Wow, it is getting more and more interesting, by talking this we may
come out with some
very useful social fishing tool :-)
Thanks Shreedhar for join in the discussion !
Judy
Shreedhar Ganapathy wrote:
> Nice thoughts, Kedar.
> I think something akin to how social news sites provide new posts by
> users with a list of existing posts that are similar in content or
> possible duplicates and ask users to confirm their posting is unique,
> might be a great way to reduce the possibilities of duplicate issues
> being posted.
>
> For example, one can try posting news item to Digg.com and see this
> list to understand the feature in abstract.
> Dont know how hard it is to get this into the Java.net issue tracker
> infrastructure. (cc'd Sonya who handles RFEs in coordination with Collab)
>
> Cheers
> Shreedhar
>
> Kedar Mhaswade wrote:
>> Folks,
>>
>> Duplicate/spurious bugs is a reality even though dealing with
>> them is a pain (we've been there). I think we need something (or
>> someone) that helps find possibly same or similar bugs on the
>> issue tracker. When someone finds a bug the first thing s/he wants
>> to do is report it :). So, let's not let that enthusiasm fade out
>> yet try our best to remove duplication *at the time of bug filing*.
>>
>> Some work needs to be done here as we already have the data of all
>> the bugs that have been filed so far.
>>
>> Let's see if issue-tracker gets enhanced magically by a big brother ...
>>
>> -Kedar
>>
>> PS - on a second read, this does sound like a mysterious e-mail, but
>> it's not without purpose :)
>>
>> Snjezana Sevo-Zenzerovic wrote:
>>> Manfred,
>>>
>>> we are generally in agreement...
>>>
>>> I want to clarify that in the quoted example (and for all other
>>> issues that I personally get to evaluate) duplicate bug will not be
>>> closed as invalid, it will be clearly marked as duplicate. In this
>>> case it is the original issue that is on its way to be marked as
>>> invalid on its own merits - it will be evaluated and (hopefully
>>> reasonable) explanation for current behaviour will be provided in
>>> evaluation.
>>>
>>> Thank you.
>>>
>>> Manfred Riem wrote:
>>>
>>>> Hi there,
>>>>
>>>> Some thoughts after reading this thread.
>>>>
>>>> On the other hand one also need to understand that the community
>>>> is not necessarily versed in what you need as a developer and while
>>>> it might be easy to close bugs as INVALID I would recommend
>>>> just send the original a short email so they don't get frustrated.
>>>>
>>>> It is a learning process for everyone. On my own projects I tend
>>>> to keep a low profile on rejecting bug reports because they are
>>>> not in the 'correct' format. If someone consistently and blatantly
>>>> ignores your attempts in making them better in filing bug reports,
>>>> then and only then would I start closing his/her bugs as INVALID.
>>>>
>>>> I hope I don't come off to strong, but the fact that you are
>>>> getting bug reports is a BIG effort on the side of your community.
>>>> It is not their job so a bit of caution in how you address them
>>>> is a prudent thing to do. Eg. the personal approach instead of
>>>> screaming ;)
>>>>
>>>> Take care and see you soon in the bug tracker!
>>>> Manfred Riem
>>>>
>>>>
>>>>
>>>>> -------- Original Message --------
>>>>> Subject: Re: (IMPORTANT) please follow bug filing procedure
>>>>> From: Judy Tang <Judy.J.Tang_at_Sun.COM>
>>>>> Date: Fri, April 17, 2009 3:57 pm
>>>>> To: quality_at_glassfish.dev.java.net
>>>>>
>>>>>
>>>>> "a good bug should has a clear description and specific steps to
>>>>> reproduce it"
>>>>>
>>>>> Totally agree with you Davis. Thanks for sharing your years of
>>>>> experience !
>>>>>
>>>>> For bug description, just one line, it needs to cover what error
>>>>> you hit in what situation,
>>>>> so others can know if they are hitting the same issue or not when
>>>>> searching for
>>>>> dup, it is an art and it takes time to come out a good one :-)
>>>>>
>>>>> Thanks,
>>>>> Judy
>>>>> Davis Nguyen wrote:
>>>>>
>>>>>> Another point I want to share here is that a good bug should has
>>>>>> a clear description and specific steps to reproduce it.
>>>>>> The more specific and detail is the better. It will help
>>>>>> developers a lot when comes time to review them.
>>>>>> So, please put your extra effort to search for duplicate bug and
>>>>>> write a detail problem, just assume if someone does not know the
>>>>>> product but be able to reproduce your problem.
>>>>>>
>>>>>> Thanks,
>>>>>> -davis
>>>>>>
>>>>>>
>>>>>> Snjezana Sevo-Zenzerovic wrote:
>>>>>>
>>>>>>
>>>>>>> I feel like Bill Murray's character in "Groundhog Day" movie....
