I think Milton Mueller and Hans Klein's paper "What to do about ICANN" is really well-informed and makes balanced suggestions which deserve serious attention.
I guess the question is - how to achieve their adoption.
Again and again, I think the paper homes in on central issues.
The irony is that although ICANN has resisted what the paper refers to as 'democratisation' (indeed, it reversed the process in 2002), if adopted, this could significantly strengthen ICANN's own credibility. It is extraordinary that ICANN does not move in these directions at a time when its own legitimacy (as a quango of the Department of Commerce of a single nation) is often called into question. Building legitimacy based on global and societal representation, rather than intra-national and governmental controls, could create a model which protects the Internet / DNS functions from too much political interference and allows for genuine self-determination by multiple stakeholders/constituencies.
I also like the model of a competitive yet shared role for ICANN and, say, ITU. After all, ICANN itself was launched in part with a view to promoting competition. Why should not ICANN itself not be part of a wider competitive market with more choice.
I accept that it would still be possible for ICANN to assert a role by itself which could have sufficient credibility if - and only if - it recognised that its credibility needs to be based on accountability and global legitimacy. In short, it needs to put registrants and internet users at the heart of its processes, and should consider the restoration of global At Large elections to a significant number of positions on the ICANN Board.
Otherwise, pressure for alternative structures will mount.
I commend "What to do about ICANN" for further study and consideration.
I hope this gets plenty of attention from all parties.
With kind regards,