If we start raising the block size it will be like that in a way: tomorrow 2MB, in one week 4MB, you understand the trick.
Simple and straight to the point. The end result would be simply that users on slow networks will be unable to catch up.
In the meantime Bitcoin would most likely have to fork back quickly somewhere between 2 MB and 4 MB else the network would be left in an unrepairable state.
I guess the current line of thinking is just to kick the can down the road. You do not implement something before it was thoroughly tested. I agree with this thinking in the sense that developers need to scale Bitcoin without risking the integrity of the whole network.
You cannot just dump bigger block sizes into the protocol and hope nothing goes wrong, because it is a Billion dollar network. Let's just approach this with caution.
Classic has no roadmap or anything. Even if we disregard the safety risk of 2 MB blocks right now, we would have this same discussion very soon (if the growth is going to increase). There's a proposal that aims to fix the validation time being quadratic; we should wait for that to be implemented.