[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: RE: [regrep-semantic] PR2: Explicit support for OWL Ontology/Class in place of ClassificationScheme / ClassificationNode
Being an admitted R/R newbie, please entertain this as a possibly dumb question: How does explicitly supporting stored knowledge representation payloads inside the R/R actually allow R/R users to do anything different or better than what they can today? IMHO, unless there is a different R/R representation (eg: model format or API) then users will always (as Carl pointed out) have to first query, then classify, then choose the right semantic tech tool -- precisely because an OWL-DL reasoner does fairly different things than an OWL-F or RDF or KIF reasoner (resulting in different usage patterns/cases). To me, the bottom line is 'what new capabilities are enabled through semantic technologies?' ... and if we are saying that users will have to nest decision break points based on the "kind" of knowledge they are getting from the R/R, then the true "semantic" part of SCM is actually occurring well outside of the R/R in actual practice. Am I missing something big here? -Jeff- -----Original Message----- From: Carl Mattocks [mailto:carlmattocks@checkmi.com] Sent: Wednesday, March 17, 2004 10:45 AM To: Zachary Alexander Cc: 'Registry TC - SCM SC' Subject: RE: [regrep-semantic] PR2: Explicit support for OWL Ontology/Class in place of ClassificationScheme / ClassificationNode Zachary: Acknowledged - if the 'Ontology payload' (OWL or otherwise) is stored in an engine outside of the R/R we should discuss what form of 'external semantic technology' could be supported. I think we have agreed that - wherever it is stored there is a need for a Registry (alone) classification (thus) discovery of semantic content capability... YES ? cheers carl <quote who="Zachary Alexander"> > Carl, > > I completely disagree with premise that RDF and OWL represent > significantly different technologies. I think that RDF and OWL > represent payloads. I think that utilization of an external > classifier and/or inference engine would represent a different > technology. I think that the utilization of an external classifier > and/or inference engine would preserve the spirit of the ebXML version > 1.06 requirements. > > I would support defining mechanism that would better facilitate the > use of external "semantic technologies" because that will allow the > application developers the option of determining how best satisfy > their customers. I think that the predisposition toward OWL is > extremely shortsighted. What is the potential that other knowledge > representations may rival OWL in the next two to three year time > frame? What is the potential that OWL may morph to better support > rules? My belief is that the reason for the payload neutrality ebXML > version 1.06 requirements is because of questions like the ones that I > just asked. > > Zachary Alexander > The IT Investment Architect > ebTDesign LLC, (703) 283-4325 > http://www.ebTDesign.com | http://www.p2peconomy.com | > http://www.itinvestmentvehicle.com > > > -----Original Message----- > From: Carl Mattocks [mailto:carlmattocks@checkmi.com] > Sent: Wednesday, March 17, 2004 12:57 PM > To: Zachary Alexander > Cc: 'Registry TC - SCM SC' > Subject: RE: [regrep-semantic] PR2: Explicit support for OWL > Ontology/Class in place of ClassificationScheme / ClassificationNode > > Thus you agree with the need for PR2.. > > <quote who="Zachary Alexander"> >> <Farrukh from Charter>"The SCMSC will identify specific Semantic Web >> technologies (e.g. RDF, OWL) that are necessary to support the >> requirements identified for semantic content management." </Farrukh > from >> Charter> I think that RDF and OWL are knowledge representations. I > think >> that knowledge representations are payloads. I would classify >> technologies as things like classifiers, and inference engines. I > think >> of semantic supporting technologies as generic not OWL specific. >> >> >> Zachary Alexander >> The IT Investment Architect >> ebTDesign LLC, (703) 283-4325 >> http://www.ebTDesign.com | http://www.p2peconomy.com | >> http://www.itinvestmentvehicle.com >> >> >> -----Original Message----- >> From: Farrukh Najmi [mailto:Farrukh.Najmi@Sun.COM] >> Sent: Wednesday, March 17, 2004 12:14 PM >> To: Zachary Alexander >> Cc: 'Registry TC - SCM SC' >> Subject: Re: [regrep-semantic] PR2: Explicit support for OWL >> Ontology/Class in place of ClassificationScheme / ClassificationNode >> >> Zachary Alexander wrote: >> >>> <Jeff> Is PR2 the appropriate context to ask questions regarding the >>> (a) "type of OWL" and (b) "where the OWL interface lies?" Or are > these >> >>> design questions that I can ask for clarity on later? </Jeff> I >>> don't know what to tell you. The direction of this subcommittee >>> seems to have changed. Originally, it was suppose to address the >>> issues surrounding