotic security policies). And from my experience, dorms not only qualify as a monopoly ISP, but typically a mandatory monopoly as well. (I've even seen colleges, public ones, that require freshman to live in the dorms.)

Multicast

[Kludge]

Speaking of reducing necessary bandwidth, when are these ISPs all going to push multicast for media delivery? Isn't this a no brainer for reducing bandwidth?

Re:Multicast

[fwr]

Yes and no. It would be a no brainer if the majority of demand were for live broadcasts, or recorded broadcasts that were played to everyone at the same time. However, it would do nothing, and probably make things worse, if you were to multicast all media, as I believe the vast majority are on-demand, like when an individual user clicks on a YouTube video. Setting up various "channels" that are broadcast on specific multicast addresses, and then using IGMP on the edge devices to join the multicast group to receive that particular live broadcast channel would of course minimize that traffic and reduce bandwidth. But as far as I know the percentage of bandwidth being used for broadcast multimedia of that type is minimal; I believe the vast majority of media bandwidth is individual on-demand transfers.

Plus, you'd have to get everyone to agree on the multicast routing protocol to use, and all of the edge devices ("consumer" cable modem router, DSL router, wireless router etc) would all have to be replaced or upgraded to provide the multicast capability. As far as I know, none of them will proxy the IGMP join messages from your inside network at home to the ISP connection.

Gotta go!

Re:Multicast

[amorsen]

Speaking of reducing necessary bandwidth, when are these ISPs all going to push multicast for media delivery?

Multicast is not going to happen. Every router needs to be aware of every multicast group (e.g. every video and radio station) that someone downstream from it wants. There is no way that can be done.

As always with the Internet, the answer is to make the end points smart and the core dumb. In this case that means the end points should pass the traffic on to other end points close by. AKA peer-to-peer.

Re:Multicast

[PKFC]

My VoIP teacher told me about someone he knew working on the Alberta Supernet. http://en.wikipedia.org/wiki/Alberta_Supernet

Huge fibre backbone in Alberta to connect the province. Supposedly after all the infrastructure and money and smarts going into the project, the conclusion of why they couldn't implement multicasting was that it was "too hard".

I don't know if that's just complaining about multicasting, a (time) cost vs benefit analysis or just simply pure scaling difficulties, but I guess multicast sucks at being usable...

Barely touched the stuff in CCNP school courses

Give us what we PAID before for now, then talk

[unity100]

You are not even providing the bandwidth you were supposed to provide with your advertisement and rates and all the public funding you got. First, give us the promised bandwidth, then talk about other stuff. You were selling products that you were not able to deliver - this passes as fraud in any country, court in the world.

Let's try it again.

[seebs]

Okay, who wants to try to come up with words which provide network neutrality, without preventing me from blocking spammers. :) (Note: I would, of course, be willing to let someone who wants to put up some money to back his claim that his mail is legitimate send me mail. After all, the problem with spam is that, since it's free, it grows without bounds. If it costs money, he's not going to send tons of crap.)

Re:Let's try it again.

[Qzukk]

Okay, who wants to try to come up with words which provide network neutrality, without preventing me from blocking spammers. You're welcome to block all the spam you want. Blocking email based on whether or not the sender shelled out cash *cough*goodmail*cough* isn't "blocking spam", just like blocking Youtube based on whether or not they shelled out cash for the bandwidth your customer was already paying for isn't a "tiered internet".

Around here, the words we use for "network neutrality" that don't prevent you from blocking spammers are "status" and "quo". We're not the ones trying to change the way the internet is run.

It's not about efficiency, stupid!

[Anonymous Coward]

It's also not efficient to have thousands of newspapers and radio shows. It would be much more efficient to have only one newspaper, one radio network, one television network, one computer company, one research and development group, one scientific union, one answer to the question about life the universe and everything, one set of rules, one bureaucracy, one queen. Elvira, where are you when we need you?

Twice what?

[AnyThingButWindows]

George Orwell warned everyone about doublespeak. Net Neutrality is the internet as it is NOW. Unregulated, untouched, and un-fucked with by the Bells.

So the headline states that we need to double the bandwidth we have now, in order for what we have now to work?
That makes no sense what-so-ever.

Re:Twice what?

[Chirs]

No, they're saying that if they introduced tiering and prioritization they could double the useful traffic without changing the overall bandwidth.

Re:Twice what?

[the_greywolf]

... by saying that they need to physically double it to keep it as it is.

I say double it.

There's a _very_ simple answer

[samael]

Charge people for the bandwidth they use.

If ISPs weren't wedded to unlimited plans for their customers then they could charge people for what they actually use and not have to worry about charging at both ends.

no sh*t sherlock

[Deadplant]

capping bandwidth reduces bandwidth requirements! film at a 11!

Re:What AT&T actually means to say is....

[Actually, I do RTFA]

I actually trust the cable company more than a phone company

And I trust my Ouija board's lotto picks over my fortune cookie's

NNs

[benhocking]

Let's take the simplest kind of NN, where literally every packet has to be treated the same.

OK, at this point I'm going to have to object. As someone who works with neural networks, I feel compelled to say that this has got to stop. It was bad enough that "neutral networks" has only one letter different from "neural networks". You are not allowed to call them NN. No, not yours. :)

Ripoff Talk

[Doc Ruby]

What this research really proves is that a neutral network needs only double the bandwidth to replace a tiered one . Doubling the bandwidth is a lot cheaper, faster rollout, and more manageable than a tiered network. And it's more scalable than a complex tiered network. Plus, it has twice the bandwidth. And it doesn't have the flexibility and openness of a neutral network.

These Net Doublecharge crooks will say anything to get their extortion money. I expect they will, because they don't care about us, just their money and political power. But why does Slashdot have to publish it? Slashdot, a big website, is a target for Net Doublecharge, which will blackmail Slashdot's servers to carry its traffic to nerd consumers.

Let's not only pay them to give us the Internet that we built for them with our taxes and scientists, and created demand for with our content and services, and also peddle their lies that are stealing the whole thing from us.

Re:Ripoff Talk

[Anonymous Coward]

http://knowledge.wharton.upenn.edu/article/1497.cf m

I think the core assertion of the anti-neutrality folks is that higher
bandwidth users are not paying their fair share. I'm not buying that
right now - I thought if you want more bandwidth, you must pay for it
- just like home users. It's always been like that! Am I missing
something?

It's more and more seeming to me that the core issue is based around
the legal right to allow and block traffic across your internetwork
based on who is sending it. If you disagree that telecom companies
should be able to do this, you support neutrality.

"According to Werbach: "There are really two issues in the network
neutrality debate: Should government step in when broadband network
owners discriminate against unaffiliated content and services, and
should there be a prospective rule mandating non-discrimination? I'm
very troubled by the possibility that network operators will act in
anticompetitive ways against application and content providers, but I
find it hard to craft a workable legal rule prohibiting such actions."

