Re: Bitcoin XT - Officially #REKT (also goes for BIP101 fraud)

The consideration given is that, unlike the original proposal, this one doesn't double the blocksize. It's a measured and more conservative increase, which will allow nodes to keep pace, which is what you want. I've even stated that the 12.5% part is open to negotiation, but it is a hard cap. It means precisely that the blocksize can't increase (or decrease) by more than that amount over a period of time. The period of time in question is also negotiable. This gives you everything you're asking for, we just need to decide what the limit should be and over how long a timeframe.

12.5% YOY sounds on par with Pieter Wuille proposal. I honestly wouldn't mind it

I'm pretty sure that we would be much better off if we had a dynamic block size limit (with the correct parameters). BIP101 is probably too big, and 12.5% YOY might be too low, both of these would eventually cause issues. As DooMAD said, it is still a hard cap which can't be changed without another fork. If we had a dynamic block size then we would not need to worry about this (unless someone tries to game the system).
I've always been against making pointless predictions as we can not know what is going to happen in the future. We can't expect any fixed solution to be the right call.

The purpose here is not to make predictions based on demand, which would indeed be pointless, but to make block size growth a function of the most conservative estimate of technological advances (mainly bandwidth). This is more manageable and a 10-15% increase would be in line with that.