[
Lists Home |
Date Index |
Thread Index
]
Roger L. Costello wrote:
> Comments? Suggestions?
Not everything has to be declared ;)
I think the issue with this approach is to avoid trying define a
REL(kilometer, mile) function in OWL, when all you really need to do
is denote the function with a name in OWL and use a programming
language (or calculus) to define it. Using MathML is pushing the
problem soemwhat around, since MathML doesn't have an MT (to my
knowledge). I could just as well provided a scripting language hook
in CDATA, or an XSLT fragment and declared /that/ as the function
(akin to providing a lambda evaluation instead of a name as Tom
pointed out).
Arguably it's just as useful to fix the name of a function and not
the function itself since not all functions have constant results.
Consider how we'd model currency exchange rates using set
containment rather than simply naming the rates.
Bill de hÓra
|
