[fitswcs] inherit vs. concatenate

[David Berry]

On Tue, 15 Jan 2002, Eric Greisen wrote:

>  > Having been working for several years with Rodney Warren-Smith's notion of 
>  > "World Coordinate Systems as Objects" (see his ADASS 97 presentation
>  > http://adass.org/adass/proceedings/adass97/warrensmithr.html plus
>  > related poster papers from ADASS 99 and ADASS 2000 ), I can't
>  > help thinking that the FITS-WCS discussions would have been a *lot*
>  > shorter if we had started from scratch, taking a more radical
>  > object-oriented approach which generalizes the whole problem of WCS
>  > representation.  The AST subroutine library which implements Rodney's
>  > ideas (http://www.starlink.rl.ac.uk/star/docs/sun211.htx/sun211.html)
>  > has a working scheme for representing arbitrarily complex WCS objects
>  > as FITS headers (it can also read and write WCS from other common FITS
>  > header schemes, obviously working within the limitations of those
>  > schemes though).
>  > 
>  > I guess the time for such comments passed several years ago though!
> 
> A very long time ago.  This paper is 10 years old and the deadline for
> its submission to the journals is approaching.  The purpose of the
> paper is to achieve a data interchange format, not to write software
> systems.  

Firsly, I agree that we've got to get the current proposals published soon
now. They allow us to be much much more sophisticated in our
transportation of WCS information, and I for one am very gratefull
to you and Mark, and all the other people who have put effort into
the proposals.

My point was that an object-oriented approach to representing WCS
information has many advantages. It allows WCS networks of arbitrary 
complexity to be built up from well defined self-contained building
blocks. It allows new building blocks to be designed and added very
easily. It shouldn't be a case of designing a software system, but of
identifying the objects relevant to WCS (an Axis, a coordinate Frame, a
Mapping, a Network of connected coordinate Frames, etc), deciding what
their significant properties are, and deciding how to represent these
objects in terms of FITS keywords. It seems to me this is a
*representation* issue, not a software issue. True, we have a software
implementation of such an idea, but that's not to say that other people
couldn't produce their own. Once the representation of the objects have
been decided on, people would be free to implement software to use the
objects as they saw fit.

Mere speculation....

David