Every problem a child has outside of school, shows up in the classroom as a barrier to learning. If you want to revolutionize education, understanding that is a very good starting point.
I think that things like living arrangements, mental health and necessities (food,etc) must be provided in a way that allows focus on learning. The larger problem seems to be a cultural one, which isn't an easy fix, but the aforementioned items are a decent start.
From my thoughts, I believe that the approach necessitates starting at a University-type level, with the self-directed learners. From the self-directed learners, we can then extrapolate flexible structure for those that want/need it and expand further to the concept of "classes" or "programs" down the line. I am interested in the fact that graduate level programs in the current model of education tend to go from the rigorous structure of K-12 to very unstructured, research driven.
My personal experience doesn't include extreme poverty, but I will make sure I consider the impact of deficits on the educational process.
I think like a lot of other HN users, you think you have disclosed your email address because you've filled out the field that lets the Hacker News curation team know how to reach you (for example, to get a new password for your username if you forget your old password). I actually can't see an email address when I view your profile. You should be able to discover how to reach me by reading my profile, and I'm happy to discuss your ideas.
Yes
"Figure out a way to 1) reliably detect the optimum education environment for each student, and 2) give it to them."
tokenadult, If you wouldn't mind discussing some ideas myself and a friend have, my email is in my profile. We're in the fairly early stages, but the above seems to be the pervading idea and feedback would be great.