Re: Why not import everything? (was: Re: getting daml:imports right is easy?)

   [Pat Hayes]
   However, I still wonder if we need ['imports']. Consider the
   admittedly naive assumption that anyone who uses a vocabulary is
   committed to using it the same way. Then we don't need importing:
   just use the same names is all you have to do. 

Here are the problems with that idea (besides the ones I already
described):

1. We have to distinguish names of individuals from names of
   properties.  It's somewhat plausible that if I refer to a property
   then I should also accept the axioms about that property; if I
   don't, there's some question about whether I'm referring to the
   property at all.  But URIs are used as proper names as well as
   names of predicates, and here things are much less clear.  If I
   want to make a statement about the United Nations, and the URI
   www.un.world/#UN points to a web page containing all sorts of
   axioms (e.g., "Zionism is racism"), am I automatically committed to
   those axioms simply by referring to the UN?  Presumably not.

2. "Anyone who uses a vocabulary is committed to using it the same
   way."  But what do we mean by "a vocabulary"?  If we have a
   well-defined notion of ontology, then a vocabulary is the set of
   terms introduced by the ontology.  If we have a vaguer notion of
   everything reachable through an RDF graph, then it's not at all
   clear what commitment I am signing on to by pointing to a place in
   that graph.

   [Pat]
   OK, so it is going
   to break down at times; but I bet 'imports' is going to break down
   at times as well, because people will misunderstand the intended
   meanings in complex ontologies, and so on. 

The consequences are different, though.  If a well-defined ontology
breaks down, you fix it.  If a flood of disconnected symbols breaks
down, then what do you do?  

   Seems to me that the
   only real purpose of having an 'imports' tag is to be able to NOT
   import some stuff you DONT want to agree to. So it might be more
   use, in fact, to have that as the primitive. 

It sounds to me like the user of a vocabulary has to draw the
boundaries around it himself, watching out for leaks.  (And what if
a piece changes?  There's no one version spec for the whole thing, so
the vocabulary can acquire new boundaries like an ameba, and every
user has to think about every new piece of membrane.)  I'd much rather
take what an ontology designer gives me, knowing that if a problem is
encountered I know who to complain to.

   ...
   This suggests a vision of an SW which is in a constant process of 
   self-repair. 

Pat, you offer the strangest mix of wild-eyed romanticism and
hard-nosed logicism.  In this message the eyes have it.  I would vote
with the nose.


                                             -- Drew

Received on Wednesday, 1 May 2002 11:32:29 UTC