Tagged: Comcast

The Lapdogs and the Carcass of Net Neutrality

The corporate lapdogs at the Federal Communications Commission are to announce this week—the week of the Thanksgiving holiday the United States—their scheduled vote on December 15 to eliminate the “net neutrality” rules that govern wired broadband Internet providers.

The timing of the announcement and of the scheduled vote is not accidental. The FCC is trying to sneak the announcement during a holiday week when the country is distracted and will take the vote on a Friday before the FCC commissioners presumably adjourn for 2017. As we know, because of Donald Trump, the FCC has three business-friendly Republican commissioners that will out vote the two Democratic commissioners. There’s every reason to expect the vote to be a mere formality.

Karl Bode posted a great essay on Techdirt about the vote predicting a strong public backlash against the FCC’s vote to kill net neutrality rules. I won’t reproduce his argument here, but I want to draw attention to the two reasons he foresees a revolt. First, the public overwhelming supports these rules because, as with the broadband consumer privacy protections the Senate killed earlier this year, this is not a partisan issue. Hardcore lefties and righties want these protective rules. Second, these rules will largely benefit broadband Internet providers: i.e., the deep-pocketed cable and telephone companies that rank among the most hated companies in America. Much like the Republican tax plans currently debated in both chambers of Congress, the benefits will go to the wealthiest and most powerful segments in our country. The rest of us will get screwed.

However, unlike Bode, I am less optimistic about a coming public revolt against this FCC and the broadband companies they are supposed to protect the public against. A lot of people don’t understand what net neutrality even is, much less other related concepts such as common carriage that are arguably more meaningful and noticeable to people on a day-to-day basis. The most immediate effect of ending net neutrality will be preferential treatment of partner services. As we’ve already seen, Netflix is fine with partnering with ISPs to ensure a clear path for its streaming video service. As long as people can still stream video on Netflix and Amazon, no one will really notice that their Internet will no longer be an open-platform.

Of course, the long-term effect will be much greater, even if its harder to identify. That’s because the next generation of Internet companies will have a harder time emerging. Someone might develop something we can’t even imagine yet that could threaten Netflix and Amazon’s dominance the same way each company all but eliminated the Blockbuster Video stores that profited with usurious late fees and the major chain bookstores that forced many independents out of business decades. But we won’t probably will never see those competitors emerge and, even worse, we may never even know they existed in the first place.

The votes that Chairman Pai has brought to the FCC over his first year as the Commission’s chairman benefit incumbents over future innovative upstarts. While this may have a short-term benefit for the large companies that employ thousands of workers and trade on the Dow Jones stock exchange, as Verizon, Comcast, and AT&T do, these actions will cost us in the long-term in lost innovation. The Internet communications revolution in the United States didn’t come from incumbent telecommunications companies. It originated from military, government, and university researchers working together—often in their spare time. Had we left it up to AT&T or RCA, our Internet would basically be AOL and what Sprint called the “wireless web.” As someone who remembers both these versions of the “Internet,” I wish I had never known they existed in the first place.

HBO Now and the Technology We Need

Monday’s Apple event wasn’t just about an ultralight notebook computer that I really want and a watch that I don’t. There were also two announcements regarding television that were quite interesting.1

  1. The price for the current Apple TV dropped to $69. This third-generation model has been on the market since 2012 and was available for $99 as recently as this past weekend. I imagine that this is an intermediate move on Apple’s part. There are clearly better options for OTT streaming devices, even for die-hard Apple nerds, and I would hope Apple plans to release an improved version in the near future for the $70 same price. (Or not… what do I know?)
  2. HBO Now will launch exclusively on Apple devices for $15 per month. In some ways, this move is big because it marks the first time HBO is available without a multichannel TV subscription and could be a threat to multichannel TV as we know it. Consumers begged HBO to offer something like this, and now we know that it will be a full-featured service, not a crippled version that only the pay TV subscribers get.

In my most recent New Media class, I addressed the second announcement as an example of a technology adapting to our needs. We were discussing how social network sites were great for, you know, networking socially. Facebook has been great for sharing with your nearby friends, distant classmates, and obsessively doting relatives. After a while, though, we learned that our unsavory activities, such as our party pics, can be found by a potential employer, a college recruiter, or that human you are trying to date. The sharing aspect of Facebook is great, but the permanence is not. So, we now have something like Snapchat, and that is where we share indiscriminately because of its evanescence. We found a technology that better suited a particular need.

