Robbing Peter to pay Paul

For the past few months, this column has explored the changing landscape of digital television — a landscape that seems to be changing before our very eyes at an accelerating speed.

Next month, we will try to sort out what is happening in Washington with respect to a hard date for the transition from analog NTSC to digital ATSC. Congress is currently trying to iron out the differences between bills passed in the House and Senate to bring NTSC broadcasts to an end in March 2009, immediately after the conclusion of March Madness (the men's NCAA Basketball Tournament).

This would not be the first hard date that broadcasters have been threatened with. The question is whether it will be any more real than the other deadlines set by the FCC and Congress, which were then rendered meaningless by subsequent legislation.

At the same time, Congress is talking about receiver subsidies for viewers still reliant on NTSC broadcasts for their TV fix (the roughly 15 percent of U.S households that watch TV without the aid of cable or satellite). And the NAB is working with several vendors to develop a reference platform for low-cost set-top boxes that would deliver DTV content to generations of legacy NTSC displays. Wrapped up in all of this is the political wrangling about multicast must carry, which some broadcasters view as a way to generate new revenue streams from their investments in the DTV transition.


Just to refresh your memory, let's look back — nearly two decades — and remember why broadcasters went to the FCC in 1987 and asked the commission for additional spectrum to deliver what was then perceived as an important new capability — HDTV. The stated purpose was to allow the over-the-air broadcast service to remain competitive in the face of increasing competition from multichannel subscriptions services. But the real reason was to stall a move by the land mobile telephone industry to share the spectrum used by television broadcasters.

At the time, it was believed that it would take two 6MHz channels to deliver HDTV. So broadcasters asked the FCC to consider giving them the extra spectrum to develop a service that would augment the existing NTSC broadcasts to produce an HDTV-quality viewing experience.

The plan worked initially, but it backfired when General Instruments demonstrated the ability to deliver HDTV in a single 6MHz channel using the same digital compression technology that the company was developing to enable the direct broadcast satellite industry to compete with cable. By 1991, the FCC changed the game, telling broadcasters that they would be loaned a second channel for digital HDTV broadcasts. Then, at the end of the DTV transition, the analog NTSC channels would be recovered.

For the next four years, the focus was on developing a new digital broadcast service that would deliver a single HDTV channel to replace the single NTSC channel assigned to broadcasters. Multicasting was off the table, as it was perceived in Congress as an expansion of the broadcast franchise.

But at the last moment in 1995, when it was clear that Congress would authorize the transition plan, the ability to deliver multiple channels of standard-definition television was added to the ATSC standard. And when the FCC established the rules for this service, it added a provision allowing broadcasters to offer paid services alongside the one required channel that must be delivered in the free and clear. The price to play is just 5 percent of the revenues generated by these new services.

Broadcasters began to look at the DTV transition in a different light. Perhaps there was an opportunity here to generate new revenue streams via their DTV broadcasts.

The cost of prime time

What is a set of eyeballs worth? According to executives at CBS, the network generates an average of 36 cents in revenue for each viewer who watches its prime-time programming. This money covers the cost to the network for the content, operating costs and profits.

Local broadcasters generate additional revenues from the ads they insert into prime-time programming — an additional revenue stream that the networks would love to control. Current ownership caps prevent the networks from owning sufficient stations to reach more than 40 percent of U.S. homes. It now appears that the networks have found the bypass technology they have been looking for and, with it, the ability to generate more revenue from the programming they control.

Last month, this column explored the realities of hardware and software video encoding and how operational constraints can influence the delivered quality of broadcast content. That column also noted the ability for consumers to download video content and consume it via portable media players (PVRs).

Even before that column was published, Apple announced that it had sold 1 million music videos and TV shows from ABC during the first 20 days after the service was launched. The networks took notice.

CBS and NBC both jumped on the download bandwagon, announcing deals, with Comcast and DirecTV respectively, to make some of their prime-time shows available for viewing via VOD services immediately after the networks broadcast the shows.

Comcast will allow viewers to watch CBS shows (with commercials that cannot be skipped) for 99 cents. DirecTV will push some NBC shows to DirecTV receivers with integrated PVRs so they can be viewed without commercials for 99 cents.

The networks all contend that these services will expand the audience for their shows rather than cannibalizing the over-the-air audience. But network affiliates are not convinced. Many see this as another way for the networks to generate new revenue streams at the affiliates' expense. And some are asking why they cannot offer download services too.

The real-time dilemma

The problems DTV broadcasters face in generating new revenue streams (such as bandwidth constraints) are not difficult to understand, but they may be difficult to overcome.

Multicasts come with their own set of problems. The 19.3Mb/s payload that can be delivered via an ATSC channel is barely adequate to deliver one high-quality HDTV program using MPEG-2 compression. As more prime-time programming is being produced and delivered using HD formats, less room is left for multicasts.

In Las Vegas — one of the markets where USDTV offers a 20-channel broadcast subscription service — some viewers are complaining about the quality of the free-to-air DTV broadcasts. If a station gives 6Mb/s to USDTV, it only has 13.3Mb/s left for itself.

Likewise, if stations get into the multicast business, something has to give. And that something is likely to be the quality of the programs that are delivered.

Broadcast competitors are beginning their own transition to such next-generation video compression technologies as H.264, which can deliver the same level of quality with about half the bits now required.

In November, DirecTV began delivering local HD broadcasts in four markets, with plans to serve 36 markets by early 2006. New set-top boxes that can decode both MPEG-2 and H.264 are required to receive these channels. And, as was the case with first-generation set-top boxes, the cost is being subsidized by subscriber fees.

Broadcasters currently have no way to collect subscriber fees, and their platform is constrained to the use of MPEG-2 compression. Only a small percentage of U.S. homes have ATSC-capable receivers. But Congress and the NAB are hoping to deploy millions of receivers over the next few years in anticipation of the 2009 shutdown of NTSC services.

These boxes will be bare bones, converting HD broadcast to NTSC. They will not support H.264. They will not have integrated PVRs. In short, the ability of broadcasters to compete will be constrained.

There is an alternative, though it is not clear that the broadcast networks will let their affiliates compete in the paid video download business. There is a huge amount of capacity during the overnight hours to download content to PVR-equipped receivers.

Collecting the fees for this content is a bit more complex, but can be managed with a number of technologies, including networked transactions and smart cards — the same smart cards used in DBS system (and now in cable-ready DTV receivers).

Unfortunately, it now appears that the networks are robbing Peter (their affiliates) to pay Paul (themselves).

Craig Birkmaier is a technology consultant at Pcube Labs, and he hosts and moderates the OpenDTV forum.