I disagree with the rusting. I've seen _many_ ferrous parts that have become a single homogenous part due to rust. I've seen the same with un-coated aluminium parts immersed in water for long periods.
I don't want to repeat that week-long argument. Suffice it to say that I strongly disagree. Iron oxide is weak and of low density. A single homogeneous part would suggest a strong bond.
Having two parts with a constrained fit can end up as two parts with a very constrained fit when the space between them is filled with that weak, low density oxide. The adjacent surfaces are acually trying to get further apart from each other, not closer together. But they are still two completely separate parts, even if you cannot remove one from the other.
Like I said originally, my point is mosty semantics.
The swingarm has been done several times but modern bikes are designed to use the squat delivered by the current system so it's actually beneficial.
There really is no reason to pivot the swingarm on the countershaft axis.
This has me intrigued. In your opinion, what benefit is there to having variable chain tension? If by squat you mean the compression of the spring (and tire) during accelaration, I would expect that to happen regardless of the details of the drive. Is it not caused by weight transfer?
Or have I misunderstood your point?
Oh, and sorry about the threadjacking.