People are excited about online TV getting big in 2016. Alon Maor of Qwilt predicts in Multichannel News that this will be “the year of the skinny bundle.” Wired echoes that sentiment. The Wall Street Journal’s Geoffrey A. Fowler said, “it’s no longer the technology that holds back cable cutting–it’s the lawyers.”
Well, I’m here to say, lawyers can’t take all the blame. In my experience, it’s the technology, too. Some of the problem is that most discussion about the future of online TV and cable cutting fails to distinguish streaming video-on-demand (SVOD) and streaming linear TV (“linear” means continuous pre-programmed and live “channels”, often with commercials, much like traditional cable).
SVOD includes Netflix, HBO Go, and Hulu. Yes, SVOD technology is very good. The major SVOD players spend millions on networks and caching so that their (static) content is as close to consumers as possible.
But streaming linear TV–like Sling and live sports–cannot be cached like static content and appears to have kinks to iron out before mass-market adoption. Read online video analysts and you realize that streaming linear and live TV online is an entirely different animal from SVOD. In August, Ben Popper at the Verge had a fascinating longform piece about the MLB’s streaming TV operation, Major League Baseball Advanced Media (BAM), which is the industry leader doing live streaming video. (HBO hired BAM for streaming the latest Game of Thrones season after HBO Go’s in-house streaming offering suffered from outages.)
Linear online TV is hard! As one BAM executive explains vividly,
What people forget is that the internet, as a technology was never designed to do something like this–deliver flawless video simultaneously to millions of people. I liken it to trying to live on Mercury. The planet is completely inhospitable. Every day all you’re doing is [fighting] a battle for survival in a place that really does not want you.
So I say this understanding the significant technical difficulties they face: in my experience, streaming linear TV is not ready for prime-time yet. (If I could find information about how firms do it, I would also distinguish pre-recorded linear TV–like Sling’s HGTV channel–from live online TV, because firms likely use different network topologies. Another time, perhaps.) Perhaps I’m an outlier, but media reports about Sling outages suggest that I’m not. I used Sling TV for the past few months and had high hopes but I just dropped my subscription. My test for acceptable streaming quality is: “Would I invite friends over to watch something using this streaming service?” Most SVOD services pass that test. Through caching and streaming protocols, SVOD operators can assure pretty consistent streaming even during times of moderate congestion.
Since linear TV typically has programming that is transmitted (nearly) live, operators can’t really do the distributed caching of content that makes SVOD function well. As a result, streaming linear TV like Sling TV and WatchESPN (a Sling subscription gives you access to ESPN’s online sports portal) currently do not pass my test. Frankly, the streaming quality varied from excellent to unwatchable. Punching into the app and casting it to my TV, I was never sure which Sling would show up that day: Good Sling or Bad Sling. On good days, the Monday Night Football game looked better than cable TV. Other days, the stream would, for instance, work well and then fail at every commercial break (perhaps the commercials were stored on another, overwhelmed server?).
Now, it’s impossible to know for sure where the network bottleneck is and why a video stream is stuttering. To be more certain that the problem was with Sling TV or WatchESPN and not, say, my Chromecast, my Wifi, or my ISP, every time Sling or WatchESPN had several severe streaming problems in a period of short time, I did an unscientific test. I would close down the Sling app, open up Netflix, and start streaming Netflix via my Chromecast. Without exception for the past three months, Netflix loaded quickly and streamed well. Keeping everything the same except the source of content suggests (but doesn’t prove) that the problem was not a local network or device problem.
There are ways of using network architecture and protocols to improve linear TV on broadband, typically with dedicated servers and last-mile bandwidth reservation. Comcast Stream TV is an example and LTE Broadcast might someday soon provide linear and live TV for mobile customers. But given so-called net neutrality rules, these services are controversial and regulated. ISPs can have their own proprietary TV service like Stream TV but, given net neutrality hysteria, probably won’t offer dedicated bandwidth to distributors like Sling. In this narrow area, the FCC’s rules are a pretty good deal for larger, vertically integrated firms that can put programming bundles together. It’s not so great for the small ISPs and WISPs who want to respond to cable cutter demands and offer a quality TV product from another company via broadband.
So I expect linear online TV to remain niche until the quality improves. A big draw of Sling is the cancel-at-any-time policy which lowers the risk if you’re dissatisfied with programming and allows single-season sports fans like me to subscribe for a season or two. I subscribed in the fall and winter to watch NCAA and NFL football. Sling recognizes that a lot of its subscribers are like me. But if Sling and other linear TV programmers want to expand beyond niche, they’ll need higher-quality streams. (And Sling might wish to remain niche so they don’t upset their programmers by cannibalizing traditional subscription TV.) Would I sign up for Sling again? Sure. Maybe the prognosticators are right, and the technology will develop rapidly in 2016. I have my doubts.
The Technology Liberation Front is the tech policy blog dedicated to keeping politicians' hands off the 'net and everything else related to technology.