Kevin, I'm seriously considering use of the twit filter.

Both YAG and Anders Hejlsberg have commented positively on the in-memory capabilities of FP. Yes YAG was a senior exec when he said it- newly promoted into the role as I recall, so your suggestions in this area are gratuitous and inappropriate.

You may not understand why, but MS clearly intends to make in-memory processing more viable in NET.

In-memory processing has always been possible in NET, but large datasets crashed the machine. This is one of the reasons why SPs were regarded as best practice.

Unless MS is planning to include 8 gig of memory with each copy of NET, how do you suppose they're going to fix the memory hogging issue? Hint- both Anders H and now MS on-the-ground functionaries continue to refer to FP's abilities in this area.

I've said repeatedly that if MS intends to achieve this via some other mechanism than automatic memory-disk spanning, that's cool as long as it performs.

All the rest of it is sophist invention by you and I'm weary of it. You may think this is intellectual discussion but it's not.

If you insist on answering, please focus on the question about "how MS plans to make in-memory processing work better".