Re: LANG, SEM: Re: more on a same-syntax extension from RDF(S) to OWL

Jeremy Carroll asks for use cases for unasserted RDF.

There are many contexts in which one wants to refer to a statement
without asserting it.  Jonathan Borden described IF THEN statements.
Others include:

I heard FOO.  (I'm not sure whether I believe it...)
Tim believes FOO.  (But I don't necessarily...)
Libby says FOO.  (I don't know if it's true.)
FOO is false.

where, in each case, FOO is a piece of RDF.

For example, in an information repository project I've been involved
with [1], we record every statement that's been made, automatically
inferred, or input into the repository.  We do NOT, in general, want to
believe all of these statements.  (For example, we collect information
from email, web browsing, etc.  We also automatically infer type
information, authorship, etc., sometimes erroneously.  Finally, users
can hand annotate information and may do so in error.)  It would be
lovely to represent all of these statements in RDF, but we do NOT want
to assert them all!

Lynn


[1] Haystack, http://haystack.lcs.mit.edu

Received on Tuesday, 5 March 2002 13:58:14 UTC