Governing Board Minutes: 2012/12/05
The OpenJDK Governing Board met via conference call on Thursday, 5 December 2012 at 16:00 UTC with the following agenda:
- FOSDEM 2013
- Process issues
- The future of the JDK 6 Project
Four Board members were present: Andrew Haley, Doug Lea, Mark Reinhold, and Georges Saab.
One Observer was present: Mike Milinkovich
The intent of these minutes is to capture the conversational flow of the Board’s discussion and also to record decisions. If you are interested only in the latter then search for the word “AGREED” throughout the text.
1. FOSDEM 2013
Mark started the meeting by asking whether the Board wished to have a repeat of the JavaOne Governing Board panel discussion in the Free Java Devroom at FOSDEM 2013, 2–3 February in Brussels, Belgium. After some discussion, confirmed attendees included Mark, Georges, Andrew, and Mike. Doug said he’ll attend via Skype, as he did at JavaOne.
AGREED: Mark will submit a proposal to FOSDEM 2013 for an OpenJDK Governing Board panel discussion.
2. Process issues
Two process topics were raised for discussion:
- The saga of the sdt probe contribution
- JEP submissions
The first issue had already been brought to the Board’s attention by Andrew. It took five months from the time that this particular contribution was first posted until it was finally integrated into the HotSpot code base. Andrew expressed his belief that everybody working at Oracle has a job to do, and that reviewing changes not directly related to their goals is of pretty low priority. As long as there are no HotSpot Reviewers outside of Oracle, this problem will continue. Mark noted that the problem is compounded by the fact that Hotspot has a requirement that all need to go through an internal build-and-test system. He apologized on behalf of Oracle for this particular case and stated that as long as HotSpot has this higher hurdle, Oracle must bear the burden of sponsoring these patches. He advised Andrew to contact him if similar problems occur in the future. Georges noted that elimination of these types of problems was a long-term goal in the score card. Andrew expected that prodding would go a long way to solving these problems, but would keep Mark’s offer in mind.
For JEP submissions, Doug requested some clarification about when a JEP is required. Mark replied that JEP 1 contains a fairly clear statement of the JEP submission threshold. He welcomed any suggestions and requested that they be sent to the general discussion list. Doug said he’d think about it. He thought that the “Funded” state appeared to mean “funded by Oracle.” Mark clarified, saying that “Funded” means that everybody who needs to do work to get the enhancement done has agreed to do that work. Depending on the feature, this may or may not include Oracle staff.
Doug asked again about the future of the CCC. Mark reiterated that the intent remains to convert it into an open process which would leverage the forthcoming public bug system.
3. The future of the JDK 6 Project
Andrew said that he could not think of a reason why someone outside of Oracle couldn’t take over leadership of the JDK 6 Project after Oracle stops maintaining it, as previously announced, at the end of February 2013. He anticipated that everything could be managed by IcedTea, with the exception of the bug system. Another minor obstacle was the transfer of ownership.
Mark observed that the ownership transfer was simply a matter of following the process described in the Bylaws to nominate a new Project Lead. The bug system was a more difficult problem. Andrew considered the possibility of using IcedTea’s Bugzilla. Mike worried that this separate system would not support the integrated viewing of existing bugs. Georges recommended creating a small side project in java.net, to take advantage of that JIRA system, with a view toward migrating the issues created there to the eventual OpenJDK JIRA system. Andrew expressed concern that neither solution would work due to existing changeset message validation. Mark explained that a new Project Lead could change that requirement. Andrew agreed that a combination of the java.net JIRA system and a change in commit-message checking should solve his problems.
At this point the Board adjourned.