Estimation and Measurement Telecon, 2009-09-11

See Weekly Meeting Logistics for telecon information.

Attendees

AndyBerner, AndrewCanham, ArthurRyman, LawrencePutnamJr, LeeFischman, ScottBosworth

Minutes

1. Discuss Milestone Plan

We need to starting finalizing the usage scenarios and draft spec. Target completion is 3Q2009.

Use Cases

Our first target is to complete the usage scenarios as described in Software Estimation Use Cases in 3Q2009. Scott said that the criteria for declaring these done is determined by the workgroup, i.e. there is no additional OSLC process.

Larry volunteered to edit the use case document and lead the review and approval process within the workgroup.

ACTION: LawrencePutnamJr will edit the Software Estimation Use Cases and review this with the workgroup. DONE 2009-09-18

There was a lot of discussion on the Reestimating use case led by AndyBerner. We need to update the use cases based on this discussion Andy will summarize this.

ACTION: AndyBerner will summarize the discussion on the Reestimating use case.

Primer

The Primer is not somewhat out-of-synch with the specification. Arthur will update it. AndyBerner volunteered to add more detail about some of the metrics.

ACTION: ArthurRyman will update the Primer sequence diagrams and XML formats to reflect the current state of the specification. DONE Sequence Diagram 2009-09-18

ACTION: AndyBerner will add more information about some metrics to the Primer. DONE 2009-09-18

Reference Implementation

We should target November-December to develop a reference implementation and test it with software estimation tools. IBM will develop an implementation of the service and PPM clients that access it. Larry and Lee confirmed this would be a feasible timeframe for them to development software estimation tool clients that access the service.

2. Fact Tables - SteveAbrams

At the last meeting Steve offered to simplify the Fact Table XML format.

Steve did not attend. We will defer this to next meeting.

3. Role Maps - ArthurRyman

At the last meeting Arthur offered to propose a format for including the mapping from non-standard to standard role. See RoleMap

We reviewed the proposed design. Lee initially had concerns about the need to express a many-to-many relation between the standard and non-standard roles, expressed as a matrix, but withdrew this after discussion.

The current designs shows the standard roles as text labels. These should be URIs to align with RDF.

ACTION: ArthurRyman will propose URIs for the standard roles

The current design shows the role map embedded in a fact table. However, in practice it is unlikely that the role map would be varied between fact tables for a given project and tool, i.e. the role map would be a global constant within a project and tool combination.

ACTION: ArthurRyman will make role maps a top level resource that can be referenced by fact tables.

4. Other Business

None.

Adjourned.

Comments

Add your comments here: