FITS compression

[William Pence]

William Thompson wrote:
>
> >http://legacy.gsfc.nasa.gov/docs/heasarc/fits/compress/compress_image.html
> 
> This is a very interesting proposal.  However, if I understand it right, only
> one column in a binary table could support compressed data, which is pretty
> restrictive.  It would be better if multiple columns could be compressed to
> better match the present capabilities of binary tables.  Otherwise, for some of
> us at least, it would be a major change in the way that binary tables are
> organized.  For example, in our own project, we use different columns for
> different wavelengths.

It is a little confusing, but this proposal only applies to compressing images
that would otherwise be stored in a FITS primary array or IMAGE extension.  The
compressed image bytes are then stored in a binary table, which serves as a way
of indexing the location of different pieces of the image.  Being able to
compress ASCII or binary tables is also an important issue, but the current
proposal does not address that topic.

-Bill Pence