I’m looking forward to “the one that makes it”.
Like most emerging technologies, it’s rarely the Best of Class that makes it but one of the more easily reproducible “good enough” models.
So I encourage experimentation in this area greatly.
I don’t believe the ‘best’ will win; but an Industrialist will pick a _practical_ one that’s cheap/fast/easy to produce and run with it.
And THAT’S when we’ll see viable change.
[in short, always remember the important role of industry; they bring this lofty stuff down to our level where we can use it]