Very interesting idea about the original Facebook roll-out method. I wonder how we could use that sort of approach with Steem...?

This is not my forté, but for whatever it's worth, here's some brainstorming:

• Start with colleges, not high schools, just to minimize potential legal/regulatory/ethical issues with recruiting students under the age of 18.
• Pick a few starting colleges
• Set up a "fast pass" to the account creation process so students from those schools can get quick/immediate account creation.
• Set up and manage Steem communities for the schools.
• Identify on-campus "influencer" students from other social media platforms or other research (maybe focus on comp. sci., journalism, communications, and other "friendly" majors?)
• Hire some of them into 1-semester part time jobs to bootstrap the schools' communities with photography, videos, or articles about campus life; to share those articles with followers on other platforms; and to recruit their peers into the Steem community.
• Provide them with guidelines about privacy, copyright, key security, accessing exchanges, etc.
• Advertise on campus, near campus, in the school newspaper, give away T-shirts & gear to students, etc...
• Make sure the communities are well-curated & moderated until they get off the ground. Keep rewards at realistic levels to manage expectations, but don't let them be ignored.
• Partner with Appics, Actifit, and/or others on the initiative?
• At the end of the semester, transfer community ownership to... someone... not sure who... faculty? admin? student? alumni assoc? endowment? Might vary by school...
• New semester, new schools. Experiment with different sizes and types of schools. Revise the program and scale up for more schools per semester as time passes.
• Hopefully, at some point, the network effect will take over and new schools will start coming online without active recruitment efforts. Not sure why, but for some reason the number "50" sticks in my head for the number of schools when Facebook hit some sort of tipping point. Maybe achievable in 3-4 years?

Don't know how useful any of this is, but that's what came to mind for me.