Cardboard's motion-to-photon latency is significantly worse than the GearVR. This Swift hack is worse than Cardboard. This is true for a number of both software and hardware reasons.
Of course, comparing Cardboard to the Vive and Rift is hardly to apples-to-apples.
Your concerns re: poisoning the well are valid. This seems to be a common divide among people working in VR: should we be making low quality (whatever that means) experiences? For example, VR devs often remark that 360 video content isn't "true VR." I don't know that I agree.
Purely as a matter of opinion, I don't see a reason not to experiment in VR, even if it's far from perfect. It's-not-VR complaints strike me a little bit like the "no true Scotsman" logical fallacy.
It's great that Oculus and Google have different approaches to the problem, and, as I'm sure you can guess, I don't believe there's a real risk of poisoning the well.
Then again, like Dennis Miller used to say: that's just my opinion, I could be wrong.
(In which case we'll all have to wait another 15+ years for another turn on the carousel.) :)