[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-devel
Subject:    Re: QA Team [JRT]
From:       Nicolas Goutte <nicolasg () snafu ! de>
Date:       2004-01-29 20:52:41
Message-ID: 200401292152.41677.nicolasg () snafu ! de
[Download RAW message or body]

On Thursday 29 January 2004 21:13, James Richard Tyrer wrote:
> First, I am willing to participate in any organized effort -- contact me by
> private e-mail.
>
> However, I also wanted to blog about this a little:
>
> Our current development model (Bugzilla) is set up as though we had a QA
> department (but we don't).

Well, it could work without one too. (Changing the bug status from RESOLVED to 
CLOSE directly without VERIFY)

But anyway we will probably never close the bugs, as it would make an email 
storm. (Or anybody tells me how to do it for KOffice 1.3 without doing an 
email storm.)

(Also be careful that Bugzilla is not really our development model. When the 
bug database had to be changed, we chose Bugzilla.)

>
> A simple cure for this is that if you report a bug you need to appoint
> yourself the advocate for that bug.  If possible make test cases but in any
> case:
>
> Until the next release, you should follow the progress in fixing the bug.
>
> When the next release is made, you should validate that the bug has been
> fixed in the release -- like the QA department does.

Yes, that can be done.

>
> Unfortunately, it is possible that developers will not appreciate this.

Yes, but that is one role of QA. It is to be a third partner between reporter 
and developer.

>
> My experience with bug:
>
> http://bugs.kde.org/show_bug.cgi?id=73379
>
> has been a little mixed.  Please read in comment #3:
> > I don't see how this can be the wrong widget. What is wrong about saying
> > the file already exists and suggesting a new name?
> >
> > IMO, this is no showstopper for KDE 3.2. Most people don't even know
> > they can divide the window in two.
>
> And comment #9.  I have always suspected that bugs are being marked as
> fixed when they actually aren't.

Yes, the temptation to close a bug that is not understood is too easy. (It 
will never be completely avoided.)

>
> Attitudes such as this are not what we need to get a high quality produce.
>
> In any case, the coders need to realize that it is the job of the QA team
> to see that the bugs get fixed -- that by doing this job that they are
> helping with the development of high quality code.

Just be careful that the QA team does not dictate what the developers have to 
do. That is also not its role and would end to be counter-productive.

>
> NOTE: most of bug 73379 has been fixed, but "Test case #1" still fails, so
> it will be retitled again.
>
> --
> JRT

Have a nice day!

>
> >> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to
> >> unsubscribe <<

 
>> Visit http://mail.kde.org/mailman/listinfo/kde-devel#unsub to unsubscribe <<
[prev in list] [next in list] [prev in thread] [next in thread]