[fitsbits] 'Dataset Identifications' postings (digest)

[Clive Page]

On Wed, 24 Mar 2004, Arnold Rots wrote:

> The scope of Tom's proposal is really quite limited:

I would like to support Tom's simple and basic proposal.  As he says, a
unique key is of great value in building and searching databases, and
there are a whole lot of cases in which the DS_IDENT will be immediately
usable and useful.  Of course we can all think of special cases in
which it will not be quite enough, but (in my opinion) this doesn't
invalidate the basic idea, which would be severely compromised by being
made more complicated.   Let's keep it simple.


-- 
Clive Page
Dept of Physics & Astronomy,
University of Leicester,
Leicester, LE1 7RH,  U.K.