Goals: (1) Link concepts of the UBL Invoice knowledge base into some upper ontology. (9VB)
(1a) deliverable: KIF-based ontology containing all of the concepts of the UBL invoice/order knowledge base plus all other concepts needed to integrate those business concepts into a general ontology. (9VC)
(1b) TimeTable: 6 months (June 30) to get preliminary ontology with all UBL concepts 1 year to have this content reviewed by UBL participants as well as Ontolog participants, and make recommended changes. (9VD)
(2) Create a demonstration application that uses the Ontolog ontology in a semi-realistic scenario/use case. (9VE)
Reference: Project analysis page -- http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologPlanning_2004 See gap analysis: "Lack a real customer and/or test application software program that uses the ontology " http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologPlanning_2004#nid0135 (9VF)
Explanation: The number of potential ways to represent concepts and the range of concepts that could be included in an ontology are very large. Different people will differ on how the ontology should be built and no objective method to decide on content (except for logical consistency) is available unless there is either: (1) a "customer" for whom the ontology is being built and who can provide specifications to determinewhat must and what must not be included, or (2) a working application program that uses the ontology whose measurable behavor will provide an objective criterion for deciding which of two or more alternatives works better for the specific purpose. (9VG)
(2a) deliverable: A functional program that uses the ontolog ontology in some typical business scenario. (9VH)
(2b) TimeTable: 6 months to decide what the specifications for a demonstration application would be 1 year to have a portion of the application operational (9VI)
(2c) Conflict How do we resolve disagreements on content? Reference "Gap" issue: Lack effective conflict resolution (how to resolve disagreements) (0127) Explanation: Since the number of alternatives in representing complex concepts can be high, there is a substantial likelihood that on many issues there will be an absence of unanimity when faced with choices of conflicting alternatives. We need a method to expeditiously resolve such disagreements, make a choice, and move on to other issues. Although in standards efforts unanimity is highly desirable, we must recognize that it may not happen for knowledge representation standards. I suggest that, after discussion of an issue, if there is not unanimity, we proceed quickly to a vote, requiring at least 80% agreement to resolve a question. (9VJ)
* * * * * * * * * * * * * * * * * * (9VK)
Drivers/Constraints: (9VL)
(1) lack solid project plans: --> work groups need specific commitments to do specific tasks within specific time frames (9VM)
(2) need agreement on what content to write and how reviewed vs just writing something --> we need to set some initial priorities as to what content should be formalized first (9VN)
18 months to have an application that can assist in evaluating a UBL-inclusive ontology. (9VO)
* * * * * * * * * * * * * * * * * * (9VP)
Strategies (9VQ)
(1) every one has a reason for doing this, but not sure that everyone has the same objectives --> create working groups for specific subtasks (9VR)
(2) lack solid project plans --> each person should propose an immediate objective and the group should select the priorities (9VS)