[fitsbits] Dataset identifications.

[Thomas McGlynn]

Thierry Forveille wrote:

> Thomas McGlynn writes:
>  > One that comes to mind is whether use of this
>  > keyword should be recommended only for the primary header of a FITS
>  > file.  If not then a file may not be associated with a unique dataset
>  > id.
>  > 
> Is that a problem, or actually a desirable characteristic? I could imagine
> a single FITS file containing data from several origins in different 
> extensions (e.g. M31 images from IRAS, ROSAT, and the CfA CO(1-0) survey)
> where different DS_IDENT would be apropriate. Or should such a merged
> dataset be given a new DS_IDENT and forget about the original ones? At
> first sight at least, retaining the original pedigree sounds like
> a good thing.
> 

Keeping the pedigree is a good idea - that would surely be
one use of the IDs.  However, I'm
concerned that if a single FITS file is not atomic with respect
to the dataset identifier, then it would generally be difficult
for software to use the ID to tie things together.  E.g., one
would need to scan the entire FITS file to find the included IDs.
If one has a derived data product that was the concatenation of
several datasets, then I'd expect there to be some link in the header
to the original datasets, but that this derived data product would
indeed have its own dataset id (assuming it had any).

My feeling is that if the two elements of data are so tightly
coupled that they would normally be put in the same FITS file,
then one would consider them to be the same dataset. At the HEASARC
datasets generally comprise more than one FITS file.  I can't think
off the top of my head of an example where a given FITS file would
naturally split up into multiple IDs.

	Tom