Today, the Wall Street Journal and others are reporting discussions about a possible merger between AT&T and DirecTV. As expected, much of the discussion centers around the size of the combined customer base and then secondarily around the impact on broadband delivery. Both of these are primary drivers of course, but there is also a technical aspect of the merger that is less visible: the combination of modalities between the true broadcast of satellite delivery, and the two-way customer by customer authorization of point-to-point or multicast internet.
I built a functional version of such a hybrid network about 15 years ago, and it is both quite powerful as well as solving some of the more vexing problems we see today with the rapid growth of widespread video streaming. For example, Verizon FiOS has a big problem Friday and Saturday nights at prime time, when their network totally clogs with Netflix traffic trying to individually serve video streams to the large audience wanting to watch the 50 or so videos that make up the bulk of the demand. The Verizon solution apparently is to abandon net neutrality, and charge Netflix big bucks for preferential access.
An AT&T/DirecTV mashup, on the other hand, could eventually implement a different strategy. Use the DirecTV satellite channel to stream, direct-to-home, all 50 movies that people want that day, and store them on the set-top boxes during low demand times (overnight the night before, for example). Now when Friday night comes along, all that individual traffic generated by streaming those 50 videos is no longer necessary, and the video performance is soooo much better, since it is being streamed locally. You want to charge for some of those videos? No problem, the AT&T internet piece has the two-way communication to do the authorization.
When I implemented this hybrid streaming approach, it was the dawn of HDTV, and broadcasters were still experimenting with over-the-air HDTV. We built a working system that pushed content over satellite links to local TV stations, who could then rebroadcast it (again, true broadcast) to all of their local PC users who bought a simple $100 external HDTV receiver. Since internet connectivity at the time was dial-up, and since credit card transactions were expensive, we aggregated micro-purchases on user-side secure chips for monthly billing. It worked, and was very elegant at the time, but the market passed on by. Nevertheless, the power of hybrid streaming modalities has stayed with me, and it’s interesting to see that possibility once again in play.