This week, reports have been circulating about a new form of eBook digital rights management (DRM) named SiDiM.
Developed by researchers in Germany, the idea is that SiDiM (German Language), unlike traditional DRM that prevents or limits copying, will instead merely track who leaked the copies that are available for sharing.
SiDiM does this by altering the text of the eBook in such a way that every copy of the book is unique. It does this by changing out words, altering punctuation and making other small changes to the book that, while detectible by an algorithm, can’t be easily deciphered by a human reader.
To be clear, the idea of watermarking eBooks is far from new. When J K Rowling launched Pottermore, she famously released the Harry Potter books without any copy protection, but with watermarking embedded in them. However, that watermarking didn’t impact the text and, as a result, were defeated pretty quickly.
By making the DRM part of the text of the book, it becomes much more difficult to defeat but that extra difficulty comes at a much higher cost. In an art form where every word, paragraph break and punctuation mark is scrutinized and carefully edited, it seems unlikely many will be happy at the prospect of an algorithm altering the work in unpredictable ways to make it unique.
In short, it’s an idea that works in theory, but likely not in practicality.
The Big Idea Behind SiDiM
Most eBook DRM currently available is platform-dependent. Books purchased on iBooks or a Kindle is locked to the platform and, more specifically, to the account. Though there are tools to break these protections and convert protected files into ones that can be copied freely, the system so far has made publishers and device manufacturers happy.
Consumers, however, understandably want more flexibility though publishers aren’t ready to completely give up on DRM, though some are dabbling with limiting its use. One such idea, as with Pottermore, is to simply make each copy of an book unique so that, if it is leaked online, it can be easily determined who bought the copy that got posted.
However, as Pottermore showed, most metadata can be easily stripped from eBooks, making them untraceable and that is what SiDiM seeks to fix. By making the watermark actually part of the content of the eBook, it’s much harder to remove without destroying the book.
However, those watermarks do not visibly alter the work. The alterations they make to the image are invisible to the naked eye. The changes SiDiM makes, on the other hand, are noticeable and, even if they only change a few words in a lengthy novel, it can still impact the original work.
Authors and publishers are not going to be keen on turning over their crafted works to an algorithm for alteration. However, that may not be the biggest problem with SiDiM, and similar systems, there’s a much bigger practical problem facing them if they’re going to be widely used.
Protection That Doesn’t Protect
The biggest problem I see with SiDiM is what happens if it works.
Suppose, for a moment, that a SiDiM-protected eBook is leaked onto file sharing networks and its owners now know instantly who purchased that copy.
Do you sue that individual? They might not have been the source of the leak. The purchase might have been a gift, their laptop might have been stolen or a million other things might have happened between purchase and leak.
But even if they are the source of the leak, it’s unlikely you’d be able to obtain civil damages anywhere near an appropriate amount from an individual.
Simply put, suing individuals is a money-losing proposition, as the RIAA showed, and that is why, outside of the copyright “troll” industry (which deals more with threats and settlements than lawsuits), such lawsuits are rare.
The other possibility is to pursue them criminally but, with the way the criminal copyright statute is written in the United States, that could be very difficult unless it can be proved they did it for financial gain.
In short, just because you have the name and address of the person who bought the copy that leaked doesn’t mean you can take satisfactory action against them.
This leaves SiDiM a copy protection tool that really doesn’t offer much protection at all.
Currently, SiDiM does not have any major publishers on board and only has two German partners signed up (it’s possible the law in Germany may make it more useful there though I do not know).
For the most part, the technology is more theoretical than it is practical. It’s an interesting idea but it’s going to be a tough sell for publishers, who are largely happy with the status quo and aren’t going to be keen to manipulate their carefully-crafted works, even in small ways, for very little additional protection.
So while the idea is interesting, its usefulness is limited.
The best thing that may come from SiDiM is the budding conversation about eBook piracy and how to best reduce it. While the conversation is important, I just don’t find many are saying SiDiM is the answer.