I really get annoyed by all this. It's not Netflix traffic, it's my traffic. It's data I requested from my ISP, and the ISP carry the traffic because their customer wants it and pays for the service. If you're not making money from me then charge me more, don't threaten and extort the services I want to use!
I agree with you to a point. The problem with this scenario is that the ISP business is built off the principle of giving users high download speeds, but with the expectation that they won't be maxing out their connection all that frequently. The higher the average/peak usage, the higher the infrastructure costs to provide good service.
It used to be that the only people who did this we're downloading games or movies off BitTorrrent: when that was the case, it didn't affect the ISP business model that much.
Enter Netflix. It's pretty ubiquitous and even your grandma can use it. Everyone loves it and everything works well. For a while. Then Netflix decides they want to push 1080p videos out to all their users instead of 720p. This doubles or triples the bitrate of the videos overnight. With adaptive bitrate players, the stream will adjust to use all available bandwidth.
Who should bear the cost of technical changes that Netflix makes? Many would say "the ISPs! Stick it to the man!" But at the end of the day, ISPs are a business. If their costs rise, they will raise prices on consumers unless they can cover those costs another way.
In the end, you're gonna pay the ISP one way or another. Which is more fair to you? ISPs raising rates across the board (including some people who aren't netflix users) to cover those costs, or passing the costs on to Netflix, who then pass them to Netflix subscribers only. It's all about who should bear the cost for technical upgrades given the way the ISP business model works today.
I prefer option 3 - ISPs charge more for people that use more. I know this would require marketing changes, and many customers wouldn't like it, so it won't happen and instead they'll just keep doing what they're doing and trying to extort money from data-heavy services.
Option 1 is unfair to some people, Option 2 puts a huge barrier to entry in place for businesses. Option 3 is the only real solution IMHO.
The problem with option 3 is people don't really want to pay for what they consume, and the state is happy to oblige by placing itself as the middle-man between concentrated benefits and dispersed costs.
Option 3 is data caps and/or metered billing. It's not exactly the most consumer-friendly option, but the ISPs would love to do it that way if they could.
I love metered billing. The problem with it is that the video streamers will discover how much more bandwidth than everyone else they're using, and AT&T and Comcast's all-you-can-eat video services that use the same wire will look a lot more attractive to them.
Ultimately, that may be a threat to network neutrality as the prices are raised on the metered internet in order to shift bandwidth to those siloed TV services, and the public won't care because they're not streaming video over the open internet anymore.
Well, the reason it works for all-you-can-eat video services is that the cable companies are providing a distribution service to the content owners. So for HBO, they provide both cable distribution and caching of video for HBO Go. They could pretty easily do the same for Netflix; considering Netflix is basically HBO that's sold outside the cable companies at this point. HBO does in effect pay the cable companies for distribution in that the cable companies take a percentage of the monthly HBO subscription fee that the customer pays, so I don't see why Netflix shouldn't.
I think a lot of geeks are in love with net neutrality because they see it as a fairness thing and think it will keep the cable companies from gouging them. In reality, it's just about cost allocation. Companies are going to take their profits under net neutrality or not; and the profit margins on cable aren't astronomically high. Netflix would never have gotten off the ground under true network neutrality; the costs have to be allocated somewhere.
If you want pure speed though; metered billing is the way to go. Look at how fast mobile data has become in the US relative to the rest of the world -- a large part of that is because it's profitable for Verizon and AT&T to have customers use more data. Businesses are going to make their profits and regulation isn't going to have much effect on the prices that consumers pay -- but under the right rules, if you create a situation where companies can make more money by providing a better service, everyone wins.
>but with the expectation that they won't be maxing out their connection all that frequently.
It's an old, antiquated expectation. It needs to die. Ever since the advent of peer to peer technology, there has been a perfect use-case for permanent saturation of available bandwidth.
>But at the end of the day, ISPs are a business.
ISPs are a utility. There's no dancing around it. It doesn't matter what the FCC says or what anyone tries to lobby for or against. They are a utility.
>If their costs rise, they will raise prices on consumers unless they can cover those costs another way.
We did pay for a national broadband network. To the tune of 300 billion dollars at this point. You know what they did? They took the money and pocketed it. We've known about this for more than a decade now, and nothing has come of it.
> It's an old, antiquated expectation. It needs to die. Ever since the advent of peer to peer technology, there has been a perfect use-case for permanent saturation of available bandwidth.
