Greetings, folks,
I've just done a copious amount of Jira issue reorganization over the past couple of
hours. Sorry for all the emails that generated for some of you. This has been motivated
by a few recent developments:
(1) At the CSDL meeting yesterday, we discussed the problem of Jira not representing much
of the development effort that is actually performed on Hackystat. In other words, people
do a lot of work "outside" of the set of Jira issues that they've been assigned. This is
not optimal, since the set of Jira issues is our historical record of what's been done on
the system for each release. We talked about having everyone "tag" their SVN commit log
messages with a string indicating the Jira issue that the commit addresses. You've seen
some of these tags already today if you're on hackystat-svn-l. Mike discovered a very
cool feature in Jira that utilizes these tags to link the Jira issue to the SVN commits
associated with it:
<http://confluence.atlassian.com/display/JIRAEXT/JIRA+Subversion+plugin>. If every commit
has a Jira issue tag associated with it, it will go a long way toward ensuring that the
set of Jira issues represents the work done. (Of course, you may sometimes find yourself
creating a Jira issue at the moment of SVN commit to represent the work you've just done,
and immediately marking it as "resolved". That's totally OK.)
(2) There is a perception that the set of issues associated with a release doesn't really
represent the actual set of expectations for the things to be accomplished for that
release. Instead, when preparing for a stable release, I simply chuck the remaining
unresolved issues "over the fence" into the next version.
(3) Cam's reporting of the missing junit.jar file motivates me to make a 7.2 release in
the near future.
So, in response to this, I've done the following:
* Restricted the 7.2 Jira issues to those that have already been resolved in preparation
for the 7.2 release.
* Instead of chucking the remaining issues into 7.3, I made them all
version-less.
* I then went through the 120 or so open issues, and individually selected which ones I
would propose that we work on for 7.3. I edited their priority and other attributes
(including Workspace) to facilitate this. About 25 issues made the cut.
I would like us all to try to make the Jira issues pretty reflective of what we're
working on. Once we're doing that, then the Hackystat analyses based on Jira issues will
become much more useful and interesting. Of course, the other benefit is that we'll know
what we're doing. :-)
Let me know what you think and if you have any other ideas for how to proceed.
Cheers,
Philip