I am more and more thinking the 'paying for higher useage' thing is a
red herring. I think that already exists.

The issue is the ability to block/allow traffic based on sender.

Re:Ripoff Talk

[Doc Ruby]

You are correct. Especially because the telcos/cablecos already have their own private research proving to them that it's cheaper and more manageable to solve network congestion problems by adding bandwith than by tiering or otherwise applying "Quality of Service" constraints on traffic. I know they have, because have interviewed for the NYC government network analysts who delivered that research to those telcos/cablecos.

The main issue is that telcos/cablecos all want to sell their own bundled apps, especially TV and phone, on their own networks. So they want to do to competitors, who must use their infrastructure in order to use "the" Internet, what the telcos did to their DSL competitors. Jack up prices and create unpredictable connection/service delays and outages to retail customers competing with them to deliver services to the same end users. The DSL industry was completely destroyed. The telcos/cablecos want to apply the same strategy to the new TV/phone industry before it ever gets a chance to compete with their huge incumbencies, but after the entrepreneurs have sparked the market with some early innovations. The last we'll ever see.

All of which is the real business reason they want the technical ability to block/allow traffic based on sender to be legal, and in fact continued to be subsidized by the government. Which means by us, the consumers in their cartel market.

The Tiered Network I Want

[Nom du Keyboard]

I want a tiered network.

I want different levels of service for different services.

And I want to be the one who sets those levels. After all, it's MY bandwidth. I'M the one paying for it!

Let me be the one to decide to put my preferred VoIP provider at the top. It doesn't use that much bandwidth overall, but response is important. My Bittorret goes at the bottom. Web browsing in the middle, and on-line gaming above that. You can guess where YouTube fits in.

I, for one, believe that if the ability to shape traffic exists, then the end user should be the one doing the shaping, and that's what we should be fighting for.

Re:The Tiered Network I Want

[PPH]

That's reasonable. You should request a certain QoS from your broadband provider with some price attached. They, in turn, should contract with a backbone provider for certain rates for each tier and at the other end of the backbone, the peer system or server you do business with should have a similar arrangement.

What the backbone operators seem to want (and I think its their major objection to net neutrality) is the ability to inspect the packets passing through their system and asses charges beyond their interfaces against service providers or users who might be in competition with one of their retail products. Lets say you use ACME ISP and your buddy uses PDQ ISP and you decide to initiate a VoIP connection. You each negotiate a price with your respective ISPs for the requisite QoS and they in turn buy a block of bandwidth through a backbone operator. Now, if this backbone operator happens to be AT&T and they sniff your packets and see that you are using VoIP instead of their long distance service, do you really think they should send you and your friend bills? They do.

Re:Wait a second...How About...?

[Nom du Keyboard]

I think at the end of the day, a clearly defined set of standards for prioritization needs to first be developed by an independent body (ICANN/ISO/IEEE?)

How about the EFF?

the problem he describes ignores the free market

[Anonymous Coward]

how about this? you get 512-768k internet access for $10 a month, you get internet access that reliably supports IPTV type traffic, for... what's the term? an economically competitive price. if apple, verizon, or blockbuster want to sell streaming movies, or movie downloads, I don't see the problem, but if the consumer wants broadband, they have to pay for it, if they want reliable 5mb so they can get near dvd quality they'll have to pay around $50 a month in the current market. if they want 36mbps, well then they're going to have to wait for the market to work it's magic. commercially viable 36mb internet to the big US markets isn't going to happen overnight, but I don't see where you need to tier the internet. newsgroups seem like the ideal mechanism for this, that or bittorrent. media companies like apple, netflix or tivo will have to start small with adaptive compression or something

Here's the difference

[benhocking]

The difference with IP as property is that, within the bounds of patent law, no one can even have property "like" it. If you have own a car, I'm free to buy a car just like it. If there were no patents on it, then I could build my own car just like it. (Maybe I could anyway, IANAL.) The analogy to it being a monopoly is not a bad analogy.

That said, I'm not against IP in theory, although I think your 30/60 numbers are a bit extreme. Unfortunately, the numbers that make sense depend on the field. In CS, 1 year/2 years might make sense. With drugs, maybe 4/8. There's a trade off between how long it would take one to make a reasonable (i.e., motivating) profit, and how long something is actually useful for. Patents were designed to encourage invention, not guarantee virtual monopolies.

Re:Here's the difference

[sid0]

The numbers were for copyrights, not patents, and yes, it will have to be studied how many years patents should apply for each field.

If I made a new type of (say) egg beater, and got it patented, then no, you can't make an egg beater just like mine. The government simply protects my right to make this egg beater exclusively for a limited time. Whether you look at it positively or negatively is your wish, and "monopoly" has negative connotations. I look at it positively. Anyone can come up with an even better type of egg beater... I'm not stopping anyone from doing so.

This even extends to non-patentable, tangible property. If I have a beautiful tract of land near a river, I don't think it would be possible for you to have a tract of land "just like it".

Re:Here's the difference

[Vancorps]

You don't think someone could take a picture of your property and build their property just like it? It's not only perfectly legal, it's not even that hard to do. Start small and look at cookie cutter housing in a lot of suburbs. Expand out and it's not that hard to make an artificial river if you're so inclined. If you already have a river nearby then it's even pretty easy. Getting trees that look identical might be a bit more difficult but I imagine it would be okay if branches were off by a few inches or feet.

With IP you simply can't do this as it would be considered infringement. Think live performances in bars for a recent example of this gone awry.

30 years for a copyright is atrocious though. If you're still making money on your copyright after 30 years then your product is big enough that the public is better served by moving it into the public domain. I would say 10 years or even 7 years would be far better terms for copyright. Of course that also depends on the field, if it's music versus software then it's usefulness will change over time. Of course that's a bad slope to be on since theoretically it would move people towards industries with the longest copyright terms. Of course I'm not sure how much money Microsoft is making on Windows 95 right now, that's 12 years old. Imagine if that copyright expired right now. The monopoly would start to crumble and Microsoft would be forced to come out with products that are actually new rather than resting on their past success.

Re:Net Neutrality-Except it Isn't!!!

[Nom du Keyboard]

Net Neutrality Positive
VOIP Packets receiving priority (because lag and bandwidth throttling reduce performance of VOIP technologies)
Prioritizing Gaming traffic of popular/well used games (IE. MMOs, FPS over internet, etc...)

Except that the only VoIP that your giant ISP will prioritize will be their own, overpriced version. It will be used to kill off all other VoIP competition.

As for the gaming, expect to pay a monthly surcharge, which might be hidden in the monthly game charges if they collect it from that end, for access to the fast lane.