It's not antiquated; it still applies to the 99% of broadband users who don't use high-bandwidth P2P services on a regular basis. Dedicated connections are available for those who don't fit within this model, but it's a lot more expensive to provide (e.g. a much lower ratio of circuits to switches/concentrators/routers). Burstable connections still exist on the backbone side as well; Netflix may have access to up to 300 mbps burstable from Level3, but they're only guaranteed 50. They make economic sense and they're not going away any time soon.
> ISPs are a utility. There's no dancing around it. It doesn't matter what the FCC says or what anyone tries to lobby for or against. They are a utility.
I disagree, but let's say you're right. Utilities are still businesses and they're not sustainable as a money-losing operation. If the FCC regulates them as a utility, they would likely move to a metered-billing model where you pay for bandwidth in and bandwidth out.
> We did pay for a national broadband network. To the tune of 300 billion dollars at this point. You know what they did? They took the money and pocketed it. We've known about this for more than a decade now, and nothing has come of it.
That $300 billion figure is very suspect if you read your own links. It includes rate increases to customers which would have happened anyway due to inflation, and is a number over roughly 15-20 years. You know that 50 mbit cable modem you like to bitch about how slow it is? In 1994, the best option for most homes was a 56.7kbps dial-up modem. 1.5 mbit ADSL wasn't even widely available until 1998 or so. Cable internet above 10mbps wasn't widely available until 2007. And as part of the tax credits, the FCC required companies to build out broadband infrastructure in a lot of places that were grossly unprofitable: the cable companies basically broke even on the credits they did get. But even if we make a vast assumption that your number is correct, $300 billion across ~5 companies over 20 years is roughly $3 billion per company, per year. The real tax subsidies were closer to $80 billion over 20 years for the entire industry (many of which went to small, rural cable providers and not the big ISPs like Comcast or Verizon.)
But claiming there has been no progress is a flat-out lie. Citing 5-10 year old articles doesn't help your case. The money was spent where it should have been; it just took a while for the build-out to happen because large capital investments move slowly.
> We did. With billions and billions of taxpayer dollars.
And the major telecom companies spend roughly $15 billion per year EACH on infrastructure. Of their own money. On which they make roughly a 10% profit (more in some years, less in others.) Good money, but hardly a windfall by corporate standards.
> For any given U.S. household, there is often only one or two choices for getting
> high-speed Internet* access and that’s unlikely to change.
> Furthermore, Internet access is often bundled with other services
> making it challenging to switch ISPs. It is this lack of
> consumer choice that leads to the need for strong net neutrality.
Spot on. The reason Comcast and others can play this game is because they're a monopoly at the local level. They've worked with the help of regulation to snuff out all serious competition. Pull back the regulation (at the local, state, and federal levels) and watch the competition spring up, and watch the Comcasts of the world change their practices when they actually have to compete.
I'm actually switching away from Verizon for my cell phone service this weekend precisely because of their role in destroying net neutrality. I expect T-Mobile to be slightly worse, or at least no better, in terms of service-per-dollar but at least my conscience will be clear. Who knows, maybe it'll even be better!
I actually switched from Verizon to T-Mo about two years ago, and while it isn't better than Verizon, it isn't substantially worse. If you live in a fairly well populated area (cities of over 5000 people or so), or along major roadways, I've never had any issues. The only time Verizon did much better than T-Mobile for me was driving through Western Colorado and Utah, where T-Mobile had no service, and Verizon barely had voice.
> Pull back the regulation (at the local, state, and federal levels) and watch the competition spring up
Unlikely. The monopolies that the cable and telcos used to have as infrastructure providers were granted because the eminent domain power of government was necessary to get the physical infrastructure, and it was considered a public good to have the infrastructure and better to have one infrastructure provider with "public access" rules (for cable) or "common carrier" rules (for telephone service).
Without government intervention, it would be practically impossible to get the property access necessary to compete, particularly at the last mile, with the vendors with existing infrastructure built with property rights acquired through government action, even if there were no longer protections for the incumbents.
"But when we ask [ISPs] if we too would qualify for no-fee interconnect if we changed our service to upload as much data as we download there is an uncomfortable silence."
That was a really weak part of the post. Settlement-free peering is when I route your traffic and you route my traffic and they're about even so there's not really anything to be gained by you paying me $1000 this month and me paying you $1000 next month. If Netflix were to start routing Comcast traffic to the general internet at the same rate that they route traffic from their data centers to Comcast, that would be a rather different story.
