TestMtg, July 9, 2007
Attending: Ronak, John, Zack, Cameron, Jim, Kim
- Discussed the B4 units and distribution. Both India and Westford test groups will get a few B4 units to add to their test beds.
- We went over the 100 laptop test. We will start setting up this week. One of each SKU (variations in hardware vendors) will go into the permanent mesh. We have added activation and updating to the tests that will be included.
Ed Russeau had sent an email last week outlining some of the areas that we should really improve on as we head towards FRS (first revenue ship). We addressed a number of those points:
- Trial-2 should really be thought of as alpha for FRS.
- MP (mass production) should be thought of as beta. We may call this Trial-3 since alpha and beta are not terms used at OLPC.
- FRS for software release is targeted for mid-October, which is when we believe the first customers will be getting the production laptops. They will be highly encouraged (required?) to upgrade. This will be v1.0. I will try to reflect these dates in the roadmap.
- Test people would like to see a more detailed schedule. Kim will create one and figure out the best way to publish it. We talked about Task Juggler.
- Each of these releases should have its release criteria.
- Ideally we would like a good list of requirements for these releases -- enumerated and testable. Kim will start by trying to get requirements for the most important features not related to Activities since that seems to be the area of least documentation (suspend/resume, wireless connectivity, etc). Next highest priority will be shipping, core activities.
- The requirements pages should be read-only and should get sign off by mgmt of olpc.
- Equipment: We would like to get a school server to Westford and to India for more complete testing. Kim will initiate the discussion on this internally.
- We should standardize as much as possible on test plan format.
- Zack and Cameron are currently responsible for getting us up on a test case management system. They will get back to me if it isn't possible or a good idea to run this from redhat.
- The test groups don't want new builds every day. We should use the current build 494 for a few days to really get to some detailed testing. We should have the next good build on Thursday or Friday.
- We want to make sure that there really is a feature freeze - no new features should be added to this stream. Kim and Jim need to follow up on this and make sure that developers who are working on new features have some place else to check them in.
We decided to tackle ideas on an ECO process next week.