Correct. And note that none of the solutions I've seen published are tolerate to partitioning (a.k.a. network fragmentation). That means Bitcoin is toast if the network fragments, because there can be a spend of every output on every network partition.

Nevertheless, there is way to absolutely solve the issue and make it provable. But it does require an adjustment to the security model assumptions, but I think the change those assumptions is superior to the assumptions we make now with existing "subjectivity" for consensus models.

Any way, I don't want to blabber on, because of course I haven't published anything yet. I just want to note that VanillaCoin has not solved anything w.r.t. to 0-confirmation double-spend attacks and we peers have already explained to them why in other threads. If they believe otherwise and want to rise above the peer review they've already received, then let them publish to a place where academics will take notice and peer review their white paper.