ubuntu-bugcontrol team mailing list archive

[C de-Avillez <hggdh2@xxxxxxxxxx>]

On Thu, 25 Mar 2010 20:21:39 +0100
Kiwinote <kiwinote@xxxxxxxxx> wrote:

Kiwinote -- I had not seen your reply *before* mine's to Gary et all.

> Hi everyone,
> 
> In order to better understand these importances, I will attempt to
> clarify some of the importances I chose. Hopefully this will lead to
> a better understanding.
> 
> On Thu, Mar 25, 2010 at 5:23 PM, Brian Curtis <bcurtiswx@xxxxxxxxxx>
> wrote:
> >
> > On Wed, Mar 24, 2010 at 10:02 AM, Kiwinote <kiwinote@xxxxxxxxx>
> > wrote:
> >
> >> 5b. https://bugs.launchpad.net/bugs/530187
> >>
> >  I don't understand why you assigned yourself to the bug report
> > here since Gary Lasker fixed it.
> >
> (I did actually fix this one.)

No discussions here. I stand corrected.

> 
> >  5c. https://bugs.launchpad.net/bugs/497109
> >>
> >  Since it's a usability issue, I think this is more of a low
> > importance bug then medium.
> >
> Ok, although usability is something that very much has the focus in
> software-center. I would have chosen the importance medium because I
> would not expect an user to find trial software in the 'free
> software' category. Especially with paid software coming in later
> versions of software-center, it is a good thing that this bug has
> been resolved. But indeed this does only affect a small fraction of
> the packages.

This is a *valid* reasoning, and acceptable. It *can* be argued that
this is an important issue, even if hitting few users: after all
(IMHO), having trial software in the free section gives the wrong
message to the users. Free is free -- now and ongoing.

> 
> > 5d. https://bugs.launchpad.net/bugs/542892
> >>
> >  A cosmetic/usability issue that does not limit the functionality
> > of an application = Low Importance. Not medium
> >
> Ok, though this bug does actually break functionality. It means
> packages like this cannot be installed through software-center. Once
> again it only affects a small fraction of the packages.

I would still think it is low.

> 
> > I notice that only two of the above need the desired importance, so
> > here
> >> are a few more bugs chosen to demonstrate deciding importances.
> >> - low: https://bugs.launchpad.net/bugs/514859,
> >> https://bugs.launchpad.net/bugs/425850
> >> - medium: https://bugs.launchpad.net/bugs/514874,
> >> https://bugs.launchpad.net/bugs/514846
> >> - high: https://bugs.launchpad.net/bugs/514875,
> >> https://bugs.launchpad.net/bugs/528051
> >>
> >  514875 - you said high, but it's only medium
> >
> Don't really agree on this. One very common use case of
> software-center is to remove software. (Software-center replaces
> gnome-app-install, which was called "Add/Remove software" in the
> menu.) In the main menu of software-center there are two items, 'get
> software' and 'installed software'. If one wants to remove software
> it is hard to explain why the software isn't listed under 'installed
> software'. I could agree with this being wishlist though.

Agreed with WishList.

> 
> >  528051 - you said high, but it's only wishlist
> >
> Yes, this is a feature request, so strictly speaking it should be
> wishlist. This feature request is however the only simple way to fix
> a major issue, namely that 31000 items are being displayed in a list
> view. Firstly this takes about 8-10 seconds to load, so many users
> will give up, thinking that software-center has frozen. A small week
> ago this took about 25 seconds, but recent updates have diminished
> this time. There is also no way to browse the list as the items are
> not in a visible alphabetical order (strictly speaking they are in
> alphabetical order of package name). Scrolling lags significantly. I
> myself believe that this would be sufficiant reason to set this as
> high importance rather than wishlist.

I agree now.

> 
> >  514874 - you said medium, but it's only low
> >
> This I could probably agree with. But although it is a minor issue,
> it does affect all packages in main and restricted. Software-center
> is giving wrong information in all these cases. I chose medium
> because although this is a small issue, it is relatively simple to
> fix, and will benefit all users.

I agree (although the reasoning surpasses simple triaging, and goes
into your persona as developer.

> 
> > 514846
> >
>  This would be similar to the case above. It is very noticeable and
> makes the difference between a good first impression and a bad one.

Agree with the reasoning.

> 
> > low importance bugs
> >
> It is probably clear that these are indeed low importance.
> 
> > Im going to say -1 because you are very close, but there are still
> > things you should probably improve upon.  Don't hesitate to contact
> > me on #ubuntu-bugs in freenode IRC if you want any help.
> >
> Ok, thanks for taking a look. It would be nice to hear from you
> whether my choices make more sense with detailed explanations though.
> 
> Looking back over my reasonings and comparing it to
> wiki/Bugs/Importance I think I can draw a few conclusions. I seem to
> give an increased importance to bugs that are small, but affect all
> users. This could be considered in line with the aims of
> software-center. I am furthermore choosing importances based on the
> current set of software-center bugs and importances in Launchpad,
> rather than strictly following the guidelines for importance. I do
> think that this is better for software-center itself, but can see
> that it leads to a decrease in consistency throughout Ubuntu. This is
> a consequence of software-center in Ubuntu being the upstream.

Yes, I thought this would also play a role. Please note that the
guidelines are pretty much it -- *guide*lines. They are not set in
stone: we are expected to use the good old common sense when triaging.

On the other hand, every Importance that is set should be complemented
by the reasoning for it.

> 
> Please let me know what you think about this.

I really wish you would have included your reasonings when you applied.
On the other hand, the wiki page did not clearly indicate we would be
considering (as one of the *subjective* evaluation criteria) the
reasons for suggesting a specific Importance setting. I have corrected
it [1].

I am glad to be able to give you a

+1

now. This also happens to be the fastest change-of-opinion I have had
on my votes, so there's a new record ;-)

Regards, and thank you very much not only for your help, but also for
following up and bearing with us.

..C..


[1] https://wiki.ubuntu.com/UbuntuBugControl?action=diff&rev2=33&rev1=32

Attachment: signature.asc
Description: PGP signature