Bandwidth is cheap. Marketing is expensive

[Animats]

Bandwidth isn't where the cost goes. The biggest costs for an ISP are in in marketing and customer support, not bandwidth. If you look at wholesale ISP rates, what an reseller pays per customer for the raw service, they're less than half the retail rates.

The key to cost control is keeping down the number of people involved. Fibre is cheap.

Re:The tiered aspect should be on the client BUT

[Nom du Keyboard]

Seriously, if you offer a base broadband package for $10/month with 2GB of download bandwidth included, and $0.25/GB after that,

Excuse me BUT, AT&T already has to offer a basic DSL connection for $10/mo with no cap. This is part of what they agreed to in order to be allowed to merge with another provider. While they hide this fact as best they can (don't believe me, go try to find it on their pages), it's already there. I'd get it for my mother, if it was just available in her area because it's all she really needs.

And metered bandwidth is bad because they'll lower the amount in increments, and jack the cost in single cent amounts, that don't sound like much, but really add up over time.

Perhaps you don't recall the day that 411 information calls ceased being unlimited and included in your basic phone bill. They started out with saying that we'll set a real high number of them before you have to pay. Something like 15 a month that will only affect the single percent of our heaviest users who are too inept to use a telephone book. Before you knew it the free calls were down to 3 a month, then you started paying for every one of them.

And the cost savings you realized on your bill from losing this previously included service: NOTHING!!

First stage complete; second stage underway ...

[jacques2cents]

The mantra of business is to make money. Otherwise, it isn't a business, just an unregistered charity. The Internet is the largest cash cow on the planet right now. Oil? Ha ha. Next week someone might "invent" cold fusion for real and the Arabs will be sucking wind (until they remember that plastic still needs oil); whew [in Arabic]! So how does one profit from the Internet? Well ... what is the Internet? In the old days, it was a few of us nerds exchanging the latest APL code or the latest dirt on some guy named Bill reaming poor Gary to make nice with IBM? Ah ... the good old [collegiate] days. But I digress. So here we have the new and improved Internet. Formerly the haunt of the techie few in pursuit of various scientific holy grails ... now the gateway to entertainment and porn ... with some science still happening in this odd corner or that. Before, no real profitable venues ... I mean who is going to pay for the data on yet another thesis on how queues at the grocery store work, or just how many penguins are at the North Pole, eh? But now! Now we have "valuable" information flowing like water over the Niagra Falls; and at the speeed of light no less, so to speak. Now we have something that people are likely to be willing to pay cold hard cash for. I mean who doesn't want to be entertained and so on and so forth. But how to profit from this? One has to have control in order to profit, eh? Stage one: Control the devices that are commonly used to access the Internet. With Vista and .NET programming firmly in place, Microsoft now has all the tools necessary to control the hardware and software which 90% of the planet uses to access the Internet. Okay, so maybe it is only 80%. Frankly it don't matter much. With the TRILLIONS to be made, even 1% is significant, eh? "What about cell phones, et al?" I know someone is going to ask. Just the fact that you asked means you haven't a clue. Please go away. "Wait, what did you say about Microsoft?" Please. Haven't you noticed that every time you touch the Internet your computer's operating system and any other Microsoft software calls home to see if there are any important upgrades it should have for your protection and benefit? Here's a prognostication. Next year, after Microsoft starts monthly billing for the use of their software and the sending and receiving of email using their software, you get rebellious and refuse to pay. The next time you touch the Internet, you're computer will die. The operating system will turn off and your applications will cease to run ... you must pay the bill my son. Just like the electric bill, the phone bill, and the water bill (I assume most of you don't pay property taxes). You're addicted. You will pay. :-) BTW, this form of control is ALREADY live and operational. Just change a significant piece of your hardware and see how fast your Microsoft operating system refuses to operate properly until you've re-registered your computer with corporate headquarters or their assigned representative. And it isn't just them. Last week I tried to convince an Adobe tech support person that the Adobe application on my computer was legit ... I mean it was 4 years ago when I registered it the first time ... but their database has been updated and now I have to actually send them a copy of the cash register receipt to prove ownership at which time they MAY send me a NEW registration number or they may just tell me to blow smoke and go buy the current version (the first suggestion from tech support BTW). Stage two: Control how these devices connect to the Internet. Wirelessly, via cell phones, et al, is already very tightly controlled by the various service providers and you pay a healthy royalty to do so. The profit margin is very nice and is enabling the service providers (ie, AT&T, Verizon, etc.) to attack the side of the business they couldn't control profitably unt

You can defend your house

[Anonymous Coward]

but you need the police to enter my house to see I'm not breaking your copyright.

THAT is why it is different.

And in any case, your answer to my question was "no". So why is it that one more little bit of government interference is worse but one little bit less is ALSO worse? How did we manage to get JUST THE RIGHT AMMOUNT?"

"Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.

It's been 45 minutes since you last successfully posted a comment"

Re:You can defend your house

[sid0]

You also need the police to enter my house to see if I'm the one who stole the goods.

Protection of property and intellectual property is a legitimate function of government. You can call it "interference". I don't. I'm not a "free-market" anarchist. ("Free-market" is in quotes because I don't think it really is a free market, without a government protecting property.)

When will people realise...

[SpryGuy]

When will people realize that what is best for the interests of an individual corporation's bottom line is NOT necessarily in the best interest of society as a whole?

Imagine if corporations took over all highways, freeways, and streets, and decided to turn them into toll roads to "manage" them, and further, decided on varying rate structures for different people with different destinations or subscription plans... which of course varied from state to state based on which corporation owned the road.

Imagine if corporations took over all fire departments and police departments, and decided to turn them into profit centers. Imagine calls for emergency services being denied because you hadn't paid your bill or because there wasn't money in it for them (i.e. it was better for their bottom line to let your home burn or be burgled).

Imagine if corporations took over all health care, and decided to turn them into profit centers. Oh, wait, that's happened. And our health care is the slowest most expensive in the world, and among the least effective society-wide (i.e. it's fine if you have tons of money and influence, but otherwise kinda sucks... go see "Sicko" if you don't believe me).

Imagine if corporations took over the news media and decided to turn them into profit centers.... again, that's happened, and our news media sucks.

I think it's clear that what is best for society is to treat the internet as a "common carrier" that must be open and neutral. Period. We've seen the disasters that result in corporate control of the common good. We need to prevent another common good from being destroyed by corporate greed.

Re:When will people realise...

[SterlingSylver]

I think it's clear, at least in your first two examples, that people *do* realize that what's good for a corporation's bottom line is not good for society as a whole. Thus our roads are public, police and fire are public, etc... There's even been something resembling a dialogue lately in the US as to whether or not health care should be privatized. As for media centers, there are plenty of arguments as to whether or not they should be public or private.

