Be careful what you wish for. The retreat of media companies from digital rights management (DRM) schemes, at least in the music business, may be followed by a growing use of digital watermarks that enable companies to track the use of files, reports David Kravets of Wired News. The watermarks, which “are digitally woven into the fabric of a download and do not restrict listeners from making backup copies or sharing music with friends,” enable companies to track the copying, trading, and transmission of files and can be programmed to include “a unique serial number that a music company could match to the original purchaser.”
Kravets reports that Sony and Universal are already including watermarks in their DRM-free files, though the companies say they don’t record information about the purchaser. Warner and EMI are not using watermarks. Microsoft has won a patent on a “stealthy audio watermarking” scheme, named El Dorado, that is, says the patent, “designed to survive all typical kinds of processing, including compression, equalization, D/A and A/D conversion, recording on analog tape and so forth.” Such watermarks could be used by ISPs to automatically block the transmission of files through peer-to-peer networks, achieving one of the aims of DRM by other means.
But this kind of spyware, which could ultimately be encoded into many different kinds of files, could be put to uses beyond clamping down on the sharing of copyrighted music, from collecting evidence for use in litigation and negotiation to providing data for marketing purposes. One expert quoted by Kravets says that the watermarking systems provide “forensic precision” in monitoring the use of particular files.
In the future, your music could be listening to you.
DRM placed real limits on what we could do with our digital purchases, be they music, e-books or movies, in terms of freely moving them between devices.
Digital watermarking does seem to be the lesser evil, and if that’s what the publishers use to track licensed copyrighted material to ensure it’s not being pirated… well, it’s a lot harder to argue against. DRM was a right pain for everyone, the vast majority of which got their digital material honestly; watermarking should only be a problem for those who steal.
As far as viewers of the content “phoning home”, one can hope that open formats and open-source viewers and OSs should (hopefully) ensure that such functionality is optional. Then again, I’m not too personally worried about the “big brother” paranoia; perhaps I should be?
This sounds like a good plan to me. Personally, I have no quarrel with the current Fairplay as used by iTunes. I have never been restricted in any way with it. I think we can take it a step further and use it in other media as well – purchased software, ebooks, video, etc. This way, if someone does post it on the intertubes, there will be a way of tracking the responsible party.
On the other hand, what if something of mine that contains this watermark is stolen? A laptop or iPod stolen could also have its media posted on usenet (a far sweeter honeypot than P2P IMO that the RIAA should focus their efforts on) and I would inadvertently get the blame. What’s one to do? Have everything encoded with my biorhythm? That would work. They are considering that with future cars. It would know my biorhythm and start for me at the push of a button. How could this possibly go wrong? Oh, instead of just a stolen car, now I have to worry about someone stealing me too, just to have my rhythm! ;-)
Watermarks are passive. They don’t send information back to anyone. They aren’t spyware.
They are tool of surveillance. They are problematic much as you suggest. But your description is misleading, and so distracts attention from the actual risks.
-t
Watermarks are passive.
You’re right. Thanks. I’ve edited the post to correct my error.
I think, though, in this context the practice fits under the general term “spyware.”
Sounds good. Thanks.
That “enable to track” runs deeper than intuition and guessing can get to, I’m guessing. So, you got me on “spyware”.
-t
I think it’s a great deal: “it works” is the biggest concern for any IT user. Whether companes will enforce it is more a question of policy—and if it doesn’t make sense now, I doubt it will ever. I’ll need more details on how Microsoft expects watermark that can survive tape-recording without being heard, though.
I don’t see this as such a terrible trade-off… if indeed it’s a trade-off at all. Without getting into all the distinctions between IP and real property, I have to say it doesn’t bother me that most of the stuff I buy has a serial number on it somewhere. ISPs aren’t going to be filtering this stuff without losing their common carrier status, and while it may provide support for individual civil cases, I think it’s already been convincingly demonstrated that litigation isn’t a working mechanism to control media sharing.
If this makes media companies more comfortable with digitizing their IP without heavy and unworkable DRM in it, then bring it on. It’s just another small step for them along the road to a different business model, one where they can view their customers in something other than an adversarial role.
I have to say it doesn’t bother me that most of the stuff I buy has a serial number on it somewhere.
Ok, but that is not what a watermark is.
Suppose that you buy a drill press. It might have an imprint on it somewhere, recording a serial number. If there is a recall on the device, pegged to certain serial numbers, you can read the serial number to see if the recall applies. If you like, you can remove or obscure the number, making it unreadable without damaging the machine. When you plug the machine into the wall socket, the power company does not interrogate your drill press to discover your usage patterns, perhaps selling that information to advertisers of wood-shop equipment.
Let us suppose that you buy a watermarked recording and wish to propogate an anonymous comment — a review perhaps — attaching a fair-use snippet of the recording. What happened to your anonymity?
What if you re-encode the recording you have purchased? Should equipment be permitted to refuse to play it, or to report a possible copyright violation?
ISPs aren’t going to be filtering this stuff without losing their common carrier status
That is naive. Deep packet inspection, rewriting, etc. are all already going on, quite a bit.
If this makes media companies more comfortable with digitizing their IP without heavy and unworkable DRM in it
Because this is workable DRM.
-t
I think that companies do need to have some ways of making sure that their content is not copied to kingdom come.
Thus the watermarking as described above, seems like the lesser of two evils.
It’s just not the same as home taping of audio and video cassettes in the 70s and 80s – as we all know, one digital file copy can easily be seeded and duplicated throughout file-sharing networks in a matter of hours.
Sure people don’t want to be treated like thieves and most people are honest – but even if 99% of people are honest, it only takes 1% of people to seed one protected copies as described above, for people to steal.
So I think that for the near future, there will always be some sort of water-marking. People are right to be concerned over privacy issues, sure – but companies do need to make money on the content that they produce.
If content becomes ‘free’ and is copied freely, then the revenue generated dwindles and thus making a new season of your favourite show becomes less and less attractive if most of it’s target audience are simply copying it for free.
Think of it this way – most people would be pretty horrified if their company suddenly turned around and told them that they had to give up half a day’s financial compensation for their time and do it for free. But they then think nothing of copying DVDs or watching MPEGS downloaded from filesharing networks…
Simon spoke pointedly of a particular concern with particular types of copyright-protected material.
Often the income generated from the creation of the property (especially music and visual media) is divided between remuneration presently paid and that which is deferred. Without control over theft, revenue generated over time can never be, well, generated…
Thank you, Simon…
It is the ‘forensic precision’ and Walt’s reply that interest me here. It is very, very difficult to copy something physical to a point where it is indistinguishable from the original, especially if the manufacturer wants to make it difficult. Normally, to an expert, there are enough tell-tale signs which allow you to say fake or proclaim a counterfeit. For a digital entity this is much harder as there are no tool marks, no style differences. There is nothing tangible to tell the original from the copy except the watermark itself.
With Walts laptop he could report it stolen and thus in large part be able to cover himself, in similar fashion to reporting stolen credit cards. I am more worried about the watermark algorithms being cracked or more likely someone with access to the algorithms/data selling off watermarks at $xxx each. If they pick one of my watermarks there will be nothing I know about it until the nasty letter in the post from the RIAA or whoever is the enforcer of the times. What kind of defence could I offer? I’m not sure what might be acceptable apart from the truth, I didn’t share and have no idea how this happened.
While I’m accepting of watermarks as advisory information, I do not think we are ready yet for them to become forensic, not by quite a way. There will need to be a period where trust in them is built by the public as well as industry.