One of my first independent consulting jobs was with Major League Baseball Advanced Media. They were partnering with Sony on something that at the time was very ambitious: a cable replacement for the PlayStation 3. They called it PlayStation Vue.
To understand why Sony chose MLB.tv as their partner, you need to know what MLB Advanced Media had accomplished. They were the first to live stream sports and do it well.
My boss at MLB described live streaming at the time as laying down train tracks as the train is coming. MLB.tv used HLS, HTTP Live Streaming, and an Apple format that streams content over the internet in six-second chunks that rebuild on the fly. They were really good at it. They streamed March Madness, the Masters, and countless other major sporting events. You might not have known they were behind it, but you definitely noticed their streaming quality was vastly superior to everyone else.
Sony's vision for PlayStation Vue was straightforward but radical: everything on cable should be available on demand. Not just some things but everything. This presented an interesting challenge. If video on demand were ubiquitous, what would the interface look like? How would the channel guide work?
Television has evolved an entire infrastructure built around TV guide data. One of the challenges of storing metadata from shows is: whose data wins. It shows when things are on, what kind of show it is, how long it runs, etc. But here's where it got the most complicated: the same movie might stream on AMC and also on ABC, but they'd be categorized according to completely different metadata. One would call it a gangster movie, another would call it a crime drama. Same film, different classification systems.
I created a taxonomy that unified all this disparate metadata, filling in gaps, and prioritizing the data that we purchased from a vendor. We were trying to keep it as simple as possible while including a reliable process for adding to it as needed. If something was labeled a “gangster” movie in one system and “crime drama” in another, we'd map both to a single category. The goal was to take these different data sources: same show, different provider, different metadata… and translate everything into one unified system.
We got the proof of concept working. We could stream both live and stored content seamlessly. The technology worked. The design worked. We'd proven it was possible.
Then reality intervened. The big five content providers didn't want everything available on demand. That was the entire vision, but it turned out the people who owned the content had different ideas. The compromise: users would have to "record" something to get on-demand access to it, mimicking the DVR model everyone was already familiar with. But this was still better than getting cable or a satellite dish installed.
Playstation Vue was ultimately sunset by Sony. The market economics for Over the Top Cable apps are challenging, and soon after, viewers began using Rokus and Apple TVs to stream versus their gaming systems.