From goodrich Mon Mar 26 09:47:31 2001
Received: from kee.keck.hawaii.edu (kee [128.171.96.130])
	by keck.hawaii.edu (8.8.8+Sun/8.8.8) with ESMTP id JAA18129;
	Mon, 26 Mar 2001 09:47:30 -1000 (HST)
Received: by kee.keck.hawaii.edu (8.8.8+Sun/SMI-4.1)
	id JAA07479; Mon, 26 Mar 2001 09:47:30 -1000 (HST)
Date: Mon, 26 Mar 2001 09:47:30 -1000 (HST)
From: goodrich (Robert Goodrich)
Message-Id: <200103261947.JAA07479@kee.keck.hawaii.edu>
To: goodrich, aconrad
Subject: just to get it written down...
X-Sun-Charset: US-ASCII
Content-Length: 2279
X-Lines: 64
Status: RO


----- Begin Included Message -----

From goodrich Thu Mar  8 18:10:21 2001
X-UIDL: fc9abe3dd8067cb4d6b3ae5d6dfe8cfe
To: goodrich, aconrad
Subject: just to get it written down...

Outstanding issues resolved from last e-mail:

subc	All is fine.  (Complicated, but fine.)

slit	We were using USER* keywords as place holders in the xshow.
	Fixed this to use the real ones.  However, whole spectroscopic
	setup scenario has not been run through completely.

Major outstanding issues:

snapi	Needs to get checked.

corona	I haven't added the positional argument.  Do we want to?  We could
	require observers to center a target on the spot by moving the field,
	rather than the coronagraphic spot.  If the spot moves too far,
	you'll get another spot (or something) coming in to the edge of
	the field, possibly.

various	Things with valid lists of positions, like "filt" and "fgr" have
	not all had those lists added to the help text.  Some have no 
	valid positions yet (like fgr and sgr).

preslit	Still potential problems with determining whether there is a
	slit in the beam or not.  I have fixed the "image" command,
	however, which did not reset "slitname", and this will
	take care of the specific instance which Eiichi documented.

sgr	Works OK, but documentation needs to be changed.

grand	Does not produce random sequences.  Al needs to find a better
	random number generator.

lastframe  Does not warn against potential overwrite situations.  It would
	   need to check the FILEOVWR keyword to see whether it is on or
	   not.  If not, then file overwriting will not occur; the frame
	   number will simply be advanced to the next highest number
	   and write a new file, negating the effects of a frame command.
	   (Note that users are not supposed to use "lastframe", but
	   rather "frame".)

New ones:

goi	Need to have a more sophisticated Ctrl-C trap.  The first one only
	catches wfg.  Need to propogate this back up to goi and pause to
	ask whether the user wants to stop the exposure in progress or not.

goibuf	Confusing the alad server's attempts to send the buffer to the
	display.  Looks like something hard-wired into the alad server?
	Currently the work-around is to link "/scratch/buf0001.fits" to
	"buf1.fits". 

Bob


----- End Included Message -----