Re: OGIP/93-013 - A list of standard strings for HE missions,instruments & filters.

["Ian M George, Code 668, NASA/GSFC, USA (Usque ad mortem bibendum)" <GEORGE@heagip.gsfc.nasa.gov>]

Jeff Bloch (jbloch@sstcx1.lanl.gov) asked a number of questions in his HEAFITS 
post with respect to keyword values for the ALEXIS satellite. The following 
'answers' are from a personal point-of-view, and do not necessarily reflect 
the view of the whole OGIP:

Jeff writes:
 > We are in the process of thinking about FITS data file formats for data
 > from our ALEXIS satellite. We would like to propose a set of keywords for
 > the six EUV/ultrasoft x-ray telescopes on our satellite. The detectors and
 > filters are fixed for each of the telesopes. The questions we have are:
 > 
 > 1. What is the maximum number of characters to be considered for each keyword
 >
.... formally I believe the limit is 68 characters (plus a ' at each end).
There has been a proposal posted to FITSBITS (many months ago) by Pence & Rots 
suggesting a syntax for a continuation character/keywords for FITS cards
(Perhaps one of this guys could comment on and/or explode the proposal again).
Nevertheless, I would *hope* that the values of the TELESCOP, INSTRUME and 
FILTER keywords would be strings substantially shorter than this (!) 
 >
 > 2. How should we weight considerations of description vs. itemized labeling?
 >   For example:  four telescopes have Al/Si/C filters. Should they all
 >   have a keyword like "Al/Si/C", or should they have their filter ID number
 >   as the keyword, such as "03479 #1", or some hybrid of the two, which would
 >   make a longer keyword? Part of the problem here is that at the project
 >   level thus far we make general distinctions between the two filter types
 >   we have ("AL/SI/C" and "LEX/TI/B") but at some point a user of the
 >   data will care about the individual calibration curves for each filter.
 >   The problem is simplified to some extent because each telescope has a
 >   single filter that will never change.
 >
.... This is a subject close to my heart, being related to calibration. 
Itemized labelling is obviously very useful prior to launch, but is rather 
cumbersome as soon as the flight filters at bolted onto the instrument. In 
this case I *think* would favour a method whereby during instrument 
calibration a hybrid scheme was used (eg FILTER = 'Al/Si/C 03479 #1'), but 
with the instrument team ID number silently dropped after launch. Thus 
user-software for "detector X" would automatically know (have hardwired 
somehow) that the filter calibration curve appropriate to 'Al/Si/C 03479 #1'
should be used. This isn't really so bad since I assume that the filter 
cal curve for 'Al/Si/C 03479 #1' would be delivered as being appropriate to 
INSTRUME = 'Detector X' anyhow.
It should be noted here that the OGIP/93-013 document is primarily aimed 
at the post-launch/data-analysis s/w phase rather than the grd-cal phase of 
a given instrument. 
 >
 > 3. The question about filter names in 2 above applies to detectors. ALEXIS
 >    detectors have serial numbers such as "AF02", but also have unique
 >    photocathode materials. Two telescope detectors have MgFl photocathodes,
 >    and four have NaBr photocathodes. In this case it might be more tractable
 >    to have hybrid names, i.e. "AF01/MGFL".
 > 
.... Given there ought to be no ambiguity as to which photocathode material 
a given detector has after launch, I dont really see the need for a hybrid 
name - but I again assume that the appropriate calibration files 
(detector efficiencies etc etc) will be re-labeled with the appropriate 
detector string prior to delivery to users/data-analysis s/w. However I have 
absolutely no objections to a hybrid name if it makes everything clearer.
 > 
 > 4. Should we stick to upper case only?
 > 
.... No, not if you prefer not to. Personally I'm very much in favour of 
NOT mixing cases in a given string, simply because it makes it so much 
easier to describe the string over the phone ! However, in the memo there 
are a few times where cases are mixed (mostly associated with the symbols 
for elements with filters). The primary concern is to make the string as 
similar as possible to the string/acronym/appreviation that (most) users
use for the quantity.

... As a final summary on the strings to be used post-launch, I think that it
*must* be highly desireable that users receive science dataset & (access to) 
calibration files which use the same naming convention for the TELESCOP, 
INSTRUME, DETNAM & FILTER keywords. The details of the precise syntax used 
for these is kinda secondary, but should be as user-friendly (obvious) as 
possible.

Regards
Ian

PS: Jeff, let me know if you have further questions; and (if you wish) I'll be 
    happy to include the values you finally decide on for ALEXIS within 
    the memo.