Apps meeting Tuesday August 16th 2005

Agenda

• Specs
• Schedule
• Engineering Issues
• People
• Other

Summary

• Plan : Merge branch in trunk. Finish all m5 tasks.
• Progress : ItemCollection conversion to sets completed. Progress on m5 tasks.
• Problems : Merge of sets work in trunk is late and will likely delay m5.

Actions

• Philippe to monitor m5 and load balance m6
• Monitor the rest of sets work and merge to trunk

Minutes

Specs

No active update in the 0.6 specs at that point unless there's something wrong.

We had one meeting on 0.7 (Friday). Likely tenets :

• Plausible dash board : lots of work around the table before getting anywhere. May be no fancy sections but new columns (like status) and sorting at least.
• Plausible e-mail : e-mail in DV, reply/forward, some elements of search, ...
• Calendar polish : react to what people dogfooding 0.6 comes up with, may be some sort of free/busy (but no invites).

Schedule

Biggest item gating m5 is sets. Update from John :

• Completed converting every ItemCollection to a set yesterday
• Added union and intersection today
• Can run Chandler now (today at least)
• Still waiting on notification (parts from Ted and Andi need to be weaved and integrated by John)
• Caution : it will take several weeks to get all the bugs out
• We should be able to test the branch seriously by the end of the week
• Will be able to make the merge to the trunk beginning of next week earliest
• People working on the branch (Ted, Andi, Morgen, John, Alec) must be running the unit the tests and the acceptance tests. Aparna to send an e-mail to set expectation to Dev.

Conclusion : it seems clear that m5 will be delayed by 1 week, may be more (depending on how the tests turn up). We'll monitor and evaluate on a day to day basis.

Individual status :

• John : see above... Trying to land the branch in the trunk asap.
• Bryan : great progress, all tasks for m5 under control.
• Alec : Moving stuff out of parcel.xml to python, 80% has been done, more after the branch merge. Alec in vacation next week but Morgen and John can do that. That work is a rearchitecture that above and beyond what we originally planned but definitely the right thing to do. Philippe and Alec to review the m5/m6 tasks left over.
• David : Refresh P1 fixed (bug #3495).
• Jed : "Busy bar in minical" (bug #2594) and "appear in collection list" (bug #2991) to be checked in.
• Donn : working from Tahoe, not available during the call.
• Brendan : working on calendar GUI bugs (6 of them logged against m5).

Philippe to revisit m5/m6 balancing for each ASAP.

Engineering Issues

i18n : No comment on BK recent e-mail.

Accessibility: Some initial concerns about 0.6 expectation following Heikki's e-mail. Just a keep in mind and do your best kind of a deal.

People

Arel's gone back to school.

Brendan last week next week.

Alec OOF next week.

---

Status

Philippe

• Monitored m5 (sets and other tasks)
• 0.7 Planning
• Evaluated Qt, XAML

John

• Worked on converting ItemCollections to sets. 37 commits to the branch.
• All ItemColllections converted to Sets.
• Work remains converting Trash and Calendar to use new sets that take more than two sources.
• Notifications work remains to be done.
• We still have outstanding Repository bugs to resolve.
• Very little testing has been done.
• On a good day Chandler launches, on a bad day it doesn't.

Jed

• Your status here...

David

• Your status here...

Alec

• Your status here...

Brendan

• Your status here...

Bryan

• Your status here...

Donn

• Changed the edit-in-place fields to use the new Type-over-text widget
  • isolates the complication of switching controls to the widget
  • some minor bugs missed during the conversion have popped up, being fixed
  • major bugs requiring this rework: 3304, 3176, 2514
• Reconciled with John on our two methods of returning results from post()
• Now working on updating the CPIA Script UI, so scripts will only be run from Script items, collected in a Scripts Set
• Looking forward to "just bug fixing" mode
• Planing to review m5 workload once the m5 timeframe becomes clear