As most have noticed, net neutrality storm clouds are gathering again and battle lines are being drawn. The flashpoint right now is Comcast’s decision to have its Xfinity streaming content not count against its bandwidth caps, but it’s not just Comcast. AT&T and Verizon are working on related ideas in which content providers would pay to have their traffic counted against data caps. While all this looks like it will develop into a purely partisan disagreement, I’m starting to wonder if there isn’t something useful buried in the idea.
Let me start by assert something that in today’s media is perhaps not obvious. The positions of network neutrality supporters and network operators are not diametrically opposed. Nor are they parallel of course, but this is not a zero sum game. At the core, network neutrality supporters want a fair, unthrottled playing field that encourages innovation and freedom of usage, while carriers want traffic growth economically linked to revenue growth so that their network investments are justifiable.
Rumors have been circling about the major wireless carriers implementing a model similar to that for 1-800 numbers in the voice business, where content providers can pay to be cap-free. I’m as suspicious as anyone and we don’t know the details of the actual 1-800-number-style plans under development, but suppose someone put out a grand compromise with the following features:
- Non-discriminatory cap exclusion would be allowed — Any app developer can choose to pay to have traffic generated by the app be excluded from bandwidth caps, but on a per-GB/TB basis across all subscribers at a fixed price. No special deals for large providers.
- A minimum threshold to encourage innovation — The first XYZ amount of bandwidth generated monthly across all subscribers for apps that have signed up would be free. This would mean that new apps have the chance to gain momentum, and hence cash flow before having to start writing checks.
- Carriers must track traffic usage reliably for registered apps — If they wish to tie revenues to this, they must be able to quantify it properly.
- Apps can choose not to participate — especially low-data-usage apps, which can convince their customers that their apps are not a threat to their cap.
- No filtering based on functionality — Carriers may only apply this regime in the context of traffic volumes. If Skype or Instagram or whoever disrupts a competing carrier product within these rules, then so be it.
Such a regime would correct several problems that have been intractable thus far:
- App developers would have a direct economic incentive to optimize the bandwidth their apps consume, whereas right now we have (in wireless) limited spectrum resources with potentially unrestrained demands.
- Carriers would gain a direct linkage between the traffic their networks support and the revenues they generate, giving them more visibility for long term network investments.
- Users would not have to worry as much about exceeding their caps, encouraging them to *use* their bandwidth rather than calling them data hogs.
- Large content providers would not have an undue advantage.
- Last mile providers would not need to start disputes like the Comcast/Level3 traffic exchange dispute.
It doesn’t solve everything (what does?):
- How are the prices determined? It’s always been unclear how to find a non-regulatory way to have fair prices across a last mile bottleneck. In this case though, perhaps the popularity of apps and devices could help balance things. If the app everyone wants is available cap-free from carrier A but not from carrier B which charges developers more, then carrier A would perhaps start to lose customers. But it’s not clear this is enough, and of course nobody wants access fees and such.
- Tracking actual traffic usage per app may not be easy for the carriers without standardized tagging and such, which would take time to work out. There are also probably privacy issues to overcome in collecting the data.
Now, I can hear the howls out there that anything with a carrier charging a new fee is bad. But we already have a new fee in place in the form of the data caps and the related overage fees. If things go as they currently are then they will just keep the caps low and charge fees anyway. That places the onus on users, and nobody should want it there – not carriers who bear the brunt of their customers anger on the subject, and not content providers who really want users to feel free to consume content without watching the meter anxiously out of the corner of their eye.
Virtually everyone is against network operators arbitrarily blocking and filtering our content. Heck even the carriers should be against that, they can’t possibly want to spend the next decade telling customers, with whom they already have an image problem, what they can’t do as opposed to what they can do. But the economics have to be aligned somehow, and right now they’re not. For network neutrality proponents to take too hard a line on all this leaves a giant gap here that would be similarly counter-productive. Why not use the need for economic balance in raw traffic growth to negotiate for freedom of the content itself?
