Difference between revisions of "Planning Council/Juno retrospective"

From Eclipsepedia

Jump to: navigation, search
(Things that could have been better)
(Comments from PC during 8/1 meeting)
 
(2 intermediate revisions by one user not shown)
Line 17: Line 17:
 
=== Positive things mentioned ===
 
=== Positive things mentioned ===
  
 +
(Especially things not mentioned in previous retrospective, such as predictability; we'll assume all those still true unless otherwise stated).
  
perceived to be smoother. projects improve with experience.
+
* perceived to be smoother. projects improving with experience. The phrase "well oiled machine" was used to describe it.
  
 
=== Things that could have been better ===
 
=== Things that could have been better ===
  
The importance (or no) of "non-greediness" requirement seems little understood.   
+
*The importance (or no) of "non-greediness" requirement seems little understood.   
  
The use of non-train projects by train projects seems confused/confusing.
+
*The use of non-train projects by train projects seems confused/confusing.
 +
:perhaps have an early "freeze" (say M6?) to be sure no "last minute" (untested) surprises.
  
  perhaps have "freeze"
+
*query2 case ... slipped through cracks.
 +
:perhaps require or suggest "one release" before joining train
  
 +
*communication:
 +
:we currently do not track or have an easy way of knowing which projects have no one subscribed to cross-project list?
  
query2 case ... slipped through cracks.
+
== Comments from PC meeting during 9/5 meeting ==
  
  perhaps require or suggest "one release" before joining train
+
* ''One issue (I brought up) was that every April or May we start to see a flurry of note/bugs about how p2 isn't good enough, or is "hard on the Eclipse infrastructure". Is there no way we can be more proactive? Do we need a "cross project team"? ''
  
communication:  
+
* ''Ian did a good job of summarizing some of the issues runtime projects have being in "sim. release" and thought it would be an ongoing discussion as to "what do they want instead, if anything". Some points made: ''
 
+
 
  who's "not listening" on cross-project list?
+
: ''p2 repo isn't all that interesting ... runtimes want to be in "maven central" (or maven.eclipse.org, I would assume)''.
 +
 
 +
: ''Some issues were "matter of degree", such as often 10 or 15 "tools related projects" often must be coordinated (such as "web tools") but for runtime projects its seldom more than a few, if that many. ''
 +
 
 +
: ''One fundamental issue (I think, if I heard right) was that many runtime clients want to stay on OLD version (even if tools advance) but OLD version is no longer in "current" repository (so, while doable ... it takes extra work, and is less "coordinated" to those users). They could (nearly) as easily use current tools repo, and individual runtime repos when needed''.
  
 
== Reference ==  
 
== Reference ==  

Latest revision as of 14:54, 5 September 2012

Contents

[edit] Juno Planning Council retrospective

These notes were collected at the end of the Indigo Release, at 8/1/2011 Planning Council call, specifically just to collect them. And not to solve issues, or even necessarily to suggest solutions, but just to capture issues (good and bad) while the release was still fresh on our minds. Where solutions are suggested below, it is intended to primarily better capture the issue discussed .. not to dictate a or pre-judge a solution. Action plans and solutions will be discussed later.

[edit] History

Juno is the seventh simultaneous release, following Callisto, Europa, Ganymede, Galileo, Helios, Indigo. The Planning Council meets regularly to come up with plans and requirements. Eclipse Foundation projects can voluntarily join the simultaneous release. For meeting minutes, see http://wiki.eclipse.org/Planning_Council.


[edit] Comments from PC during 8/1 meeting

These rough notes were captured from the "brainstorming" session. We will refine in future meetings.

[edit] Positive things mentioned

(Especially things not mentioned in previous retrospective, such as predictability; we'll assume all those still true unless otherwise stated).

  • perceived to be smoother. projects improving with experience. The phrase "well oiled machine" was used to describe it.

[edit] Things that could have been better

  • The importance (or no) of "non-greediness" requirement seems little understood.
  • The use of non-train projects by train projects seems confused/confusing.
perhaps have an early "freeze" (say M6?) to be sure no "last minute" (untested) surprises.
  • query2 case ... slipped through cracks.
perhaps require or suggest "one release" before joining train
  • communication:
we currently do not track or have an easy way of knowing which projects have no one subscribed to cross-project list?

[edit] Comments from PC meeting during 9/5 meeting

  • One issue (I brought up) was that every April or May we start to see a flurry of note/bugs about how p2 isn't good enough, or is "hard on the Eclipse infrastructure". Is there no way we can be more proactive? Do we need a "cross project team"?
  • Ian did a good job of summarizing some of the issues runtime projects have being in "sim. release" and thought it would be an ongoing discussion as to "what do they want instead, if anything". Some points made:
p2 repo isn't all that interesting ... runtimes want to be in "maven central" (or maven.eclipse.org, I would assume).
Some issues were "matter of degree", such as often 10 or 15 "tools related projects" often must be coordinated (such as "web tools") but for runtime projects its seldom more than a few, if that many.
One fundamental issue (I think, if I heard right) was that many runtime clients want to stay on OLD version (even if tools advance) but OLD version is no longer in "current" repository (so, while doable ... it takes extra work, and is less "coordinated" to those users). They could (nearly) as easily use current tools repo, and individual runtime repos when needed.

[edit] Reference

See also last year's retrospective