Here's what I don't get about inBloom: Is there supposed to be just one national data store? If so, why?
They've promised to make the software available under an open source license, and so far they're off to a good start in that department.
Ten years ago, this would be A Good Thing, but of limited practical value, because actually running a system of that scope would require a fairly intensive, customized IT infrastructure, perhaps relying on some proprietary components. Today, you can use the cloud and a free software stack as a standard platform, and, according to NYC Public School Parents at least, inBloom is just going to run on Amazon Web Services. If that is true they should be able to make it fairly easy to spin up your own instance of the inBloom data store which is indistinguishable in performance from any other inBloom.
It seems to me that running one big national inBloom store -- which seems to be the idea but they're not very explicit -- serves only to make the commercial data mining somewhat more efficient, at the cost of putting a lot of power in a few private hands. If large districts and states are running their own inBloom instances, the whole thing is a lot less scary, and it would work just as well from the user's point of view, with a lot more accountability.