[AstroPy] FITS files and column descriptions

[Paul Kuin]

Dear Yannick,

You seem disappointed by the FITS stuff.

The FITS standard is a compromise. It has in it what is required to make it
work as an exchange format.  In the early 1990-ies there was some attempt
to go further and standardize other keywords, like units. This was
unsuccessful, but organisations like HEASARC defined their own unit
convention, and keywords they require, likewise, CDS did, and I suppose
NRAO.  That led to some keywords being registered, so their interpretation
would not get diluted by different conventions using the same term. I'm a
bit unclear in my memory what happened to that registry and what it is
called.

I think a description of a column is nice-to-have, not a requirement per
sé. However, it could be one of the reserved keywords, I suppose.  But
you're not the first one to preach for the global adoption of some FITS
keywords. Unfortunately, you must get everyone to agree.

I think FITS has been quite a success so far and has not run out of
momentum. It may eventually be overshadowed by HDF where you can put in
metadata and also links, which is kind of nice.  But for now we have to
make do.

Anyone can add stuff to the FITS headers, so those column descriptions
could be added by you. I'm not sure if you easily can with Astropy.ascii

But I hope this helps you understand why things are the way they are.

Paul







On Mon, Nov 11, 2013 at 6:36 PM, Perry Greenfield <stsci.perry at gmail.com>wrote:

>
> On Nov 11, 2013, at 1:31 PM, Yannick Roehlly wrote:
>
> > Le Mon, 11 Nov 2013 08:51:05 -0500,
> > Perry Greenfield <stsci.perry at gmail.com> a écrit :
> >
> >> What TOPCAT is doing is in no way part of the FITS standard.
> >
> > Hi Perry,
> >
> > As far as I understand, even the TCOMM keyword is not part of the FITS
> > standard and they're is nothing in it to store column descriptions.
> > Right?
>
> Yes, I believe so. That is to say, nothing prevents you from using such
> keywords, but the standard doesn't give them any special meaning with
> regard to columns that anyone else is forced to recognize.
>
> At least as keywords, they are more easily accessible from standard FITS
> readers (that don't need to do anything special to access them, unlike the
> meta data that is put into the primary HDU.
>
> Also, I understand the reason they did that. It's there for those willing
> to go to the trouble to extract it. We are doing similar things for HST and
> JWST WCS information. But we don't assume that all other FITS readers must
> be able to access it. It's mainly for our software to use.
>
> Perry
>
> _______________________________________________
> AstroPy mailing list
> AstroPy at scipy.org
> http://mail.scipy.org/mailman/listinfo/astropy
>



-- 

* * * * * * * * http://www.mssl.ucl.ac.uk/~npmk/ * * * *
Dr. N.P.M. Kuin      (n.kuin at ucl.ac.uk)
phone +44-(0)1483 (prefix) -204927 (work)
mobile +44(0)7806985366  skype ID: npkuin
Mullard Space Science Laboratory  – University College London  –
Holmbury St Mary – Dorking – Surrey RH5 6NT–  U.K.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.python.org/pipermail/astropy/attachments/20131111/ec57e266/attachment.html>