"The discussion so far has been focused on your physical (scaling,
lag) goals, but you also seem interested in some logical (model,
datatype, interface) improvements over existing RDBMS's. Is that
something you're also interested in discussing here, or would you (or
other list members!) rather avoid turning this into a feature-request-
fest? If you're going to start from scratch anyway ..."

I'd like to keep it somewhat focused. I guess it is also a matter
of 'layers': some bits are low level, debating the guts of the
system, other bits are about how the guts will interface with the
client and yet other bits are about candy for the app designer.

Right now I'm trying to define the field and sort out real issues
from non-issues. To be honest, when the discussion moves from the
general to the specific it will become very time consuming, and I
fear it will start to fizzle at that point.

"As to disks: [...]"

I don't mean to cut you of, but in my view this discussion is about
handling workloads that exceed the processing capacity of a single
node -- even if that node has 100x the capacity of today's high end
Sun boxes.

Paul