>>>>>>>
>>>>>>> While I realize that in some cases it may not be that easy to
>>>>>>> tell if an issue is duplicate, in this case it should be rather
>>>>>>> obvious that
>>>>>>>
>>>>>>> https://glassfish.dev.java.net/issues/show_bug.cgi?id=7913
>>>>>>> is duplicate of
>>>>>>> https://glassfish.dev.java.net/issues/show_bug.cgi?id=7731
>>>>>>>
>>>>>>> (incidentally, issue will eventually be marked as invalid, but
>>>>>>> that's different story and requires evaluation).
>>>>>>>
>>>>>>> Once more, please do take couple of minutes of your time to at
>>>>>>> least skim the list of existing bugs. Also, please do not file
>>>>>>> updatetool bugs in upgrade_tool subcategory - they should be
>>>>>>> filed either in update_center subcategory, or, as Tom Mueller
>>>>>>> was kind to point out previously, in their source project issue
>>>>>>> tracker:
>>>>>>>
>>>>>>> https://updatecenter.dev.java.net/servlets/ProjectIssues
>>>>>>>
>>>>>>> Thank you.
>>>>>>>
>>>>>>>
>>>>>>> Judy Tang wrote:
>>>>>>>
>>>>>>>
>>>>>>>> I talked with SJSU Professor Gao who has been very supportive.
>>>>>>>> And we agreed on the following:
>>>>>>>>
>>>>>>>> Students will use school bug system to report bugs found by
>>>>>>>> negative testing. If students find
>>>>>>>> P1/P2 bug from positive testing, they will send email to
>>>>>>>> GlassFish quality alias, SQE and developer
>>>>>>>> will let them know to log a bug or not. We really like to see
>>>>>>>> those good P1/P2 bugs filed by
>>>>>>>> students and we like to encourage students to test deeper and
>>>>>>>> wider to find those good bugs.
>>>>>>>>
>>>>>>>> I have also communicated with a few students to point out those
>>>>>>>> dup/invalid bugs to them, they
>>>>>>>> are very welcome to my input and they are working on to
>>>>>>>> improve. It is true that we don't
>>>>>>>> have a very good way to query for dup bugs, one area to improve.
>>>>>>>>
>>>>>>>> As always we appreciate students/community testing GlassFish
>>>>>>>> very much ! A good communication
>>>>>>>> is the key to make the community testing working. We welcome
>>>>>>>> your valuable input :-)
>>>>>>>>
>>>>>>>> Thanks,
>>>>>>>> Judy
>>>>>>>>
>>>>>>>> Judy Tang wrote:
>>>>>>>>
>>>>>>>>
>>>>>>>>> Thanks Snjezana and Anissa for the valuable input. And thanks
>>>>>>>>> Manfred for the suggestion.
>>>>>>>>>
>>>>>>>>> To encourage students learning at the same time not
>>>>>>>>> introducing dup/invalid bugs to the system,
>>>>>>>>> I would like students to do the following:
>>>>>>>>>
>>>>>>>>> (1) Please do not file negative test bugs in Admin GUI,
>>>>>>>>> Installer, Update tool since most cases are
>>>>>>>>> already reported and when developer fix one place, they most
>>>>>>>>> likely will take care of the
>>>>>>>>> same situation for other places, some time the fix in one
>>>>>>>>> place will address other cases. Just to give
>>>>>>>>> you an example last night and this morning Yifeng and Anissa
>>>>>>>>> processed more than 70 dup/invalid
>>>>>>>>> bugs. This need to stop !
>>>>>>>>>
>>>>>>>>> (2) Please try more main functional/positive end to end
>>>>>>>>> testing, when you see any issue, please send
>>>>>>>>> email to quality alias first, SQE and developer will evaluate
>>>>>>>>> the issue and let you know a bug
>>>>>>>>> is needed or not.
>>>>>>>>>
>>>>>>>>> It is first time we have so many students testing GlassFish,
>>>>>>>>> we are very happy to have
>>>>>>>>> you here to learn and to help, because this is first time, we
>>>>>>>>> also have a lot lesson to learn together on how
>>>>>>>>> to handling dup/invalid bugs. We will keep improving. With
>>>>>>>>> your help we will :-)
>>>>>>>>>
>>>>>>>>> Thanks,
>>>>>>>>> Judy
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>
>>>>>>>>> Manfred Riem wrote:
>>>>>>>>>
>>>>>>>>>
>>>>>>>>>> Hi there,
>>>>>>>>>>
>>>>>>>>>> Maybe the professor should assign a team lead that files
>>>>>>>>>> the bugs. So no filing before the team lead has collected
>>>>>>>>>> them?