Whether such a 1-800 model for last mile bandwidth can fill the current gap or not, sooner or later a compromise will need to be found. And this feels much better than other ideas we have seen in the past. There must be a more robust linkage between bandwidth usage by apps and users on the one hand and the capital costs for the network that must carry that bandwidth. And there must be the freedom of content and innovation too.
Neither side should win this battle outright, because both are correct on many fronts. It would be disastrous for the carriers to gain full, arbitrary veto power over content on their networks, but it would be equally disastrous for the economics driving infrastructure investment to become dysfunctional. We want an environment where so-called ‘data hogs’ are instead ‘valued customers’, where innovative products that use less bandwidth to create more value are celebrated, and where needing to add more capacity is a cause for celebration all-around. Utopia? Perhaps, but we can do better than the status quo.
What do you think? Would such an effort be a step forward, backward, or sideways?
If you haven't already, please take our Reader Survey! Just 3 questions to help us better understand who is reading Telecom Ramblings so we can serve you better!
Categories: Government Regulations · Internet Traffic
Join Our LinkedIn Group
Rob, interesting, but you’ve essentially just reconstructed the switched access regime for the internet, exactly what ILECs would love. The switched access regime, through an artificial cross-subsidization scheme did nothing but assure the ILECs (i.e., Baby Bell’s) monopoly on local phone service until the passage of the ’96 telecom act.
I say, abandon all access structures. Make it all about COST CAUSATION. Plain and simple, if I and some subset of my Internet provider’s subscribers are consuming too much content/bandwidth, CHARGE US!!! Do not charge the content provider.
If my ISP (e.g., Comcast, VZ or T) doesn’t want me and a subset of its subscribers because we consume too much data, some other carrier will. Some other or new entrant will make the investment to win me and these other data hogs over to their network. Burying costs at other points in the packet stream is a monopolists dream because it keeps new entrants out of the market.
Personally, (and this will sound crazy) I believe the ILECs want metered usage LESS than its users do. Why? Because metered usage leads to two things — (1) unmetered usage and, worse, (2) competition from new entrants
Competition causes unmetered usage. Recall back in the early AOL dial-up days, there was a peak and off-peak rate — unlimited during off-peak and hourly rate during peak. Then facing competition from the likes of Netcomm, Compuserve, UUNet, PSI Net and others, AOL, the industry’s largest ISP launched their flat-rate service.
And metered usage causes new investment from new entrants. CAPs and CLECs TCG, MFS, ICG, Intermedia, Covad, Eastern Telelogic, MCI Metro, and countless others jumped into the telecom market between 1990-1996 to chip away at the switched access regime by offering competitive special access services.
Dear Comcast, I’m beyond tired of hearing about how you need to recover your costs from the likes of Netflix or Disney or Youtube. None of those content providers are pushing that data through to subscribers unsolicited. The cost causer in each and every case is ME and my friends. We asked for the data, the videos, the music, etc. You spent years luring me to your network with all kinds of inducements. Now I’m here and you don’t want me? That’s fine, just tell me and I’m sure someone else will.
hmmm…I guess I’m no longer wondering why Comcast seems to have stopped talking, at least publicly, about metered usage and net non-neutrality.
It’s nice to see that capitalism is still alive and well. Sure, it would have been nicer had it been a start-up (i.e., David) taking on a monopolist (i.e., Goliath), but (to horribly mix a metaphor) it’s still fun to watch a whale swim into another whale’s waters and steal it’s food supply.
I hope the regulators are getting this. Next time Eyeball Owning Network Operators (EONOs) like VZ, Comcast and AT&T cry to regulators that they need net non-neutrality relief or must have a switched access packet scheme because they can’t recover their costs, regulators should tell them to quit whining.
Google has a lot of money that could have easily been allocated to alternate large investments in content, software, hardware, etc., but they obviously saw excellent returns building a competing facilities-based ISP in the KC market.
