TPTP-PMC-20080917

Logistics

Attending: Oliver, Eugene, Joanna, Chris, Paul

• Regrets: Harm

Any objections to last weeks summary?

• No

4.5.1 status update

Thanks especially to Asaf for the extra unanticipated hours spent. Thanks to Oliver for his team's support too.

4.5.1 update. Joel is preparing the deliverable.

• We have what we believe to be a good build
• We are testing and have high hopes
  • Testing on IBM side on full force since yesterday
  • Joanna has not heard of any significant breakage.
  • Paul has not seen 4.5.1 issues. Most critical tests are pretty much done at this point
    • Paul will finish by ~3 this afternoon.

We think that we are on schedule to meet the delivery.

Oliver notes that Harm has asked us to think about how to get a knowledge transfer to folks on project

• There has been a fair amount of project turnover.
• IBM realizes that TI and the profiler are somewhat sensitive
  • Jonathan has spent some time with the St. Petersburg engineers reviewing the AC
  • Knowledge of probekit is somewhat weak at this point
  • Realistically (except for residual time from Asaf), profiler has had 100% turnover from original authors
• Question about whether Asaf has pointers to more documentation?

Team rampup

We discussed the turnover in the project and how to make sure that the team is fully ramped up. This is not exclusively the Beijing resources but we did discuss the Beijing resources a bit. Oliver mentioned that if it were OCS he would get a senior person to prioritize some bugs for them to fix concentrating on bugs that will grow experience rather than highest priority defects

• After fixing those ~5 bugs would have knowledge in this space.
• Chris noted that before AlexA left the project he had created such a list
  • Joanna went and looked this up
  • About 12 into defects were assigned to Beijing engineers.
  • Two of the assigned defects were fixed during transition in conjunction with Russia team
  • The remainder have not had too much traffic 241085 has had some discussion

Joanna noted that she hasn't communicated with them as much yet. Eugene has had technical communications.

• Because of time of day, Beijing engineers cannot participate in weekly calls and promoting a separate syncup will be desirable
• Getting into a mode of tracking statuses weekly with team might be helpful
• Joanna would like to see at least 2 sets of eyes reviewing code patches for profiler especially because of the brittleness that we have seen recently

Oliver asks who has action item to prioritize defects for Beijing engineers

• Eugene has been helping do some of this. He will do more.

Oliver asks if we want to have the "only" profiler center of knowledge in the Beijing team or if we should also decentralize with skills at IBM

• Given the fact that consumers at IBM are actively stressing this component, Joanna has had some discussion already about need for possible investment here.
• Joanna plans to follow up here.

4.5.2

4.5.2 plan (and XML roadmap themes file)

• File is due at end of this month. We need to allocate some focus on it
  • Joanna has not yet identified document owner.
  • Oliver asks that when 4.5.1 is completed later this week that we have an email discussion on this.

We need to plan what defects will and wont get fixed for 4.5.2

• Beijing team will probably spend most of their effort on ramp up bugs for 4.5.2
  • We will probably start measuring bug thruput
• Joanna will put in a bulk request for deferral of all remaining defects targeted at 4.5.1
• Defer to future, then start pulling in critical/proposed ones to 4.5.2.
  • We will prioritize ones consumers mention as well as ones that team members believe important
  • There are no objections to this approach

Joanna noted that in the past we have pulled in enough P1s for everyone to be busy 100% booked for the whole release

• We then did not complete them because of all the priority issues and related issues that appeared under more intense testing
• Suggest taking a more minimalist approach here.
  • Pull in only the really critical ones to the release
  • Pretty much guarantee that these will be done along with additional priority items that will most likely arise
  • Oliver likes this idea.
  • AI: project leads to do some prioritization by early next week for discussion at next PMC

(added 9/19/2008) Harm notes that prior to 4.5 we always budgeted only 60-70% of peoples time for fixes and enhancements in order to allow for other community work and web content. Then only part of the 60-70 was allotted to P1 committed work. This troubled some consumers, but it kept the plan commitment churn down and allowed us to over achieve and absorb special situations. This also forces us to get commitments on resource for the following cycle. Perhaps we should take a serious look at how we assign P1 and P2 in light of resources rather than demand, so that consumers get an up front warning on the plan. Maybe we can trial run this for 4.5.2 and 4.5.3.

Misc

Oliver asks if we still plan to do community outreach. We somewhat got sidetracked with the critical defect at end of 4.5.1

• Euene says that if everything in test for 4.5.1 goes ok, he thinks 4.5.1 should be stable enough to do it
• We will make go/nogo decision next week.

AC person sitting with jonathan meeting/code review for larger defect

We discussed travel a bit. Chris asked if we should consider trying to arrange some face time for Beijing team with Asaf and/or Toronto team (to sit and work together)

• General thought was that that might be premature but that it might be desirable at some point
• The idea of a virtual livemeeting style discussion was discussed as well
• Oliver raised the question of whether some face time to correspond to EclipseCon might be desirable
  • Joanna mentioned that it is a bit early to really discuss EclipseCon travel budgeting.
  • Chris mentioned that it will be good if we plan face time/justification well in advance to maximize chances of travel budget