Re: ISSUE-124 (datarange complement): The complement of a datarange is defined relative to the whole data domain

On May 21, 2008, at 5:33 AM, Boris Motik wrote:

> I really do not expect people to use complemented data ranges  
> directly.

Over and over, in many different technical situations over the years,  
such predictions have been shown to be wrong, and therefore I don't  
think this reasoning should be used in considering resolving this  
issue. Last such in the OWL space: No one will want to make large sets  
of classes disjoint.

If the feature is there, it will be used. If the common situation is  
that it will be error prone, then I think it's worth considering  
either a) figuring out how to find a way to not expose it or b)  
provide and document a mechanism for avoiding those errors.

-Alan

Received on Wednesday, 21 May 2008 13:10:10 UTC