Skip to main content

Notice: this Wiki will be going read only early in 2024 and edits will no longer be possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.

Jump to: navigation, search

COSMOSF2F-08-May-07

The focus will be on refining the design of the data collection framework and prioritizing its delivery.

Here are a few agenda items to get us started - please add your own

Logistics

May 8 - 10, 2007

Computer Associates has been kind enough to host this meeting

Training Room,

8th floor, 125 Commerce Valley Dr. West,

Thornhill, Ontario, Canada, L3T 7W4

  • Need to have regular updates at the end of each day

Requirements

  • Requirements for Persistence Framework
  • Requirements for additional framework features
    • Out of box experience
      • Initial user experience
      • Install/packaging
  • Requirements for additional components
  • Add detail to all requirements to facilitate planning and design
  • Resource Modeling
    • Review design for repository API
  • Data Visualization
    • Discussion around the deployment and programming models for the user interface.

Architecture

  • High level design of functional enhancements
  • High level design of exemplary components
  • Alignment of query interface with TPTP and Data Reporting
  • Review of the SML definitions for Data Collection and Managed Component resources ( RM project, requested by Valentina )

Planning

  • Prioritization of component and feature delivery
  • Tentative assignment of resources to deliverables
  • Discuss features for post 1.0 releases

Process

  • Planning for the next release. Define schedule and project direction for COSMOS 1.1 ( or 2.0 ? )
  • Release engineering/build - clarify who is responsible for leading this component. I understand Hubert is just helping with setting up the build until a replacement is made available. Do we have any plans moving forward?
  • Define an iteration testing process. During the test pass cycle executed at the end of every iteration all projects are expected to run testcases and report results. We should clearly describe the process and the iteration test pass exiting criteria. At what point can a project declare that the testpass has been completed? (Is it when all testcases are run and execution is logged in CVS? Or when all testcases are marked as pass? )



Minutes

A first pass of the detailed minutes has been posted. File:COSMOS F2F-May 8-10 2007-minutes.zip



Minutes from Tuesday

Attendees

From IBM:

  • Mark Weitzel
  • Harm Sluiman
  • Valentina Popescu
  • Ali Mehregani
  • Sheldon Lee-Loy
  • Hubert Leung

From CA:

  • Chris Craddock
  • Charles Connelly
  • Wilson MacDonald
  • Jimmy Mohsin
  • William Muldoon
  • Martin Simmonds

From Compuware:

  • Don Ebright
  • Joel Hawkins



Agenda For Wednesday

8:30am -11:00am

  • Use Cases
    • What does the user look at?
    • What happens when a data collector comes up?

11:00am – 1:00pm

  • COSMOS Domain
    • Bootstrap / Startup

1:00pm – 4:00pm

  • define what SML/WSDL/RMD/XSD documents we need at each step
    • look for convenience API to sit on top of “generic” API
    • where possible, look to standardize on the return types, e.g. CBEs, SDMX stat crap,
  • Define the interfaces for query API
    • repository API

4:00pm – 6:00pm

  • Release plan
    • Post June plans
  • Discussion on how Tue/Wed conversation mapped into what we have today
    • What the heck to do with the control interface?
      • Model in SML compliant XML instance Documents (SXID)
  • Define the work items (bugzillas)

6:00pm Summary Call


Attendees

From IBM:

  • Mark Weitzel
  • Valentina Popescu
  • Ali Mehregani
  • Sheldon Lee-Loy
  • Hubert Leung

From CA:

  • Chris Craddock
  • Charles Connelly
  • Jimmy Mohsin
  • William Muldoon
  • Martin Simmonds

From Compuware:

  • Don Ebright
  • Joel Hawkins

Agenda for Thursday

Thursday 8:30 - 3:00pm COB

  • Governance—eclipse development process
    • code reviews
    • test passes
  • Release engineering
  • User interface programming model
  • define how we collaborate with TPTP on new blue statistical stuff
    • common database schema for stats
    • common db for events
    • resource repository API


Attendees

Thursday

From IBM:

  • Mark Weitzel
  • Valentina Popescu
  • Ali Mehregani
  • Sheldon Lee-Loy
  • Hubert Leung

From CA:

  • Chris Craddock
  • Charles Connelly
  • Jimmy Mohsin
  • William Muldoon
  • Martin Simmonds

From Compuware:

  • Don Ebright
  • Joel Hawkins

Back to the top