The reinvention of video-on-demand
Technology and competitive shifts are defining next-generation solutions
The first round of modern video-on-demand (VOD) deployments began about seven years ago, and two companies – SeaChange International and Concurrent Computer Corp. – emerged as the two dominant suppliers. A second round is about to begin, marked by new technological variations on the VOD theme. In general, elements of a VOD system are becoming commoditized (notably video pumps), and some resources can now be shared (e.g. universal edge QAMs).
The marketplace has changed: there has been significant consolidation, new requirements have surfaced, and competition is using a different playbook and completely different technologies. Some service providers are looking at VOD for the first time. Of those who have been providing the service, many are starting to replace systems that are already seven years old, such that within the next two to three years a majority of the top VOD service providers will have replaced or upgraded their VOD systems in all their key serving markets.
The signs suggest impressive growth for VOD, but speculating on who the victors might be is pure guesswork. The best anyone can offer is a look at how VOD is changing and how service providers can exploit the changes.
VOD has come a long way since the early attempts by John Hendrick’s Your Choice TV (YCTV) in 1992 or the US West Omaha trials in 1994. Today, watching television is increasingly becoming an on-demand type of activity that requires numerous precisely orchestrated technologies to deliver a quality user experience to subscribers. The industry seems to be moving toward a point where nearly everything one could watch on TV may appear as though it is on-demand.
Among these orchestrated technologies are the emerging next-generation systems from VOD vendors who, after the glory days of 2000 through 2002, have fallen on tough times. SeaChange International and Concurrent Computer, who represent the two most prolifically deployed platforms, are lingering at or near their 52-week lows on the stock market, while their main competitors have been scooped up by larger cable equipment manufacturers (Entone by Harmonic, Broadbus by Motorola, Arroyo by Cisco, C-Cor by Arris).
Between an equipment upgrade beginning and new business teeing up, VOD equipment vendors are anticipating good times ahead. Still, service providers are hesitating as they try to evaluate three evolving technology shifts.
The most important of these shifts involves the separation of hardware and software. VOD vendors are moving off their proprietary hardware and file systems to focus on becoming “pure” software companies relying on commercial off-the-shelf (COTS) hardware and open API software to maintain some reasonable portion of their previous margins.
Achieving pure software solutions is becoming critical. With VOD hardware components becoming commoditized, and the increasing adoption of Linux, the only place to add value (and justify margins) is in software. The increasing use of the standardized advanced TCA hardware platform is only going to contribute to driving down costs.
The second major shift is the migration from proprietary bundled content management to a stand-alone content delivery network (CDN) approach for VOD content propagation. CDN-based content management systems for VOD will enable more rapid ingest and content propagation to any device from any location – preferably the nearest available source while using transcoding, as required, by the defined capabilities of the device being served.
Table 1: Hardware VOD offerings.
The last major shift is the increasing importance of the VOD back office to support not only VOD session management but also other similar sessions such as switched digital video (SDV) and on-demand content delivered to wireless devices. Expanding the back office to manage these other sessions is creating a very lucrative upgrade path for back office vendors, to the point where VOD back office systems could very well become the most expensive part of any VOD system.
There is clear evidence of these emerging technology shifts in VOD. The following are some examples.
VOD’s change in focus from specialty hardware provider to pure software vendor is readily apparent in the latest video pump product offered from Concurrent. Its MediaHawk 4500 features their own real-time variant of a Linux operating system along with its video pump software that runs on a short list of approved COTS hardware platforms.
The importance of CDN-based content management systems for VOD becoming a stand-alone product rather than a feature of back office software is evident from recent work taking place at Harmonic, whose acquisition of the VOD server assets of Internet protocol television (IPTV) set-top vendor Entone has given it access to one of the most interesting content management systems available. The process of pulling out its CDN content management system creates a stand-alone solution which will be trialed at a customer site beginning Q4 of 2007.
Strong evidence of the VOD back office gaining value is evident from SeaChange’s recent overhaul of its Axiom platform. SeaChange Axiom is one of the most widely deployed VOD back office systems; it now supports video pumps from multiple VOD vendors as well as managing streams to mobile devices.
New VOD systems emerging
Just as these new products become available, new methods of VOD delivery (primarily download) are emerging – sometimes accompanied by equipment, sometimes not. These include services such as Amazon/TiVo, Blockbuster/Movielink, Netflix, and Satellite/DVR solutions that are challenging previously accepted delivery models.
These emerging solutions combine access to deep libraries of content with the broad appeal of an extensive online marketplace and appear only to be limited with respect to their last-mile connection to the living room. However interesting these solutions appear, they pale in comparison to more mature living room VOD threats from those listed in Table 1. These provide HDTV-quality VOD titles on state-of-the-art hardware platforms plugged directly into the television with back office distribution systems which are built for low-latency real-time interactivity.
All downloadable VOD solutions are not really on-demand; rather, they are “near” on-demand solutions dependent on the speed of their subscriber’s Internet connection. What makes them a threat is the increasing ubiquity of faster Internet connections along with their connectivity to the television in the living room.
VOD requirements are changing
Increased competition for VOD is driving new functional requirements. Next-generation VOD systems attempt to solve problems such as:
- Further marketing or promoting deeper libraries of content (see October 2007 CED Click Here) – essentially a need exists to expose all titles within the VOD library so that users are aware of all available titles, able to easily find them, and rent them.
- Addressing delays in propagation of content throughout the network – current VOD systems require one to three hours to ingest and 11 to 15 hours to propagate content to make it available for users. New VOD systems need to shorten this delay to enable real-time ingest and propagation of content to, for example, allow viewers to watch in its entirety a football game that just concluded, without being dependent on having first recorded it on their DVR.
- Subscription VOD and the electronic program guide (EPG) are not integrated – current systems require tuning to channel 1 to access VOD, but next-generation VOD needs to be more thoroughly integrated into the EPG such that navigating through available video programming and selection of VOD titles represents a seamless experience.
- Encryption of VOD streams is currently program-based rather than session-based. As a result, SVOD offerings can only be sold as a bundle rather than tiered. Next-generation systems will require bulk encrypters to allow expanded VOD offerings.
- No business rules on stream availability such that the availability of VOD streams cannot differentiate between someone wanting free content and someone wanting paid content. Since the build-out of VOD does not allow everyone on any given segment of the network to watch VOD, the availability of VOD titles needs to be cognizant of the availability of resources such that if a certain threshold is met, only paid VOD content is available.
The Achilles heel of even the most serious VOD threat is download delays. However, since these threats are banking on the continued fierce competition among service providers offering high-speed Internet, delivery delay is diminishing, and eventually could become negligible enough to be moot.
As bandwidth wars continue, resulting in increasingly faster connections, the opportunities available to the likes of Apple, Microsoft, Sony and others to expand their delivery of near real time services could catapult them into becoming legitimate video service providers.
Meanwhile, more experienced providers of video content know that providing quality video entertainment is not all science – there is also a certain degree of art to it, and it is this degree of art which separates IPTV and download VOD solutions from the real deal – namely, high-quality streaming MPEG video.
Expect to see IPTV and traditional VOD merge into a hybrid that combines their respective benefits, rather than the two solutions remaining and evolving independently. As this plays out, the infrastructure components that differentiate VOD from other video programming will be increasingly hard to discern to the point where offering VOD above and beyond other video programming will be a relatively small incremental cost over what today’s solutions fetch. VOD service providers who can keep their existing VOD deployments on life support for at least another year will benefit most from these developments as current next-generation VOD solutions aren’t quite there yet.