[mmaimcal] Re: subarrays

[Bryan Butler]

the argument that the only drivers for designing the array are 
whether "significant science that cannot be done without some
specific feature of the instrument" is a specious one.  this
is not the way we should design this instrument.  sure, we 
can get by with only 2 subarrays (in fact, we could get by with
only 1 - just demand that the entire array shuts down whenever
we need to redetermine baselines, pointing, delays, etc...),
but if this is going to mean that it will take us significantly
longer to do some (or all) kinds of things, then we need to
weigh that against the cost of having more subarrays.

since you have provided us with no cost estimate, we cannot have
a serious discussion or study of the tradeoff.

	-bryan