Re: Time for Software

[<aah@nofs.navy.mil>]

I see no major problem in everyone writing their
own software for processing the data, and as both
Mike and Michael have stated, there may actually
be an advantage in that cross-checking of reductions
can be made.  However, I think the data acquisition
code should stabilize into two driver camps (Linux and
Windows) with just those two control programs used,
and the output from all of the reduction programs should
use a common file format.  How that output gets merged
and put into a database is yet to be determined, however.
Michael and Chris had lots of problems with Postgres
and the Mark III, so I am hesitant on suggesting a
similar exercise with 100x more data.  As I've said
before, I intend to make a 'clean' database here with
just those objects with both V&I detections, and will
make it available on our ftp server on a weekly basis.
The 'unclean' database will just be starlists, as I
have no interest in detections in only V or I, and
don't want to spend time merging the starlists into
some additional database.  Those starlists will also
be available on a weekly basis.
  We've been through this exercise before.  I expect
to have 500K-1M double detections per night, and
200 usable nights/year for a yearly total in the few
100M range, just from my site alone.  Multiply this
by some factor for the additional sites and see if
your database manager can keep up.
Arne