Archive for the ‘watermark’ Category

Digital Watermarks: Can they save copyright?

Saturday, August 8th, 2009

Everyone knows that illegal downloads can’t be stopped (except, maybe the MPAA, but they’ve been deluded for a while about speech issues). That could seem kind of depressing if you are a filmmaker who’s just maxxed out a few credit cards and hit up every friend you have making a movie with no obvious hope of recouping. Some people have said that we should just dump copyright altogether since it’s unenforceable. But copyright was created to protect artists who put their original ideas and execution into a work so that they could control how money is made from the work. This doesn’t seem like an idea that should be abandoned just because technology has changed.

One potential solution might be digital watermarking. Instead of DRM, which tried to determine where the content can be used, watermarking just tries to track where the content is going and what is its source. The watermark, which is invisible, can be embedded with other useful information like film title, cast list, synopsis, etc. so that it is desirable to retain for those exchanging the file. It’s a way for the filmmaker, distributor, or content provider to mark otherwise anonymous files as having a maker. The Digital Watermarking Alliance, an advocacy group for watermarking technology, commissioned a study that claimed that using a digital serial number rather than DRM would cause “Active sharing via file-sharing applications [to] decline by one-half overall, a little less among
BitTorrent users, a little more among P2P network users.”

Watermarks have their problems. They are relatively easy to break or “attack” though some argue that there is less reason to destroy a watermark than DRM since it doesn’t impede the user experience and in fact can enhance it. But once a file with a broken watermark is released to the downloading stream, it will propogate.

The ideal watermark would not ever prevent the viewer from watching the file. If a user attempted to remove the watermark, the file might become degraded. The same thing might happen if the file were altered. Of course, anything you can do to something digitally can be cracked. But if there wasn’t a motive for the average user to get rid of the watermark, it seems like it could gain traction.

In the future, it seems likely that the per-user costs really will be free or subscription-based. But commercial uses, i.e. the content providers, cable companies, Netflix, etc. who are making money from having the content, should still be on the hook. Watermarks might be a way to preserve authorship while keeping the files flowing.

Are the days of artists owning their work over, when work can be infinitely replicated? Should content producers just look for new revenue sources?

Politics II: Watermarkworld

Thursday, June 12th, 2008

Tim Lee has posted his response to Rasmus Fleischer’s proposal to ditch copyright law at Cato Unbound, making the hardly strident but accurate point that copyright law is still functional outside the digital realm.

In that domain, a reader comments on my earlier post:

I think the general consensus among folk who study this stuff is that watermarking — and a variety of schemes have been floated for years now — isn’t really going to be that helpful. The large-scale distribution content firms worry about, as on p2p networks, typically involves skilled geeks who can strip away every form of watermark yet devised with minimal trouble, so watermarking won’t stop untraceable copies from getting into circulation there (all you need is one clean copy). On the other hand, you have small-fry casually sharing a song or DVD with a couple friends, who might leave the watermark in, but are unlikely to get on the copyright enforcement radar screen.

That has probably been true in the past, but as we go forward it’s possible a new model may emerge that will be more conducive to the watermark- that of audience profit participation. In an ad-based profit model, for example, one way of incentivizing tracking and data collection in general may be to let the viewers actually get paid (or rewarded in some way) for watching films.