I've seen your posting before, but don't see the relevance. Our indices are a trivial portion of our data size, and even if they were zero bytes in size it wouldn't make an appreciable difference to the data size.
Though you are right, the architecture we tried earlier, when we were doing a proof of concept, would not have helped with compression. Back then we were trying to play nice with mongodb's data format and keep its legacy indexes around. It turned out that for many reasons (concurrency, recovery, and compression, and a few others), it was a lot better to replace all of the storage code, so that's what some of us have been doing the past couple months.
What we are doing now is replacing all of mongodb's storage. So all of the data and all the indexes are in fractal trees. This means we can compress everything (including the field names!) and keep everything nicely defragmented.