My Latest Thoughts On The FCC’s Statements and The Netflix/Comcast Dispute

I appreciate so many people being interested in what I have to think about the latest comments by the FCC and the latest hair pulling fight between Netflix and Comcast. Since I’ve gotten so many emails and questions on the topic, I figured another blog post was needed. On the FCC and net neutrality topic, I really don’t have any comments on that yet, as it’s too early and we don’t have any concrete info on what exactly is being proposed. While FCC Chairman Tom Wheeler published a blog post on Thursday, it does not say much. The post is entitled “Setting the Record Straight on the FCC’s Open Internet Rules,” but it doesn’t do anything of the sort.

The post is vague, with no real definitions and is full of words and phrases like “unreasonable” and “commercially reasonable”. We don’t have any idea what those words mean, how they are defined, or the scope of what they will or will not cover. I see people making a lot of guesses or assumptions on what will take place, but I’ll save judgement on that until I have something tangible and detailed to review. Without it, my guess is just as good, or bad, as anyone else’s on what may or may not transpire.

I will add that personally, I’m tired of hearing so many consumer advocacy groups quoted in the media telling everyone that as a consumer, I need to be saved from the harm that will come to the Internet. They don’t even have the details on what’s being proposed be it the business terms, how compliance will be monitored, who exactly will be in charge of it, or how any of these rules will work. And many of them, can’t tell you the different between peering and transit or speed versus capacity. Just because they call themselves a consumer advocacy group, does not mean they know the subject they are being quoted on. Personally, I think the Internet has worked pretty well over the past twenty years that I have been on it, so I’ll save judgement on what good or harm will come from this until I have more details.

On the Netflix and Comcast dispute, honestly, it’s getting really tiring watching two billion dollar companies try to use the media, and consumers, to convince us of their stance on the subject. Netflix is standing behind blog posts and letters filled with vague, high-level terms, which many times are inaccurate. I say stop all the posturing, show the data you have and let all of us decide what makes the most sense. Stop deciding for us what you think is best and let the consumer, or those who track the industry, decide. Netflix and Comcast both have a lot of data on what’s actually taking place on the backend, from a technical and business level, yet they won’t use any of it when they make their arguments.

If Netflix wants to solve this, make points with facts, real data and numbers, not assertions based on opinions. It’s fine if Netflix has a different take than Comcast or anyone else, but then treat it as such and don’t call things “unprecedented” just to exaggerate and get people all worked up. If either company wants us to takes sides with them, show some transparency as to what’s taking place, especially when it comes to the numbers. I don’t care what anyone says, this debate is all about big business, with both Netflix and Comcast wanting to protect their bottom line. It’s about dollars and cents and who can spend less and have better margins. Any other implication by either side is simply not genuine.

While it is about business for both sides, Comcast has the leg up in this regard as their argument is that interconnection deals have been taking place for a really long time, and they are right. This is how the Internet has always worked and anyone who builds their own CDN, like Netflix has done, has costs associated with connecting to any ISP, if it does not fall within the ISPs peering policy. If Netflix wants to argue that this is no longer the Internet of 2000, or that an ISP like Comcast is now too big and has an unfair advantage, then that’s a fine argument. But it won’t go far without facts and the burden of proof is on Netflix to prove this is bad for everyone.

Netflix has been very vocal to say it’s bad for the whole Internet and consumers, but the one point I hardly ever see anyone talk about is why Netflix is all alone in their corner? Why aren’t other large content owners/distributors backing them up? Why isn’t there a big contingent of content companies rallying around Netflix? Where is Google? Apple? Microsoft? Amazon? They all have a stake in this, if Netflix’s argument is to be believed, especially because they all operate their own CDNs. I find it odd that no one really seems to mention that to date, Netflix is standing alone on this topic, being backed up by no other major content owner.

It also does not help Netflix’s cause that they are all over the place in their statements. When the media and others were stating, with no proof, that Comcast was throttling Netflix’s stream, Netflix’s CEO said he didn’t think that was the case, and we’ve seen no proof to show otherwise. Netflix then signs a deal with Comcast saying it was “a mutually beneficial” agreement, but then weeks later says it isn’t and they did the deal because they had no choice. Netflix said it was a mutually beneficial agreement because in reality it is, from a cost and quality standpoint. The only thing more beneficial to them, would be if they didn’t have to spend any money with Comcast, but still got the improvement in quality.

As I have said before, if Netflix thinks the economics of the CDN market need to change, I’m all for that debate and listening to their reasons. But it seems, at least to me, that Netflix isn’t saying they are paying too much, but rather they should not have to pay anything at all. Am I the only one who feels like every time Netflix makes their argument they sound like someone who always wants everything for free? BOTH sides should bear the cost to improve video quality as it benefits both companies. If that’s what Netflix is proposing, then they need to detail what those numbers are and how BOTH sides pay for it. You can’t try to debate something in a public forum, but then hold back all the data that’s needed to solve the problem.

For those not following the latest happenings between Netflix and Comcast, and I don’t blame you if you aren’t, it’s very time-consuming, and frankly, getting tiring. On Thursday, Netflix’s VP of Global Public Policy sent a letter to Sen. Al Franken. In the letter, he once again reaffirmed Netflix’s statement that Comcast should not merge with TWC as it would be bad for consumers and content owners. Comcast then replied with a blog post about the letter, saying Netflix isn’t being honest about its cost-shifting strategy. Netflix’s person responsible for their CDN, Ken Florance, whom keynoted my CDN Summit show last year, published a post on Netflix’s blog, signaling out Comcast as the one troublemaker. The post points out that “Comcast alone sets the terms and conditions for access to Comcast subscribers,” which is true, because it’s Comcast that owns the Network.

