New DRM 'text watermarks' link pirated ebooks to their original uploader

Jesse

Posts: 358   +42

Through a program called SiDiM, researchers at the Darmstadt Technical University in Germany have developed a new form of digital rights management (DRM) to counter piracy and trace a pirated work back to its original source.

The method relies on what the researchers call “text watermarks” to make each copy of an ebook unique, TorrentFreak reports. The technology makes small adjustments to the book’s text, effectively giving it a digital fingerprint of sorts so that if a copy is uploaded to the internet, the publisher is able to trace it back to the original purchaser.

"Users are encouraged to take responsibility [for] their copy and it will deter illegal file-sharing, as copies can be traced using these marks,” said Dr. Martin Steinebach, a researcher with SiDiM.

The DRM alters word order, inserts synonyms, changes paragraph format, adjusts punctuation, and does so in an unobtrusive way as to not influence the meaning of the text, according to the researchers. TorrentFreak gives an example of the DRM swapping the word “unsympathetic” for the phrase “not sympathetic.”

drm pirating digital rights management ebooks

This technology raises a number of concerns, one being the reliance on an automated system to make these adjustments effectively without being obvious. Additionally, changing prose as described by the researchers could lead to dampening or destroying the nuances of the text, or even breaking common literary techniques like alliteration and pace through punctuation crafted by the author.

Traditional DRM acts as a pseudo-gatekeeper, barring access without the proper key. While its creators are describing this new technology as unobtrusive, it arguably could be among the most obtrusive if it augments the very creative work it was intended to protect.

Permalink to story.

 
"The DRM alters word order, inserts synonyms, changes paragraph format, adjusts punctuation, and does so in an unobtrusive way as to not influence the meaning of the text, according to the researchers." How does one change all these things and do so unobtrusively exactly? Altering punctuation alone will make something unreadable, imagine that combined with altered paragraph format and synonyms for no reason. Might as well just buy the book, it will be easier to read and you don't have to worry about making copies of it, you could just lend it to your friends when your done reading it.

Unless what is trying to be conveyed is all these alteration will appear if someone who isn't the original purchaser tries to read the document after illegally downloading it, that would be more logical.
 
This isn't anymore effective than just putting the persons name and order information obtained from when the person purchased the book into the ebook. If you put someones name and address on every page, they probably won't leak it.
 
Stupid and easily thwarted. The sheer intrusiveness of the code will make it easy to locate and strip out.
 
This isn't anymore effective than just putting the persons name and order information obtained from when the person purchased the book into the ebook. If you put someones name and address on every page, they probably won't leak it.
They could just strip that info out of the ebook? E.g. use an OCR reader to rip page content.
 
This isn't anymore effective than just putting the persons name and order information obtained from when the person purchased the book into the ebook. If you put someones name and address on every page, they probably won't leak it.
They could just strip that info out of the ebook? E.g. use an OCR reader to rip page content.

You can't strip embedded content out of a flattened pdf file. Everything appears as one layer almost like images but the text is still recognizeable.
 
This isn't anymore effective than just putting the persons name and order information obtained from when the person purchased the book into the ebook. If you put someones name and address on every page, they probably won't leak it.
They could just strip that info out of the ebook? E.g. use an OCR reader to rip page content.

You can't strip embedded content out of a flattened pdf file. Everything appears as one layer almost like images but the text is still recognizeable.
That is what an OCR is for :) it takes a picture of a paper and turns it into editable text. One screen shot and an OCR program and you have a copied book not hard at all just takes a bit of time.
 
All these I***t researchers got nothing better to contribute toward making a better world instead of researching on useless DRM for a bunch of greedy corporations.
 
All these I***t researchers got nothing better to contribute toward making a better world instead of researching on useless DRM for a bunch of greedy corporations.

Profit isn't greed. Get off your lazy butt and work for a corporation and you will see that they provide some of the best benefits to employees in the world.
 
This is horrible. What about audiobooks? What the narrator says will be different if you add synonyms or add words like "not". This is a REALLY bad idea and I hope it never happens.

I just got a Tablet for reading ebooks and this is what I have to look forward to? Automatically altered text for the goal of fighting piracy. Such a horrible idea. I foresee this backfiring and causing overall piracy to increase.
 
All these I***t researchers got nothing better to contribute toward making a better world instead of researching on useless DRM for a bunch of greedy corporations.

Profit isn't greed. Get off your lazy butt and work for a corporation and you will see that they provide some of the best benefits to employees in the world.

Benefits like laying people off by the thousands? You're either a puppet or a CEO.
 
Get off your lazy butt and work for a corporation and you will see that they provide some of the best benefits to employees in the world

You meant be a SLAVE for a corporation.
 
This is a good way to counter pirates, not even taking pictures will protect people this time, I know some drm techs are bypassed by taking low res pics of the pages, but not with this tech, I like it :)


and the reason I like it, is because if I write a computer book I dont want 10 actual buyers and 1000 free loaders
 
Okay,
So I buy a few e-books, then somebody steals my Kindle Fire, starts seeding my legally purchased books to the internet until it has been downloaded a million times, then what? They trace the book back to the original purchaser and then what? "Okay, we know whose books these are. That was fun. But who actually committed the crime?". When a thief steals a car and does a hit and run they know who the vehicle belongs to but how do they know who actually committed the crime?

For even more hypothetical fun, lets say I lose my tablet/ebook/kindle and somebody finds it but they wanna snoop all around my tablet contents first. They see some cool books they might wanna read so they copy them to a USB flash drive and THEN return the property to a lost and found box at some college or a Wal-Mart store. I get my device back but the crime has already been committed. Seriously?

My grade for this new DRM technology? F for FAIL.Okay, next DRM idea?

I've got one of the most fool proof ideas on planet earth for preventing piracy. Stop selling your **** like it's solid gold. I can buy a multi-million dollar budgeted video game on Steam for pennies on the ******* dollar but these greedy e-book publishers wanna charge $30 and $40 a pop for books that have been out for a long time? Stop trying to rip people off and you won't have to worry about piracy. I highly recommend purchases from APRESS. They have daily sales at $10.00 a book, which still seems a little high sometimes but at least they are trying. I won't pay over $10.00 for an e-book especially ones that are outdated 6 months after they are released.

How about that indie game offering of 8 games recently where they let the CONSUMER set the price of the purchase? Solution to piracy. Solved. End of story.
 
A group of pirates individually buys 3 copies, uses OCR to get the text, and has a computer run through the text to find differences. Then either the original copy is generated from finding what is the same in more than one copy, or if more copies are needed for that, at least the watermark is destroyed by making one text from picking and choosing between the differences of the 3 to result in a unique 4th text. Perhaps each copy is so distinct that a combination of 3 texts results in a watermark that incriminates all 3 pirates? Or would each copy being so distinct result in it being easier to produce the original with fewer copies? The whole idea seems flawed, but the devil is in the details perhaps.
 
One might also take into consideration that once this will be introduced and well polished the next step must be to burn all physical copies. Or will everybody have to sign one that buys with his ID number and every single book will differ in some way ?
 
I just had an Einstein moment (oh wait it's eureka moment, I'm such a analpfhapet), here it goes:

why won't you lower price to combat piracy ?
 
I just had an Einstein moment (oh wait it's eureka moment, I'm such a analpfhapet), here it goes:

why won't you lower price to combat piracy ?

Dude don't bring your logic into this convo how dare you.

Don't you know how the world works....

they would rather pay researches hundreds of thousands of dollars to make this "hack proof" system just so some 12 year old can crack it in his basement the following year.

Do what you suggest the obvious is to easy.
 
Back