Solving Piracy
The digital watermarking concept in my previous post sparked an idea. I’ve written before that music retail (and eventually video and book retail) as we know it is in its death throes. The oh-so-twentieth-century “grocery store” model of bins of physical media (CDs, DVDs, books) won’t last. Music retail is not not doomed because of file-sharing, which I think will turn out to be just a blip in the history of how we deliver and consume content, but because it’s just bad business: why pay the enormous overhead of manufacturing, shipping, warehousing inventory, rent and utilities, employee salaries, etc. to sell and distribute content that, thanks to the digital age, can be reproduced and distributed for a miniscule fraction of those costs?
The primary reason we continue to do so that the mega-corporations that control the content also own, control or are invested in a many of the manufacturing and distribution channels. For the last several hundred years, content has been a manufacturing industry, with big factories that churn out physical media that just happen to have this abstract stuff called content cut, stuck, or burned on them. Likewise, some of the distribution channels, e.g. Wal-Mart, are as powerful (if not more so) in our culture than the media conglomerates.
Eventually greed eliminates even a powerful status quo, though. The media industry will find a way to divest themselves of the physical distribution infrastructure, because, as the software industry alread realized, you can make better margins selling just the bits. Which brings me back to the digital watermarking.
The challenge of piracy to the content distributor is that, currently, content is pretty much untrackable. I can rip a CD, share it, and once it’s shared it’s nearly impossible to track that particular file back to me. Big value (e.g. free content), little risk (e.g. hard to track me down), leads to lots of file sharing.
However, suppose you go to an online store, enter your account information, and every song you download is stamped with a digital watermark identifying you as the purchaser? What if you didn’t purchase a CD that was a duplicate of 5 million other CD’s, but instead went to a retail kiosk, entered your account information, and had a CD custom-burned for you (or even more likely the files transferred wirelessly to your MP3 player/personal content storage device), again with a digital watermark identifying you as the purchaser? Suddenly it becomes much simpler for the copyright infringement to be tracked back to you, thus providing a disincentive to illegally share the file, but — and this is the important part — this custom digital watermark does so without squashing fair use.
That model requires some pretty drastic changes, not the least of which is the eradication of the retail content industry (record stores, video stores, and eventually book stores), but also probably some advances in the speed of burning/transferring content (I need to be able to get 50 songs nearly as quickly as a clerk can ring up 5 CDs), advances in digital watermarking, and a significant mind-shift among consumers. Also, it only works moving forward — back catalogs are already lost to file sharing because they were never watermarked (but, for the most part, it’s the new content where the money is made).
If, for the purposes of this thought experiment, I can fiat all those assumptions into existence, I think the custom digital watermarking idea might work.
I can’t be the first person to have thought of this, can I?
> I can’t be the first person to have thought of this, can I?
No, you aren’t. There have been quite a few atttempts in the watermarking direction. See http://www.digimarc.com/ for one that is still alive. Doing it in music is non-trivial, but feasible. Some of the problems include maintaining the watermark as the digitized music is compressed (MP3) and not compromising quality.
That said, this has been a tough business. Many watermarking schemes have been coupled with attempt at DRM enforcement, which the consumer market rebuffs. If there’s no active (executable) component to the scheme, then usability and acceptance goes up, but then there’s not much reason for music producers to adopt one particular scheme. I also wonder (I don’t know) re the legal standard of proof one would require to actually pursue an apparent ‘leaker’ of a song, rather than someone actively sharing the file.
Custom digital watermarking is in use now, sort of. The AAC (what Apple is using, an open standard that’s the successor to MP3 audio, but with the capability for DRM) and WMP (Windows Media Player) formats both have degrees of watermarking. Apple is leading this crusade in a way, having put forth the first usable product for downloading files ($1 a song, no subscription charge, and reasonable rights to use the music). There’s now four players in the market (including Apple, who should be announcing the Windows version of the service today).