Assuming success, which is still uncertain, Google’s KC experiment will be the case study in regulatory forbearance. Just because the monopolist (or oligopolist) cries poor to the regulator don’t make it true.
http://www.businessinsider.com/heres-what-google-fiber-is-like-2012-11
Am I correct in reading your conclusion as indicating that either large-scale providers capable of raising the funding or newer entities that don’t have older legacy-network cost baggage would be favored in these circumstances?
Wouldn’t the argument be that this is what led to the problems of the last decade of telecom and the “race to the bottom”? Thanks for the post.
I think you’re reading too much into my point. Plain and simple, charge metered rates. If technology exists to let a new entrant serve customers at a rate lower than what the incumbent is charging, a new entrant will get funded, build and compete.
My belief is that burying the cost elsewhere in the network distorts investment opportunities.
If I’m watching 100 high def videos per month, I AM the problem, not content provider. I asked for the data. If I know there’s a consequence — for example, a higher rate — I will decide whether or not to consume it. At the same time, an investor may look at that situation and say, if I can get the unit costs down to $x/gb, I will happily invest $xxx million to build a network. That competitive network will not be built if subscribers are cross-subsidized so that subscription rates are kept artificially low. This can only happen if the network provider is recovering that cost somewhere else.
My objective here is to ensure transparency and to avoid investment distortion. I AM NOT SAYING COMCAST OR VERIZON DON’T HAVE A RIGHT TO RECOVER THEIR COSTS. I’m merely saying they should recover that cost from the cost causer. The cost causer is their subscriber, not the content provider.
But if a content provider *wants* to pay for its users traffic consumption in order to foster greater use and gain marketshare or some other advantage, would you forbid it to ensure the dominance of the model you prefer?
Then the content provider can pay the subscriber directly. If the content provider is the same as the ISP (CP-ISP) — for example, Comcast — then Comcast should send the subscriber a separate “foster greater use” check at the end of every month.
If there’s a subsidy going on, let’s make it explicit and transparent. This will ensure that content providers know what they’re up against as they create new content and that potential new network operators know what they have to compete against.
Excellent post Rob and I think this is a very well thought out concept. The Comcast news had also triggered me to hypothesize on a model that works (see below link), but I think your structure is one that would likely work better if there could be some agreement on the terms.
http://www.ciena.com/corporate/blog/Did-Comcast-just-unveil-the-future-of-bandwidth-caps.html
One question though… the current 1-800 system is for North American voice traffic…could the same model work for a broadband world where international traffic is much more common?
You are forgetting something VERY different about wireless and the switched network. With wireless, these companies are operating over a PUBLIC RESOURCE. There is NO WAY for me to start a competing wireless phone company. There WAS a way for me to start a new LD phone carrier, and many did. This is why this model will NEVER work.
Anon, it’s not clear to me who your comment is aimed at.
In any event, there are MVNO models that could be implemented that address your concerns.
CCL: You are going to FORCE the networks, be it wired or wireless, to sell the pipe in the ground or their wireless at a fixed cost to whomever? GREAT! We have solved telecom. I will allow Comcast/Verizon/etc to do WHATEVER they want to do if I can use their network for a fixed cost. My point is, the carriers would NEVER agree to this. This is why Google is doing what they are doing in Kansas City, to show why an OPEN network makes sense. I am ALL for companies being able to do whatever they want, as long as I have the same access to that network…because I sure am not going to get the same government subsidies and rights that they did in running that network.
Anon, apologies, i think we are definitely talking past one another because I’m not sure I get your point.
CCL: I was replying to your point about MVNO being the solution. I don’t think you can force the carriers who are going to be putting these non-neutral products into action to allow MVNO’s or MVNO-like customers on their networks for a fair price.
Very interesting article — thanks for sharing your thoughts. Our company Box Top is already in discussions with numerous global carriers to develop 1-800 data on mobile networks. The platform is called FreeBand and uses paying party generated, on device apps to track and reverse bill for data consumed on the device, allowing dynamic pricing, load management, and free / secure bandwidth for the end user. If any of your readers would like to learn more, we have numerous tutorial videos at http://www.boxtop.tv. Thanks!
very cool application!! I’m guessing that the mobile operators love your application.