After all, who gets to decide what is "best" for society?

What this really maens

[Big Bipper]

What this really means is that on average the ISPs are screwing over half their customers, who are paying for something the ISPs are not providing. What the ISPs really want is to screw over three quarters of their customers, and charge the other quarter twice as much.

Confusing Terminology

[PPH]

Why do these studies and articles confuse net neutrality with tiered service? There is nothing that I can see about net neutrality which rules out tiered service. IMHO, net neutrality is all about providing equal access to these service tiers to all users at uniform prices.

You want a fat pipe run to your house or business? You pay more than the people with dial-up. If a backbone provider (AT&T, for example) offers differing QoS routing through its system, it should offer a consistent pricing structure to each ISP or customer at its periphery based only upon volume, QoS, etc. and without regard to whether these parties are partners or divisions of the corporate parent.

If we can't get that kind of level playing field voluntarily, perhaps its time to split carriers and content/service providers apart by regulation.

It isn't.

[jd]

The study is flawed from a number of perspectives. First, as you mention, the current costs to consumers vastly exceed the costs to provide this mythical doubling in bandwidth. Secondly, network neutrality (to me) means that everyone gets the same experience and access as everyone else. This means that if you implement user-neutral traffic management (packet dropping schemes to minimize retransmits, throttling streams at congested points, fair service to ensure no connection is stagnant, smarter routing algorithms that avoid segments that are basically dead in the water) and user-neutral methods of improving data distribution (web caches, multicasting, SRM) then you gain virtually all of the benefits AT&T claim for their non-neutral system - and more besides - without harming a single user.

The above remedies would give all of the smoothing at peak times on heavily loaded routers, but in a manner that is entirely equitable and - get this - doesn't actually reduce the service provided to anyone. The peaks that kill the backbones are not particularly long-lived and contain a vast number of unnecessary retransmits, inflating the traffic levels. Schemes already exist that can potentially halve the retransmits and diffuse the load over just enough time that it can be handled. Other schemes already exist that can eliminate unnecessary repeat transmissions from source, massively reducing the load on the most burdened segments.

None of these require that any user be given priority or special privileges. None of these require that neutrality be compromised. Yet none of these require that either services or end-users experience any detectable delays (at worst) - and most of the time, both services and end-users will experience a much faster, smoother Internet.

Of course, you'll never get AT&T to admit that the reason they can't do any better is that they're not only greedy but also technologically incompetent. Nonetheless, that is the reality of the situation. It is also something missing from said "study".

Tiered network could be OK.

[weston]

Sure, develop a tiered network. Charge a premium for the higher speeds and QoS. Let your customers and peers decided if they want to pay you to use it.

Just don't try to shake down the customers of your peers. You have a problem with traffic coming through a peer, you work it out with them.

OVERPROVISIONING THE NETWORK

[Anonymous Coward]

DUH!!! that's why it's called overprovisioning the network; increasing the bandwidth ad infinitium is the only genuine way to provide for increased use. grow the pie, dammit!

Aren't the costs the same in the long run?

[swb]

Assuming a tiered internet, you have telcos investing in the manpower and equipment to tier their services, and them passing these costs onto the sites paying the tiering fees, who in turn raise their prices to cover the tiering fees. In the end, you have $N spent.

Assuming an untiered internet, you have telcos investing manpower and equipment increasing bandwidth. These costs are passed onto bandwidth consumers, who in turn raise their prices to cover the cost of increased bandwidth. IN the end, you have $N spent.

From a very rough economic perspective, it looks the same to me, but one leaves the telcos with the ability to raise prices arbitrarily and squeeze supply artificially.

Corporate Research

[vertinox]

Corporate sponsorship of research doesn't automatically invalidate that research

No, but it sure does include the high possibility of it being biased.

An amazingly small multiplier!

[Tired and Emotional]

Twice the bandwidth is an amazingly small multiplier. Its the sort of growth factor that one can reasonably imagine being accomodated with improvements in technology and build-out over a short period. So one has to ask what was the motivation for coming up with that number.

Seems to me what they are thinking is that all the managed stuff will fit within existing capacity and then the unmanaged stuff requires new capacity. Or, to put it another way, all the available capacity needs to be managed.

So the real statement here is "we need to close down the internet as it exists today so we can repurpose the network in order to generate greater revenues".

Not to mention

[hey!]

It's actually a rather boneheaded comparison.

It's like saying a tractor-trailer requires an engine with 20x the torque of a family sedan. Well, yeah, because they do different things.

A net neutral network provides a level playing field on which content providers can enter without barriers and compete against anybody.

A non-neutral net does not provide the ability of content vendors to enter the market on an equal basis without setting up a special deal with a network bandwidth provider. In practice, this means content is bundled with service, and that the network providers provide a poor selection of content.

If you want to see what a non-neutral net looks like, look know further than cell phone companies, who have all lame proprietary multimedia and information services. You'd have to be a fatuous rich person to spend good money to watch Verizon videos on your phone (apologies in advance to any fatuous rich /. readers). The main thing these offerings do is make it much more confusing to compare prices between competing wireless vendors. How much does that cost the market every year? We'll never know.

And to top it off, what is the most exciting new phone product in years? The iPhone. A device whose multimedia features (iPod, YouTube access) are not tied to the bandwidth provider.

So, yeah, I'll take the tractor trailer if I'm moving 50,000 pounds of bananas to market. I'll take the neutral net if I want to have a vibrant content market.

Devil's advocate

[benhocking]

Whether you look at it positively or negatively is your wish, and "monopoly" has negative connotations. I look at it positively.

Well, I choose to look at monopolies positively. Have a monopoly allows a company to deliver the highest quality goods to its customer by removing inefficiencies inherent in competition. :)

This even extends to non-patentable, tangible property. If I have a beautiful tract of land near a river, I don't think it would be possible for you to have a tract of land "just like it".

That's a valid point. However, I think I would be able to (legally) have a tract of land a lot more like it than I would an egg-beater like your new invention. Another major difference between IP and other property is that if I were to copy your egg-beater design, you would still have your original design. To the extent that I cannot have a tract of land "just like yours", it's only because doing so would deprive you of it.

IP laws should be designed to foster invention, and not hamper it. The point of IP law is not to "protect" the inventor; it is to encourage invention. Although the details about how to best encourage invention are arguable, I presume that we agree on the what the goal of IP law historically has been and what the goal should be again in the future. Correct?

