Will the means of personalizing learning be just?

Let’s consider how things become personalized.  There are right ways and wrong ways, complicated processes and simple processes, human-driven methods and technological methods, superficial strategies and in-depth strategies, just means and unjust means.

Think, for example, about how your experience with amazon.com is “personalized”.  Some people may be annoyed with the way amazon.com “big brothers” you as it tracks, records, interprets and assesses your book-buying habits, whether or not you ask for it, just so it can give you “personalized” recommendations.  I may not care if Amazon knows what I read, but I may care about how I am being classified in the process of returning personalized recommendations.

How can we endeavor to “personalize” learning efficiently and effectively and ensure we are not missing the mark, “big brothering” students, or creating crudely oversimplified categories from which to disseminate personalized learning streams (i.e. all category B students get personalized learning packet #7.)?

What are the ethical considerations of the processes by which learning will be personalized?