[News]
Digimarc suggests: mandate will expand (just add watermarking)

The Digimarc Corporation, a watermark vendor, has suggested that the scope of a digital television mandate is likely to expand in the future.

In a message from Digimarc VP Reed Stager, the company suggested that the broadcast flag by itself is probably technically insufficient to prevent unauthorized redistribution of digital TV broadcasts. Although there are several reasons for this conclusion, Stager presents only one: the "analog hole", or the fact that an analog video signal can easily be digitized.

Stager notes that the original proposals on which the BPDG's work is based considered both a "broadcast flag" and a "broadcast watermark". (Possibly in the interest of speed, most companies involved decided to drop the watermark proposal for the time being. Earlier submissions from watermark vendors showed that they considered this unfortunate, and hoped to reverse that decision in the future.)

Stager argues that the broadcast watermark is important and that, if it's not to be included in the BPDG's standard, a notice ought to be provided that the broadcast watermark may be standardized in the future. He proposes a new section X.12 with a notice to manufacturers that they can expect the mandate to expand:

The requirements anticipate that they may be expanded in the future to include a broadcast watermark to carry the "no redistribution" state, especially in regard to securing analog outputs, and that these requirements and the definition of Marked Content may be modified accordingly.

In general, Digimarc's submission argues that "closing the analog hole" -- identified as one of the MPAA's major legislative wish-list items -- is essential to the success of a digital TV mandate. It's presented, not as a separate issue, but as a key requirement for making what BPDG is doing work. The "broadcast watermark" would likely be implemented by a mandate that analog-to-digital convertors include circuits to detect and respond to it, so that consumer equipment couldn't digitize copyrighted broadcasts or other copyrighted video signals.

Digimarc has an obvious business interest in the adoption of a watermark mandate, but the technical arguments it presents are not without merit. Essentially, watermark vendors argue that the mandate currently being proposed is insufficiently broad and insufficiently intrusive to be genuinely technically effective. Without endorsing the conclusion that the mandate therefore ought to be far broader, we can see merit in this contention.

Indeed, we've made a parallel argument ourselves: This proposal can't even be minimally effective at preventing infringement unless all analog-to-digital conversion can be controlled. Perhaps we shouldn't be giving certain watermark vendors ideas.