[fitsbits] New draft of WCS Paper IV

Greetings,

It has been quite a while since WCS Paper IV, "Representations of
distortions in FITS world coordinate systems", first saw the light of
day; though dated 2002/03/12, it was only generally announced on
2003/04/29. Mostly our limited WCS time has been consumed by Paper III
and other matters.

Whereas Sect. 4 and beyond of the original draft consisted only of
placeholder text extracted from early drafts of Paper II, all sections
of this draft have been updated, except for those that exist only as
section headers (Sects. 3.2, 3.3, 4.2, 5.1, 5.3, 5.4, and 5.5) and also
the Summary.

Changed text is marked in red in the PostScript documents (but not in
the PDF) available at

http://www.atnf.csiro.au/~mcalabre/WCS/

You will see that Paper IV has undergone significant change,
particularly to Sect. 2.4, particularly concerning a new "record-valued"
keyword type. Reasons for the changes are given in the paper.

The translation of a DSS header in Sect. 5.2 shows what can be done with
the new formalism: the DSS header is first approximated as a classical
"six-constant" plate solution using only the methods of Papers I & II,
then a distortion correction is added using the methods of Paper IV to
complete an exact translation. This is a much more sophisticated
treatment than the original TAN+polynomial described in an earlier draft
of Paper II.

Of course, the formalism described in this draft is provisional; it must
not be used as though it was an accepted standard.

Mark Calabretta
ATNF