[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: [ubl-ndrsc] Minutes for 15 May 2002 UBL NDR SC meeting
Minutes for 15 May 2002 UBL NDR SC meeting 1. Roll call (quorum is 8) Straw poll on local vs. global * Bill Burcham YES L * Mavis Cournane YES L? * Mark Crawford awol * Fabrice Desré YES G(Q) * Matt Gertner regrets * Arofan Gregory YES G(Q) * Jessica Glace YES G(Q) * Eduardo Gutentag YES (dropped y:20) * John Larmouth regrets * Eve Maler YES ? * Sue Probert YES (x:58) ? * Lisa Seaburg YES G(Q)? * Gunther Stuhec YES (y:02) L? * Paul Thorpe YES ? Phil Griffin has dropped to observer, at least for now. Quorum reached. 2. Acceptance of minutes of previous meetings 1 May 2002 http://lists.oasis-open.org/archives/ubl-ndrsc/200205/msg00002.html 8 May 2002 http://lists.oasis-open.org/archives/ubl-ndrsc/200205/msg00007.html Both accepted. 3. Adoption of agenda Adopted, with addition of #4a. 4. Status review Everyone: - A: Review the NDR document. DONE Lisa: - A: Provide examples of metadata enumerated lists. DONE Mavis: - A: Tag structure/NDR changes. IN PROGRESS Eve: - A: Code list paper. IN PROGRESS Bill: - C: Role model paper. Arofan: - B: Forward the latest version of the xCBL versioning paper. - B: Containership paper. Gunther: - B: Facet paper. DONE - C: Date/time paper. Fabrice: - A: Make proposal on local vs. global elements. Summary of work to be done: A Code lists IN PROGRESS A Modnamver IN PROGRESS A Review the purchase order schema design IN PROGRESS A Local vs. global elements -- round 2 A Officially decide elements vs. attributes DONE A Officially decide empty elements DONE A External recommendations for changes to CCTS DONE A Global attributes DONE B Facets B Tag structure IN PROGRESS B Containership B Fixed vs. varying types C Dates and times 4a. Arofan report on EIDX liaison committee Electronics industry group that is cross-referencing the core fields in every important document type from their business perspective, and has produced a comparison matrix. Arofan will bring or mail copies of it to the F2F. 5. NDR document comments It's in good shape, but needs to incorporate new code list and modnamver decisions, as well as mention the implicit rule about *ContentType simple types inside complex types. 6. Code lists We voted on accepting the "namespaced type hybrid method". Accepted with one abstention from Jessica. We agreed that the instance extension method should still be described as a (failed) contender. 7. Modnamver "Wake up" the following topics: - Separate RT/CCT module There seems to be some interest in breaking down the UBL "core" into multiple core-ish files, for both memory management reasons (the C1 folks experimenting with Xerces report this) and for reasons of reusing only the parts one wants (some verticals seem to want to reuse the built-in ebXML CCT semantics in a neat package). There's a question about whether such a low-level module needs its own namespace, but it needs one if you are worried about memory management. There's also a question about what we would call this module: Is "CCT" incorrect, given our comments on CCTS? "Leafy things" is too informal. :-) They are sort of "built-in UBL types"; would this be a good name? But other UBL types will be built in to UBL too, by definition. We agreed on "common [UBL] leaf types" (CLTs or CULTs!) for the CCT-ish (basic) stuff, and "[UBL] common aggregate types" (CATs) for the aggregate stuff. - How many root schemas We began questioning whether we need both a functional area module and an instance root schema that defines only the top-level element (the message root). We voted on combining the instance root schema and the root schema, such that a small handful of message-level (top-level) element declarations will be included in a single functional-area root schema module. Accepted with Eduardo abstaining. 8. Local vs. global elements Fabrice's assessment of our original decision is that it was based on some false assumptions. For example, we seemed to treat "qualified vs. unqualified" as if it meant "prefixed vs. unprefixed". Our straw poll indicates that we're open to reconsideration. Arofan's recent experience with his engineers is that they really want global elements because it's hard to handle fragments without them. We don't want to reopen our decision about making our types global and referenceable. But if we go back to global elements, we need to have rules about which "global thing" to reuse and when. Our ground rules in making this reconsideration: Any new proposal needs to fit itself into the picture created by our existing decisions, including our element naming decision. Some potential requirements: - Parse fragments of documents easily - Avoid invasive changes to XPaths when repurposing XSLT from operating on standard UBL to customized UBL - Content model must be able to include two elements of same type - Facilitate reuse of UBL components with the context methodology (this is fuzzy) - Make Matt's life difficult by making firm decisions without any substance behind them :-) 9. Update the schedule through May May 22: modnamver (incl. versioning), local vs. global, facets May 29: approve NDR document for distribution June 3-7 (the F2F meeting itself, if all goes according to plan) Who is known to be going to the F2F? Mavis, Arofan, Jessica, Sue (mostly in LC), Lisa, Gunther, Eve, Mark. It looks like we'll have quorum. 10. Adjourn Adjourned y:47. -- Eve Maler +1 781 442 3190 Sun Microsystems XML Technology Center eve.maler @ sun.com
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC