[fitsmime] Thoughts on FITS MIME types.

[Clive Davenhall]

10/12/02.

Steve Allen wrote:

> I am hoping to see some input from the VO community here, but at
> present my impression remains that the RFC should register
>        image/fits
>        application/fits
> and that the rest of the issues raised here are best solved by
> starting up classification efforts within the FITS community.

I agree.

The proposed classification parameter has much to commend it (and it
could probably be implemented as a FITS keyword, a MIME parameter or, 
indeed, both).  However, it is also potentially a can of worms and working
out a set of permitted values seems likely to be a difficult and protracted
process.  Any standard certainly needs to emerge through the FITS community
and perhaps also through the wider astronomical community (we are really
talking about the classification of astronomical datasets rather than just
FITS files, though of course most astronomical datasets are FITS files).  I
also suspect that there is a small but finite possibility that the effort
would get hopelessly bogged down and nothing useful would emerge.

For these reasons I don't think that we should delay the work on the MIME
types because of the classification parameter, or make the MIME type
depend on the classification parameter.

There is some convergence between a dataset classification scheme and
other work which has been going on in (broadly) the VO area.  Over the
past few years the CDS has developed the UCD (Unified Column Descriptor),
which is a classification scheme for the columns in astronomical tables,
but which might also have some wider applicability.  AstroGrid (the UK
VO Project) is considering developing a so-called astronomical `ontology'
(basically a classification scheme for technical terms) for use in some
aspects of the VO.  These efforts obviously have some commonality with a
classification scheme for datasets, but will take time to come to fruition.

So, to reiterate, a classification scheme for datasets seems a good idea,
but it is a long-term project and the registration of MIME types should
not be delayed because of it.

-----------------------------------------------------------------------------
Clive Davenhall                                      Institute for Astronomy,
e-mail (internet, JANET): acd @ roe.ac.uk        Royal Observatory Edinburgh,
fax from within the UK:   0131-668-8416            Blackford Hill, Edinburgh,
fax from overseas:     +44-131-668-8416                    EH9 3HJ, Scotland.