the query and life cycle management of semantic >>> objects. Now it has become about how best to support OWL Explicitly. >>> When I > see >> >>> terms like explicit, I think that the result will be hardwiring. (a) >>> The discussion have centered on the most popular forms of OWL which >>> appear to be OWL DL. (b) I think that this discussion is suppose to >>> lead to modifications to the ebXML Registry which will eliminate the >>> need for an OWL interface. >>> >> Our charter is posted at: >> >> > http://www.oasis-open.org/apps/org/workgroup/regrep-semantic/descripti > on >> .php >> >> Part of the charter I quote below: >> >> " >> The SCMSC will identify specific Semantic Web technologies (e.g. RDF, >> OWL) that are necessary to support the requirements identified for >> semantic content management. " >> >> I believe PR2 is very much within the spirit of the original charter. >> Please recall that the "P" in "PR" is for "Proposed". We are >> brainstorming on requirements. >> >> Please understand that I do not have any hidden agendas here. I do >> not have an OWL implementation or product I am looking to peddle. > I >> >> am just doing the best I can to keep >> ideas and discussion flowing within the SC. >> >> -- >> Regards, >> Farrukh >> >> >>> Zachary Alexander >>> >>> The IT Investment Architect >>> >>> ebTDesign LLC, (703) 283-4325 >>> >>> http://www.ebTDesign.com | http://www.p2peconomy.com | >>> http://www.itinvestmentvehicle.com >>> >>> -----Original Message----- >>> *From:* Jeffrey T. Pollock >>> [mailto:jeff.pollock@networkinference.com] >>> *Sent:* Wednesday, March 17, 2004 11:26 AM >>> *To:* 'Farrukh Najmi'; 'Zachary Alexander' >>> *Cc:* 'Registry TC - SCM SC' >>> *Subject:* RE: [regrep-semantic] PR2: Explicit support for OWL >>> Ontology/Class in place of ClassificationScheme / ClassificationNode >>> >>> Farrukh- >>> >>> Is PR2 the appropriate context to ask questions regarding the (a) >>> "type of OWL" and (b) "where the OWL interface lies?" Or are these >>> design questions that I can ask for clarity on later? >>> >>> Specifically: >>> >>> (a) discussion of the tradeoffs and consequences between OWL-F and >> OWL-DL >>> >>> (b) if the regrep gets queried as usual (and returns an OWL ontology >>> as a 'blob') or if there are extensions to allow a reasoner to >>> interface the regrep directly (allowing inferencing against the > regrep >> >>> APIs). >>> >>> Thanks for your guidance and clarification. >>> >>> -Jeff- >>> >>> -----Original Message----- >>> *From:* Farrukh Najmi [mailto:Farrukh.Najmi@Sun.COM] >>> *Sent:* Wednesday, March 17, 2004 7:47 AM >>> *To:* Zachary Alexander >>> *Cc:* 'Registry TC - SCM SC' >>> *Subject:* Re: [regrep-semantic] PR2: Explicit support for OWL >>> Ontology/Class in place of ClassificationScheme / >> ClassificationNode >>> >>> Zachary Alexander wrote: >>> >>> Farrukh, >>> >>> I think that this violates ebXML version 1.06 requirements. The >>> ebXML registry should be payload neutral. I think that this > should >>> trigger a change in the charter of this subcommittee. I think the >>> charter should be changed to explicitly state that this >>> subcommittee is dedicated to creating an OWL based ebXML > Registry. >>> >>> I said nothing in the PR2 about how the requirement is met. In no >>> way does the requirement imply hardwiring OWL in ebRIM. >>> Lets focus on teh requirement and not how it is going to be >>> addressed at this stage. >>> >>> Zachary Alexander >>> >>> The IT Investment Architect >>> >>> ebTDesign LLC, (703) 283-4325 >>> >>> http://www.ebTDesign.com | http://www.p2peconomy.com | >>> http://www.itinvestmentvehicle.com >>> >>> -----Original Message----- >>> *From:* Farrukh Najmi [mailto:Farrukh.Najmi@Sun.COM] >>> *Sent:* Wednesday, March 17, 2004 9:04 AM >>> *To:* Registry TC - SCM SC >>> *Subject:* [regrep-semantic] PR2: Explicit support for OWL >>> Ontology/Class in place of ClassificationScheme / >> ClassificationNode >>> >>> *PR2. Explicit support for OWL Ontology/Class in place of >>> ClassificationScheme / ClassificationNode* >>> >>> Allow use of an OWL Ontology in ebXML Registry wherever we use >>> ClassificationSchemes in Version 3. >>> Allow use of an OWL Class in ebXML Registry wherever we use >>> ClassificationNodes in Version 3. >>> >>> *Motivation: *Enable multiple-inheritance which was not possible >>> in ClassificationScheme. Enable use cases 4,5,6,9 >>> >>>-- >>> >>>Regards, >>> >>>Farrukh >>> >>> >>> >>> >>>-- >>> >>>Regards, >>> >>>Farrukh >>> >> >> >> >> > > > -- > Carl Mattocks > > co-Chair OASIS ebXMLRegistry Semantic Content SC > CEO CHECKMi > v/f (usa) 908 322 8715 > www.CHECKMi.com > Semantically Smart Compendiums > (AOL) IM CarlCHECKMi > > -- Carl Mattocks co-Chair OASIS ebXMLRegistry Semantic Content SC CEO CHECKMi v/f (usa) 908 322 8715 www.CHECKMi.com Semantically Smart Compendiums (AOL) IM CarlCHECKMi
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]