>>>>>>>>>>
>>>>>>>>>> Manfred
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>> -------- Original Message --------
>>>>>>>>>>> Subject: Re: (IMPORTANT) please follow bug filing procedure
>>>>>>>>>>> From: Anissa Lam <Anissa.Lam_at_Sun.COM>
>>>>>>>>>>> Date: Thu, April 16, 2009 1:59 pm
>>>>>>>>>>> To: quality_at_glassfish.dev.java.net
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> I second that. Please use your common sense when filing bugs.
>>>>>>>>>>>
>>>>>>>>>>> Last night, a student filed 43 Admin GUI issues, all of
>>>>>>>>>>> them are either duplicate, no clear description of what is
>>>>>>>>>>> wrong, not reproducible or invalid. It takes both QA
>>>>>>>>>>> and Eng. hours to clean that up late last night, only to
>>>>>>>>>>> find yet another similar batch come in.
>>>>>>>>>>> Judy has send several guidelines and there are umbrella bug
>>>>>>>>>>> available for adding comments. PLEASE follow the guidelines.
>>>>>>>>>>>
>>>>>>>>>>> thanks
>>>>>>>>>>> Anissa.
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>> Snjezana Sevo-Zenzerovic wrote:
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>>> Plea from frustrated and swamped GF engineer:
>>>>>>>>>>>>
>>>>>>>>>>>> * Please go through existing bug list before filing your
>>>>>>>>>>>> issue. Time I am spending evaluating and closing issues
>>>>>>>>>>>> which are very obvious duplicates of existing issues could
>>>>>>>>>>>> be much better used fixing some of genuine issues.
>>>>>>>>>>>> * Consider filing umbrella bugs for similar issues or
>>>>>>>>>>>> issues that are likely to have the same root cause. For
>>>>>>>>>>>> instance, if you cannot install several packages from Admin
>>>>>>>>>>>> GUI, you don't have to file separate issue for each one of
>>>>>>>>>>>> them.
>>>>>>>>>>>>
>>>>>>>>>>>> Thank you.
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>>> ---------------------------------------------------------------------
>>>>>>>>>>>>
>>>>>>>>>>>> To unsubscribe, e-mail:
>>>>>>>>>>>> quality-unsubscribe_at_glassfish.dev.java.net
>>>>>>>>>>>> For additional commands, e-mail:
>>>>>>>>>>>> quality-help_at_glassfish.dev.java.net
>>>>>>>>>>>>
>>>>>>>>>>>>
>>>>>>>>>>> ---------------------------------------------------------------------
>>>>>>>>>>>
>>>>>>>>>>> To unsubscribe, e-mail:
>>>>>>>>>>> quality-unsubscribe_at_glassfish.dev.java.net
>>>>>>>>>>> For additional commands, e-mail:
>>>>>>>>>>> quality-help_at_glassfish.dev.java.net
>>>>>>>>>>>
>>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>> ---------------------------------------------------------------------
>>>>>>>>>>
>>>>>>>>>> To unsubscribe, e-mail:
>>>>>>>>>> quality-unsubscribe_at_glassfish.dev.java.net
>>>>>>>>>> For additional commands, e-mail:
>>>>>>>>>> quality-help_at_glassfish.dev.java.net
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>>>>>
>>>>>>> ---------------------------------------------------------------------
>>>>>>>
>>>>>>> To unsubscribe, e-mail: quality-unsubscribe_at_glassfish.dev.java.net
>>>>>>> For additional commands, e-mail:
>>>>>>> quality-help_at_glassfish.dev.java.net
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>> ---------------------------------------------------------------------
>>>>>>
>>>>>> To unsubscribe, e-mail: quality-unsubscribe_at_glassfish.dev.java.net
>>>>>> For additional commands, e-mail: quality-help_at_glassfish.dev.java.net
>>>>>>
>>>>>>
>>>>> ---------------------------------------------------------------------
>>>>> To unsubscribe, e-mail: quality-unsubscribe_at_glassfish.dev.java.net
>>>>> For additional commands, e-mail: quality-help_at_glassfish.dev.java.net
>>>>>
>>>>
>>>>
>>>> ---------------------------------------------------------------------
>>>> To unsubscribe, e-mail: quality-unsubscribe_at_glassfish.dev.java.net
>>>> For additional commands, e-mail: quality-help_at_glassfish.dev.java.net
>>>>
>>>>
>>>>
>>>
>>>
>>> ---------------------------------------------------------------------
>>> To unsubscribe, e-mail: quality-unsubscribe_at_glassfish.dev.java.net
>>> For additional commands, e-mail: quality-help_at_glassfish.dev.java.net
>>>
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: quality-unsubscribe_at_glassfish.dev.java.net
>> For additional commands, e-mail: quality-help_at_glassfish.dev.java.net
>>
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: quality-unsubscribe_at_glassfish.dev.java.net
> For additional commands, e-mail: quality-help_at_glassfish.dev.java.net
>