That yes/no image is incorrect.
If core is the longest chain and someone is stupid enough to mine an XT block that core doesn't allow, then all XT nodes will follow the XT block fork.
Thus if core continues to be the longest chain (which it will) then XT clients will not follow the longest chain possibly until later.
That would only 'be useful' if people were ready to double-spend the shit out of inputs to the doomed block.
Someone who cares enough might set up a system which keeps an eye out for XT fork double-spend opportunities and gives those who can capitalize a timely informational saying that their opportunity had arrived.
Previous similar situations (like the BDB fork) resulted in only a few screw-jobs which people were willing to absorb. If there were many more of these and they were undertaken deliberately, it might creates such a cluster-fuck that Hearn would feel justified in just forking XT (with his 'checkpointing' if necessary) rather than to try to clean up the mess.
To understand why I would welcome such a thing, one needs to understand that I see a fair amount of utility in having an XT fork which siphoned off a sizable number of 'bitcoiners' who are deeply ignorant about the principles of the system and are mostly MultiBitch-class users who add no value. They'd stop even being users if their transactions were not deeply subsidized anyway, and the only way that is sustainable is if large corporates took over infrastructure operation. I say let Hearn have them...their loss would only strengthen Bitcoin.
