[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Raw chat text of 2008-03-10 telcon
Khanderao: scribe:Khanderao anish: approval of april 3rd 2008 minutes: http://lists.oasis-open.org/archives/sca-bpel/200804/msg00012.html Khanderao: Action Item : Editors to apply changes in the doc: Michael Rowley anish: Agenda: http://www.oasis-open.org/apps/org/workgroup/sca-bpel/event.php?event_id=16249 Khanderao: No update from Detier on Action Item 33. Khanderao: Action Item 33 transferred to MikeRowley Khanderao: Action Item 33 and 34 assigned to Mike Rowley Khanderao: No update from Detier on Action #0034: Editors to revise some spec sections to apply RFC 2119 terminology while using concrete conformance targets . Tentative ETA next week. Khanderao: Re: Action Item #0036 MikeE to raise an issue with the liaison subcommittee regarding RelaxNG/Schematron Khanderao: Action Item 36: Done Khanderao: Issue Discussions 13 and 18 Khanderao: Issue 16 postponed for future Khanderao: b) Issue 15 http://www.osoa.org/jira/browse/BPEL-15 TITLE- Define Conformance Targets SUBMITTED BY- Martin Chapman Khanderao: Anish: Do we need to approve conformance target ? anish: martin's original email: http://lists.oasis-open.org/archives/sca-bpel/200803/msg00000.html anish: my response: http://lists.oasis-open.org/archives/sca-bpel/200803/msg00014.html anish: 1st meeting minutes: http://lists.oasis-open.org/archives/sca-bpel/200803/msg00048.html anish: 2nd meeting minutes: http://lists.oasis-open.org/archives/sca-bpel/200804/msg00001.html Khanderao: MartinC : As a conformance test we need define BPEL component type generation rules in terms of RFC-2119 . Specs does not say/should not say who/which tool does it. Mike Edwards: the compoennt type is testable however Mike Edwards: even though it is virtual Khanderao: Anish: Conformance target could be BPEL file or ComponentType if it exists Khanderao: MIkeE: Even if Component Type virtual, it is testable. Khanderao: MikeE: It is unwise to assume "how" generation is done Khanderao: MikeRowley: Testing is output or generator of the output? Mike Edwards: +1 Khanderao: MikeRowley: what is tested? Anish's example gives examples of tests like generated artifacts. Testables should be take input and generate testable constraints. We do not need to say "where" but should be able to say that my software does it. Khanderao: MikeE: We should concentrate on consequence and not the means. Mike Edwards: it could be done in a number of places, there is no one entity that would take a BPEL process and derive its componentType Khanderao: MartinR: We are defining fine grained conformance constraints. Hence we should be able to test constraints like "statement x results into y". Khanderao: Issue discussion: No conclusion. However, Anish suggested MikeR to get pick up a section that talks about componentType and propose examples of conformance constraints and target. Discussion would continue over email and future meeting.
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]