When we were preparing for XT, we also went and talked to the Chinese miners. They told us that the original 20mb limit Gavin proposed was too high, but that they could accept 8mb. So we compromised and went with 8 + a growth function. Then after XT was launched they changed their mind and said any growth after 8 at all was totally unacceptable. Now they're telling the Classic guys that 2 is the most they could handle. Did the Chinese internet border really get 4x worse in the span of 3 months? I doubt it.
Western miners aren't much better. One told me quite clearly they'd start voting for BIP101 back in November (though: voting in such a way that it wouldn't actually activate!). But they didn't. When I followed up, they again said it was on their todo list and they'd start really soon. But they didn't.
The miners have proven over and over again that what they say they will accept and what they actually do accept is not aligned. So right now I'm seeing some excitement (maybe more like desperate hope) that Bitcoin Classic will solve anything. Maybe now the "Scaling Bitcoin" conferences have come and gone and Core's reputation is much worse, they'll have better luck, but even then the best case scenario is that Bitcoin gets a 2mb limit. That isn't nearly enough and big backlogs will still occur.
More to the point, even in the best case scenario, the community will essentially accept that Bitcoin is controlled by the Chinese government and grows or shrinks at their whim.
I supported BIP101, but your unwillingness to compromise - they offered 4mb doubling every 4 years I believe - played a great part in it's eventual failure.
The situation now is very much different with almost 100% of miners saying they will support 2mb and some 50% already supporting bitcoin classic with more to come.
So, I share your concerns, but unfortunately mistakes were made, some of them grave mistakes, mistakes from which we learned, and are thus now well placed to move forward.
On your point about the current dominance of China in mining. Two years ago people were flipping out that Ghash.io might have the ability to perform a 51% attack and now they barely register a whole number percentage share of the hashrate. Things change.