[Libreoffice-qa] do not update MAB without a comment - was: Re: [Bug 37361] LibreOffice 3.5 most annoying bugs

Hi dE,

first, thanks a lot for helping with the bug triage. We need more hands
here. I have just a small note.


[hidden email] píše v St 18. 04. 2012 v 03:39 +0000:
> https://bugs.freedesktop.org/show_bug.cgi?id=37361
>
> dE <[hidden email]> changed:
>
>            What    |Removed                     |Added
> ----------------------------------------------------------------------------
>          Depends on|                            |42720
>

You added dependency on the bug 42720 but I do not see any explanation
in any of the two affected bugs.

Please, always mention explanation in the meta bug (fdo#37361). I has at
least two advantages:


1. The comment is send to the developers mailing list. Human readable
notice increases the chance that it triggers a bell and the bug will be
fixed soon by an interested developer. If you send just a bug number,
developers will ignore it.

By other words, it is easier to read one sentence than opening bugzilla,
searching for the bug, reading crazy comments from the initial reporter.


2. It helps to understand why it is so important and reduces later
fighting about it.

For example I am not sure why you added the bug 35158 :-) The comment #5
says that the primary problem is solved in LO-3.5. It does not say much
about the layout problems. People will not download the 1.4MB document
just to see how it is serious. Have you tried it? Does it affect many
users?

Note that the bug is assigned to a developer, so we are avare of it. The
question is if it is really that important that we should fix it ASAP.
There might be other bugs that more annoy more users.


Best Regards,
Petr

_______________________________________________
List Name: Libreoffice-qa mailing list
Mail address: [hidden email]
Change settings: http://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

Petr Mladek píše v St 18. 04. 2012 v 10:21 +0200:
> For example I am not sure why you added the bug 35158 :-)

I forgot to say that I agreed with the other nominated bugs. Great work!

Thanks for help with triaging.


Best Regards,
Petr

_______________________________________________
List Name: Libreoffice-qa mailing list
Mail address: [hidden email]
Change settings: http://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

On 04/18/12 13:59, Petr Mladek wrote:
Don't you think bugzilla's bug priorities should be used instead of this
meta bug?

Also making these kind of things public is bound to cause controversy, I
don't think it's a good idea.
_______________________________________________
List Name: Libreoffice-qa mailing list
Mail address: [hidden email]
Change settings: http://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

dE . píše v St 18. 04. 2012 v 20:23 +0530:
Good point! Well, we can't use the bugzilla severity/priority at this
point. Though, we should update these fields when triaging the bugs. It
will allow to use it in the future. Please read
https://bugs.freedesktop.org/show_bug.cgi?id=37361#c285 for more
details.


> Also making these kind of things public is bound to cause controversy, I
> don't think it's a good idea.

I sometimes speak too much. On the other hand this is open project and
we are improving the processes together. The reasons for adding comments
in the most annoying bug were newer discussed here. I though that it
might be useful for others as well. What exact thing from my mail could
cause controversy? :-)


Best Regards,
Petr

_______________________________________________
List Name: Libreoffice-qa mailing list
Mail address: [hidden email]
Change settings: http://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

On 18/04/12 16:53, dE . wrote:
in my experience the priorities in freedesktop.org bugzilla are utterly
useless because everybody can change them, which leads to some reporters
setting their own pet bugs to the highest possible priorities, which are
then never corrected.

due to this problem i have learned to ignore the priorities; imho to
prevent this abuse the priorities should be settable only by experienced
QA or developers, not by everybody.

the MAB is already a step in the right direction (because a lot of
reporters don't know that it exists, and also because QA people actually
seem to pay attention to what gets added), but what really gets my
attention these days is a "regression" keyword.

_______________________________________________
List Name: Libreoffice-qa mailing list
Mail address: [hidden email]
Change settings: http://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/

Michael Stahl-2 wrote

imho to
prevent this abuse the priorities should be settable only by experienced
QA or developers, not by everybody.

Having done such mistakes in the past (still doing?) I couldn't agree more.

Michael Stahl-2 wrote

but what really gets my attention these days is a "regression" keyword.

Excellent :) Regressions are indeed the worst form of getting people unhappy. Getting stuck on something that used to work and doesn't any more is more frustrating that something that never worked :)

El 18/04/12 20:19, Pedro escribió:
+1
I have remarked sometimes about this.

Regression are the worst, specially those what change the results
without notice, more critic in calc. And those what leads a crash.

I believe the regressions must be have the high priority, and IMHO solve
the regressions make easy the development, because avoid crossing bugs
with new improvements.

Miguel Ángel.
_______________________________________________
List Name: Libreoffice-qa mailing list
Mail address: [hidden email]
Change settings: http://lists.freedesktop.org/mailman/listinfo/libreoffice-qa
Problems? http://www.libreoffice.org/get-help/mailing-lists/how-to-unsubscribe/
Posting guidelines + more: http://wiki.documentfoundation.org/Netiquette
List archive: http://lists.freedesktop.org/archives/libreoffice-qa/