[fitsbits] Question on angstrom and erg units in FITS standard

[Lucio Chiappetti]

On Tue, 4 Nov 2014, Thierry.Forveille at ujf-grenoble.fr wrote:
> Quoting Demitri Muna <demitri.muna at gmail.com>:
>> > Given FITS' rule of once FITS, always FITS

>> the idea that all past software being able to read any future FITS file 
>> is a sure path to obsolescence.
>> 
> The latter has never been the case, with the only requirement being that 
> old FITS files remain readable.

I agree with Thierry, and even more, I would say that a requirement of 
past software dealing with future data is unrealistic and substantially 
insane :-)

Also to be realistic I see nothing against the fact a FITS reading s/w 
written for a PARTICULAR purpose refuses to read some legal FITS, 
deprecated or not, ... it is sufficient it is able to DETECT it as an 
"unsupported feature" (if one definitely needs to read an unsupported 
file, one would look for another piece of s/w or a sort of converter).

Examples: a custom FITS viewer for private use can just recognise and skip 
the deprecated random groups (surely I would not expect that MYSELF I'd 
ever have the NEED to read such files)

Or a custom thumbnail image viewer may reject as unsupported some (or 
most but a small list) of WCS (or at least not do any pix<->sky 
conversion), if one has no time or will or need to support them.

-- 
------------------------------------------------------------------------
Lucio Chiappetti - INAF/IASF - via Bassini 15 - I-20133 Milano (Italy)
For more info : http://www.iasf-milano.inaf.it/~lucio/personal.html
------------------------------------------------------------------------
Do not like Firefox >=29 ?  Get Pale Moon !  http://www.palemoon.org