Offering to just accept and blackhole traffic is silly, and everyone involved knows it. I'm really disappointed that Netflix chose to present that argument in their post, because it's deliberately misleading about the nature of settlement-free peering.
Netflix originates a lot of traffic, but it doesn't accept non-Netflix traffic for routing. That is, I can't send a request for gmail.com to my ISP, who then sends it to Netflix, who then sends it to Google. Netflix does not have a neutral network which routes traffic regardless of origin or destination - they just accept requests for Netflix and serve responses from Netflix.
I mean, literally, that Netflix traffic is valuable to Comcast's customers (it's their movies) and not valuable to Comcast (it competes with their other offerings).
Isn't it valuable to Comcast as well in the long run? I know several people who have updated their internet packages to ensure high quality Netflix streaming. I have to imagine Comcast's profits on that upgrade far exceed the delivery costs.
I was puzzled by that argument as well. It's not an apples-to-apples comparison. Netflix is not a network provider. They do not route traffic. They are not a "peer". What traffic from Comcast are they going to offload, and to where?
The one thing which bothers me about the current ISPs is how their networks are designed for downloads. All cable, ADSL, and even some fibre connections all have greater download compared to upload.
How is any company which deals with a residential ISP expected to be a peer?
I wondered if this was an implicit threat to explore bittorrent like protocols for Netflix clients. In general, client to client sends between residential customers would make it more difficult for ISPs to block. In addition, at Netflix's scale, it would likely cause the balance of upload/download from residential clients to shift dramatically. ISPs would have to route a lot more uplink traffic which they generally avoid.
This is hilariously wrong. Netflix is paying ISPs because they dumped their CDN providers and started doing it on their own. They did it because it was cheaper for Netflix to cut deals than to use a CDN (who, by the way, also pay ISPs).
> imagine the plight of smaller services today and in the future.
This is flat out a lie, smaller services uses CDNs.
This situation has not been caused by an action of Netflix's, it has been caused by an inaction of the big ISPs. Interconnects are a boon to a distributed internet, and as Netflix points out in the post, ISPs routinely employ them - for free - in other situations. The main difference is that they typically can't extort the customers on the other end of the connection so easily; i.e. other ISPs and companies whose products don't visibly degrade when subjected to interconnect bottlenecks.
> This is flat out a lie, smaller services uses CDNs.
The reason Netflix is targeted here is not because they aren't using CDNs. It's because it's the first popular service that could easily be blackmailed in this way. The end result is that Netflix's cost have gone up in order to get access to the same 10Mbits/sec that their customer has already paid for. Those costs will be borne by Netflix (and its customers) and the ISP has effectively double-charged. Don't kid yourself by thinking the ISP won't turn that into a regular practice for any company that is successfully reaching its customers with a latency/bandwidth-sensitive service.
Why shouldn't Netflix be their own CDN if they think they can do it better? Why should any CDN have to pay ISPs to deliver bandwidth that the ISP's customers already paid for?
By that logic, why consumers should pay anything to ISPs? ISPs don't have obligations to sign free peering agreements to CDNs or other large network owners, like Netflix, if it isn't advantageous to them.
Large network owners agree to peer without charges when it is mutually beneficial. That usually boils down to whether the traffic exchanged between the two networks will be roughly balanced and managed reliably.
One of the issue here is that Netflix doesn't operate an Internet backbone. Therefore, ISPs can't use Netflix's network to exchange traffic across the wider Internet. So this leaves the ISPs two choices here:
1) Agree to a free peering agreement with Netflix and build more capacity in their network at the ISP's expense to support Netflix's egress traffic;
2) Request Netflix to sign a paid peering agreement to support the construction and maintenance of the needed extra network capacity.
ISPs who opt for the first choice believe that improving their access to Netflix will make their customers happy and thus allow them grow their business. The other ISPs who opt for the latter choice believe, on the other hands, don't see the free peering as an investment; they want rather to extract more money from their existing customers. So, it is not that surprising to see the incumbent ISPs, who have locked-in their users already and killed-off most of the competition, going against free peering.
That said, I just can't wait to see Google Fiber kick the behinds of these greedy fools.
Really ISPs have a choice between peering with Netflix and reaching Netflix over transit. Since most broadband ISPs would have to buy transit, peering with Netflix (at any price, even free) is a better deal.