Re:Devil's advocate

[sid0]

Have a monopoly allows a company to deliver the highest quality goods to its customer by removing inefficiencies inherent in competition. :)

Lifted that out of The Godfather, didn't you? ;)

As for IP law: it is meant *both* to foster innovation and to protect the rights of the inventor. Ripping off a product as soon as it is invented will not protect the rights of the inventor, nor foster innovation in the long run. I really have no problem with copyrights, patents, trademarks and trade secrets: all foster innovation and protect the rights of the inventor in different ways. The current implementation isn't really great, though.

Simple answers

[einhverfr]

I am actually in favor of tiered network on ISP sides, provided that remote endpoint is not considered in routing priority (though bording network may be subject to tiering agreements between providers) and provided that it is not used to crush competing services.

I think that it is possible to have the major benefits of network neutrality and still allow tiering policy routing on ISP's and other traffic management techniques.

Here are my answers to your questions:

The question is: How do we decide what traffic is more important on the Internet? Not a valid question. The needs of the application should determine what is more urgent, not what is more important. THe issue usually is latency, not bandwidth.

For example, VOIP traffic is always more urgent than email. THerefore it is a possibility for tiering agreements.

On the other hand, certain problem protocols (like Kazaa) might be deprioritized so that they don't interfere with other customers' use of the internet.

Now, note that I said something about anticompetitive uses before. I think that if an ISP is a part of a business entity which provides voice service, they shouldn't be allowed to block or deprioritize VOIP. They may still be able to sell priority routing agreements, however.

Who pays? Who pays more? If an ISP wants to sell priority routing agreements for VOIP, that is fine with me. Do you want your Vonage call quality to go down just because some idiot is flooding the network with Kazaa? If not, buy a tiering agreement for that VOIP traffic.

I do not think that ISP's should be allowed to deprioritize traffic from content providers as a way of extorting money, however. Nor do I think that ISP's should be allowed to prevent you from running services that compete with their offerings (such as email and web hosting), and that any ports that they choose to block should be unblocked upon request and at no extra charge.

I also think that one should target the desire to extort money or crush competition rather than the means. In the technology world, the means will change frequently but the desires may not.

Re:Simple answers

[Em Adespoton]

The problem here is that companies like AT&T own the lines, which means they are EVERYONE'S ISP at one point or another.

This is similar to a road system: say that AT&T owns a ferry system to get people quickly from point A to point B.

They also happen to own a few hundred yards of the roadway that is the alternate route. Only Comcast busses are allowed on that road, as they own most of it and won't let anyone else drive on it. However, AT&T has set up a toll booth on the bit of road they own, and they charge the same amount to cross their patch of road that they charge for the ferry service, hoping that this will encourage passengers to take their ferry instead of the Comcast bus.

See the problem? In this instance, AT&T is Comcast's ISP for a short distance.

Now let's introduce the Vonage tour system. They want to route passengers from A to B through the AT&T/Comcast morass. They consider that they should have higher priority than John Q Individual trying to get on a bus/ferry to make the trip, so they rent entire ferries from AT&T and busses from Comcast, degrading the service of the average individual. Eventually, AT&T notices that Vonage is making a handsome profit in its business, and decides to get in on the deal.

What do they do? They could provide a competing service, but this smells of anticompetitive monopolistic actions. However, all they really have to do is raise the toll on the Comcast road. Comcast will then charge Vonage more for their bus rentals, and Vonage will be more likely to rent ferries. This will allow AT&T to raise ferry rates in order to supply customers with more ferries. Meanwhile, they can also raise the rates they charge Vonage for ferry rentals, and add in a "group plan" for individuals that is better than individual tickets, but not as good as the Vonage rate.

Sounds good, right? Only problem is, they've still blocked off traffic on the Comcast road, and have priced Vonage out of the market, without any direct competition to either service.

All the end user sees is degrading service and higher rates; their transport providers all say "well, there's nothing we can do about it -- the problem is with the people providing us access." Those people just shrug and say "we can't do anything about it; it costs money to maintain this stuff, and as you can see, the demand is outstripping our ability to supply dependable routes to all transport providers. We're doing what we can. Oh, and here's our bundled service; you can buy a pass from us that lets you use both our ferries and our roads for a reduced bundled rate!" Meanwhile, Comcast has to cut rates to survive, resulting in potholes in their roads, slowing traffic even further.

I know, there's a lot wrong with this metaphor, but with the state that switched packet data transfer is in today, it's not an easy solution no matter what you try to do.

Re:Simple answers

[xappax]

This is similar to a road system: say that AT&T owns a ferry system to get people quickly from point A to point B.

Look, I don't know how many times I need to say it: The Internet is not something you just dump something on. It's not a big truck.

Re:Simple answers

[Em Adespoton]

You seem to be confusing the Internet with the hardware used to transmit packets between locations. The Internet is a concept, not a thing; AT&T owns some of the "thing"s required to make the concept a reality.

In other words, IP packets on a fiber-optic line ARE millions of little trucks and need a transport layer to get anywhere.

Re:Simple answers

[xappax]

The Internet is a concept, not a thing;

Listen sonny: It's a series of tubes. And if you don't understand those tubes can be filled and if they are filled, when you put your message in, it gets in line and it's going to be delayed by anyone that puts into that tube enormous amounts of material, enormous amounts of material.

Re:Simple answers

[Em Adespoton]

No, honest! The internet is a public transportation system for little information-carrying vehicles!

Two teirs = two separate pipes

[Riskable]

What the big ISPs want isn't just a two-tiered Internet where some traffic gets priority over another. They want two distinct Internets. One were you have control and another where they have control. They'll probably share the same tier-1 backbones but everything below that will be separated (imagine a router configured to send packets from their sources directly to you via a hyper-speed backbone whereas all other traffic gets routed through a dozen or so more hops on the "economy" backbone).

If you want a practical example of precisely how they they plan to violate network neutrality look at the DOCSIS 3.0 spec. It reserves about 80% of the bandwidth on the coaxial cable for video and telephone services that are exclusively provided by the cable company (i.e. no one else is allowed on). The other 20% of the bandwidth is provided as general Internet access (with the usual limited upload speed). This way they can be the gatekeeper for high-bandwidth content (i.e. video) and low-latency applications (i.e. VoIP) while every other business that wants access to their customers has to either pay to get on their high-speed channels or get stuck with the slow lane.

The telephone companies are already rolling out technologies that divide up fiber connections in a similar fashion. The "big plan" is to get paid extra for that exclusive, high-speed and low-latency channel into people's homes. It is a hugely anti-competitive situation.

If you provide streaming video to anyone on the Internet you will not be able to compete with the speed and quality of the video coming over Comcast's, AT&T's, and Verizon's dedicated pipes. If you're a VoIP provider that provides telephone service to anyone on the Internet you will not be able to compete with the low-latency and high quality of the big ISP's dedicated pipes. If you provide *any* service over the Internet all it will take for you to be crushed out of existence is for the big ISPs to start offering the same service on their dedicated, exclusive channels.

