[fitsbits] further reopening of Public Comment Period on the CONTINUE convention

[Lucio Chiappetti]

ANNOUNCEMENT:  REOPENING OF PUBLIC COMMENT PERIOD

This is to announce the reopening of a formal Public Comment Period on the 
incorporation of the long string keyword convention (aka CONTINUE) in the 
FITS Standard.

A first PCP was held in June-July (started on 19 Jun 2015), and a second 
one in February-March (started on 26 Feb 2015).

As soon as the IAUFWG was passing to the votes, a serious objection was 
raised and for this reason we are now re-opening the PCP until the issue 
is solved and agreed. The issue is explained below.

1 the original (optional) CONTINUE convention did not recommend its "use
   with reserved or mandatory FITS keywords (e.g., TTYPEn or EXTNAME, or
   other commonly used keywords ... [but only] for new application-specic
   keywords, the values of which are not critical ..."

2 the text proposed in June strengthened this statement as "it must not be
   used with the XTENSION, TFORMn, EXTNAME, TTYPEn, TDISPn, and TNULLn
   keywords, and it is not recommended for use with any other string-valued
   keywords that are defined in this Standard."

3 arbitrating among the arguments raised during the discussion, the text
   submitted to the second PCP removed any explicit restriction, since the
   technical team felt that it was generally sufficient to simply trust
   that future FITS file creators will use the CONTINUE keyword judiciously
   and only in cases where it is truly needed.

   The edited text which was being submitted for vote is reported here
   http://sax.iasf-milano.inaf.it/~lucio/FITS/Conventions/continue-rev2.pdf

4 it was questioned in particular by Mark Calabretta (whom I encourage to
   submit any clarification and correction to this summary) that the
   removal of all restrictions underestimated the impact on existing s/w
   (in particular the WCSlib API), and it was de facto introducing a new
   datatype

   a the "Single record string keywords" defined in 4.2.1.1 correspond
     to the "string" datatype as used so far.

   b the "Continued string keywords" defined in 4.2.1.2 define a new
     type which can be called "long-string"

5 it was proposed to add to the text a restriction 'CONTINUE may only be
   used for keywords that are declared to have a value of type
   "long-string".  It must not be used for keywords of type "string".'

6 the exact final wording is to be agreed, and this is the purpose of this
   PCP.

I will avoid to give here my own suggestions and opinions in this 
introductory summary, but will post them separately. I encourage anybody 
to review the issue carefully and post any suggestions, alternatives and 
comments on the FITSBITS mailing list.


-- 
Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy)
for IAU FWG (prorogated)