In the letter Netflix sent to Sen. Al Franken, Netflix said they are “committed to sharing facts with policymakers to increase their understanding of this issue.” That’s a bit disingenuous to me. Netflix is using the public to their advantage, getting consumers to complain to Comcast, but then aren’t willing to share “facts” in a public forum. If you stand behind your argument, show what you have, give us the data, and state your case with the evidence you have. This is the same request I made in some of my previous blog posts, but I’m not expecting Netflix to do it, which is a shame.

This whole argument between Netflix and Comcast is pretty childish. Two billion dollar companies who can’t come to an agreement for the benefit of consumers and their own customers. I’m amazed that Netflix would rather the government get involved, which is crazy, considering what kind of track record they have in solving problems. And before I get hate mail saying I am talking down on our country, I’m not. I spent six years in the military, I respect my country, but our legislators don’t have a great track record when they start making policies around technologies they don’t understand.

I’ve got to get some sleep, but the comments section is open and I’ll try to get to any questions when I can.

  • Michael Fiumano

    Hoo-rah!I think you’ve got it quite right Dan.

  • Vaughan Read

    So is Netflix’s VP of global policy completely lying in these statements to Senator Franken?

    “Netflix developed an entire CDN architecture, called “Open Connect” based on settlement-free peering. This no-fee interconnection norm avoids the gamesmanship and blackouts that plague cable carriage and retransmission-consent negotiations in the traditional video space. Indeed, Netflix is directly interconnected with ISPs all over the U.S. and internationally without any exchange of payment from either side. Our agreement with Comcast is the first time that Netflix was forced to pay an ISP for what amounts to access to their subscribers… Prior to our agreement to interconnect directly with Comcast, Netflix purchased all available transit capacity into Comcast’s networks from multiple transit providers. Every single one of those transit links to Comcast was congested (even though the transit providers requested extra capacity), resulting in poor video quality for our members.”

    Netflix keeps talking about settlement-free peering, but this just isn’t (nor ever has been) the case when it comes to transit providers connecting with ISPs, has it? My understanding was that transit providers may have settlement-free peering with each other, but definitely do not have settlement-free peering with last-mile ISPs. At least not the ones carrying Netflix traffic. They pay if they’re delivering a disproportionate amount of traffic, yes?

    And though details of the Comcast deal haven’t been released, don’t you think the most likely scenario was that when Netflix was negotiating the terms of entry for their CDN into the last-mile, Comcast must have asked for a little extra money on top of whatever it costs Netflix to bring their servers in?

    Clearly Netflix is misleading the media and everyone else, but I’m trying to understand what they’re really, actually upset about. Because, like you said, they’re not stupid. Or maybe this is all just a bunch of rent-seeking… distorting the truth simply to secure the most favorable set of regulations in the future… which is what happens when you’ve got a regulator that doesn’t know what its own mandate is!

  • Dan, there are a bunch of moving pieces here, but they primarily get down to:
    -revenue/payment models
    -bandwidth consumption
    -cost of transport and switching/storage

    Starting in reverse order, WAN transport cost/minute of voice is $0.0000004 based on 2 year old OECD data, while the cost of non-competitive MAN transit/termination is $0.001. That’s a big spread. Google Fiber has introduced a new MAN pricing model that moves the decimal right by 2 or 3 zeros; or much closer to WAN pricing.

    Video consumes typically 10-50x voice capacity. Longform video consumes a lot of capacity in aggregate, but Netflix spends a lot to transcode any given video 125 different ways to be viewed across 1000+ different screen/network contexts; making the viewing experience more ubiquitous and the transport costs down. The bit rate is typically just 1/10th of what an end-user buys for downstream bandwidth. (One could argue that BB providers should give Netflix a marketing vig for driving consumers up the bandwidth and ARPU scale). Youtube video clips are typically much shorter in duration and the quality much much lower. Video ads are also mostly 15-30 seconds, and often skipped after 4 seconds.

    Netflix has a flat-rate AYCE model which appeals to a large pool of customers. I’ve written in the past about why these models will go the way of the dinosaur. But its a risk Netflix is currently managing and profiting from well. Its customer base is highly targeted and known and its taking advantage of the above WAN transport costs. Google’s advertising driven web ecosystem, while it has in aggregate, 3-5x the amount of traffic Netflix has, it is far more dispersed and a varied user base. The ad supported revenue model is tied to discrete consumption based on engagements. The edge access providers also have an AYCE model, but their margins suggest that their pricing is well above average or marginal cost.

    Google and the web is more dispersed than Netflix. So they would have different short-term considerations wrt peering and CDNs. But Netflix uptake is a serious disruptor for the LinearTV models of the both sets of incumbents; particularly as we start moving to a 4K future. In the present and future it is inefficient for Netflix streams to be transported hundreds of mile by 3-4 monopoly edge providers to markets outside the top 20-50. Netflix can statistically multiplex the consumption better from its core vantage point, and it can use 1-2 competitive transport providers to move the cloud to the edge. So in effect, Netflix becomes the locomotive that continues to scale everyone else’s cost.

    But Comcast is preventing this by keeping the WAN/MAN demarc (the cloud) closer to the core. This should concern regulators (federal, state, local) because it moves out the dates when universally cheap 4K, 2-way HD telepresence, mobile BB, and the internet of things happen. Instead of years we are talking decades. The latter 3 trends rely much more on upstream capacity and performance and latency than do 1-way HD or 4K video streams. Therefore it makes sense to move peering points farther to the edge (trade off of layers 2 and 3). So people may not realize it (even incumbents) but Netflix is the best bet to get to a far more robust networking future than the fictions the edge access providers would have us believe.