I think you've got a lot of great answers here. And I'm pretty much in agreement with the majority - but particularly coming from a design research background I

I think you've got a lot of great answers here. And I'm pretty much in agreement with the majority - but particularly coming from a design research background I would say that very often, the best bet for something like this - that definitively can't have a single right answer - is to just prototype and test. Prototypes can be as simple as writing a short piece of what one vision could be and then putting it on a blog to get people's comments. It could be creating a storyboard type of comic painting that alternative future. It could be making memes that aim to get people's responses to the idea. The important part is to just start the conversation and slowly start to uncover all of the messy and complex systemic factors. And then continue to gather the information, divide it into smaller parts, talk to specialists about the parts that are outside your immediate expertise, and definitely do what others suggested and apply cross-domain thinking to it. The solutions to some challenges already exist in some form in other unrelated domains. But as long as you can treat what you create as 'disposable' you won't go wrong with a design(make) first approach.