FCC Moves Ahead with Internet “Fast Lanes”
In a 3-to-2 vote along party lines, the Federal Communications Commission has decided to proceed with a preliminary Open Internet proposal that would permit ISPs to charge companies for faster access to their networks under “commercially reasonable” terms.
The proposed rules are the FCC’s latest attempt to create a legally enforceable framework protecting net neutrality, wherein all lawful Internet traffic would be handled with equal, best-effort priority, regardless of origin. (See “FCC Hopes Third Time Is the Charm for Net Neutrality Rules,” 20 February 2014.) The new proposal will go through a four-month comment period, and the FCC hopes to have finalized rules in place by the end of 2014. Right now, no laws or regulations protect net neutrality principles in the United States.
Protecting net neutrality sounds good, right? Except for the seeming contradiction in the FCC proposal allowing ISPs to set up separate “paid prioritization” deals so deep-pocketed companies can get better, faster access to customers. Big companies would be able to pay to be more equal than others.
Examples might be Netflix’s recent deals with Comcast and Verizon (which Netflix claims it has been forced to make to work around deliberate congestion), but anyone trying to deliver high bandwidth or low latency services to Internet subscribers might feel the pinch. Google, Facebook, Xbox Live, Amazon, Steam, Aereo, Hulu, Yahoo, PlayStation Network, security and health monitoring services… the potential list goes on and on.
FCC Chairman Tom Wheeler says the proposed rules are all about preserving net neutrality. However, in practice the proposed regulations are trying to walk a fine line between net neutrality principles and the business concerns of American broadband providers. And while everyone seems to think net neutrality is a good idea — publicly, at least — almost everybody also seems to hope the FCC’s latest effort will fail.
Can an Open Internet Have Fast Lanes? — Fundamentally, the idea behind net neutrality is that any Internet user should be able to access any legal content or service — using any legal device — in a non-discriminatory manner. That is, network operators shouldn’t be able to block, degrade, or play favorites amongst content and services based on their own business interests or other factors. This essential equality — that traffic from multi-billion-dollar companies like Apple is treated with the same priority as a goofy snapshot from your kid sister — has been a major factor in making the Internet an economic and cultural force, not just in the United States but
worldwide.
But “open” doesn’t mean “free of charge.” The reality is that many parties set down serious cash to bring the Internet to consumers, including (but not limited to) Internet firms building data centers, backbone providers laying fiber and wireless links, middle-tier providers caching and distributing content, ISPs bringing copper, fiber, or wireless connections to homes, and end users who typically pay a monthly bill.
ISPs provide what’s called the “last mile” of connectivity to consumers, although the largest (like Comcast, Verizon, Charter, Time Warner, CenturyLink, AT&T, etc.) also operate massive nationwide networks and have invested many millions (often many billions) of dollars building those networks. ISPs basically argue that if a service comes along and eats up a good portion of their network bandwidth — Netflix currently accounts for a whopping 34 percent of North American Internet traffic, according to Canadian network services company Sandvine — ISPs shouldn’t have to reach even deeper into their own pockets just to support that other company’s business. Basically, they don’t think they should be required to
give that other company what they consider to be a “free lunch”.
Pure network neutrality principles, on the other hand, say yes: ISPs must give companies like Netflix a “free lunch.”
The FCC has worked in some wiggle room over the years. The FCC’s 2010 Open Internet Order (which came after Comcast got the FCC’s 2005 rules overturned) allowed for ISPs to discriminate against some traffic in the name of “reasonable network management.” In other words, if an Internet service or company was degrading an ISP’s network, the ISP could do something about it without getting in trouble with the FCC. However, the FCC also included a transparency requirement: ISPs had to disclose when they were blocking or degrading traffic so consumers could make informed decisions about service.
In essence, the FCC’s current proposed net neutrality rules aim to strike a compromise between the two sides. ISPs must allow consumers to access any legal content or service using any legal device — the core of net neutrality — although ISPs can engage in network management so long as they’re transparent about it. However, ISP’s can also work out paid prioritization deals with content and service providers where it’s “commercially reasonable” to do so — and they’d have to be transparent about those paid deals, too.
Pros and Cons — The FCC’s latest net neutrality proposal does not lack for critics.