HBO Now is intended to provide stream HBO to those without pay TV subscriptions. But a lot of people I know already stream HBO programming with HBO Go. They just use someone else’s credentials to access it. As intrepid cord cutters, we already figured out how to get the product we need without HBO offering it. I’m not going to guess whether HBO Now is going to succeed. It could be a decade-long lifeline as iTunes was for the music industry or it could be as negligible as News Corp making an iPad-native Daily newsmagazine.

The Technology We Deserve

Net Neutrality plays an important role with this OTT service. Chris Morran at Consumerist speculates about the power of an ISP without net neutrality rules, where one could theoretically “throttle HBO Now while still allowing HBO Go to get through at full speed, effectively saying that the only way to get a decent HBO streaming service is if you have a cable package.” But with net neutrality rules, the two services should operate at the same level of performance. I won’t need to have my TV polluted with reality TV shows and shitty reruns in order to watch The Wire in 1080p.

The only not-neutral thing about the HBO Now service is that it will initially launch on only Apple devices, such as Apple TV, iPad, and iPhone. Roku, Amazon Fire, and Chromecast users will have to miss out on Game of Thrones until midsummer. Perhaps Mark Cuban was right when he said that we shouldn’t worry about the power of ISPs and that instead we should “worry about Google and Apple” because they make the operating systems of our mobile devices.

The Technology We Don’t Need

A few years ago, only the most devoted Internet libertarians were cognizant of that an ISP could potentially throttle or block a service it didn’t “like.” Today, even casual Internet users are skeptical about the control their ISP could potentially wield. For example, after the HBO Now announcement, Comcast subscribers were unable to access the http://hbonow.com website. It turns out that it was a technical problem caused by a DNS issue on HBO’s part, not some sinister shenanigans at the hands of Comcast. But because everyone hates their cable companies and because Comcast is as big as a cable company gets, the Internet reflexively blamed Comcast.

https://twitter.com/Peekaso/status/575110519369129984

It’s hard to shed a tear for Comcast though. As I mentioned earlier, when a technology fails to meet our needs or desires, we move on to something else. Cable television as we know it evolved from two converging technologies: Community Antenna TV and satellite cable.2 CATV was a demand-side technology. Starting in the 1940s, CATV operators piped TV signals to TV set owners living in areas where an over-the-air signal wasn’t available. CATV was almost a necessity for people living in a valley, such as in rural Pennsylvania, or in densely populated and overbuilt area, such as Manhattan, because the terrain blocked the radio signals necessary for TV reception. Satellite cable, on the other hand, was a supply-side technology. Beginning in the 1970s, it allowed national distribution for emerging television channels, such as Ted Turner’s WTBS-TV in Atlanta and the Manhattan-based Home Box Office. In both cases, CATV and satellite cable—eventually merging as the modern cable TV industry—enabled TV viewers to get what the broadcasters were failing to provide them. Today, however, the cable companies are the ones failing to provide us what we want, and that’s why we’ve migrated to something that does.

And, I’m sorry, Apple. As of right now, I still can’t figure out why I need a smartwatch.


  1. All that health stuff was not interesting. 
  2. Read Megan Mullen, The Rise of Cable Programming in the United States: Revolution or Evolution? (Austin: University of Texas Press, 2003) for a great account of the evolution of the cable TV indusry. 

Why Streaming Media Buffers and Why It’s Better To Take the Train

This week, FlowTV has a great lineup of articles on digital media. One noteworthy post comes from Kevin Hamilton, who distinguishes the difference between throughput and latency as the components of broadband. He compares throughput to traffic moving on a roadway. As a highway can only carry so many cars in a given amount of time, a broadband connection can only carry so many megabits per second. Once that limit is exceeded, there’s congestion, which we colloquially call “traffic.”

The second part of broadband is latency, which he compares to pneumatic tubes. Although he abandons the highway metaphor because he didn’t find it fit his understanding of latency, the pneumatic tube example didn’t work for me. As I’ve always understood it, latency is the delay in starting a transfer. Perhaps, if we could go back to the highway metaphor, latency could be compared to those traffic signals at freeway on-ramps that I remember seeing as a child in Los Angeles. (If you are driving with one or more passengers, you can bypass the signal, but who would ever do that?) At any rate, you can’t travel on the highway until you actually get on the road. Incidentally, the purpose of the signal is to reduce highway congestion, but I don’t think it makes anyone’s overall trip any faster.

Carpool Lane On Ramp with Traffic Signal

Planes Are Fast… Sometimes

Speaking of transportation metaphors, my favorite way of comparing latency and throughput is in terms of airline travel. It can take hours to move a few miles to get started on your trip. For example, you take a taxi to the airport, wait in line to get check a bag (if you still do that), clear security, wait for your flight to begin boarding, board the plane and get to your seat, wait for the flight crew to secure the aircraft, and wait for your pilot to queue up for take-off. At the end of all that, you’re finally airborne. All that waiting is latency, and it’s why it’s much faster to take a train to Philadelphia from New York than to fly there, unless you’re just traveling from one airport to another.

