Through protecting “innocent infringers”, the Digital Millenium Copyright Act has done more to encourage Internet piracy than the invention of the file server. It’s no surprise to anybody that recently revealed court filings show YouTube’s founders were well aware of the massive infringements on their website, and in the case of founder Jawed Karim, actually uploading pirated videos. Perhaps, as many have argued, today’s Internet piracy is a disease of immaturity and poverty. But I think it’s more likely a symptom of the DMCA which puts the burden of proof on the victim, over and over and over again, with no penalty or shame falling on the serial “innocent” infringer. A law that protects reckless and illegal behavior by businesses is a fair definition of a moral hazard.
If Google didn’t index these piracy sites, the DMCA ensures that the other search engines would, and that many people would defect to them. The same holds true for the “legitimate” file sharing sites like ScribD and YouTube whose early growth was supported through large-scale copyright infringements. As long as enabling copyright infringement on a massive scale is viewed as an innocent activity, any Internet based corporation foregoing the opportunity for free content is ignoring its fiduciary obligation to its shareholders. Setting up a website for pirated eBooks (while protesting one’s innocent intentions with a DMCA link) is a great growth industry. The graph below shows a fairly new download site I never heard of before last week. It will soon be one of the top 500 websites in the world, according to Alexa. A good chunk of their traffic comes from search, some of which is supported by illegal copies of my eBooks.
The damage done DMCA in encouraging the widespread dissemination of illegal copies of digital works goes beyond the erosion of rights-holders earnings and the moral hazard for infringers. Innocent people using Google to search for books can be steered to pirate sites without understanding that they are infringing on copyright. That’s why I still go to the trouble of filing DMCA takedown notices with “legitimate” sites like ScribD, where a normal person could believe they are downloading an eBook that the author and publisher decided to offer for free. The version of my laptop workbook that I requested ScribD remove last week had been edited to remove my name, which had been replaced with:
and the entire copyright and license agreement page was replaced with:
Another problem with the moral neutrality of search engines is the indexing of file sharing sites which are likely to infect the user’s computer. You might think I would find it amusing for somebody trying to steal my computer troubleshooting books to end up buried in malware, but I don’t like seeing my work used as a stalking horse. I have too much respect for the technical talent at Google to believe that they couldn’t filter out these results in real time with their own malware site detection algorithm, which is why I attribute those results to the moral hazard of DMCA and the abdication of responsibility to do right. The results with the double exclamation points were judged potentially dangerous by AVG. Sometimes, not being evil just isn’t enough:
Any author or publisher of eBooks with an audience can do a search to see how many sites are using the lure of pirated versions of the eBooks to attract traffic. Just Google the title of an eBook in quotes, followed by the word “download” and tell Google to limit the results to the last 24 hours through the "options" link. During the occasional waves of piracy that sweep over my eBooks, the new notifications for potential downloads reach into the hundreds per day. But it’s a little trickier to find out how many people are intentionally (traditional spelling) searching for pirated eBooks. One rough measure is simply to start typing the title of the eBook and see what suggestions Google’s auto-complete function comes up with, as in:
Unless you’ve chosen to release your eBook for free, seeing the title paired with “torrent” or “rapidshare” is a bad thing. But to get some hard numbers, though dated from some previous month, I suggest the Google’s Adwords KeyWord tool. The results below are from the month of February, before the wave of new infringements that caused the number of webpages referring to a downloads of the laptop eBook to jump. That search gives an idea of the number of people who know exactly what they are looking for, vs the ones who might have heard about the book somewhere and want to learn more about it or even to make a legitimate purchase:
Ironically, freedom loving piracy sites never make the PDF versions of eBooks directly available to search engines for indexing. The reason? If search engines sent seekers directly to the downloadable eBook, the pirate site would lose the opportunity to show advertising and end up paying for bandwidth with no financial return. Pirate sites can’t embed advertising in the stolen eBooks without taking the legal risk that the DMCA would no longer protect their infringements. I say “risk” because it’s possible that by automating the addition of advertisements to all files uploaded, a sharing site could to maintain a gauziest veil of legal innocence under DMCA.
In the short term, I look at the piracy issue from the standpoint that the Internet giveth and the Internet taketh away. Whether or not I’ll continue bringing out new editions of the books that are popular piracy targets in the future depends in part on whether sales of the paperbacks and the legal eBooks justify the investment. If piracy impacts sales to the point that new editions aren’t logical, I’m not publishing for the sake of advertising supported piracy. And while I don’t laugh at people getting viruses in the process of ripping me off, I might find it amusing to Google around a bit in a couple years and see how many piracy forum wish lists I can find where people have requested the new edition of Rosenthal’s book. They may have a long wait.