- From: Deborah McGuinness <dlm@ksl.stanford.edu>
- Date: Wed, 05 Dec 2001 16:20:26 -0800
- To: Jeff Heflin <heflin@cse.lehigh.edu>
- CC: hendler@cs.umd.edu, webont <www-webont-wg@w3.org>
i dont know that i have the "answer" but i have a few thoughts. - i agree with jeff that contradiction/inconsistencies is a requirement for interoperability however it is really a requirement of any application that wants to do any kind of verification and/or inference. this would include services applications. i think contradiction detection is more like search in that it is a technical issue that would be needed across most applications. - it feels to me like the first 2 categories jim (and later jeff) suggested are different from the next 1 or two categories. It seems to me that basically any web application will need to address issues of content interoperability thus topic 3 is also an area that will be addressed across all applications. I would claim that content interoperability is more a "technical issue" as i interpret jim's use of the phrase rather than an application category. Jeff Heflin wrote: > I think we can divide the use cases into three classes: > > 1) Web services > 2) Information management (this includes archives, catalogs, large data > sets, etc.) > 3) Content interoperability > > Note that I left out Jim's fourth category adapation of content to > user/device. I think that this is really just a form of dealing with > content interoperability. As I see it, the content exists in some form > (ontology), and needs to be translated to another form (ontology) for > use by a different user or device. Thus this is about the > interoperability of content. > > Real time/sensors (the remainder of Jim's fourth category) seems to me > to be closely related to web services. For example, Ned Smith described > a use case in which a sensor is embedded in a network and interacts with > other devices in the network. I think this is analagous to agents using > and providing services to other agents. Sensors may also fit under > content interoperability, particularly if you're dealing with sensors > that have different ontologies. > > Conceptual open-hypermedia seems to me to fit best under my information > managment category. From what I understand in Nick Gibbins' recent > message, it is a technique for locating information in large data > sources. > > Finally, I think that condradiction/inconsistencies is really a > requirement of content interoperability, and not a use case in its own > right. When we integrate heterogeneous content, then we need a means for > detecting and resolving inconsistencies. > > Any comments/thoughts? > > Jeff Heflin > Lehigh University > > Jim Hendler wrote: > > > > WOW-Gers > > > > I have taken a stab at a categorization of the use cases - I find 3 > > natural categories, 1 almost a natural category (two ideas that might > > be related, but don't fit totally) and only a couple of loose ends I > > could not figure out exactly where to fit. In addition, some > > technical issues that might cross cut seem to come out. There are > > listed below. > > I would be happy if we could end up with 4 total groups -- winnowing > > my 4 categories down to 3, and developing the "cross cutting > > technologies" to become more focused (to be the base for the sort of > > "requirements" Peter Patel-Schneider argued for earlier) > > > > Please discuss and let's see if we can resolve by Thurs. > > Jim H > > > > ========== > > [Note WOL is only an acronym for Web Ontology Language at this point- > > does not represent commitment to this name] > > > > Use cases - rough categorization proposal: > > > > 1) Web Services > > WSDL is only a starting place, seems to stress interconnection, but > > not content. WOL has potential to be used for better advertising > > (via hierarchy/classification), for connecting advertisements to > > ontologies, and for exploring the compositionality of services. > > > > 2) Archives/CatalogsLarge data (or image) sets/web site management > > All of these areas focus on the use of ontologies in the management > > of large scale information sources. Includes need for matching, > > classification, default reasoning. "Traditional" metadata (i.e. > > document markup) would fall in this category. > > > > 3) Content Interoperability (a/k/a/ agent markup) > > RDF has advantage over XML in allowing easy merging of content found > > on different sites/resources, and the use of the combined sources. > > Use cases include linking of databases (DB schemas), coupling data to > > pages, linking instance data to ontologies. Also allows linking of > > ontology to ontology for mapping of vocabulary, etc. > > > > 4) Adaptation of content to user/device (real time/sensors?) > > Use of ontologies to help determine what info to show to whom when, > > or to be the information shown (i.e. in PalmDAML the user can browse > > the semantics separate from the source pages). The real-time and > > embedded sensor area seems to fit in best here - but I admit I'm > > pushing a little. > > > > Loose ends: > > open hypermedia > > contradiction/inconsistencies > > > > Technical issues that could be address in some or all > > versioning > > ontology-based search > > domain-mapping/ontology linking (how much is commited to by a link) > > ontology querying > > rapid creation of large ontologies ? > > > > -- > > Professor James Hendler hendler@cs.umd.edu > > Director, Semantic Web and Agent Technologies 301-405-2696 > > Maryland Information and Network Dynamics Lab. 301-405-6707 (Fax) > > AV Williams Building, Univ of Maryland College Park, MD 20742 > > http://www.cs.umd.edu/users/hendler -- Deborah L. McGuinness Knowledge Systems Laboratory Gates Computer Science Building, 2A Room 241 Stanford University, Stanford, CA 94305-9020 email: dlm@ksl.stanford.edu URL: http://ksl.stanford.edu/people/dlm (voice) 650 723 9770 (stanford fax) 650 725 5850 (computer fax) 801 705 0941
Received on Wednesday, 5 December 2001 19:16:41 UTC