[fitsbits] reopening of Public Comment Period on the Green Bank convention

[Mark Calabretta]

On 25/05/2016, at 22:18, Lucio Chiappetti <lucio at lambrate.inaf.it> wrote:

> Since the main idea was to simply *document* the existence of the historically significant Green Bank convention in the FITS Standard, we are now just falling back to the exact wording that was used in the WCS paper, which implies NO CHANGE to the standard.

(Responding to this and subsequent replies on this thread.)

In retrospect, I think it was a mistake to mention the Green Bank convention in WCS Paper I without being able to reference a proper definition for it.  (The same for pixel lists.)

Likewise, I think it would be unwise to repeat that mistake in the formal FITS standard document.

The description of the GB convention in the registry of FITS conventions omits important details, such as precluding its use with the many "structural" FITS keywords such as XTENSION, NAXIS, BITPIX, EXTNAME, TFORMn, END, COMMENT, etc.  It is also written largely from an historical perspective, referring to the (non-standard) use of primary image WCS keywords in tables rather than their bintable forms.  As such, it is inadequate for current purposes.

Consequently, if the GB convention is to be mentioned in the text of the FITS standard, then I think it should be described rigorously in an appendix.

Secondly, I am sympathetic to complaints that this seems to be an attempt to incorporate the GB convention into the FITS standard by fiat.  It was not the intention of the WCS papers to promote the GB convention as standard usage.  On the other hand, I do think that if images are to be stored as array column elements, then the GB convention is virtually indispensable.  It is also extremely useful otherwise, e.g. for DATE-OBS, where it varies from row to row.  It is of more than just historical significance.

Regards,
Mark Calabretta