Net neutrality proponents argue that allowing ISPs to create “fast lanes” via paid prioritization means all other Internet traffic would necessarily be in a “slow lane.” Moreover, once fast lanes exist, ISPs would have little to no financial incentive to upgrade their standard network performance. Letting the “slow lane” deteriorate and fail to meet consumer needs would only drive more companies to seek paid prioritization deals. There’s also an innovation argument: having to pay for priority access to consumers could keep the next Amazon, Netflix, or Google from getting off the ground — and that empowers incumbents, stifles innovation, and hurts the U.S. and global economies. Some venture capitalists are already backing away from bandwidth-intensive startups due to the FCC proposal.
ISPs argue that if they aren’t allowed to work out deals with content providers and services that dump extraordinary amount of data into their networks, they can’t afford to support those services very well. Yes, it’s ironic when nationwide ISPs and network operators plead poverty — many are quite profitable — but building out broadband networks to consumers is an expensive endeavor, and the companies have to justify that expense to banks and investors before they can even begin. Being able to generate additional revenue from data-intensive services means a quicker return on investment, which means more broadband to more people more quickly. Just as most people agree net neutrality is a good idea, most people agree
increasing broadband availability and capacity in the United States is also a good idea.
One factor complicating both perspectives is broadband competition — or, rather, the lack of it. According to data compiled by the FCC last December, fully one third of American households have only a single choice for broadband Internet access, or no access at all. If more Americans had realistic choices between broadband providers, competition could keep ISPs from engaging in predatory pricing practices and effectively holding their customers hostage.
ISPs, conversely, generally don’t see a lack of competition, at least in major markets. Sure, many places might only have a single cable or DSL provider, but ISPs view competition from 3G and 4G mobile broadband services as a major threat everywhere they do business. After all, why would consumers pay for cable (or fiber, or DSL) if they’re already paying for 4G Internet access on their cell phone bills?
Why are ISPs scared of mobile Internet — particularly when some of them are mobile operators themselves? Back with the 2010 Open Internet Order, the FCC explicitly excluded mobile Internet from net neutrality requirements, feeling that the emerging technology was too new to regulate. So, mobile operators can discriminate against any Internet app, content, or service they want at any time, so long as they disclose the action. The FCC’s latest rule proposal also excludes mobile Internet from net neutrality requirements, although the FCC says it’s open to reconsidering that.
The Legal Foundation — The FCC is basing its new proposed rules under authority in Section 706 of the Telecommunications Act of 1996, which gives the agency authority to regulate “advanced telecommunications capability.” When Verizon got the 2010 Open Internet Order struck down at the beginning of 2014, the court noted the FCC could reformulate the rules under Section 706 authority — and that’s exactly what the FCC has done. (The court also affirmed the FCC could reclassify ISPs as common carriers under Title II of the Communications Act — we’ll get to that in a minute.)
Here’s the thing: where the 2010 Open Internet Rules were struck down in court, Section 706 authority has survived a court challenge. Back in 2011 the FCC used Section 706 to mandate that mobile operators strike roaming deals with smaller carriers. Verizon challenged that requirement on grounds very similar to its arguments against the Open Internet Order — and lost.
Does that mean net neutrality rules based on Section 706 authority are bulletproof in court? No. ISPs could make the same arguments they did with the 2010 Open Internet Order, saying that net neutrality requirements amount to treating ISPs as common carriers without classifying them as such. Or ISPs could bide their time, pushing the envelope on what’s “commercially reasonable” for paid prioritization deals, then take the FCC to court if the agency tries to declare a particular deal “unreasonable.”
And what is “commercially reasonable?” Nobody really knows — and the FCC isn’t spelling it out. The FCC says it would determine whether a practice is unreasonable based on a broad range of factors and the “totality of the circumstances.” Legally, that’s the best the FCC can do under the Section 706 authority that has been upheld in court. There would be no clear line defining what made a paid prioritization deal unreasonable, and ISPs could discriminate between (or against) customers, offering them different rates and terms so long as the deal wasn’t somehow “commercially unreasonable.” In other words, it would be a free-for-all until somebody stepped over an invisible line.
What About Common Carriage? If net neutrality advocates hate the FCC’s proposed new rules so much — and ISPs aren’t big fans either — why doesn’t the FCC exercise its authority to reclassify ISPs as common carrier telecommunications services (like phone companies) and mandate they handle all lawful traffic without discrimination?
For one thing, it would kick up a political firestorm. Right now, the American political landscape is highly polarized, and Republicans have indicated they will vigorously oppose any effort to impose additional regulation on Internet providers. (Many are even opposed to reformulating net neutrality rules.) We’d not only have a regulatory battle over net neutrality: we’d have a political battle as well.