It isn't about prioritizing traffic. It is about dividing it up and destroying the free market that is Internet access in people's homes. It is literally "divide and conquer".

So you DO want interference in the market

[Anonymous Coward]

So why is NN *not* allowed to remain? Why is THAT a step too far? All you've done is say "yeah, but property is owned by the government" you've never said why continuing network neutrality is NOT to be allowed?

"Slashdot requires you to wait between each successful posting of a comment to allow everyone a fair chance at posting a comment.

It's been 1 hour, 17 minutes since you last successfully posted a comment"

Re:So you DO want interference in the market

[sid0]

Did I ever say anything about NN? I'm torn on it too, like the OP.

On one hand, the US telcos are really government granted monopolies. The situation, not ideal, can call for regulation if necessary. (Consider Muslim drivers being forced to carry passengers against their wishes. Ideally the matter would have been private and they would have gone out of business or been fired for such a stupid move. HOWEVER, since there are a limited number of licences granted by the government, the regulation was necessary.)

On the other hand, a POTENTIAL possibility is taken to be a DEFINITE result. It is POSSIBLE, hell even PROBABLE, that the telcos will misuse it. However it is not CERTAIN.

artificial limitations

[stenn]

artificial limitations reported by those with much to gain by a tiered solution. yea, i'll believe their reports.

what ever happened to all that dark fiber?? hmmm...

Let them have it

[billcopc]

I say let them have their stupid tiered internet. When the common peon realizes that the internet has been downgraded to a TV-like ad marathon with scraps of cliffhanger content occasionally thrown in, maybe then we'll have enough motivation to start a better network, one that doesn't depend on a handful of megacorps laying down cheap wiring all over the continent. I'm thinking a wireless uber mesh. Hell I'd even get dirty and lay my own damned fiber all over the neighborhood.

Web 2.0 has shown common folk the value of the internet as a democratic medium. It won't be so easy for the big guys to take it away anymore.

Bandwidth aside

[Z00L00K]

if users are locked into the services of their ISP then the usage and usefulness of the internet will be out the drain. Of course those who use it will bet better bandwidth, but for the other people that want cheap reliable long-distance communication they will lose.

This doesn't mean that ISP:s can't install caching servers for the most commonly used services to ease of some of the bandwidth.

wrong, what about the gazillion adverts bandwidth?

[Anonymous Coward]

you know a 'tiered' network controlled by corporations will fill the top tier
with adverts so much for a 'smart' network.

This is Misinformation designed to scare you

[Liath]

The current copper network that was built 30 years ago may "not be able to handle" bandwidth from everyone emailing, talking, surfing, and gaming, however fiber optic networks have so much more capacity as to utterly reduce this argument to fud. Lawrence Lessig discusses this blatant scare-tactic in his books (the future of ideas especially, as it is totally focused on net neutrality).

Article I, section 8 of the Constitution

[benhocking]

As for IP law: it is meant *both* to foster innovation and to protect the rights of the inventor.

Yes. However, where those two goals conflict, the intention (at least per the US Constitution) is to show preference to fostering innovation. From the Constitution:

To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries;

Note that in the Constitution they stated two things: (1) the goal is to promote progress, and (2) these exclusive rights are "for limited times".

Ripping off a product as soon as it is invented will not protect the rights of the inventor, nor foster innovation in the long run.

Absolutely.

Neutral net is more valuable

[Geof]

I agree completely. But I have three points: 1) the paper doesn't make a general claim about efficiency, only a specific claim when certain requirements are set; 2) it is expensive to make a network non-neutral; 3) the neutral network is more valuable because it has more bandwidth.

First, let's be careful: the paper does not claim that a "Class of Service" (non-neutral) net is more efficient in general. It lays out a specific objective, then determines how much more bandwidth a neutral net would require to achieve it. If your goals are different (e.g. you're willing to accept a different level of performance assurance), their conclusion may not hold.

We compare the capacity requirements of a Diffserv environment providing service for applications that require delay or loss assurances in comparison to a network that provides classless (i.e., besteffort) service and still has to meet the same performance assurances.

As your post suggests, this is not a general claim about efficiency. It's a claim about efficiency for a spefic purpose. If you're #1 non-negotianble requirment for the net is video-on-demand, their analysis may be correct. But then you may build a network that is less efficient for other purposes, such as reading Slashdot, sending instant messages, or buying books on Amazon.

Second, a non-neutral network does not cost the same as a neutral network: it costs a whole lot more. Prioritizing traffic costs bandwidth, infrastructure, maintenance, etc. Some have claimed the cost of implementing QoS exceeds the cost of increasing bandwidth. Furthermore, there is an opportunity cost: the more complex infrastructure may be more costly to maintain, guard against attack and failure, and may be more rigid in the face of new applications.

Third, a neutral net with twice as much bandwidth has twice as much bandwidth! It is, by definition, more valuable than the non-neutral network with half the bandwidth. That extra bandwidth is not wasted; it doesn't exist solely to provide the quality of service for video on demand. It can - and will - be used for other things.

Re:Neutral net is more valuable

[arminw]

......If you're #1 non-negotianble requirment for the net is video-on-demand, their analysis may be correct.......

This is a requirement that the Internet was never designed for in the beginning. It is something that is being bolted on and it doesn't fit all that well into intrinsic design and purpose for which the Internet and its protocols were intended for. It was never designed to transmit time critical video and audio. It original purpose and design was to accurately transmit COMPUTER data from A to B even if the data path was at times not all that good. The computers at each end order and sort the data and reassemble it without error (mostly). If an error does happen, the receiving computer ask for that fraction of the data to be sent again and then stuffs the corrected data into the proper spot. for live video and audio, that doesn't work. All the data or at least most of it has to get there in the proper order and on time.

Today, a requirement is being imposed on the present Internet for transmitting a kind of data for which is, at its very core, not all that well suited. Looking at and prioritizing data is and always will be a stop gap measure to try to get around a basic, intrinsic limitation of the Internet at its very core. Maybe some entirely new procedures and protocols need to be invented in order to transmit real time data over existing networks.

Fiber is layed....

[tinkerghost]

Most of the fiber to aleviate any network congestion has already been layed. It's just not lit. In the boom days of 'Everyone's getting fiber' this stuff was going down with 3-400% surplus in city loops.

As it was explained to me, if your growth projections put you at needing 100 strands over the course of the next 5 years, you plan to lay down 120 strands to cover defects/damage. However, since it's hard to open up the channel but easy to lay the fiber, a lot of companies were laying down 300-400 strands because the bean counters figured it's cheaper to amortize the fiber over the course of 10 years, than to re-open the channel to lay down the next 100 fibers in 6 years. This is the whole issue of dark-fiber. So there are places like DC & NYC that have huge bundles of dark fiber (often many times the volume of lit) waiting to be lit. They don't need to 'run fiber', just connect it to the network. Part of this buildout is why the telcos were given $BB in tax credits over the last decade anyway.

toll roads

[rejecting]

This is the equivalent of saying a toll road gets you there faster!

