From: Richard Fikes (fikes@ksl.stanford.edu)
Date: 09/10/01
> >I appreciate the effort of specifying the language as a DAML+OIL > >ontology. > > Well, I wonder what the point of this effort was, and would like to > raise this as an issue for discussion. Ok. In addition to Dan's comments, I will add mine. As I said in the discussion during last week's telecon, my motivation for expressing the query language and query results language as a DAML+OIL ontology was in support of the notion that DAML+OIL is to be used as an interlingua for agent communication in the Semantic Web (as XML is being used in today's Web). So, I envision a query being sent to a query-answering service as an instance of class Query represented in DAML+OIL, and the service returning its result(s) as instance(s) of class QueryAnswer represented in DAML+OIL. Now, as we also discussed, DAML+OIL is certainly *not* an appropriate language for describing the query and query result languages to people. Also, just as we expect Web site designers of the future to be using tools that enable them to create DAML+OIL KBs without having to type all the detailed XML syntactic forms that will be in the resulting source files, I would expect there to emerge query asking tools for human use that provide a friendly means of asking queries and viewing results. Regarding the specific points you raised -- > (a) completely pointless, in the strict sense that it provides no > useful functionality or understanding of the language (ie the query > language) to describe its syntax in DAML+OIL as opposed to, say, BNF. > All that this enables a hypothetical DAML reasoner to do is to parse > the expressions of the query language. Using an ontology language for > parsing seems a very poor design decision; at the least, one that > should be discussed on its merits rather than simply assumed to be > somehow a Good Thing. I would expect a DAML+OIL reasoner to be able to read and write DAML+OIL KBs, and so those capabilities would be being used in this case to read a query and write query results. You refer to DAML+OIL as "an ontology language". It is. However, since it supports the description of individuals (i.e., instances of classes), it can also be used as a representation language for an entire knowledge base. The notion here is to use DAML+OIL as the knowledge representation language for a knowledge base containing a specific query and a specific query result. Remember that my argument for proceeding with a query language for just DAML+OIL rather than waiting for adoption of a rules language (and for building DAML+OIL query answering systems) is that DAML+OIL can be used as both an ontology language and a Web site knowledge representation language, at least for demonstration and experimental Semantic Web applications, and that we should be proceeding with such demonstrations and experiments to promote the Semantic Web. > (b) actively misleading, in the sense that it suggests that the > purpose of DAML+OIL is to be a syntax specification language, which > as far as I was aware wasn't ever even close to the intended goal of > the project; As you know, one way of creating a new knowledge representation language is to simply add an ontology to an existing knowledge representation language. That is what is being done to produce DAML-S, that is what I expect we will do to produce whatever query language and query results language we decide on, that is what was done to produce the OKBC frame language (where the base language was FOL), etc. There is no new syntax in such an activity, only new vocabulary. > (c) based on a basic misapprehension about the nature of descriptive > languages, in that part of the very idea of a *syntactic* > specification is that it describes domains of recursively defined > finite entities to which results such as the second recursion theorem > apply, whereas descriptive (assertional) languages like DAML+OIL (and > RDF and FOL) have an extensional semantics which is (because of > Goedel incompleteness) inherently unable to fully capture the notion > of finiteness or recursion. I don't have sufficient familiarity with the elements of the argument you are making to understand it. I am open to hearing more about it. I would only repeat my comment from above that there was no intention in the strawman proposal of specifying new syntax. The intention was only to specify the ontology that would be used to express queries and their results. Richard
This archive was generated by hypermail 2.1.4 : 04/02/02 EST