On what issues could I help clean up for JDK 9

Patrick Reinhart patrick at reini.net
Tue Dec 6 15:27:05 UTC 2016

On 2016-12-06 12:44, dalibor topic wrote:
> Looking at the issue, it's a rather poor choice for a "cleanup" issue
> for newcomers, as the only time the word 'cleanup' shows up in its
> description is in the logged output of a failing tests. That means
> it's not actually a cleanup issue at all.
> In addition, reproducing the test case failure requires installing two
> additional proprietary products (a third party database, and a third
> party database driver). Each additional software installation
> requirement introduces additional challenges for newcomers, such as
> purchasing such third party software.
> It presents the same challenges to potential reviewers, as well.
> That's a good example for the kind of contribution that would most
> likely be a net negative: it would end up using up valuable time both
> on the new contributor's side and the side of existing contributors,
> while the novice would only have a tiny chance of being successful.
> As Patrick pointed out, mentoring is important. It's also very hard to
> do well. One of the hard parts of mentoring is picking the right
> challenge for the novice to tackle.

That's exactly the reason I started the discussion here.

As of the result of the many calls for help on different presentations
at JavaOne and Devoxx events. I was approached as a existing contributor
to have a introductory talk about contributing to the OpenJDK. And at
the end of it the question emerged, what would be parts, that help could
be needed besides of testing new features, sending in new bug reports,
writing blog posts and so on.

At this point I realized, that if I just point them to the existing
documentation about contribution [1] or pointing them to the wiki [2]
contains a lot of good information, but does not actually answer the
question of small things to clean up from the JBS.

> Someone unfamiliar with JDK development attempting to mentor other
> novices will likely pick the wrong ones, such as this one, as will
> someone picking out issues randomly from a bug tracker search, etc.

For me it is still hard to find an issue that may fit my knowledge due
the basic lack of finding the correct module/submodule for a certain
class. Also I first of all do not want to introduce new issues in the
first place, but fix existing ones to reduce the total number of issues
if possible...

> Instead, the better way of picking out such cleanup tasks is to follow
> along the mailing list threads to understand what kind of cleanup
> activities are already ongoing or being discussed for JDK 10, or other
> OpenJDK Projects.

Here I'm a bit confused to find the correct mailing list to pick to get
hold on those discussions. I registered myself on the jdk10-dev list
already, where the actual discussion is more about the new repos.

> Accordingly, a more useful contribution to this Group would be to
> track such cleanup effort related mailing list threads on our wiki,
> for example, so that novices can find them, read through them, and
> understand if that's an effort that they would like to get involved
> with, that matches their skill levels.

That would be very good thing and it seems to me, that exactly this is
one of the missing pieces right now. There could also be a list of 
who would host a webrev on cr.openjdk.java.net in case the contributor
has a little bit bigger patch as I did in the past.

> And of course, not every cleanup effort is suited for everyone. Please
> don't attempt C compiler warnings cleanups if you are not a practicing
> C developer^W language standards lawyer, for example ...

I fully agree with this point.

> cheers,
> dalibor topic


[1] http://openjdk.java.net/contribute
[2] https://wiki.openjdk.java.net

More information about the adoption-discuss mailing list