[fitsbits] Chandra's WCS tour de force

[Steve Allen]

The Chandra folks at Harvard have assembled a beautiful image of Cen A at
http://chandra.harvard.edu/photo/2002/0157/index.html
and this image starts me wondering.

The press release (PR) images are all (and naturally so) typical
commodity image formats which are standardized by various agencies and
registered as MIME types.  The PR images are beautifully composited.

Each of the input images was undoubtedly a FITS image.  Three of them
should have been 2-dimensional "classical" FITS images, and one of
them may have been a FITS table containing a "pixel list".  It is
possible to imagine combining all of these into a single FITS file
consisting of 4 individual HDUs.

Each image could have its own set of instrument specific keywords
indicating its provenance.  And they all should now be able to have a
set of WCS keywords that conform to the WCS papers that are submitted
to A & A -- each one with its own scale and rotation parameters.

But the PR images have neatly combined each of these images onto a
common scale and rotation.  Each one is assigned a color from the
palette of blue, green, orange, red, and is neatly stretched to span
that color from dark to light.

The existing FITS standards provide good grammar of multi-HDU FITS
files, but that grammar consists only of a set of nouns.  We don't
have any verbs.

And that leads to my questions.

When should FITS attempt to create a standard document which would
tell an image display how best to display such a FITS file?
Who is best qualified to draft such a document?
Would the keywords for the combination be kept in the PHDU?

These questions are very relevant for Don Wells effort to create MIME
types for FITS.  The current drafts of FITS MIME are really not
complete.  There is no means of directing the recipient of a multi-HDU
FITS file just how it should attempt to present that FITS file to the
requestor.  (Of course the natural result of adding verbs to FITS is
that the MIME type will have to have a non-null section on security
issues of performing operations directed therein.)

--
Steve Allen          UCO/Lick Observatory       Santa Cruz, CA 95064
sla at ucolick.org      Voice: +1 831 459 3046     http://www.ucolick.org/~sla
PGP: 1024/E46978C5   F6 78 D1 10 62 94 8F 2E    49 89 0E FE 26 B4 14 93