But doesn’t this add a cost to the application provider that wasn’t there before?
For example, if my application is advertiser supported, won’t this increase the cost of advertising on a boxtop.tv delivered app because I now have to recover my end user’s usage fee?
Also, isn’t there a risk that mobile operators will see this intermediary as a cash register and keep jacking up the bandwidth costs?
As you add more and more intermediaries into the packet flow doesn’t playing, for example, Words With Friends start to look like a mortgage settlement contract with countless intermediaries taking a small slice? Anyone who’s ever bought a house knows that there is this endless stream of fees — e.g., title search, title insurance, loan origination fee, FHA fee, flood determination fee, property survey costs…
Won’t more intermediaries added to the packet flow raise the cost of developing new applications?
The switched access scheme has been a complete nightmare. It’s hard to embrace a structure that seeks to replicate it.
Finally, switched access spawned countless arbitrage schemes over the years — e.g., leaky PBX, Local 800, interstate traffic posing as local for recip comp, revenue fraud, etc., etc. How will these be avoided?
isn’t there a risk that all this intermediation will spawn a whole new cottage industry of fraudsters?
Thanks for the comment. We have spent alot of time working on the model from the perspective of the content creators, app creators, carriers, regulators, and consumers. All parties benefit so long as the service is provided as an additional and voluntary service. You don’t want to use a mobile app per-bundled with free broadband, that’s fine — just pay for the bytes using your out of pocket bytes purchased from a carrier. What happens is that over time users will choose which content or apps are not good enough to “spend” their bytes on vs those that are. The former should therefore come to you free (just like a mail catalog or Netflix DVD in the post). The key is to make it 100% the end users choice. From there the market just responds to demand, sometimes bundling free byte delivery into the content (putting the stamp on the stuff that is mailed to you) and sometimes making you pay for the delivery cost. Today you pay all the delivery cost — no choice. We think once users figure out they can have much of the delivery for free this will be a very big and positive development enabling better, more dynamic pricing of bandwidth and empowering more people to get connected at lower costs (which regulators really like). As for the carriers, the can finally sell bytes to the content providers directly, charging them (or revenue sharing on ads, etc) and sell at different prices based on when the downloads occur (load management for selling bandwidth more cheaply during off peak hours. It’s all based on basic real market pricing that you see in every industry other than telecoms — but making the technology accessible via the app byte inspection capabilities (not expensive, slow, and cumbersome in-network DPI techniques).
Anon, I understand your point now. At some point don’t regulators have to regulate? Either regulate net neutrality or foster rules that permit others to compete.
If eyeball owning network operators (EONOs) know regulators aren’t going to intervene on either net neutrality or on anything else in this space, is there really a commercial incentive for any solution other than those that clearly favor Comcast, VZ, T, etc?
Although this is not the classic antitrust tying arrangement, doesn’t it look like Comcast is bundling its CAP-FREE streaming service with their internet service for the sole purpose of squeezing out competitors like Netflix, HULU, youtube and others?
I am Level 3’s General Counsel for Regulatory Policy. While Level 3 believes this is an interesting proposal, we think it is a step sideways as opposed to a step forward.
For many months, consumer Internet Service Providers (ISPs) like cable and phone companies have looked for ways to charge Internet companies for the content their own broadband subscribers request and have paid for, such as on-line movies, TV shows or streamed sporting events. The underlying assumption implicitly advanced by these ISPs is that they need more revenue to augment their networks to deliver the services they sell their subscribers. Noticeably, there is little evidence supporting this assumption. In fact, analysis filed with the FCC by Netflix showed that incremental last mile bandwidth costs are minimal compared to the amounts ISPs want to be paid for each extra gigabyte of data delivered to their subscribers. And cable company returns on invested capital are on the rise, which would not be true if their access network costs were growing at unsustainable rates.