For another thing, ISPs argue that reclassifying their businesses as common carriers would introduce new costs and stifle their ability to create new services, since they’d have to ask for government permission in many cases. They also argue that Title II reclassification would make them less attractive to banks and investors, costing jobs and hurting the economy. Although others have noted that ISPs actually did pretty well before 2005 when they were regulated under Title II provisions, it’s certain the FCC doesn’t want
to give ISPs any reason to reduce investments in their networks, or discourage newcomers like Google Fiber from trying to disrupt the ISP market. That’s no way to improve the poor state of broadband competition in the United States. So Chairman Wheeler looks at reclassifying under Title II as a last resort — although he’s quick to mention the FCC’s Title II authority if this attempt to create enforceable Open Internet rules fails.
What Can You Do? The FCC will be taking comments on the proposed rules until mid-September 2014 in two phases: in the first phase, anyone can chime in, and the second phase allows for responses to the first. The FCC will then roll through the comments — as well as input it gathers at public meetings and other sources — and decide what changes, if any, it wants to make. Public comments can lead to changes in proposed FCC rules (although they might be subtle), but it’s rare for the agency to scrap proposed rules entirely and start over. However, the process is fully public — the FCC publishes all
formal comments — and anyone can participate. You can also call Chairman Wheeler’s office directly (1-202-418-1000), and of course contact members of your congressional delegation to express your opinions.
It’s probably fair to say that the FCC’s new proposed open Internet rules represent Chairman Wheeler’s last, best hope to preserve the open Internet without the regulatory, judicial, and even legislative battles that would come from reclassifying ISPs as Title II telecommunications services. If only everyone would act in good faith, Wheeler hopes stakeholders can agree on a reasonable set of regulations that ensure the Internet remains an open, vibrant medium for commerce, culture, and communication, but also enable network operators to extend and enhance broadband Internet in the United States.
That seems to be what everyone wants. But it’s still looking like a long, uphill battle.
Geoff, this was an outstanding neutral report and, contrary to the "public press" brought forward that there are many players between the "traffic generators" and the "last mile" deliverers. Thanks
Being older than dirt I do recall that it was illegal to attach an item to the handset (microphone / speaker) of the desktop phone so you didn't have to raise your shoulder to hold the telephone handset to your ear / mouth. Oh and you didn't want black? not allowed
Please folks, don't let us degrade Internet to the situation that governed "regulated telephone service". Innovation was illegal back then
Again to Goeff, thanks for a great report
Jerry
Thanks for the kind words - glad you found the article useful!
For folks who aren't familiar, Jerry's reference to it being illegal to attach a shoulder rest to telephone handsets has nothing to do with hands-free or distracted driving. Instead it's about FCC Tariff 132, which stated "No equipment, apparatus, circuit or device not furnished by the telephone company shall be attached to or connected with the facilities furnished by the telephone company, whether physically, by induction or otherwise." I don't know offhand when it went into effect (just after WWII?), but it was the law of the land until 1968.
Because AT&T was a monopoly - and all telephone equipment was sold by AT&T through its subsidiary Western Electric - that meant attaching *anything* to a phone was illegal if it wasn't supplied by the phone company. That included utterly simplistic things like the Hush-a-Phone (a plastic cup that fitted over a mouthpiece so users could speak more quietly, like they were cupping their hand over the mouthpiece).
The device that finally broke Tariff 132 was the Carterfone, which basically attached a short range radio to a telephone handset to extend phonecalls out to unwired locations. It was made for oil fields in Texas. Of course AT&T went after Carter, but he stuck to his guns, and after more than a decade of back and forth with courts and regulators, the FCC loosened up Tariff 132 to allow for any "lawful" device to be connected to telephone equipment.
And, basically overnight, the United States public was able to buy all sorts of wacky-shaped handsets in fluorescent colors, attach shoulder rests to handsets so they didn't hurt their necks being on the phone all day, and buy these funky little devices called "answering machines."
You know what devices also came on the market as a result of the change in Tariff 132 rules? Modems.
Isn't the underlying problem that ISPs are selling all-you-can-eat at "best-effort" speeds to everyone at roughly the same price? If they need more bandwidth to handle all the streaming, they should have customers pay for it. You want to stream 20 GB/month? Pay for it. You just want to read email, go cheap. Your choice. Not the ISP's, not the content providers'.
