[fitsbits] CDELTn

[Doug Tody]

This is essentially right, although I was probably the one originally
responsible for this.  In the case of CDi_j the more logical default
is probably the unit matrix (diagonal elements 1.0, off-diagonal 0),
but the more global default for generic missing keywords was/is zero
if not present.

	- Doug


On Fri, 21 Sep 2007, Mark Calabretta wrote:

> On Thu 2007/09/20 11:51:24 -0400, Phil Hodge wrote
> in a message to: Mark Calabretta <mcalabre at atnf.csiro.au>
> 
> >I presume you meant CDELTia rather than CRVALia.  I've never understood 
> >why, but WCS Paper I actually specifies that the default value for any 
> >CDi_ja keyword is zero (although if all of them are absent the PC form 
> >is assumed).
> 
> This was the de facto interpretation of CDi_j before it was made
> standard by Paper I.  If I remember correctly (not wanting to search
> through 1000s of old emails) it was supplied by Frank Valdes and
> Lindsay Davis by interpretation of the IRAF code which was considered
> to be the defining instrument.  So effectively it's for historical
> reasons.
> 
> Cheers, Mark
> 
> _______________________________________________
> fitsbits mailing list
> fitsbits at listmgr.cv.nrao.edu
> http://listmgr.cv.nrao.edu/mailman/listinfo/fitsbits
>