It looks like Scribd, one of the biggest document sharing website of the Internet, has upgraded to a new, fantastic, hyper-efficient copyright management system, able to recognised illegally uploaded copies of copyrighted works.
Great, you say? No, because…guess what? This system does not work. I’ve received for the fourth time in less than a month a document removal notification because according to their system I would have uploaded protected material. Have a look by yourselves:
So it turns out that Micromégas by Francois Marie Arouet aka Voltaire is a protected document and that I violated copyright uploading it on Scribd, even though Micromégas has been published in 1752 and of course is not subject to any form of copyright.
The same happened a few days ago with “Persian Letters, Volumes 1 and 2” by Montesquieu, Candide by Voltaire and “The Subjection of women” by John Stuart Mill.
Hope is not necessary to point out that none of these documents is protected by copyright.
This is how copyright enforcement does (not) work, this explains why a copyright-checking software cannot work: because it would be unreasonable to think the opposite.