From: <Chris.Phillips_at_email.protected>
Date: Wed, 8 Dec 2004 17:22:58 +1100 (EST)

Hi all,

Simon's message has reminded me an email I was going to send during the
VLBI week. There was a lot of talk about making sure schedules were ready
with plenty of time. I agree having schedule arrive at the last moment
(and I'm an offender myself) is bad, but this will not fix a large
fraction of the problem that we had but these problems could have been
fixed is PIs took a little more care with their schedule.

Sched produces all sorts of good stuff when it is run. Basically all PIs
should as a matter of course carefully do the following:
 - Look through the sched.runlog file
 - Use the "sumitem" option in the .key file
     (I use sumitem=el1, el2, slew, early, dwell, tpstart)
   and then carefully look through the .sum file. This tells stuff
   like elevation of the sources during the observations, slew times,
   tape changes etc.
 - Look though the output vex schedule (the .skd file), mostly the
   $FREQ section and the $SCHED section. Once you have looked at a
   couple of these files, you will understand them pretty well.

If PIs look at these files before submitting their schedule this would have
eliminated a significant fraction of problems in the last run.

Cheers
Chris
Received on 2004-12-08 17:23:27