On Mon, Jan 23, 2017 at 3:06 PM Ullrich Hafner <[email protected]> wrote:
> Am 21.01.2017 um 15:07 schrieb Baptiste Mathus <[email protected]>: > > +1 for no assignee by default and adding the NEW status > > Having a state to basically say: "this is issue is new, but has not been > reviewed, so may be invalid/incomplete/whatever for whatever reason" makes > sense IMO. And that the Triage team could use for, well, triaging. > > Also, I'm trying to think about what the "default assignee" way of working > right sends as message. It may push back some ppl willing to contribute, > > > Why would this push back people? A default assignee (as Atlassian defines > it) is a person who is responsible for an issue, i.e. the owner (not > identical to the person fixing it). > I mistakenly assumed that "assignee" meant "the person who is working on this bug" or at least "the person who will eventually work on this bug". When a bug was assigned to someone, I was hesitant to start work on that bug without acknowledgment from the current assignee that they are not actively working on the bug. If the bug is unassigned, then I am more confident that no one is actively working on that bug. I think I can get the same information from reports if I can trust that people use the state "In Progress" to indicate they are actively working on a bug. I'm happy to start using "In Progress" as well. There are currently 574 bugs "In Progress", with approximately 10% of them updated in the last month. Based on that, "In Progress" does not seem to be a widely used status value. Can you provide a pointer to that Atlassian definition of "assignee"? I'd like to understand their intended use model. Likewise, can you further expand on your definition of "responsible for an issue"? Is "responsible for an issue" - the lead maintainer of the component? - the person implementing the fix? - the person verifying the fix? - the person checking the bug can be duplicated? - the person automating tests of the bug? or "send" the wrong message with essentially newcomers (which are probably > the majority) thinking basically when seeing an assignee "cool, someone is > looking into that" which is generally (90% of the time?) wrong. > > > I think the opposite message is also wrong: ohh, no assignee, that means > nobody cares about my bug report:-( I think the majority of users reporting > a bug is not interested in fixing it on their own. > (And a side note: as a user of a professional software system (especially > one that is used to improve the quality of software) I would expect that > most of the issues are going to be solved by the team, not only 10%. But > this would be a topic for another discussion: how can we close the gap > between new and resolved issues...) > > I disagree that "most of the issues are going to be solved by the team", unless you and I have radically different definitions of "team". I define "team" as "active maintainers of that plugin", which makes the number of people quite small. I hope that many, many issues are investigated and resolved by a wide collection of infrequent contributors who discover and fix something important to them. Mark Waite > > I love the idea of the TRIAGE team, thanks a lot Slide for that. I will > try to help a bit (though I should already start by being back to more > activity on the HOSTING project front). > > 2017-01-21 0:32 GMT+01:00 Ullrich Hafner <[email protected]>: > > I see, the discussion seems to indicate that we have (at least two) > different kind of development processes. I’m not sure if I can summarize it > properly: > > We have plugins with one main developer who takes the responsibility for > all new issues. Even if the issue is not solved immediately these issues > should be assigned here to the component lead (which should be a good > practice to set via the IRC bot anyway) to mark the responsible person for > this issue. > > Then we have plugins (and core?) with one or more developers: here nobody > is responsible for an issue in the beginning. If there is time for fixing > an issue available then a team member is picking an interesting issue with > no assignee, assigns the issue and starts work on it. All unassigned issues > are waiting to be fixed by either a team member *or* a volunteer. > > So I think the only way to make a bug reporter think that somebody really > cares about an issue is to introduce this new status in the beginning > (before Open). Then the triage team or the component lead can verify this > issue and ping the reporter for additional information. If this issue will > be finally accepted then it is moved to Open. In this step we can either > remove the assignee or not (depending on the component lead). Then the > reporter sees that his bug report has been accepted: if there is no > assignee set, then the reporter also sees that nobody yet has the time to > fix it and it would be good to provide a PR by someone else (including the > reporter). > > > > Am 20.01.2017 um 14:39 schrieb Stephen Connolly < > [email protected]>: > > > > On 20 January 2017 at 10:29, Ullrich Hafner <[email protected]> > wrote: > > > Am 20.01.2017 um 08:16 schrieb Stephen Connolly < > [email protected]>: > > I also would love if we cleared out the assignee setting entirely. > > > How do we identify who is responsible for the issue (or who is the owner)? > If there is no assignee then nobody gets notified about new bug reports or > issue updates (if you are not watching an issue). > > > There are loads of plugins were somebody else has taken up (great) but I > still get assigned the issue. > > > Then we should set the default assignee accordingly. > > I'd rather use assignee to track actually picking up the issue. Status to > track triage state. > > Esp in credentials (which I have been repeatedly and actively scrubbing) > use of assignee also helps identify when somebody starts working on the > issue. > > > There is a status in progress which we have activated in Jira for such a > use case. > > > Ideally the status would have a state to indicate that the JIRA has been > accepted as a bug. > > > This makes sense. Currently we have no idea if a developer has accepted a > bug report as valid. Since no assignee is attached to most of the new > issues the reporter has no clue to see if the bug has been recognized or > not. > > We are using in several other projects an additional initial Jira status > ’New’. This status indicates that the bug has been created, but it has not > yet been confirmed by the developer team that it is valid. From ‚New‘ there > is a transition to ‚Open‘: this transition can be used by our new triage > team to indicate that the issue has been reviewed and accepted as a bug. > This transition should only be started if the issue is reproducible and if > all information are provided. The triage team then could have a filter on > all ’New’ issues to see what needs to be reviewed. > > > > Another status to indicate that it is ready to be worked on would be > awesome > > > 'In Progress' is already available: > https://confluence.atlassian.com/adminjiraserver071/working-with-workflows-802592661.html > > > `In Progress` != `Plugin maintainer has blessed this issue as one that > somebody can pick up` > > Rather `In Progress` means - to me - that somebody is actually working on > it > > > We just need to clarify for everyone what those statuses mean (if they are > not self evident) > On Fri 20 Jan 2017 at 00:21, Mark Waite <[email protected]> wrote: > > On Thu, Jan 19, 2017 at 1:59 PM Ullrich Hafner <[email protected]> > wrote: > > I think this is not the common practice. Wouldn’t it be better to use the > in progress transition for such issues? > > When I create an issue and see that there is no assignee it gives me the > feeling that I should not have spent the time in creating the issue since > nobody actually is interested in fixing it (or responding to it). (As a > user I don’t know that the component owner does use the assignee field > differently then the rest.) > > > In my case, I'm one maintainer with 400+ bugs assigned across the two > plugins I maintain. If someone assigns all the bugs for the git plugin and > the git client plugin to me, "assignee" will no longer be a useful field to > me and I will ignore it. I already ignore severity and several other > fields, so that isn't a big problem. I'd then maintain a record of my > "working set" somewhere else. It will probably be less visible to others, > since I won't necessarily try to use Jira to maintain it. I'm fine with > maintaining my "working set" elsewhere, I only use Jira for the working set > because I'm already there reading bug reports. > > I acknowledge that I'm an exception (since the git plugin and the git > client plugin are second only to the Subversion plugin in the total count > of bugs open against them), but since Jesse noted that others use the same > assignment process which I use, I may not be as much of an exception as you > think. > > Thanks! > Mark Waite > > > Am 19.01.2017 um 20:50 schrieb Slide <[email protected]>: > > If that is a common practice, then we can skip setting the assignee and > focus on component assignment and reproducibility, or we can have a > "whitelist" of plugins that we set assignee for. The goal isn't to make > things harder for maintainers, we want to help as much as possible by > funneling things to the correct place. If there is something else that > would be more helpful, or an additional scope, please bring it up. > > On Thu, Jan 19, 2017 at 12:48 PM Mark Waite <[email protected]> > wrote: > > On Thu, Jan 19, 2017 at 12:11 PM Jesse Glick <[email protected]> wrote: > > On Thu, Jan 19, 2017 at 12:34 PM, Slide <[email protected]> wrote: > > > > The outcome of a triage on a specific issue would be that the correct > > > > component(s) and assignee were there. > > > > > > Do we really need to set an assignee? For