Quote:
Originally Posted by GreenMonkey
As an I.T. guy, I'm with bigtext on this.
It wouldn't be hard to track at all, since it's apparent that Amazon and Kobo and the like are tracking things like who is highlighting what and how many pages people read.
|
They are tracking highlighting and pages read on
books purchased at their stores. There's no indication that they're tracking those details for sideloaded books, and there's no *purpose* for them to do so--they have no way of collecting useful data about those books unless they've got someone assigned to read & analyze them.
Tracking the metadata of sideloaded books is useless; that info can be changed on the user end. Worse, the books themselves can be changed without touching the metadata--I have no qualms about opening up an ePub in Sigil and adding an extra chapter with my personal notes about the book, or making corrections, or removing broken links. I have an epub on my reader that's collecting a set of blog posts; I update it regularly, but the UUID stays the same.
I grant that not many people do this--but that possibility, combined with the ability to read text files and downloaded PDFs with who-knows-what in the metadata fields, means that tracking sideloaded data involves collecting huge masses of data they can't use--which has to be filtered out before they get to usable data.
"We'll just track data for the books with UUID's that match those sold in our stores" might get useful information, but it's not info they could use to make marketing pitches, because they can't be sure the metadata matches the contents. And "it's probably close enough" is a good way to either kick off a failed marketing campaign and lose a whole department's jobs, or get sued for fraud, if they make the statements in public.