I think one thing we should consider before all the rest is Jiras issues with patches. Those are more valuable than others and patches have a
disconcerting tendency to become stale, we should avoid that.
I have a rule for that (I did not invent it, thanks Internet): the 10 min rule. If I (expect I) can do it in 10 min, they I do. Of course sometimes
you happen to shave the yak, and you can't use the rule all the day long.
Also I guess the reason why most Jiras are of major priority is because don't care when they create it. Like they don't care when choosing between
"Fixed", "Done", "Implemented" when closing, which is obviously of a less importance
My 2cts
Jacques
Le 05/09/2016 à 13:06, Sharan Foga a écrit :
Hi Taher
+1
You have raised some good actionable points here. I think you are right that
the Jira guidelines need be completely separated out into something that is
clear and can be used as an easy reference.
I will create a wiki page to start pulling the information together.
Thanks
Sharan
On 2016-09-03 07:56 (+0200), Taher Alkhateeb <[email protected]> wrote:
Hi Everyone,
As of this writing, we have 3 blocker, 6 critical 733 major, 521 minor and
50 trivial open JIRAs and I note:
- I think none of the blocker and critical JIRAs belong in that category.
- It is also unrealistic to have 733 major issues. it is like saying we
have that many serious problems
- I think a realistic distribution would have a pyramid count with a max
(pyramid base) being trivial JIRAs.
- A substantial number of jiras is old and no longer applicable.
- A substantial number of jiras is vague and not understandable, poorly
written, without enough details to work on.
- A substantial amount of jiras are placeholders for something to do
without proper thinking and planning, which would make them unrealistic or
not applicable.
- Some JIRAs are extremely granular. For example a task broken to subtasks
each for a tiny amount of work that is not worth the time and effort of
making so many tasks.
So I think our issue tracking system is not properly used and we have many
JIRAs that are not useful and distracting. I propose the following actions:
- Add a wiki page (if one does not exist) documenting:
- Guidelines for writing JIRA mentioning clarity, provision of solution.
- A clear definition of priorities (blocker, critical, major, minor,
trivial) with some examples.
- A description of meaning of assignee and how to use it
- A description of other metadata and how to properly use it (tags,
components, affects version, etc...)
- we need to close all old, not applicable, vague and poorly written JIRAs
as per wiki guidelines. Alternatively authors can rewrite JIRAs to comply
with guidelines.
- We need to fix priorities of all remaining JIRAs as per wiki guidelines
This would give us a realistic view of the _real_ issues in OFBiz instead
of drowning in so many JIRAs many of which are hindering rather than
helping.
What do you think?
Taher Alkhateeb
