So you think a developer of a small-scale game should spend their limited resources porting a massive game engine to an unsupported platform rather than creating a game? UE4 is huge, and porting it over would require a significant time investment by a team of competent programmers.
And even assuming they successfully port it, they'll then by limited by the target hardware.
What experience or knowledge are you basing this assessment on? I have developed (small-scale, but complete and working) programs in both UDK and UE4, and if you want to make full use of UE4 (and I don't mean this just in the "AAAA graphics" way, but also in the ways in which it improves productivity) it's simply not that easy. And if you don't use the full capabilities in order to ease a downport then well, that proves my point doesn't it?
I wouldn't be surprised if some tooling exists to ease UE3 -> UE4 transitions. I'd be quite surprised by the other way around.
No. I said I would reduce my pledge, because the port makes the project less focused. A reduction is not a "pull", and a loss of focus is not "destroyed integrity".
I still haven't seen an argument that additional platforms don't reduce the focus of development. The best case scenario is an after-the-fact port by an external studio, but even that could influence design decisions (like e.g. the skills in Diablo 3 were designed not to require accurate positional play out of consideration for a later console port).