It will never be 100% solved due to CAP theorem. There will always be a window of time where it could be applied, even if its short.
With our consensus and ledger, a double spend is very difficult to do due to the logistics of catching the available window, and the costs associated with being in a position to actually influence one. But, its still possible, just highly unlikely.
I'm not an expert by the language of John's Whitepaper he seems to think he has solved it.....not just made the likelihood of it happening negligible.
Are you saying John is wrong or that he never really technically said its '100% solved'
I'm saying I'm too tired of debating this and the next step for having it reviewed is to put it out in something more widely read. 30 seconds with Google suggests something like
http://ieeexplore.ieee.org/xpl/RecentIssue.jsp?reload=true&punumber=71 but there may well be better suited journals.
That said, fairly certain a major netsplit (i.e.
https://en.wikipedia.org/wiki/2008_submarine_cable_disruption) would render parts of the network unreachable from other parts, creating a fork there's no way of reconciling until the network is repaired. I didn't read it as he was claiming that's solved, but it's still a double-spend risk.