It is important to keep in mind that most ISPs have dominant positions in that they control the access to their subscribers. In its Open Internet Order, for example, the FCC observed that as of December 2009, nearly 70 percent of U.S. households had only one or two firms providing advertised Internet download speeds of at least 3 Mbps —the closest observable benchmark to the minimum download speed of 4 Mbps the FCC has used to assess effective broadband deployment. The FCC also reported that 60% of U.S. households have only one firm providing more than 6 Mbps. Accordingly, consumers should not count on any meaningful competition between ISPs to ensure the fees that they would charge for third-party high-bandwidth applications delivered “over-the-top” would be fair, equitable and non-discriminatory.
Further, many of these ISPs also own proprietary content (like Comcast’s ownership of NBC Universal) or other businesses (like Verizon FIOS or cable TV) that compete directly with online video. Comcast, for example, had revenues in 2011 in excess of $55BN, approximately 35% of which analysts report came from NBC Universal (its newly acquired media division) and 65% came from its Cable Communications business, which serves 22.3MN video subscribers and includes its Interactive Media businesses (Comcast.net, Fancast, the Platform and Fandango). Analysts also observe that Comcast is seeing increasing pressures on pay-TV subscriber growth as over-the-top video outlets such as Netflix and Hulu continue to expand their viewership, while video streaming to iPads and other mobile devices gain more prominence. They also list “cord cutting” (eliminating cable TV service in favor of content delivered over the Internet) as a rising concern. All of this provides ISPs an enormous financial incentive to discriminate against competing over-the-top offerings, which affect their very livelihood.
The proposed solution would allow these ISP to unilaterally set charges that would be required for an Internet company to avoid having its over-the-top services subjected to a bandwidth cap. Under these circumstances, it is hard to imagine a regime where the payment would be “nondiscriminatory”. It appears very unlikely that such a regime would preserve a level playing field where new content and applications (the next Facebook or the next Netflix) would get a fair shot at succeeding based on their innovation and the value they bring to end users, as opposed to whether they were owned by an ISP or competing against it. If Comcast, for example, were permitted to charge an over the top video provider such as Netflix a fee for uncapped access, it would be ineffective and meaningless to ask Comcast to also “pay itself” to ensure its Xfinity streaming service or its NBC Universal content are subjected to the same charges.
We believe that ISPs should be required to interconnect on reasonable and nondiscriminatory terms that are no less advantageous than the terms effectively provided respecting their own or their subsidiaries services, and that they should not be allowed to charge for local access to their subscribers. We believe such policies are best for competition and the continued openness and evolution of the Internet.
Mr. Mooney, do you along with your Level 3 constituents believe all internet pipes are “dumb” deserving no portion of advertising revenues tied to the distances its content travels in meeting the eyeballs seeking it out?
I remember discussing this crucial question with one of the Fathers of the Internet, who agreed with me that a “sharing” relationship between content owners and infrastructure providers must be sought and agreed to if the internet would sustain itself ultimately.
This makes sense when fully understanding that “broadband” in the longer term is a “scarce commodity” including but not limited to “inefficiencies” in delivering it to the marketplace.
Because sharing is caring in order to maximize long term benefits of valuable resources, this Father of the Internet believed with me that, the advertising component of the internet revenue pie should also be explored.
Is Google ready to ensure that the dumb pipes carrying their content to end users around the globe has its advertising revenues shared according to certain distance formulas by infrastructure owners meeting their end users’ demand as part of the carrying equation?
http://finance.yahoo.com/news/google-tries-again-google-tv-100402567.html
What’s good for the goose is good for the gander.
A gander is a group of geese. So the saying means, if it is good for one person, its good for everyone.
ISPs should be required to interconnect on reasonable and nondiscriminatory terms that are no less advantageous than the terms effectively provided respecting their own or their subsidiaries services, and that they should not be allowed to charge for local access to their subscribers.
When will American end users stop letting the ISPs (the likes of at&t, Verizon, and the cable guys) from bulling them around and slowing down the delivery of content to the screens.(buffering – throttling) Please, someone in this country stand up and stop the old guard from blocking competition and open up the Internet, so innovation can evolve.
couldn’t the same argument be made for Level3 to open up access in the LH fiber….?
or is that implied in their response?
or is that even a problem these days?