From: Pat Hayes ([email protected])
Date: 01/13/02
>In this message, I will reply to your comments on my overview of what's
>in a query and in a query response. I will comment on your proposed use
>of continuations in a following message.
OK, and sorry the response is so late. Im catching up slowly :-)
> > >WHAT'S IN A QUERY
>> >
>> >* Knowledge Base - I think we are all agree that a query is posed with
>> >respect to a DAML+OIL knowledge base. Thus, a query needs to include a
>> >reference to a DAML+OIL knowledge base. I am referring to that
>> >knowledge base as the "query KB".
>>
>> It would make more sense to call it the 'server KB'.
>>
>> Do we in fact want to assume that there is a unique KB for each
>> query? Eg consider a 'services' setting in which a query can be
>> published, meaning 'any site that can prove this, give me an answer'.
>> The RDF core WG considered such a possibility, where one might
>> publish a piece of RDF that said, in effect, please prove that I can
>> get flowers from you amounting to this quantity before this date at
>> less than this price (and then you and I have a deal).
>>
>> This might well be a natural way to deal with 'queries', in fact, in
>> a commercial B2B context; the logic is the same, so why not allow it
>> as a possibility? In other words, such a publication is a kind of
>> open-ended query in which the KB - ie the identification of the KB -
>> is itself part of the answer.
>
>Interesting. I have been thinking of a query as asking what is entailed
>by a given logical theory.
Right, me too.
>I think you are suggesting that we expand
>that notion to include asking what is true in some domain of discourse
>like our consensus reality.
Well, you might phrase it that way, but what I had in mind was
something more restricted, ie allowing logical theories to choose
themselves to 'match' a query.
>That would mean, for example, that a query
>could ask "Who is the Chief Justice of the U.S. Supreme Court" (to take
>a random example), or as you suggest, ask a query about a Web service.
>If we do that expansion, what constitutes a correct answer to such a
>query.
Well, the answer is still what it was before, but now it includes a
specification of the KB that is providing the answer. The only
difference is who chooses the KB: is the query always directed to a
particular KB, 'chosen' by the query itself, or can a query be
directed to a community of KBs, and thereby invite many, possibly
different, responses?
>What if a server said that I was Chief Justice of the U.S.
>Supreme Court?
Well, if that is what it said, then it could presumably be held
responsible for any valid conclusions from that assertion.
>When a query is about entailment in a given logical
>theory, we know what the test is for correctness of an answer.
I wasn't suggesting changing that, only allowing the query to be
answered by several different 'theories'. This is a web context,
after all, and presumably we have to expect that there will be many
ontologies in use.
>I don't
>know what a query is about when it is not with respect to a given
>logical theory.
Well, it is exactly the same, but it invites any theory to prove it.
The logic is the same.
>Of course, an easy case is where a knowledge base is
>not specified in the query but is specified in a query answer so that
>the answer specifies a sentence that is entailed by the knowledge base
>that is referred to in the answer.
Right, exactly. That is what I was suggesting.
>
>> >* Premise - I have proposed that a query optionally include a premise to
>> >facilitate if-then queries while still remaining within the
>> >expressiveness of DAML+OIL. Specifically, I have proposed that a
>> >premise be an arbitrary DAML+OIL knowledge base. There has been no
>> >formal agreement on whether or not DQL will allow a "query premise".
>>
>> I would vote not, in the first draft. It smacks of tiptoeing into
>> 'rules' territory, and it ought to be definable in any case by
>> querying a KB containing the premise and an import of the previous KB.
>
>Well, we have discussed this before.
Right, and I still think that the answer should be not, in the first
draft. I would like to keep the query language as simple as possible
at first, and solve the problems a few at a time.
> The primary motivation is to allow
>queries to be stated using only DAML+OIL (no rules) that hypothesize and
>describe objects and then ask a question about the hypothesized objects.
That seems like a potential minefield to me, and one that we should
venture into carefully.
>
>> >answer will include a binding for each distinguished variable. I am
>> >referring to the variables in the query pattern that are not
>> >distinguished variables as "non-distinguished variables".
>>
>> undistinguished variables?
>
>From a quick check on the Web, I find them being called
>"nondistinguished variables".
>
>> >WHAT'S IN A QUERY RESPONSE
>> >
>> >* Query - The query to which this is a response.
>> >
>> >* Server - The server that produced this response.
>>
>> ? This seems rather like having a piece of code sign its name to
>> everything it does. Surely, if I am querying a KB, I already know
>> what the query was. Why do I need to be told this again?
>>
>> BUt in any case, what exactly *is* the 'server' here? You seem to be
>> assuming that servers are genuine things on the web, but that seems
>> to be something that we havn't really decided on yet. How does DAML
>> refer to agents, so it can express this response? (Or indeed to
>> queries, for that matter)?
>
>I was assuming that a server has a URI.
OK. But then we need on ontology of servers, right?
>
>I suppose it is not critical that a query response contain a pointer to
>the query to which it is a response, but it certainly needs to contain a
>pointer to the server, since the answers contained in a response are
>server-specific in that those are the answers that that server did
>produce to the query. (Yes, each answer is correct or not regardless of
>who produced it, but the set of answers is server-specific in that those
>are the ones that were produced by the server out of all the possible
>correct answers.)
>
>> >* Answers - Zero or more answers to the query.
>>
>> Right, but how is 'zero answers' indicated?
>
>I don't see that as being a problem. A query result would necessarily
>have some sort of collection (e.g., a list) of query answers, and if
>there are zero answers then that collection would be empty.
OK again.
Pat
--
---------------------------------------------------------------------
IHMC (850)434 8903 home
40 South Alcaniz St. (850)202 4416 office
Pensacola, FL 32501 (850)202 4440 fax
[email protected]
http://www.coginst.uwf.edu/~phayes
This archive was generated by hypermail 2.1.4 : 04/02/02 EST