example for most > > > `{workflow,pipeline}-*-plugin` components there is intentionally no > > > default assignee. If and when someone intends to work on a fix, they > > > can assign to themselves. > > > > > As further support for what Jesse says, please don't assign me as an owner > for bugs in the git plugin or the git client plugin during the triage > process, unless you're willing to accept that I'll immediately remove that > assignment and return them to "Unassigned". > > I only assign bugs to myself when I want to indicate that I'm working on > them, or intending to work on them "soon". I use the list of bugs assigned > to me as a reminder of active work, not as another way of expressing the > component name. > > Mark Waite > > > > > > > -- > > > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > > > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CANfRfr0SRoe%2BWirJPscSjeLb_kXszYVFxTqpQDOCNjio86-TbQ%40mail.gmail.com > . > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > -- > > > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > > > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtH_W8iHnQUL_xjV2fA-pCibkzfOkkU2P%2BzUNrdqV0H43w%40mail.gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtH_W8iHnQUL_xjV2fA-pCibkzfOkkU2P%2BzUNrdqV0H43w%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > -- > > > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > > > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CAPiUgVfcK4R8i4W%3DGs8mW94Agzn5jdyudQk_zpt_Yg3XC9%2BbuA%40mail.gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/CAPiUgVfcK4R8i4W%3DGs8mW94Agzn5jdyudQk_zpt_Yg3XC9%2BbuA%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > > > > > > -- > > > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > > > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/4C16E3F0-8F9D-43F2-854B-C2E0FD29D6D2%40gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/4C16E3F0-8F9D-43F2-854B-C2E0FD29D6D2%40gmail.com?utm_medium=email&utm_source=footer> > . > > > For more options, visit https://groups.google.com/d/optout. > > > > > > > > > > > > -- > > > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > > > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > > > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtE685zds-5QdsWhWR59%3D_MmL%2B-AZseAH%2B-dpkcaNqCmng%40mail.gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtE685zds-5QdsWhWR59%3D_MmL%2B-AZseAH%2B-dpkcaNqCmng%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > > > For more options, visit https://groups.google.com/d/optout. > > > -- > Sent from my phone > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CA%2BnPnMwGf466pLG_mnRQ8e0jYGB%2B%3DX-pOKpwfo32dDixnUbhSg%40mail.gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/CA%2BnPnMwGf466pLG_mnRQ8e0jYGB%2B%3DX-pOKpwfo32dDixnUbhSg%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > For more options, visit https://groups.google.com/d/optout. > > > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/78782FEB-70BD-4396-AE7D-7951877B4765%40gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/78782FEB-70BD-4396-AE7D-7951877B4765%40gmail.com?utm_medium=email&utm_source=footer> > . > > For more options, visit https://groups.google.com/d/optout. > > > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CA%2BnPnMw0OH2Wb9gz-dRVNTFX%2BUh5Oa7gpZrC%2B7Xd15W0u6cfvg%40mail.gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/CA%2BnPnMw0OH2Wb9gz-dRVNTFX%2BUh5Oa7gpZrC%2B7Xd15W0u6cfvg%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > For more options, visit https://groups.google.com/d/optout. > > > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/56A2B757-0333-4BD0-B5BF-B07BA0DB16A4%40gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/56A2B757-0333-4BD0-B5BF-B07BA0DB16A4%40gmail.com?utm_medium=email&utm_source=footer> > . > > For more options, visit https://groups.google.com/d/optout. > > > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/CANWgJS6SETyf1-Jit0vny_pv0KsNtpt%3DijtOBqoNG2E6isUDWA%40mail.gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/CANWgJS6SETyf1-Jit0vny_pv0KsNtpt%3DijtOBqoNG2E6isUDWA%40mail.gmail.com?utm_medium=email&utm_source=footer> > . > For more options, visit https://groups.google.com/d/optout. > > > -- > You received this message because you are subscribed to the Google Groups > "Jenkins Developers" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/jenkinsci-dev/899E86B0-CB76-4F38-94FE-B588C5FF131F%40gmail.com > <https://groups.google.com/d/msgid/jenkinsci-dev/899E86B0-CB76-4F38-94FE-B588C5FF131F%40gmail.com?utm_medium=email&utm_source=footer> > . > For more options, visit https://groups.google.com/d/optout. > -- You received this message because you are subscribed to the Google Groups "Jenkins Developers" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/jenkinsci-dev/CAO49JtFKis8eO0YH0a4TH%2Bp%3DMWmVESAkcpbqeag_LjF3AXBXQg%40mail.gmail.com. For more options, visit https://groups.google.com/d/optout.
