Can the USPTO build fences around each patent and each claim in arguing their position? It would seem to me that is what may be happening here. All attention is directed to specific words/phrases in a vaccume.

It is interesting to note that when you look at the 148, 749, 584 and 336 patents, they all claim exactly the same benefits. They were one application originally. Actually the 749, 336 and 584 (and possibly others) are what allows the 148 to achieve the benefits as described in all the patents. It is not a matter of getting there with just one of the patents. I'm confident it can't be done. Which is why the original application included A+B+C+D = E. It was the USPTO that told the applicant to split the architecture out into multiple patents. Can the USPTO now claim that it is not an architecture and just individual unrelated patents?

CM devised an architecture which utilizes cheaper/slower DRAM in a novel way which achieves the claimed benefits. The architecture requires the 148, the 336, the 584 and 749 (and possibly others). A+B+C+D = E. E being the stated benefits.

The USPTO seems to be saying any two of the above can be found in prior art, hence a combination is obvious. But they fail to consider whether the combination provides the benefits outlined in each of the patents. If the combination of pieces of prior art does not provide the invention's benefits, doesn't that impact on the question of obviousness to combine them? I certainly hope CM can use that kind of argument. All IMHO. Other thoughts/comments welcomed. Opty