A Frictionless Response To Piracy?
11th February 2012Ugh. It seems like a bad time to be a writer, just as digital piracy is expanding now that people are starting to get into ereaders in a big way. The rot is spreading from film and music to the written word. It's bad enough that a typical writer makes about a quarter the wage of a paperboy.
That's just about bearable if authors can console themselves that it's all about the art. But as markets shrink so does the willingness of publishers to invest in anything that's not a safe bet. Observe what's happened in music and film: endless X-factor popularity contests and re-hashes of movie franchises that were successful 30 years ago. If piracy goes unchecked we can expect more Celebrity Get Me Out Of This Sexy Jungle Dance Competition memoirs, and endless TwiPotter re-imaginings with the details changed. Edgy fiction? Forget about it.
I don't have the answer, because I don't think there's one solution that will magically make book piracy vanish. It's a complex problem that demands a multi-layered response. But wait, wait, don't click away just yet! You don't solve burglary by fitting good locks, you need police and CCTV and a dog, and those laser thingies if you need to protect expensive jewels from Tom Cruise.
If I'm going to extend that burglary analogy, this is the CCTV solution. And best of all, it doesn't inconvenience honest readers one little bit.
For your eyes onlyThe basic idea is simple: give everyone who buys a book a completely unique version. This is much easier to do with text than it is with film or audio files, where you'd end up degrading the experience. The crude way to do this would be by tagging a serial number on to each file, but it's a trivial matter for anyone to find that and strip it out. However it's much less trivial if the number is encoded within the text so it looks like part of the book.
How might this workSay you want an 8 digit serial number for each copy. That's a good number, because unless you're selling the next Bible 100 000 000 numbers will have you covered for digital sales. Each digit of that serial number stands for a word, a phrase, a font, or any other element you might want to change, so in each case you need to come up with ten alternatives. These variable elements are used to generate a unique ebook, so that should the book be illegally shared the publishers can examine the file and compare it with the secret master version to find out the encoded serial number.
Obviously it's no good changing anything important. So this method is particularly good for the kind of non-fiction that contains bullet-point lists, where the order of these lists doesn't matter. 1000 Ways To Shine A Sheep, that sort of thing. All you need to do is find eight lists to shuffle and you're all set.
For fiction this is harder, but not impossible. You may still find lists you can shuffle. If Dave, Bob, Jane and Linda enter a room, you could have Bob, Jane, Linda and Dave instead, and so on, and it wouldn't affect the meaning of the story. Over the course of a 40 000+ word novel there are always trivial details somewhere, pieces of scenery effectively. "They overtook a green Audi." In that sentence the vehicle could become
I've done a similar thing with my short story, Pulpulous. There are 14 580 000 000 000 unique versions, but each one is basically the same story. In that case I set out to make the differences between them as noticeable as possible, and they're squeezed into a 1600-word tale, but what struck me when I was building it was just how hard it was to make these differences stand out. Spread over the length of a novel few people will notice 8 trivial changes, particularly if some of them are to do with layout rather than meaning.
Will this prevent piracy?Somewhat. Put yourself in the shoes of a potential pirate with one of these completely unique ebooks. It's possible to compare it with other copies of the work and figure out what's changed between each version, just as it's possible to strip DRM from a file. No security system is absolute. But it's very hard to tell which words are just words and which words or phrases stand for part of the serial number. In any random set of numbers there's roughly a one in ten chance of the same number coming up for each digit, so our pirate needs more than two of these ebooks for comparison. So if he uploads his copy to a file-sharing site he risks identifying himself unless he can strip out most of the variable words. It's not impossible, but at this point uploading starts to look less like a fun thing to do and more like a job.
And what of actually catching someone at it? Pirates could easily alter their uploads enough to make them look like someone else's copy, of course. But if they're careless enough times with enough different ebooks it could be traced back to them, and that's information that lawyers can use. One ebook upload could be an innocent caught in the crossfire, but several establishes a pattern that's hard to deny.
Technical difficultiesI'm aware that what I'm suggesting sounds like hard work. It's very easy to create tons of variations on a text file, but setting up the systems to automatically generate these files and record the unique serial number for each customer and title is more of a challenge. Someone has to set up a system to scan any pirated files to extract a serial number. And then there's the small matter of having authors find an extra 72 words where 8 would have done, or finding other bits of the book that can be safely varied, and getting used to the idea of not having one definitive version of any book.
Would such an effort be worth it to save modern literature? Would it even work? I'm just throwing this idea out there because I want to know what other people think. Discuss it with me on Mastodon.