Thanks for the very thoughtful reply.

Let me start looking at it from bottom up:

1) Make it "will be fun years", it won't happen just this year....

2) MIPS: This is a NO from my side. Why? I did read with quite some amusement other posts in this thread debating performance details(e.g. ARM vs x86). These discussions are missing two MEGA points: First, there is less and less differentiation on the low-end via performance; and second, there is a graphics engine on the same chip as the compute cores in NVIDIA's approach! And I'm not emphasizing the 'how to display' aspect of the graphics engine (which is nice to have anyway), but the 'how to compute' aspect - think of CUDA or OpenCL. This has the potential to become the next level of acceleration step similar to MMX/SSE2 - but on steroids. I'm not aware of a similar thing to be in the making for MIPS. Intel has the same opportunity btw...it will be interesting to see how they play it.

3) Microsoft: The availability of Mirosoft SW for ARM will pump significant amounts of $$$ from the market in this area. This will result many systems, which in turn will stimulate non-Windows implementations (e.g. Android). We should not underestimate the power of the invested Giga-$ in Windows know-how. Why are so many folks still using MS-Office, while there are free alternatives out there? So yes, I see potential for other SW (e.g. Android) to become more visible, but I don't see MS dominance go away. Linux did not make MS go away...

4) I'm not arguing whether ARM is better than MIPS, or Atom or [insert your favorite here]. I'm just saying that they are here at the right time with the right stuff for the market to jump on it. This means income/profit for many companies - money which is spent partially to have even more exciting stuff in the same area as we go forward. This creates huge market dynamics. You might remember the old 'DOS sucks in comparison with Atari OS' discussions - the market nevertheless jumped on it and made it prevail. There are strong indications that the same will happen with ARM.

Last-not-least, to clarify: What I sketched above will unfold in the 2012 - 15 timeframe, not now. But by then, there will be not a stone left standing given the market disruption that will go along with it.

Will