Draft of FITS standard revision 1.2 available for comment

[Clive Page]

In article <Pine.OSF.3.95.980417173858.28733T-100000 at poseidon.ifctr.mi.cnr.it>,
Lucio Chiappetti  <lucio at ifctr.mi.cnr.it> wrote:

>would it be the right time to freeze and discourage use of some of the earlier
>extensions (random groups and ASCII tables) which are no longer justified ?

I would support this, except for the fact that ASCII tables seem to be
alive and well.  For example the CDS at Strasbourg has an unrivalled
collection of source catalogues and other tabular data.  They hold them in
some internal form, but produce FITS tables for export.  But these are
always ASCII tables.  I asked CDS staff last year why they didn't use
binary tables (because more efficient, no loss of numerical precision,
etc);  they vigorously defended the use of ASCII tables.  Unfortunately I
don't recall now all the reasons they cited - perhaps someone from the CDS
reads this group and can enlighten us?

-- 
-- 
Clive Page,
Dept of Physics & Astronomy,        
University of Leicester.