[iaufwg] IAUFWG review of the new draft FITS Standard

[William Thompson]

William Pence wrote:
> Arnold,
> 
> Are you suggesting that we define a new binary table TFORMn data code 
> (e.g., 'T') for ISO 8601 date strings?  In this case, I presume the 
> repeat count value (as in TFORM1 = '19T') would give the width of the 
> single date value, in bytes, and would not mean that the column contains 
> a vector of 19 ISO data values.
> 
> If this is what is being proposed, then it sounds to me like a major 
> change to FITS that should be discussed as a separate proposal, rather 
> than including it as part of the current exercise to simply update and 
> clarify the wording of the FITS Standard.

I agree.  I think that time needs the same kind of careful analysis that was 
given to spectroscopy in the 3rd WCS paper.  It's a very complex subject.

Typically, what we've done with time as a dimension of an array is define it as 
an offset value in seconds from the DATE-OBS keyword.  For example, if we're 
taking a series of exposures 3 seconds apart, then the axis values would be 
0,3,6,9,...

-- 
William Thompson
NASA Goddard Space Flight Center
Code 671
Greenbelt, MD  20771
USA

301-286-2040
William.T.Thompson at nasa.gov

(Note changed email address)