[Elfwreck]

Originally Posted by bigtext

But Amazon could be sued by a publisher or some industry representative group like the RIAA is for music to get access to the information.

It's entirely uncertain how much access Amazon has to sideloaded & personal document reading info. If it wasn't purchased through Amazon, they may not have a way to track it--and more, they have much less right to collect or sell that information; the Kindle TOS gives them rights to certain data related to digital purchases, but its rights to your other data are a lot blurrier.

Now let's say you bought all the books on the New York Times bestseller for over a year. What is the probability that you got access to a DRM free copy legitimately for ALL those titles?

Much of that depends on whether it's illegal to strip DRM for personal use, an issue that has yet to hit the courts. If that's legal, they can easily all be legit.

And the issue of "what's the likelihood that a person got permission for 85 bestsellers to remove DRM" is irrelevant in court... the issue will be "Did, or did not, this particular person have permission... and did you know that before you accessed their data?"

In other words the more DRM free books you have in your collection the more it looks like you either pirated your collection or removed the DRM yourself.

Heh. All books in my collections are DRM-free. (And wifi free, so it's a fairly moot point for me, but for those who deal with cloud-storage reading, it's a topic worth some consideration.)

In a certain sense this is all hypothetical to illustrate the problem that your decision to strip DRM may not be as private as you think it is.

I suspect that checking the details of non-DRM'd/sideloaded content, even if it's allowed by the TOS, is far too much of a bandwidth drain. I don't mean "of the wireless service," but of the company itself... to find out if I'm reading The Game Of Thrones by George RR Martin, acquired by stripping the DRM from a purchased copy, might be minor--the UUID could be used, perhaps. (Not sure if Calibre assigns a new UUID or not.) However, finding out if Song_Ice_Fire.PDF is "Martin's book, downloaded from a torrent site" or "the quickstart rules for Green Ronin's RPG, renamed" or "the map from tor.com" or "a blog article about the books" or a fanfic or some other content with a filename that sounds somewhat *like* Martin's book title, is hellishly complicated.

Amazon (nor anyone else) doesn't have the time to put into sorting out what personal content has similar filenames to works in copyright that are currently only distributed with DRM. I don't think any retailer is going to start tracking use of personal documents, just because there's so *much*, and the potential return to them is so little, even if they do have the legal right to read those documents, which I'm not at all sure they do.

It's likely that anything sideloaded is invisible to them (after all, you didn't give them permission to track use of that content at all), and personal documents sent through their convert-and-email aren't likely to be tracked. They don't want to be bombarded with requests to sync random user content, especially when that could mean viruses etc getting into their system. They *really* don't want to be asked about lending ebooks not bought from them.