Thanks for the link. I think everyone wants to build the community but I agree with Dims that this is hard.
I think doing some integration with Geronimo, OpenWebBeans, and maybe the Aries project might be some possible communities to interact with to expand functionality with. Are there any others? For 2), I don't think we ever agreed to a Review then Commit policy nor do I think we should. Sometimes I put up patches (especially with performance) for the reasons you discuss since eventually the multi-threading gets more complicated or uses "non-standard" Java patterns. The rest I also agree with. On Wed, Sep 23, 2009 at 9:35 AM, Kevan Miller <[email protected]> wrote: > > On Sep 7, 2009, at 10:59 AM, Davanum Srinivas wrote: > >> Team, >> >> To get out of the incubator, we need to diversify the base of committers >> which means we need to build up a good base of users and then turn the power >> users into committers. Also crucial is to make sure that the set of >> committers should be from different employers as well. >> >> Please do think from the shoes of potential users and think about >> different scenarios and integration with other projects. >> >> This is IMHO the hardest part of graduation... > > Haven't seen any follow-up to this. So, want to second Dim's point. This > needs to be a high priority for the community. You may find the following > useful -- http://incubator.apache.org/guides/community.html > > A few additional community observations: > > 1) There's a lot of communication that is occuring within Jira's. Jira's are > great at tracking issues and tasks. Not necessarily so great for > communicating with the community. I encourage you to make more use of the > mailing lists. > > 2) I'm seeing a lot of patches from committers being reviewed within Jira's > and then applied by another committer. Is there some form of formal review > process going on? Most Apache communities operate under a Commit-Then-Review > policy. If someone is learning a new component, is unsure of their change, > or thinks a particular change might be controversial, then reviewing a patch > can be very useful, community-wise. However, I would expect that all > committers would be encouraged to directly contribute their changes. Unless > there is a Review-Then-Commit policy in place. In which case, I would expect > the policy to apply to all community members. > > 3) Somewhat related -- so far this month, there have been commits from only > two committers. This general pattern extends further back in time. I > encourage you to broaden your active committer base... > > 4) Also related -- when committing someone else's work (even if you've > altered the work), please be sure an acknowledge the contribution in your > svn commit message. > > --kevan > -- - Bryant Luk
