its primary purpose is to lay the foundation for “modular hivemind” (an application framework for 2nd layer applications)

That is a killer - 2nd layer has lot of potential - literally app can store anything in JSON and process as needed - once it's ready, we should plan for publishing some ready to run sample codes in various languages, that can help developers to set up a second layer app in few hours. I hope the data that's going to be stored in second layer, is not going to hurt the performance of the chain, right ? Also is there a limitation on the size of data that can be stored in second layer ?