Android for STBs: What every pay TV operator should know

Launching a pay TV platform typically requires 2-3 years’ development to create a customized user experience based on proprietary middleware with continuous updates throughout the STB lifetime. It’s slow, expensive and requires specialist developers, but operators get the UX they choose.

Google’s AOSP and Android TV are shaking up this status quo. Android rose from zero to global dominance of mobile markets in under seven years, so it’s little wonder this disruptor is being taken seriously.

But many operators remain confused about the options available for hybrid STBs. So, let’s compare the facts.

AOSP stands for Android Open Source Project. Operators use it for free to create completely custom user experiences. Early examples included Amazon’s Fire devices.

Built-in support for Widevine DRM enables live and on-demand OTT. And because Android’s Java-based source code is so widely used, operators wont struggle to recruit qualified developers.

But it doesn’t come with an App Store or third-party apps like Netflix, so operators must create their own. This means the average AOSP-based service takes 18 months to launch, but the operator has control over every app on their box.

The alternative is Android TV, a smart TV platform with a standard look and feel that operators “paint” with their own brand.

Android TV is also free, but operators must license Google Mobile Services (GMS) for essential features like messaging, location services, and in-app billing. It supports PlayReady DRM in addition to Widevine.

The Google Play Store is included, giving access to thousands of ready-made apps including games, streaming services (Netflix, Amazon, Hulu etc.) and Google Apps like maps and YouTube. But, crucially, the operator can’t limit the range of apps to keep direct competitors off their box.

Voice search, PVR and recommendations are built-in. So is Picture-in-Picture, and a TV Input Framework (TIF) that seamlessly combines live feeds from separate sources such as DVB-C*, IPTV and OTT in one UI.

Navigation is via games console controller, the Android TV mobile app or a standard remote.

Direct Carrier Billing allows VOD and apps purchased in the Google Play Store to be added a subscriber’s TV bill, on a revenue share basis.

Like AOSP, there’s a wide pool of developers and operators could realistically launch a new STB in as little as 6-9 months.

They can also rely on Google integrate future developments like Virtual Reality and the Internet of Things.

Fundamentally, the choice between Android TV and AOSP is one of speed and convenience versus customization and control. Neither includes the pay TV stack which the operator must develop.

Most early adopters are telecoms companies familiar with the open Android environment on mobile devices. But analysts OVUM predict Android TV will capture 35% of the Smart TV OS market by 2019.

Will Android kill middleware as we know it?

Google’s Android operating system took just a few years to go from upstart newcomer to dominating the global mobile market. Can it do the same for TV?

Telcos like Swisscom, KDDI and Telecom Italia swiftly adopted AOSP and Android TV, but pay TV operators have been cautious, fearing Google’s influence on their business. What if a future strategic shift sees Google exit the STB market, leaving the operator with an unsupported platform?

Operators seem increasingly willing to overlook such legitimate concerns because Android’s benefits are so great (and because middleware providers can also exit the market).

First among Android TV’s charms is the potential to slash both costs and timescales for developing and launching a new STB UX compared to middleware-based boxes. Off-the-shelf support for voice search, recommendations and an app store stuff with third-party apps are a big attraction.

And although it’s hard to predict what technologies lie ahead, operators expect Google to integrate innovations such as virtual reality as standard, reducing the need for their own technology roadmap.

Android TV’s ready-made app store may be great for time-to-market, but it means operators can’t limit the choice of apps on their box. Subscribers are free to download a direct competitor’s app on an operator’s subsidized STB.

This concern leads many operators to favor AOSP, but others have simply accepted that consumers are now in control. If a subscriber wants Netflix, they will get it. Even if it means buying a separate OTT device which could eventually supplant the pay TV box altogether. Operators tell us it’s safer to allow rival apps on their managed STBs and retain control of the subscriber’s HDMI1 port. Some are even willing to accept cord-cutting in the short term because it’s easier to re-connect a subscriber that’s still using your box.

Of course, UX development isn’t the only cost consideration for a STB platform. Android TV has typically required a pretty high-end chipset to achieve the necessary graphics performance. But Google has halted incremental rises in chipset specs and leading manufacturers have already responded with middle-priced chips to support Android TV. Much cheaper compatible chips will surely follow, adding to the appeal of Android.

So here is my prediction: within five to 10 years, no newly developed STB platform will use proprietary middleware. Operators launching hybrid STBs will choose Android or another open source middleware. Examples include RDK, developed by Comcast and friends, which requires more UX development effort than AOSP, and Frog by French company WyPlay. Low-cost markets without reliable or widespread broadband access will stick with middleware-free zapper boxes without PVR or hybrid features.