Throughput, on the other hand, is the time you’re in the air, en route to your destination. In airline travel, that’s relatively fast, but as anyone who flies in and around New York knows, there’s plenty of congestion which slows your travel time. And because so many flights go through New York, it wreaks havoc on the nation’s air traffic. The same thing happens in inclement weather in other airports, such fog in San Francisco and snow in Chicago. In those cases, the capacity of each airport diminishes because fewer flights can get through.

Not Net Neutrality

Understanding these terms help explain the recent Netflix-Comcast agreement that some critics hailed as the death knell to net neutrality, but as I wrote here some months ago, the agreement has little to do with net neutrality. Thompson explains that it was not a preferred throughput lane that Netflix bought, instead it moved into a crash pad closer to the airport.

By most technical accounts (and even these may still be wrong), the recent agreement between Comcast and Netflix seems to have addressed a throughput problem through an effort at latency reduction. Where many net neutrality advocates worried that Netflix was paying Comcast to give them faster throughput, the agreement is more oriented around removing a middle agent that was introducing latency into the system. Cogent, the company that transported Netflix data to Comcast for delivery to consumers, was not keeping up with demand – not enough staff in the tube transfer room, so to speak – so Netflix worked out a deal to tie their system more integrally to Comcast’s.

Another way of thinking about it was that Netflix cut out some steps towards getting on the plane. It no longer checks a bag, it got PreCheck to go through security faster, and it maybe even gets a ride on a Mercedes right up to the aircraft. Oh, and it always fly nonstop. All that makes the trip faster, cutting down on latency, but it doesn’t get you to the next airport any faster because as far as we know, the Netflix-Comcast peering agreement doesn’t include increased throughput.

At least, not yet.

Not Neutrality

It’s been a busy month for those of us who love streaming video and are suspicious of the cable or telco companies that function as our broadband providers.

First, there was the announcement that Comcast is acquiring Time Warner Cable to increase its position as the largest provider of multichannel video and broadband Internet in the United States. Second, Netflix released the second season of House of Cards over President’s Day weekend led to some quality of service issues for Verizon customers in the Washington, DC area. It was such a popular story that it was the basis of a sales pitch for a Roku Streaming Media player, warning that the days of streaming video might be numbered. Third, over the weekend, a Comcast announced that it had struck an interconnection agreement with Netflix. All of this news comes within the context of a DC district court ruling that the FCC lacks the authority to enforce net neutrality rules over wired Internet service providers.

Dan Rayburn, who writes for the insightful Streaming Media blog, takes issue with the coverage of the Netflix-Comcast deal.

Many of these same people are also implying that because Netflix has to pay Comcast, consumers will foot the bill for this as Netflix will have to charge more for their service. This could not be further from the truth. Those stating this have no clue how Netflix delivers their content today or what costs they already incur. If they did, they would know this is not a new cost to Netflix, it’s simply paying a different provider, and it should be at a lower cost. It should actually be cheaper for Netflix to buy direct from Comcast, and they also get an SLA, which also improves quality and that’s a good thing.

Netflix and Comcast: Get a Room

I’m still trying to figure this out, but it appears that with this agreement, Netflix connects its own routers with Comcast’s. That obviates the need for Netflix to use and pay a third-party content delivery network, such as Cogent or Level3, to route its traffic directly to the ISP. Instead, Netflix will have a direct connection to the Comcast’s pipes and to its broadband customers.

One of the crazy facts about the Internet is that is very much like the telegraph network of a century and a half ago. Yes, both disrupted our conceptions of space and time, but they also share two technical details: both are based on a binary code, and both rely on relays for interconnection. Almost every single transaction you do on the Internet is converted to binary, grouped together as a stream, and then broken up into packets[1]. It then goes through a series of routers that transit your packets to your intended destination. In the days of the telegraph, your message would transit from one telegraph office to another until it finally arrived at its final destination. Messages were priced according to length and distance because you were using up more resources than someone sending a short message to the next town.

A popular Internet service such as Netflix doesn’t have just one server to distribute all of its video content. It has dozens of them. They are located in places that are nearest to their customers. The aim is to minimize not only the distance each data packet has to travel but to also minimize the number of hops. Netflix contracts with companies that provide this service, each is known as a CDN. It appears that this deal merely cuts out the “middle man,” a third-party CDN that transits the data in this case, and allows Netflix connect directly to Comcast’s routers and have access to its customers all the way to the final mile without a third-party intermediary. Comcast subscribers should have a better experience streaming video from Netflix.