Yep.

[Belial6]

Of course the same research could also show that free internet could be given to every person in the world. They would just have to reduce the priority to all traffic to zero. There would be no packet loss, not a single packet would have more than 10% of the current latency, and the cost to run the network would be 100% less expensive than it is today.

IOW: Corporate Welfare Economy not Capitalism

[OldHawk777]

All our politicians favor a Corporate Welfare/Socialist Economy like communism.

Communism was social theory applied to economics governing.
Corporatism is economic theory applied to society/culture governing.
Either way it is a plutocracy/oligarchy but not Capitalism or Democracy.

So, Capitalism, Democracy, People, Patriotism, god, values ... are all
spun to political agenda-crap demagoguery. No meaning, all truthyness,
complete special interest dogma by the few to oppress the many. Sounds
like governments of 3 millennium ago .... Who could possibly believes in
evolution there must be a cruel god playing war and hope as practical jokes.

FOLLOW THE WEALTH VALUES, because there are no ethics, moral, and justice
values for corporatist/plutocrats, just greed-value for those fools/losers.
Without honor, ethics ... you are a fool and loser ... just like King George.

King George and the corporatist aristocracy depend on net-nepotism.
Only US Citizens support Net-neutrality for capitalism and advancements in
quality of life, education, democracy.

Re:IOW: Corporate Welfare Economy not Capitalism

[Ana10g]

You forgot to cite your source, Parrot!

Re:IOW: Corporate Welfare Economy not Capitalism

[OldHawk777]

I need no source to state rain is wet or that politicians (many not all) are corrupt and amoral.

I suspect that you would not notice rain is wet, and you would remain standing in the rain till struck by lightning.

PLEASE, always note my sig-quote: "Reality is self-induced hallucination."
It is my way of saying; I may not, and you may not be suffering from the same mass-hysteria.

That I think you're an idiot for attempting to sound dogmatically-smart by requesting source and evidence,
when source and evidence is something you would reject if it conflicted with your limited understanding of
reality, but you would be in excellent company with Bush, Chaney, Hitler, Goring, Lenin, Stalin, Mao, Chang....

NOTE: I may be wrong and you may not be an idiot, but I think, I am right, you do not see what a frail, weak,
and meaningless species we are in relationship to everything. When we go extinct, we will be at fault for
allowing self-serving leaders to self-destruct everything that humanity has achieved.

If George Bush decides god told him to (when it was Chaney wishing and whispering) press the button,
the repercussions is all our fault from Tokyo, to Moscow, to Mecca, to Johannesburg, to Berlin, to
London, to Quebec, San Diego, to New Orleans, to NYC, to Washington DC... and George, US, EU ... will
suffer the consequences. Okay, it ain't the fycking button ... it is global warming, cancerous chemicals,
German/Cambodia/Africa... Genocide. Humanity has some great and wonderful art, science, stories, hopes and
dreams, philosophies ... all that can end because of a greedy and delusional plutocracy supported by
dogmatist that suspect they are intelligent, because they can memorize then regurgitate/parrot/ape political,
mythology, economic, doctrine ....

Next time show me you are intelligent and think for yourself, or go-pound-salt and give Bush a BJ in the O.
Source and evidence abounds, you and others chose to ignore/deny at great peril to all and posterity.

Of course you can just wave and call me a doomsday nut ... a few years ago I stopped saying I told you so, it ain't fun no more.

Besides it's a meaningless statistic

[sterno]

Net neutrality doesn't prevent a network provider from providing service sensitive networking. They can prioritize the network however they want to maximize their bandwidth utilization. What they would not be able to do is provide different levels of service based on who was paying them money. If they want to make sure that VOIP, or video is as lossless as possible and lower the priority of say file sharing traffic, that's fine.

Where it becomes a problem is when they start saying, "well since Google paid more, they get higher priority." Or more specifically the problem is they'd prioritize their own services over everybody else and make them seem cheaper and faster by comparison when really it would be them crippling the competition,

"Dumb" networks? I spy AT&T word games.

[bmh129]

Now, if I support network neutrality (AT&T's "dumb" network), then does that make me a "dummie"? This is pure AT&T propaganda, brought to you by the people who sold your private communications to the federal government, paid for with your tax dollars. What a nice deal, eh? You paid to be spied on, and AT&T profits from it.

You paid for Internet construction and access, and AT&T is arguing that it costs them money.

What a bunch of bullshit, AT&T. Come on! How "dumb" do you think we are?

MOD PARENT UP!

[Anonymous Coward]

ah yes, dark fibre; I'm surprised anyone else on slashdot is old enough to remember this historical fact. bravo!

Yes. AND??

[manowar821]

They have the money and technology to do it, they just don't want to. They're in this for the money, and they're going to try to get as much money out of us as they can. CEO needs another mansion, look out!

Telenor disconnected themselves from NIX

[l33t gambler]

A blow to net neutrality here in norway, major ISP Telenor disconnected themselves from NIX (Norwegian Internet eXchange) and requires content providers like Schibsted to pay.

Telenor is the former state tele monopoly, and are not very popular with recent pay-per-MB and P2P bandwidth throttling.

Telenor blaimed NIX for allowing content providers connect directly to NIX instead of through an ISP, and getting charged too little. The truth is more likely Telenor has been overselling their bandwidth.

http://www.google.no/search?hl=no&q=telenor+nix&bt nG=Google-s%C3%B8k&meta=

Yes, yes it does.

[Anonymous Coward]

"Corporate sponsorship of research doesn't automatically invalidate that research"

Yes, yes it does. Corporate sponsorship of this kind of "public relations" research does automatically invalidate the research. As we all know from experience, it's certainly possible to construct a very convincing argument that a proposition is true when it is actually false. Why on earth do we have advertising when everyone on earth, when asked whether or not their buying decisions are affected by advertising, answer an emphatic, "No!"

Simple way to solve this...

[Anonymous Coward]

Anyone with any knowledge of data/packet networking knows that the statment "open Internet will require double the bandwidth" is completely fraudulent from an engineering perspective.

Partition the bandwidth.

Here's an idea. Take the available bandwidth on any outbound link at any chokepoint, divide it in half, divide that half by the active inside addresses on the outbound queue. Repeat the sequence again by using the outside addresses for the outbound queue. This is child's play compared to what they are attempting for QOS, what they will be asking you to pay for.

