Re: [eclipse-dev] Planning Meeting Notes - Feb 4, 2009

[Mike Wilson <Mike_Wilson@xxxxxxxxxx>]

I appreciate the sentiment Kevin, but as the PMC member who ok'ed the splash screen change it was my responsibility to ensure that no problems resulted.

In any case, the conversation wasn't about finding fault. We were just reminding ourselves that *any* change that late in process can cause vastly disproportionate amounts of grief, which is something we sometimes forget. In fact, this particular example is just about perfect; I think everyone I talked to said "It's just updating the splash screen. That can't *possibly* break the build". Well, now we know better. <g>

Unfortunately, when we do forget, it's Kim who bares the brunt of the pain. :-(

For future reference, Pascal was on the right track this time: It was ok to *try* to get the new splash in, but as soon as we discovered there were problems, I should have said "promote the previous build". The fact that the team discovered, and helped to resolve the signing issues was great, but we shouldn't have put Kim through PTMD to do it.

McQ.

Kevin McGuire---04/02/2009 14:40:59---Kim, I wasn't on the arch call so I can't speak to what was said but from my perspective:

From:	Kevin McGuire/Ottawa/IBM@IBMCA
To:	"General development mailing list of the Eclipse project." <eclipse-dev@xxxxxxxxxxx>
Date:	04/02/2009 14:40
Subject:	Re: [eclipse-dev] Planning Meeting Notes - Feb 4, 2009

---

Kim, I wasn't on the arch call so I can't speak to what was said but from my perspective: 

I was the one who got the splash in late.  So I drove part of the chaos, my fault, not yours.  Thanks for accommodating.  And I'm glad you discovered and helped resolve the resigning jars problem. 

I've always found you extremely helpful when it comes to botched submissions and rebuilds.  So, thank you! 

I think we could *all* be doing a better job (me included) of ensuring our builds are more reliable.  It seems we often have build or test failures for one reason or another, with a resulting spotty success record for weekly integration builds.  That unevenness comes to the fore on milestone builds, like this one. 

Regards, 
Kevin

Kim Moir/Ottawa/IBM@IBMCA
Sent by: eclipse-dev-bounces@xxxxxxxxxxx

02/04/2009 12:45 PM

Please respond to
"General development mailing list of the Eclipse project." <eclipse-dev@xxxxxxxxxxx>

To	"General development mailing list of the Eclipse project." <eclipse-dev@xxxxxxxxxxx>
cc
Subject	Re: [eclipse-dev] Planning Meeting Notes - Feb 4, 2009

I've been thinking about the discussion in the arch call today regarding the M5 retrospective.   

The PMC approached me on Friday morning  and said that we needed a rebuild to include the new Galileo splash screen.  It was my understanding that the splash screen needed to be correct for Eclipsecon.  The rebuild broke and I spent many hours troubleshooting the problem.  In the process, we discovered several issues around resigning jars which were valuable for the community.   If I didn't fix this problem on the weekend, the builds would have all been broken this week. 

I could have just promoted the older build without the splash screen fix.  However, I was trying to promote a good M5 and avoid a milestone rebuild.  Everyone laughes about M5a.  However, milestone rebuilds waste a LOT of my time.    Time that could be better served by making progress on bugs for the next milestone.    Not only is it costly for me personally, it wastes a lot of time for the community as they have to adjust to a new milestone.   It doesn't make our team look great either  :-( That being said, I understand that unforseen issues arise that sometimes require a milestone rebuild. 

My point is that I made the best decision I could given the information available to me. 

Kim

Mike Wilson/Ottawa/IBM@IBMCA
Sent by: eclipse-dev-bounces@xxxxxxxxxxx

02/04/2009 09:07 AM

Please respond to
"General development mailing list of the Eclipse project." <eclipse-dev@xxxxxxxxxxx>

To	eclipse-dev@xxxxxxxxxxx
cc
Subject	[eclipse-dev] Planning Meeting Notes - Feb 4, 2009

--------
Status 
--------

JDT APT:
- newsgroups, inbox, Bugzilla, conf calls
- testing for M5

JDT UI:
- no pending 3.4.2 fixes
- 3.5 M6 planning
- improved selection behavior of Quick views (Ctrl+O, Ctrl+T) for
enclosing types
- continued work on auto-sort of working sets in Package Explorer
- fixed problems with Go Into in JDT extension of Common Navigator
- fixed Sort Members to avoid moving initializers (bug 263173)
- inbox tracking

Platform Text and JDT Text:
- no pending 3.4.2 fixes
- 3.5 M6 planning
- continued work on 'Open Implementation' hyperlink detector
- bug fixing
- inbox tracking

Platform Search:
- no pending 3.4.2 fixes
- inbox tracking

JDT Core:
- 3.5M5 bugs verification and N&N
- removed deprecated NamingConventions.VK_CONSTANT_FIELD
- keep investigating problem with indentation in ast rewrite
- resumed work on search to provide matching regions
- activated memory trace while running model tests to track:
- https://bugs.eclipse.org/bugs/show_bug.cgi?id=260739
- fixed performance regression during startup

Debug:
- contributed 3.4.2 fix for console deadlock (bug 259107)
- bug fixing
- 3.5 M5 test pass
- M6 planning

PDE:
- contributed 3.4.2 fix (bug 242516) 
- 3.5 M5 test/fix pass
- bug fixing
- M6 planning

User Assistance:
- milestone testing
- investigating server performance
- intro accessibility work

Workspace:
- inbox tracking
- 3.5 M5 fixing, testing, N&N 
- working on 3.5 M6:
- http://www.eclipse.org/eclipse/platform-team/team3.5/plan.php#m6
- vacation

Rel. Eng.:
- M5 delivered
- slowly recovering from PTMD (post-traumatic milestone disorder)
- 3.4.2 endgame
- testing M5 bundles in build, new DS bundles in SDK, composite repos,
ant task to create list of orbit bundles to exclude from signing,
fix categories in repo
- bug fixing
_______________________________________________
eclipse-dev mailing list
eclipse-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipse-dev
_______________________________________________
eclipse-dev mailing list
eclipse-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipse-dev

_______________________________________________
eclipse-dev mailing list
eclipse-dev@xxxxxxxxxxx
To change your delivery options, retrieve your password, or unsubscribe from this list, visit
https://dev.eclipse.org/mailman/listinfo/eclipse-dev