> Really ISPs have a choice between peering with Netflix and reaching Netflix over transit. Since most broadband ISPs would have to buy transit, peering with Netflix (at any price, even free) is a better deal.
They have a third choice, which is "not reach Netflix at all". Which, for ISPs that are also selling streaming video services (e.g., pretty much all of the big ISPs, which tend also to be digital cable with "on demand" video streaming providers) with which Netflix competes, is a pretty attractive option.
It's especially galling with Comcast throwing $45 billion at acquiring another company. You really couldn't afford to upgrade your infrastructure with $45 billion burning a hole in your pocket, Comcast?
> That said, I just can't wait to see Google Fiber kick the behinds of these greedy fools.
I think there is also a fairly strong argument that free peering with Netflix directly is a net neutrality violation. You are giving Netflix data a huge advantage over other video services. Netflix talks about the little guy, but the little guy doesn't get to just dump his traffic onto the ISP for free.
Okay... Your implied statement here is that it's OK for ISP's to charge CDN's/Netflix/anyone...
Is that a statement that you agree with?
Regardless of whether Netflix is portraying the situation with some bias(on their own blog who would have thought) this is in important issue that should get attention.
If we let the free markets work we wouldn't even be discussing net neutrality. ISP monopolies and duopolies are the cause of this problem. If there was competition consumers would choose the ISP that gives them proper, fast access to the services they use (eg Netflix).
It's just not practical to dig up an entire cities worth of streets to lay a second transmission line. You are legally allowed to compete with the cable co and telephone co. But it's rarely attempted because it's not profitable. They already have a network, building one is very expensive.
Competition is actually our problem. In countries with only 1 telecom choice, the government regulates it. It often lets ISPs pay to rent the transmission line that the telecom owns.^1 Instead America lets the cable company compete against the telephone company and we get two poor choices.
If you want nationwide fiber? Give ATT and Verizon their monopoly on ISP lastmile networks, but force them to lease your line to indie ISPs.
Just calling for more competition sounds like a good idea, but it won't work.
^1you can actually do this now for DSL in America.
To be fair (and probably pedantic :), natural monopolies don't happen in any free market given enough time. It's more like, certain very specific markets (such as ISPs and desktop operating systems, apparently) have built-in dynamics that lead to natural monopolies. Most markets aren't like this and can remain competitive without regulation.
My desktop OS example requires some clarification- there were plenty of things Microsoft did in the '80s and '90s where they exploited their position in the market to increase market share. Per-processor licensing fees are a prime example of that- a PC vendor that wanted to license a Microsoft OS would have to pay a licensing fee for every computer they sold, regardless of whether or not that computer had an MS OS on it. However, one could argue that this was a consequence of MS already having a monopoly (50% market share can be considered a monopoly in some markets), and they were exploiting that monopoly to become even more deeply entrenched. I honestly think that the desktop OS market in the pre-web days had very strong tendencies towards a natural monopoly regardless of whether or not any given player in that market was engaging in unfair business practices to achieve that monopoly.
Any time discussion about competing products revolves around which product is ultimately going to "win", there is an implicit assumption that the market those products compete in tends towards natural monopolies. It may not be true, but the people participating in those discussions think so even if they're not fully aware of it.
I agree that some markets have a greater tendency for forming natural monopolies. I think the reason for the desktop OS monopoly tendency lies in the fact that developing software and drivers for multiple operating systems is expensive. This, and the desire to have marketable skills, drives developers to one OS. And also, applications you do your work in were typically incompatible between platforms, thus creating a drive for users as well to band together on one OS... Some markets have a huge tendency for natural monopolies, perhaps mostly due to this cost of switching (or sitting in two or more chairs, so to speak).
But even in a market where you can easily switch one supplier for another, there often remains a helpful factor for grouping - everything costs less in large quantities ;) And many administrative costs can be reduced. Not to mention that perfect competition would, strictly mathematically speaking, drive profits towards zero...
> Most markets aren't like this and can remain competitive without regulation.
It's clear that some markets are like this, and some aren't. It's not at all clear to me that most markets aren't. In fact, i believe the opposite - most markets are like this. I believe that in most markets, economies of scale, cost of entry, and the coercive power of large participants outweigh the incremental efficiencies of new innovators.
If there is any logic or data which supports the contrary position, i would be interested to hear it.
This is the result of free markets. Some markets can be completely free and still result in monopolies. Some markets have natural tendencies towards monopolies, and will inevitably become dominated by monopolies unless (ironically) regulations artificially create conditions under which competition can exist.
