[fitsbits] NOST 100-2.0

[William Pence]

If one goes back to the older NOST 100-1.0 and NOST 100-1.1 documents, 
section 5.2.1.1 begins with this sentence: "Principal mandatory keywords other
than SIMPLE are required in all FITS headers".  Also, the caption to Table 5.1
(which lists the following keywords: SIMPLE, BITPIX, NAXIS, NAXISn, and END)
reads "Principal mandatory keywords".  In NOST 100-2.0 that first sentence
(now in section 5.4.1.1) has been deleted and the caption to Table 5.1 has
been changed to "Mandatory keywords for the primary header".

So, it looks to me like the term "Principal Mandatory Keywords" was originally
intended to include the following list of keywords: SIMPLE BITPIX, NAXIS,
NAXISn, and END.  The Appendix F (or D depending on the version) also includes
the EXTEND keyword in this list, which I think is technically incorrect.

In any case, "principal" is not synonymous with "primary"; instead it was
intended to define the "principal" set of mandatory keywords that must be
present in every FITS header (except SIMPLE).   I don't remember the details
at this point, but I think the NOST technical panel that drafted the 100-2.0
version of the document found the use of the term "principal" to be confusing
and removed it from several places in the document, but didn't remove every
instance of it (it probably should have).  Table 5.3 explicitly lists all
these keywords as being required in conforming extensions, so I don't think
that the term 'principal' retains any useful meaning in this context.

-Bill Pence

Steve Allen wrote:
> 
> 'tis probably a bit late to mention it, but I've just noticed
> something in NOST 100-2.0 which leaves me wanting for a clarification.
> I'm looking at the version dated March 29, 1999, which is, I believe
> the most recent version.
> 
> Section 5.4.1.1 is titled "Principal".  The word is used again in the
> paragraph that follows, but it is not defined anywhere that I've seen
> in the document.  It is used interleaved with the term "primary", which
> is defined.
> 
> Principal is also used in point 18 of Appendix E.  This usage appears
> simply to be maintaining self-consistency, and adds nothing to the
> comprehension.
> 
> Most confusing, however, it is used in Table F.1 in a location where
> I would expect the use of the word "primary".
> 
> What is "principal" supposed to mean, and how is it different from "primary"?

____________________________________________________________________
Dr. William Pence                          pence at tetra.gsfc.nasa.gov
NASA/GSFC Code 662         HEASARC         +1-301-286-4599 (voice)     
Greenbelt MD 20771                         +1-301-286-1684 (fax)