Why OWL Unit Ontology has no Log/Exp (Was: Re: OWL ontology for unit)

[David Berry]

On Tue, 2 Sep 2003, Brian Thomas wrote:

> I agree. In fact in the original draft that Ed showed to me we discussed
> whether or not the mathematical functions should be included. I believe
> the last thought was that, no, these should be separate, and perhaps
> belong in another schema/ontology or library. The log and exp functions
> were seen as being part of said "other half".

It was just that your document explicitly included some functions
("PowerOf", "InverseOf") but didn't include "LogOf" or "ExpOf" - I don't
understand why they are treated differently.

The analysis of a textual unit specification should identify the basic
units involved, and the functions used to combine them together (as your
ontology does). You then need a method which, given two such analyses,
works out a recipe for transforming values from one system of units to the
other (if possible). And finally you need code which will do the actual
transformation of the numerical data values by interpreting the recipe
produced by the previous step.

> WCS probably is a good point
> to start looking for requirements on this set of mathematical functions
> needed for full unit represenation.

The FITS-WCS paper I section on units is a general scheme for representing
any astronomically relevant units in FITS - it is not restricted to
units used in the the representation of WCS information (admitedly it
is odd that this general unit description should be in a paper
specifically describing WCS!).

David