"show only duplicates" in advanced search options

towsonu2003, you're not supposed to change the status of a bug report after marking it as a duplicate. There's no benefit in doing that, and it makes extra work if the report is reopened later. (We should make it impossible, but haven't gotten around to it.)

Is there any other reason you want to be able to show only duplicates in search results?

towsonu2003, would this be solved by bug 3796?

> towsonu2003, you're not supposed to change the status of a bug report
> after marking it as a duplicate. There's no benefit in doing that,
> and it makes extra work if the report is reopened later. (We should
> make it impossible, but haven't gotten around to it.)
>
> ** Changed in: malone (upstream) Status: Unconfirmed => Needs Info
>

Then could you please document it in a place that one can see easily,
until you make it impossible.

As far as I know, normal procedure for marking a duplicate involves:
1. find the original
2. mark bug X duplicate
3. close bug X duplicate
4. let the reporter of bug X why the bug is closed

To reopen a dupe:
1. unmark duplicate
2. reopen duplicate

This is what I see in gnome's bugzilla and openoffice's issue tracker.
They do not say anything about closing a dupe as extra work. I'm not
sure how it creates extra work either (1 more step for each procedure
doesn't seem bad if you're not on dial up).

So far, the open dupes confuse me (for example, one is marked
"confirmed" even though there is the notice "this is a dupe" just above
the status... what's that now?..)

and they screw up bug statistics for Ubuntu...

thanks :)

>
> Is there any other reason you want to be able to show only duplicates
> in search results?

no.

> towsonu2003, would this be solved by bug 3796?

yes.

On 18-Oct-06, at 10:47 AM, towsonu2003 wrote:

>> towsonu2003, you're not supposed to change the status of a bug report
>> after marking it as a duplicate. There's no benefit in doing that,
>> and it makes extra work if the report is reopened later. (We should
>> make it impossible, but haven't gotten around to it.)
>>
>> ** Changed in: malone (upstream) Status: Unconfirmed => Needs Info
>>
>
> Then could you please document it in a place that one can see easily,
> until you make it impossible.

I appreciate the suggestion, though I think only a fraction of users
will ever read any Malone documentation (and who can blame them? :)
Prioritizing a fix for bug 3796 is the best next step, IMHO.

Bugzilla and OpenOffice's issue tracker have a verification process for resolutions, where bugs/issues can be (for example) VERIFIED DUPLICATE. Launchpad doesn't have this, though, primarily because nobody has asked for it.

If there's no other reason to restrict a search only to duplicates, I'll reject this for now.

I would prefer this to stay as a dupe of bug 3796. this wouldn't be a problem for me at all if bug # 3796 was fixed. as Brad Bollenbach mentioned, somehow "prioritizing a fix for bug 3796 is the best next step, imho" :)

it's just confusing and kinda wrong to see a bug that is marked both as fixreleased, unconfirmed, needsinfo, or confirmed *and* duplicate... that was why I wanted to find those bugs and fix them (and doing so caused some troubles with the ubiquity developer).

thanks :)

Bug 3796 is about presenting duplicate bugs differently. This report is about being able to search for bugs by duplicate status. The requests are completely different, even if their motivation was the same.

In the future we may want to allow filtering by duplicate status for different reasons, in which case this bug will be valid regardless of whether bug 3796 is fixed.