In the case of ISPs, the most pragmatic resolution is not to artificially create a free market but to classify them as utilities. This is an acknowledgement that they exist in a market which tends towards regional natural monopolies, and allows them to enjoy those monopolies, but prevents them from exploiting their status too much (for example, by using their natural monopoly in one market to affect another market).
I don't even know who their hypothetical competition would be- my understanding (which could be wrong) is that there is nothing preventing such competition, it's just very difficult to go up against an entrenched incumbent. Google Fiber is great, but if that was Google's only business they wouldn't last very long.
The only counterexamples I can think of are the legal battles where ISPs have effectively shut down attempts to establish municipal broadband, which is terrible, but I don't think that's the kind of competition you're referring to.
Yes, local governments are complicit in the ISP monopolies, but only government regulation (or ownership) will be able to force companies to share the existing last-mile infrastructure, which is the only way competition can truly exist.
Net neutrality I can see being a very important topic to Netflix. I could definitely see ISPs charging extra for Netflix services due to "load" or some BS like that (when really it would be to make up the damage to providers cable/satellite offerings).
Net neutrality is good for everyone, all links and all bytes should be equal.
This is where the beauty of popcorn time just shines because it's the same amount of data but it's coming from everywhere.
If Netflix could implement something like this and charge for it ISP's asking for a kick back would be lost.
Unfortunately the MPAA and RIAA would never go for that since they don't get to put their DRM everywhere.
Atually the MPAA and RIAA should be backing Netflix on this one since if Netflix fails the most likely replacement for it will be popcorn time and then the won't be making any money at all.
Then again Netflix will probably destroy net neutrality by complying with their demands before it fails.
> "This is where the beauty of popcorn time just shines because it's the same amount of data but it's coming from everywhere."
The net neutrality decision not only opens the door to monopolists 'shaping' traffic by peer, but also 'shaping' traffic by protocol. [1]
Even if your data is coming from 'everywhere', they can squeeze it down to almost-nothing. The MPAA is likely popping champagne over the prospect of finally having a tool with which to kill casual bittorrent use. [2]
[1] And shaping traffic by device is also on the table. I hope game console- and platform-makers are preparing for this.
[2] Yes, people could tunnel that traffic through a proxy or VPN. But the ISPs only have to kill casual use to significantly harm the network. And throttling all VPN/SSH traffic and using that as a cudgel to extract fees from more companies is also not only feasible, but perfectly in line with their incentives.
True however if it was used for legitimate proposes like by Netflix what's your excuse for throttling if it's not piracy.
Plus I suspect that if they ever go to such draconian lengths there would just be a lot of random protocols popping up and it will turn into another race of cat and mouse until the filters get so bloated it can't handle the traffic anymore.
> True however if it was used for legitimate proposes like by Netflix what's your excuse for throttling if it's not piracy.
Without the Open Internet rules that were struck down, the excuse is "we're a for profit business, and (1) Netflix is a fat stream of revenue that we are enabling and we don't see a reason to do that without getting a big cut, and (2) Anyway, Netflix is competing with our streaming video services which are one of our major sources of revenue, so if our customers have degraded access to Netflix, that just makes our service that Netflix competes with more attractive, and brings us more revenue. So, considering both of those factors, its win-win for us to provide Netflix the choice of (1) give us more of your money, or (2) have your service work less-well, if at all, for our customers."
This is, fairly exactly, the kind of harm that Open Internet order was designed to prevent -- to the extent that broadband providers that are also video (cable TV or internet streaming) providers discriminating against competing video services was one of the things called out to illustrated what was prohibited by the non-discrimination provision.
They don't need excuses anymore. That's the point.
They can just do it, and when Netflix says "hey, we use that legitimately" the ISP can say "and if you pay us, we'll loosen the restriction, for you, for the term of the contract"
"Big ISPs aren't paying money to services like online backup that generate more upstream than downstream traffic."
Maybe not paying money, but if you're running an offsite backup provider (ahem) and you're not getting special cut rates on your all-inbound usage ... you're not going to be in this business very long.
Remember when Netflix did the price hike and got a consumer backlash? Hey maybe if they could indirectly price hike through IPs, then no one would notice ;)
Did they ever mis-sell their service as being Linux compatible? I don't see how you could start a lawsuit because Netflix don't provide a service they never claimed to...
