Skip to main content

Notice: this Wiki will be going read only early in 2024 and edits will no longer be possible. Please see: https://gitlab.eclipse.org/eclipsefdn/helpdesk/-/wikis/Wiki-shutdown-plan for the plan.

Jump to: navigation, search

CBI/July25 2012

< CBI
Revision as of 14:02, 25 July 2012 by Andrew.ross.eclipse.org (Talk | contribs) (Executive Summary)

Meeting

  • July 25, 2012, 1pm EDT
  • Eclipse Conference facility

Attendees

Andrew R., Igor F., Thanh H, Paul W., Krzysztof D., Mike W., Mike M., David W.

Executive Summary

  1. The CBI build for the Eclipse platform has made considerable progress. It is now possible for someone without prior experience to build Eclipse for themselves in a matter of minutes.
  2. Juno SR1 (Sept. 28th) was set as the date for the Eclipse Platform to cut over to start delivering via. the CBI build rather than the legacy PDE build.
  3. The team agreed SR1 is not feasible. Will be picking a new Kepler milestone to target cut-over of both Kepler and Juno.
  4. Now is the time for interested parties to invest in CBI. Special thanks to Red Hat for demonstrating leadership in this area and making significant contributions. Others are encouraged to get involved.

More details and the minutes from a related discussion are below.

Minutes

Andrew: This particular call is about people power and schedule. I called it because I'm concerned, and learned others are also concerned whether we are on a credible track to be able to switch the Eclipse platform to the CBI build for SR1 (Sept. 28th 2012). Looking at the work on hand (~15 bugs) and factoring vacations, etc. It looks dubious. What do you think?

McQ: Cutover to the CBI build for SR1 is not possible. In many ways the PDE build still has significant areas that are not well understood. The people that created the build left the team 3 years ago. Naturally this isn't a comfortable place to be. In case it wasn't clear, we very much want to cut over to CBI.

Andrew/McQ/Paul: The logical course of action is to just keep working as fast as we can and keep up the sense of urgency. Still need to be aggressive.

Mike M.: Are Juno & Kepler seen as distinct in this regard?

McQ: No, they're one in the same.

Mike M.: It makes sense to pick a Kepler milestone, say M2, and make it the point to cut over.

McQ: Agreed. (Paul will take the action to figure out which milestone makes sense)

Andrew: I like the idea of two streams built on CBI soaking in parallel.

Igor: Is the SR2 (Feb. 22nd 2013) date a hard date or soft date?

Mike: We've not done so before, but Juno SR2 cut-over to CBI is slip-worthy. This is far too important for LTS.

McQ: CBI is making it significantly easier to build. This is a great thing that we want very much.

Paul: e.g. Someone approached seeking to build for themselves. 2 hours later they were done & happy. That's phenomenal.

McQ: We'd like to see others contribute as well.

Andrew: Credit to Krzysztof and the team from Red Hat. They really have been doing a ton of good work. Thank you!

Paul: Agreed

Krzysztof: Fedora will be switching to the CBI build in any case. Will the platform team help us if we need it?

McQ: We want to help. We'll do our best. Naturally priorities have to be considered.

Back to the top