Re: ALMA Scripting: A VLT S/W Developer's Comments

[Barry Clark <bclark@aoc.nrao.edu>]

Erik Allaert raises an interesting question, whose answer is implicit
to what I have been writing, though I haven't mentioned it.  Because
real-time systems are so hard to work on, I think the real-time system
should be written with utmost generality.  It should be able to accomodate
to the requirements of an analysis of the requirements of the data flow
through the system, as currently definable, but should not be limited to
them.  As much as possible, the on-line system should be limited only
by the limitations of the hardware (an impossible ideal, I know).

And Brian's reply raises another of my implicit assumptions.  The hard
part of designing this interface is deciding where to draw the line
on a lot of needed features.  I am going to argue that Doppler tracking 
belongs in the observe file preparation tool, rather than in the real-time 
system itself.  The argument, and a rather weak one, I admit, is that the 
capability is certainly needed in the observe file preparation tool,
and not providing it in the real-time system means that you don't have
to worry about a mechanism to make sure it gives the same answer in
both places.  (Note that I come to the opposite conclusion for precession/
nutation from J2000, but to the same conclusion for B1950=>J2000.  A
tricky business indeed.)
Date: Thu, 27 Jan 2000 11:40:39 +0100 (MET)
To: lucas@iram.fr
From: Majordomo@web3.hq.eso.org
Subject: Majordomo file: list 'alma-sw-ssr' file 'alma-sw-ssr.000115'

--