I understand there is not a lot of documentation on projections (yet), but I want to make sure my understanding of their intended purpose.
Should I expect projections to obsolete the need for my own read data store?
From what I understand, Event Store projections are for temporal queries or joining streams. You will still need a solution for read models.
The problem, as I understand it, is doing [set] queries over projections.
Thanks 
I realized just after I hit send that I was limiting the scope. That’ll teach me to quickly post while waiting for a build to finish.