Re: [alma-sw-ssr] [Fwd: ALMA-SSR review]

[Bryan Butler <bbutler@aoc.nrao.edu>]

one small comment on this...

>(3) TIMETABLE FOR DYNAMIC/AUTOMATED SCHEDULING
>We should probably expect that the initial phase of ALMA operation will be 
>scheduled in interactive mode.  Apart from the need to debug the hardware, a 
>true automatic scheduler will probably require at least a year (allowing for 
>seasonally varying weather patterns) of archived data.  Otherwise, there will 
>be no past results from which to predict future performance-- a problem which 
>is especially severe when evaluating how to take advantage of the best (and 
>therefore rarest) weather conditions. 

while it may be true that we won't know absolutely everything about
conditions at the site, and performance of the array, until we have
some bulk of data in hand from it, i must say that we already know
an awful lot about conditions at the site, and will know even more
by the time the array is fully operational.  so, whether a year's
worth of "real" data is needed before an automatic scheduler can be
constructed is a point that could be debated either way, i think.

now, whether we have the resources to construct the automatic scheduler
sooner than that is another issue...

	-bryan