Dynamic balancing of data traffic has been around for decades. What it basically means is that bandwidth hogs will get restricted, but for those people that are running normal levels of traffic, they're unaffected.

This still provides "open" access to everyone. Everyone gets a fair chance to the PUBLIC Internet. AT&T (or any other ISP) does NOT own the Internet, and as such, it is not up to them to regulate it's access. They can provide the line, and they can provide fair share, but anything other than that and they will run straight into a series of governmental tie-ups with nearly every highly connected nation on the planet. You don't want to go there.

On the other hand, you don't want to be in a bandwidth race condition. "Pure" Open Internet will cause exactly that scenario. Put in Dynamic Balancing by IP address, problem solved, move on to a more important issue to humanity, something like global warming (since the Internet will remain open, there will be some poor starving scientist who will figure out how to cool the sun, because we kept the public Internet open).

Leave it open, and balance the chokepoints. Anything else will end up costing you more than it's worth. (from every standpoint)

robluce1 yahoo.com
www.shepluce.com

Money, Money, Money

[Anonymous Coward]

The kickbacks err, research grants must be pretty good from the NSA, err AT&T.

Sounds great to me!

[the_greywolf]

... And the problem is...?

We all know AT&T has the money to pull this off by themselves, and given that major parts of the network are owned by several other major brokers, I really don't see how this is a problem. A few basic bandwidth management techniques (i.e., what they do already) a few more routers, and fiber to the premises ... all can be done on their existing budget.

Fire the CEOs. They're a waste of money, a waste of resourcs, a waste of my air, and a waste of our time. I want fiber. Now.

Why not "Who pays more"?

[Ogemaniac]

Just about everything you have ever purchased became yours precisely because you were willing to pay more than anyone else that the previous owner could find. It has nothing to do with anyone anywhere deciding what was more important.

Capitalism works, my friend. I am sorry to disappoint you.

I WANT to be able to pay to keep my traffic ahead of the lazy college students downloading mountains of porn and mp3s off bit-torrent. My money says my claim is more important. What do you offer in return?

Re:Why not "Who pays more"?

[timeOday]

Nothing about net neutrality says you can't pay more to get more bandwidth. Nobody objects to that, and it's the only market dynamic necessary to drive growth of the Internet. In fact, net neutrality doesn't preclude burst-shaping bandwidth so that, among people paying equally, those who've used a lot of bandwidth recently get lower priority than those using it only sporadically. This is all very standard practice in CPU scheduling for the last 5 decades or so.

And again...

[Anonymous Coward]

I know the US gave us the internet (to some extent) but I can't believe that the corruption of your big business and government are, very likely, going to screw up one of the greatest inventions of the 20th C.

Source is unreliable

[Anonymous Coward]

Of course *AT&T* says that.

Appropriately enough, the CAPTCHA is "drones".

Not really

[einhverfr]

There are actually a number of tier 1 ISP's and I don't think either AT&T or Comcast are among them. However, in the last mile, you are right to a point.

However, I think your case actually fits another area I would like to see some action from the government regarding. Unfortunately the current FCC policy seems to be targetted towards the idea that a duopoly is sufficient competition. This is making it harder to force companies like AT&T to lease their lines to others.

I think that your concerns would be better addressed by pushing for more competition in the free market and this would be handled by forcing last-mile providers (like AT&T and Comcast) to lease lines to the competition. I think that if we approach this with the idea of protecting and expanding the free market (and essentially regulating the physical lines only in order to accomplish this), we would see better results.

Now, I know I will get flamed from both sides. Some will say I am proposing socializing the infrastructure and that companies like AT&T and Comcast should have the moral right to control their physical infrastructure. As long as their infrastructure doesn't use public telephone poles, or other public land rights, they might have a point, but the instant they start running lines to your house or my house, that infrastructure becomes, in my view a public good.

In short, carrier neutrality (essentially declaring the last mile lines to be common carriers) makes a great deal more sense than targetting a specific network management technology.

We both see the need for government action, but we disagree what form that action should take.

Okay

[obeythefist]

So, we remove net neutrality, charge everyone much more for internet access as they previously paid, push the prices up enough that your demand drops to half what it was, and surprisingly, you only need half the bandwidth.

In other news, demand for inexpensive, useful products remains unchanged.

Let's be clear here...

[Talgrath]

There's nothing wrong with regulating internet traffic based on what is being sent, so long as the website it's sent from doesn't matter; throttling and controlling the speed that something is sent based on what is being sent is fine, video needs more speed than e-mail, after all. What people oppose is websites having to pay to get their content through faster; the study is talking about the former while the latter is what is being opposed, it's a trick that AT&T and others have used to drum up opposition to Net Neutrality.

yeah

[nanosquid]

Twice the bandwidth at 1/10 the price and cost. Seems like a good deal to me.

(Remember: this is the same AT&T that was trying to tell us that packet switching networks would never work for voice and video and that we should all be investing enormous amounts of money in expensive circuit switching technology. Now they are trying to shoehorn their hare-brained ideas into QoS guarantees on the Internet.)

I Don't Care

[l0rd.47hl0n]

I don't care what studies/articles imply. I want a neutral net, and I don't care if it means double-the-bandwidth of a tiered network. The internet wasn't designed for machines, it was designed for people. I'm not a second-class citizen if I can't pay a higher premium for tiered net access. My connection latency and throughput (up to my bandwidth cap) should be the same as that of my neighbor (household hardware dependent, or course), regardless of the packages we pay for. If our ISP cannot deliver a video-on-demand service (for example) without improving network capacity, the cost of that upgrade should be born equally by every customer the ISP has.

Telcoms only want to line there pockets

[toyotabedzrock]

If they want to tier the Internet then they should do it based on the type of communication, that is the only way in which any increased performance will be realised. Trying to charge for the added speed will only make it worse as competing companies will prioritize different traffic and only cause more bandwidth problems. There solution only serves to put money in there pockets. Heres my idea the telecoms need to stop being cheap and actually strengthen the backbone of the Internet. Throwing added processing and equipment into the mix will only hurt overall performance. Even prioritised traffic will slow down a bit since they are adding the overhead of sorting which packets get priority and then adding priority tags to them, which then need to be read and sorted at every switch in the backbone of there network and eventually removed if the packet needs to travel onto another telecoms network.

An amazing observation of the obvious

[bigpurpledick]

"Twice the bandwidth."

Duh!! Of course it will take twice the bandwidth to support
what you *want* to do on the internet, things you are doing today,
as the amount it would when they throttle back everything that
they can't charge you for.

It seems that the providers that are against
a free, unfettered internet are just trying to mimic the
model that the music industry started in the '80s:
pump out a limited amount of mediocre crap at minimal cost,
and fleece the public, for whatever they'll bear, to see it.

And look at the state of the music industry.