Version 100-2.0 of NASA FITS Standard released

[Eric Greisen]

Peter Teuben writes:
 > Having been on the original NOST-1 panel, and working with
 > FITS Random  Format files almost on a daily basis I've of course
 > also been weary about what "deprecate" really means. You can
 > even find a definition in the current NOST manual.
 > 
 > We (the panel) meant to tell the community that new data should
 > not be using this format, since "better" ones are available
 > (binary tables). However, existing ones should indeed fall under
 > the "once fits, always fits" chapter.
 > 
 > 
 > In an older version we defined it as:
 > 
 > "To express earnest disapproval of. This term is used to refer to   obsolete
 > structures that ought not to be used but remain valid."
 > 
 > However, there has been some discussion that the word "ought" in
 > that definition may be too strong.
 > 
 > 
 > I remind you that we deprecated 3 FITS features:
 > 
 > - random groups
 > - the BLOCKED keyword
 > - the EPOCH keyword 
 > 
 > but again, it remains true that old data will not be deprecated,
 > one will be able to read it. It's meant to refer to new data, but
 > does not disallow it, but may be frowned upon.
 > 
 > 
 > Bill Cotton wrote:
 > 
 > >    I note with dismay that this version of the NASA FITS standard
 > > still tries to deprecate the FITS random groups format.  This format
 > > is in widespread use in the radio interferometry community.  The
 > > attempt to deprecate this format is in violation of the fundamental
 > > principle of FITS, "once FITS, always FITS".  Clintonian weasel words
 > > on the subject are not helpful.  I am forced to question the
 > > responsiveness of the NASA committee to the non-NASA FITS community.
 > > 
 > > -Bill Cotton
 > 
 > 

    The real problem is the word "deprecate" itself.  That word has
been given a meaning by the ANSI which has been used then with Fortran
and other languages (at least).  The word means that the deprecated
feature will be no longer supported at some future date.  I suspect
that we cannot use an arbitrary definition of the word which we make
up and spell out.  The word has a meaning to the standards business
and we do not mean that meaning.  I think we should find a new word.
The problem is that I do not know what that word is.  Does anyone
anywhere have a suggestion?

Thanks,
Eric W Greisen