Hudson Renames to Jenkins
Hudson, the popular continuous integration and build server, will be renamed to Jenkins this week, following a decisive vote in which community developers voted 214 to 14 to rename the project. More information on the back story can be found in earlier InfoQ posts.
A post to the community-backed
HudsonJenkins-labs confirms the decision to rename the existing mailing list and GitHub infrastructure to point to the new locations:
- hudson-labs.org -> jenkins-ci.org
- @hudsonci -> @jenkinsci
- http://github.com/hudson -> http://github.com/jenkinsci
- hudson-dev -> jenkins-dev
- hudson-users -> jenkins-users
- hudson-commits -> jenkins-commits
- hudson-issues -> jenkins-issues
At the time of writing, these have been set up but are not yet in use. The plan is to have a phased transition in which a changeover date is announced so that the majority of people know about it ahead of time; to populate the existing infrastructure before the vote (other than registering key names) would have been presumptuous. The trademark "Jenkins" does not appear to have been applied for yet; ironically, there is an EU-based trade-mark and patent company called Jenkins who may be able to assist.
An interim governing body will be set up to look after the initial development of Jenkins pending a full governance being in place. The initial members are Kohsuke Kawaguchi, Andrew Bayer and Dean Wu, who takes the place of Winston Prakash who was offered but declined a seat on the board. Dr Winston Prakash is a former Sun engineer, now Oracle, who was involved in the Java Store and whose current assignment is to get involved and contribute to Hudson. Oracle's current position is to continue to work on Hudson. InfoQ asked Oracle whether they will be participating in Jenkins and was told that they're in the process of working out the details. Winston Prakash has also posted an update to the Hudson home page:
Many of you may be aware that Andrew Bayer and Cloudbees (where Kohsuke Kawaguchi now works) have confirmed their intent to fork Hudson into a new community. We want the Hudson community to know that we have a dedicated team that includes people from Oracle, as well as some of our partners and current Hudson community members who will continue to build and grow the Hudson project and community. Your current installations will continue to run fine, and you won't need to change any code. You can still come to http://www.hudson-ci.org/ for all things Hudson
Our primary goals over the next months will be to fix high-priority bugs in the current edition of Hudson and formalize a development and release process that will allow users of Hudson to have more insight into what is being changed, why it is being changed, and when their bugs will be fixed. We will also continue the discussion that I started last month about what the next version of Hudson (v2) looks like. I look forward to working closely with the community on where to go from here.
Response to the vote has generally been positive, with H-Online referring to it as a community landslide and a number of comments on reddit. There have been some questions on the user list as to whether Hudson and Jenkins will diverge, and what effect this will have on the already widely installed user-base.
At least in the short term, this rename/fork is unlikely to have any significant difference. Until Jenkins makes its first release, there won't be any noticeable difference for users of Hudson. Even then, existing Hudson installations will continue to work with Oracle providing on-going support and development, though the relative pace of development on the Hudson and Jenkins projects will ultimately determine the future popularity of the projects.
What do you think of this change of project name? Is it just a rename, or is it a more serious fork? If the latter, will you follow the Jenkins fork or the Oracle Hudson fork?
In my organisation we'll stick to our current Hudson release until the dust has settled. The trail of git commits and mailing list discussions will automatically lead us to the most active and maintained distribution and that is where we'll go :-)
Above article does not mention how this landslide will affect Sonatype's  commercial implementation . It would be interesting to hear their insights since they've been rather quiet during the whole discussion and are nevertheless a significant stakeholder. They have developed internally several features from what I can see and were looking to contribute some of it back. So i'm wondering now which side of the fence these contributions  will land ?