Is This Neutral?

With this deal, Netflix on Comcast will be better than Netflix on Verizon. Netflix on Comcast will also be better than another comparable video service (YouTube?) on Comcast that doesn’t have an interconnection agreement. But it should not impact upstart content providers. One of the major concerns of net neutrality is whether smaller players will have their traffic treated equal to the major players. In this case, it appears that the answer is no.

That argument however doesn’t consider Netflix’s size and its footprint on broadband networks. If it truly accounts for nearly a third of all Internet traffic in the US during primetime, it has few, if any, peers. Netflix has joined a new tier of Internet content providers, something like:

  1. Websites, blogs, and commercial services hosted on one server. If you’ve ever paid for a web hosting account, chances are you did this. You can get away with doing this because you are not serving much content and can stay within the bandwidth limits of your hosting plan. A dedicated server would be next option once you outgrow your shared hosting account.
  2. Websites, blogs, and commercial services hosted across different servers. Once a website becomes even more popular, it could outgrow that one server, and will have to shift its content across a number of servers. Many sites pay a CDN for the trouble of locating servers across different locations because it can be expensive— think rent and utility bills.
  3. Netflix. It is in a class by itself, connecting directly to the ISPs network. It has outgrown the third-party CDN model and struck a deal with the ISP itself. It is both a content provider and a Tier 2 network.

Again, I’m still trying to understand all of this, but it appears that net neutrality is still important even if it doesn’t apply to this case. Content providers in the first category above are safe because their volume of bandwidth is relatively low and won’t overwhelm ISP networks. Content providers in the second category should also be safe should they not overwhelm ISP networks. However, content distributors might have cause for concern if too many of their customers are like Netflix and begin overwhelming their peer networks. It’s just hard to determine what the line might be and how much it will cost to join this elite class of an upper-tier network. It’s also hard to predict if peer agreements effectively constitute preferential treatment.


  1. This is largely metaphorical. Forgive me if my use of this terminology is imprecise.  ↩

It’s Not TV, It’s Broadband

Late last night, news broke that Comcast is buying Time Warner Cable, a company that was recently spun off from its Time Warner media conglomerate parent. The two companies rarely compete against each other because cable television distributors, also known as MVPDs, operate as regional monopolies. Moreover, cable television companies see as competitors over-the-top services, such as Netflix and Hulu. They insist that while there will be fewer firms in the MVPD space, there will still be plenty of competitors in the TV distribution space.

This merger might be about getting MVPDs having better leverage during retransmission disputes. But the masterminds behind this acquisition wouldn’t be earning their money if they weren’t aiming to consolidate broadband Internet access. Free Press points out that “Comcast is the country’s #1 cable and Internet company and Time Warner Cable is #2. Put them together and you get a single giant controlling a massive share of our nation’s TV and Internet-access markets.” It’s hard to imagine that they wouldn’t put the squeeze on those same over-the-top services that they see as competitors.

We need to keep the pressure on regulators to impose net neutrality–type restrictions on these broadband providers. Or regulators can block the merger altogether.

Happy Hour is Safe, Hulu’s Fate is Not as Secure

While the New York Post received Mayor Bloomberg’s nomination for the Pulitzer Prize in fiction, it was another story that sent a few chills down my spine. Will Hulu be tied to a pay TV subscription?

the move by Hulu toward the new model — called authentication because viewers would have to log in with their cable or satellite TV account number — was behind the move last week by Providence Equity Partners to cash out of Hulu after five years, these sources said.

The major networks founded Hulu to prevent everyone from circumventing revenue-producing way of watching television. The move has been great. It has been a formidable way to get TV shows, and I have been a subscriber of Hulu Plus since the service launched. The appeal has always been that I can watch the television shows I want to watch and when I want to watch them.

This move could threaten the a-la-carte model that Hulu provides in favoring of bundling a package of channels that I’m never going to watch. While the Post article indicates that Providence Equity Partners is behind this strategy, I question that thinking for two reasons.

  1. the fact that one of the owners of Hulu is Comcast, the biggest and least favorite cable MSO in the United States, and a deal like this really help retains its customers; and…
  2. Hulu was up for sale last year, but then the sale was pulled due to its “unique and compelling strategic value”.

Free Press has launched an online petition to save Hulu from its possible fate as some “TV Everywhere” service that will only inflate cable bills.

(Via TV News Check.)