History   View   Links to this page 2013 May 1 | 08:30 am

Time: 11:00AM Eastern US (contact MichaelFiedler if you’d like to participate)

Agenda

Minutes

Attending: Michael Fiedler, David Liu, Martin Pain, Stephen Rowles, Nick Edgar, Jing Qian, Charles Rankin, Paul McMahan, John Abbott, Heather Fraser-Dube

  • New workgroup participant
    • Nick Edgar introduced Heather Fraser-Dube from the IBM Rational Team Concert build and automation team
  • V2 Specification issues
    • Nothing new this week

  • V3 Scenarios

    • Temporary deployment scenarios
      • Martin walked the workgroup through the flow diagram
      • Several questions and discussion points were raised
        • Is the lifetime of a result contribution generally the same as the lifetime of the result itself? Not necessarily - the result may be a more or less permanent record of what happened during execution while the contribution (log file, build binaries, deployment, etc) could be cleaned up when no longer needed.
        • Charles asked if there is a distinction between interest in a result vs. interest in a result contribution (or a new resource representing a deployment or other result contribution). For consideration.
        • Stephen raised the question of how much behavior described in the flow should be enforced. That is, how much of the behavior in the scenario is a MUST requirement - or at least a MUST once the provider has opted in to this scenario. Tradeoff between requiring too much of the provider vs. making life harder for the consumer.
        • How is a “teardown action” represented? Possibility: a result has contribution representing a deployable entitly with link back to a teardown automation plan - does this cause artificial creation of resources (extra automation plans/requests/results)? too heavy for something that might be simple?
        • Comment from Charles that there might be relationships between this scenario and the orchestration/lifecycle scenarios.
      • TODO: Charles will try to find time to put some comments on the list
      • TODO: Michael Fiedler will try to summarize a secondary teardown scenario not related to deployment

  • Next meeting: 2 May at 11AM Eastern US