Also, content providers are by no means getting a free lunch. The consumers of their content paid for the content (iTunes for example) and for the pipe from their homes to the internet (e.g. Comcast). If that doesn't pay for it, charge more. Don't act as if you're giving them everything for next to free and then go behind their backs to slow them down while crying bloody murder at the content providers.
One size does not fit all. It's time to stop acting like it should.
Well, if one size fits 95% then it works pretty well. Most customers don't have a problem with caps, because 99% of them fit under them. The others get charged for the data, or have to move to a business connection which doesn't have the cap.
But the entire argument for breaking neutrality hinges on the ISPs' claim that they can't afford sufficient bandwidth extensions.
Instead of trying to force content providers into subsidizing their networks they should rather charge their own customers according to their bandwidth consumption. If I understand you correctly, that would only affect 5% of the customers anyway.
It should be noted that content providers are ISP customers, too. If their ISPs need more money because of all the streaming, maybe they aren't charging enough for the bandwidth they supply to the content providers.
Either way, both issues can IMHO be solved by a free market solution, without any sleazy backdoors, i.e. breaking neutrality.
The ISPs costs are not determined by a small number of data gluttons but by peak use. Everyone's watching online video in the evening so ISPs should pay for fatter pipes to accommodate them. It's the reason why long distance phone plans used to have peak and off-peak rates, to give customers an incentive to call less during peak call times; an off-peak call cost the phone company almost nothing, the same is true for off-peak Internet use.
Good point.
In principle I don't see why ISPs shouldn't try to attack the issue the same way. So far it was them who chose to charge everybody the same regardless of how much bandwidth they actually used or when they chose to use it. It's up to them alone to change that.
Good point. Is there a good reason why ISPs could not also charge differently for peak and off-peak use? Electricity providers do that for their big customers.
Now that the customer expectation is set, making such a change would be difficult. Per-household metering technology is also poor. Plus, one doesn't have complete control over bandwidth use, devices are using it all the time; e.g. OS X downloads a big OS update during a peak time. Peak electricity metering would be difficult too if your microwave would turn it self on at all hours.
I don't think peak and off-peak metering for bandwidth is a good idea anyway. Since most people are using it during peak times anyway, you might as well spread the cost by raising everyone's monthly rate. Or they could improve their bandwidth and be just a little less profitable; crazy, I know.
Or perhaps they should make bandwidth "caps" only apply from 8 am until 12 midnight. I'm not sure on the exact figure, but 90%+ of customers are well within caps. Then the torrenters etc, could just leave their stuff running at night.. You know, like the old days of BBSes where you'd queue up a bunch of downloads then go to bed and hope that they all completed properly!
But really this isn't the problem what we're seeing.. Mostly that ISP's aren't ensuring that they're able to handle their customer's peak bandwidth.
I find the lack of pragmatism from both sides annoying.
I'm a huge supporter and very long time customer of Netflix. (Seriously, when I joined their service they charged me late fees.)
I like the Openconnect CDN boxes that they've put out. However, I think its a bit naive of them to expect that they should get ports, power, space, some network admin from the ISPs for free. Yes, this does save the ISPs a bit when they peer with other networks, but from the ISP's perspective its an increase in costs, with little direct benefit for them.
What I do think is that ISPs should have to provide:
a) connectivity to all Autonomous Systems (AS) on the internet. Such that, during peak times a website will load without timing out.
b) a tariffed, regulated, service that provides direct access to X thousand customers. This'd include the rack space, an internet port, electricity, appropriate service from the NOCs, and reasonable margin above costs, say 10%.
Part A would ensure that all the internet is accessible. Part B would allow Netflix, Limelight, Akamai, and other CDNs to provide improved service. This helps shift costs a bit toward the direction they belong, without making my grandmother pay for my streaming video via a more expensive internet connection because she only wants to check email and little flash games.
I've had some people decry this as extortion, but if Netflix wants to use up multiple ports on an ISP, shouldn't they pay for them?
Here's my suggestion for a pragmatic solution.
Step 1. Netflix pays for their uplink. According to the required bandwidth.
Step 2. You pay for your downlink. According to the required bandwidth.
Step 3. There is no step 3.
I think we're in agreement.. We've just stated it differently.
Geoff, thanks for this very informative article. I've read a good bit about the net neutrality debate, but had not come across the fact that Netflix was taking a third of the bandwidth. That one fact puts the whole issue in a clearer perspective.
Now if someone can explain Bitcoin to me....