All the sessions from Transform 2021 are available on-demand now. Watch now.
Last week, The Verge published a piece saying Mark Cuban wants to “kill” the open Internet. The piece was based on a Washington Post interview in which Cuban said he is in favor of creating “fast lanes.” He said that the fast lanes would guarantee the quality of video for specific applications.
One of the main arguments The Verge makes against Mark’s idea, and one we have heard before, is that startups would not be able to compete in today’s market if fast lanes were allowed. The Verge says that such fast lanes would have been a “death blow to the very startup that made him so rich in the first place.”
While I don’t feel the need to stand up for Mark Cuban, as he can easily reach more people than I can via his blog, it is important that people understand the actual facts of what’s taking place. And the Verge piece got the facts wrong in many ways.
Fact 1: Startups don’t deal with ISPs, they use third party CDNs
The Verge piece says that “If ESPN and the NFL can afford to pay for priority lanes to deliver their data, what chance does a young startup with limited cash really stand?” While that sounds like a scary proposition, implying that fast lanes would kill all promising young startups is not accurate.
In fact, startups and large media companies alike today have access to the same quality when it comes to delivering video: They’re called content delivery networks (CDNs), and they help ensure that traffic arrives at customers’ screens in a timely way. CDNs are available, for a fee, depending on how much traffic your company needs to support.
ESPN and NFL don’t own or operate their own CDN and no young startup would have to worry about paying any ISP, as no startup has the desire to build their own CDN to deliver video. Building your own CDN does not make sense from an operating expense or capital expense standpoint unless you get really, really big. That’s why today only a handful of big companies like Netflix, Apple, Microsoft, Google, and the like have their own CDN for video delivery. Most companies, by contrast, use third party CDNs such as Akamai, Amazon, EdgeCast, Limelight, and others who already have the necessary connections in place, both paid and free, with the ISPs.
Even today, ESPN and the NFL use these third party CDNs to deliver 100 percent of their video. No content owner using a third party CDN would have to “pay” an ISP for priority lanes as they would not be dealing directly with the ISP.
When Netflix itself was using third-party CDNs for all of its video delivery, you didn’t see anyone talking about performance issues with Netflix streaming. You also didn’t hear any talk about Netflix paying any ISPs. That’s because third party CDNs are in the business of delivering video with excellent performance, at a low price, and on a massive scale. That works for someone as big as Netflix, or as small as a one-person startup. No startup would try to replicate that on their own since they would not be able to get better performance, at a lower price.
The argument that startups would be affected in how they deliver video, based on any new rules the FCC enacts, isn’t valid. Those rules would only apply to ISPs, whom the startups don’t deal with directly. Even if the FCC changed the language to cover interconnect agreements, there is still no impact to startups since they don’t operate their own CDNs and have no need for interconnect agreements.
Fact 2: Paid access is not the same as paid prioritization
A lot of confusion around the net neutrality topic also has to do with people defining the term “paid prioritization.”
The definition of prioritization is to “arrange” something “in order of importance.” Many use the Netflix and Comcast paid interconnect agreement as their argument, even though both Netflix and Comcast have made it very clear that no prioritization of Netflix’s packets is taking place on Comcast’s network.
The Verge article gets this wrong as well when it says that Mark wants “preferential treatment” for his content, because he said he would be willing to pay T-Mobile to “subsidize their consumers getting [Dallas] Mavs games streamed live over their phones or to mobile home routers, without impacting their data caps.” At no time did he say that his data should be given priority over any other data or delivered in such a manner that it will stream faster. All he said is he would cover the cost of distributing the content. That’s not the same as asking for preferential treatment for his data, or a “fast lane”, which is what The Verge piece implies.
Fact 3: Cuban’s startup did actually face competition
The Verge piece also tries to argue that Broadcast.com didn’t have competition from a large competitors like today’s Pandora or Spotify, which might have struck deals with ISPs that would have effectively slowed down Broadcast.com traffic. This also isn’t accurate, as RealNetworks was one of the largest competitors to Broadcast.com at the time and had a distinct advantage over them in that the very server software Broadcast.com had to purchase, RealNetworks owned. Competitors did exist at the time that were larger than Broadcast.com, but The Verge article misses that point.
Fact 4: Most Internet video depends on proprietary protocols, not open standards
There is also the inaccurate statement in the piece stating that Mark built his business “using a set of free, open protocols.” That could not be further from the truth. All of the video Broadcast.com delivered at the time was via servers running software from RealNetworks and Microsoft. None of that was “free” and they all had their own proprietary streaming media protocols.
Even today, there are no “open” protocols being used for video delivery. A lot of video delivery is moving towards the HTTP protocol, but we still have RTMP, RTSP and other protocols that are anything but open. There is no standard when it comes to video on the Internet.
What people really mean by fast lanes
When people use the term “fast lanes” it generally refers to something taking place inside the ISP’s last mile network that reaches down to the customer’s home. Many have made their opinions known that they don’t want any ISP to be able to give one type of content special treatment over another by allowing one content owner to be able to pay to have their content delivered before their competitor. While that’s a fair argument to make, to date, we don’t have a single example of any ISP having done this — and Comcast legally can’t.
Many also want what takes place outside of the ISP’s network, at the interconnection point, to also be regulated by the FCC, so that no ISP can charge other CDNs to connect to their network. This is the argument Netflix has been making. Netflix wants free unlimited access to the entry points to the ISPs’ last-mile networks. But Netflix’s deal with Comcast and other ISPs isn’t a “fast lane”.
It’s simply a business agreement made between CDNs and ISPs to connect their networks without having to use a transit provider. These paid deals have been taking place in the U.S. for many years between large CDNs and ISPs and they don’t prioritize traffic or put one type of content before another. It’s simply a smart way of building and connecting networks that benefit the CDN, ISP, and the end user. Paid interconnects are a common network management practice in the U.S., and one of the reasons the Internet continues to work so well, even with all the traffic growth we see each year.
The Internet is already a level playing field
With all the cloud based services on the market today, startups are not at a disadvantage over companies like Netflix, even if the FCC comes out with new rules that allow for true “fast lanes.” Companies that sell content delivery as a service are in the business of making sure they have all the connections they need to the ISPs, be it free or paid. They are selling content owners a service that guarantees, in writing, a level of quality and performance for the content being delivered. New FCC rules could impact these CDNs’ costs of doing business, but they would not impact startups.
Throwing around terms like “fast lanes,” “priority,” and the like, without understanding what they actually mean, is dangerous. Any discussion around fast lanes should start with a definition of what it is, how it works from a technical level, which ISP is actually doing it, if any, and the reason for doing it. Many are quick to simply call an idea a “fast lane” and argue how bad that would be for startups and how it would kill innovation when in reality, that’s not the case at all.
Dan Rayburn is EVP for StreamingMedia.com and has his own blog at StreamingMediaBlog.com. He is often referred to as the voice of the streaming media industry and has been quoted in more than a thousand news items by nearly every major media outlet over the past nineteen years. He is also a Principal Analyst at research firm Frost & Sullivan, where he publishes market sizing data and trends on many video technologies and trends across mobile, broadband and last mile networks.
VentureBeatVentureBeat's mission is to be a digital town square for technical decision-makers to gain knowledge about transformative technology and transact. Our site delivers essential information on data technologies and strategies to guide you as you lead your organizations. We invite you to become a member of our community, to access:
- up-to-date information on the subjects of interest to you
- our newsletters
- gated thought-leader content and discounted access to our prized events, such as Transform 2021: Learn More
- networking features, and more