From: Dan Connolly (connolly@w3.org)
Date: 01/09/01
Frank van Harmelen wrote: > > Deborah Mcguinness wrote: > > > The namespace error in this document results from the default > > namespace being set to the daml+oil URI instead of the daml+oil-ex > > URI. > > [...] > > I suggest that the error be corrected by changing the default > > namespace to the URI of daml+oil-ex.daml. In addition, any > > usage of resources defined in daml+oil.daml should specifically use > > the daml namespace. > > Jessica, > > Thanks for spotting this and providing us with a fix. Quite. > I will update the running copy of daml+oil-ex.daml as you suggested, but have two worries: > [1] Stefan, how come your parser didn't spot this?, and Because the bug is with the translation from intended-semantics to RDF syntax; there's nothing wrong with this RDF syntax; it just doesn't mean what you probably wanted to say. Jessica wrote: |Person is not defined in |daml+oil.daml. As a result, Adam is |declared to be of a type that does not exist. Er... that's sort of a closed-world reaction to this situation. Adam is declared to be of a type that's not mentioned in the daml+oil.daml document. There's nothing wrong with that. > [2] Now there is pretty much no unqualified symbol in sight in the entire document, and it's beginning to look pretty ugly. Do namespaces have a habit of always getting in the way? In a sense, yes. But I hope beauty of the syntax is not job #1 here; ultimately, deployment of this should result in folks looking at angle-brackets only in development/debug mode, not in ordinary use. -- Dan Connolly, W3C http://www.w3.org/People/Connolly/
This archive was generated by hypermail 2.1.4 : 04/02/02 EST