108514 - the description of resolution INVALID should be clarified

Status: RESOLVED WORKSFORME

(bugzilla.mozilla.org :: General, defect)

Description

[Diego Biurrun]

There has been some debate in bug 108416 about resolving a bug as INVALID versus
resolving it as WORKSFORME. In the following private discussion we came to the
conclusion that there is some room for improvement in the wording of the
description so that outdated bugreports or those for outdated builds should also
be covered by the INVALID resolution, since it is current practice for testers
to resolve bugs against outdated milestones as INVALID.

As a minimum we should adopt the wording from the bugs page
http://www.mozilla.org/bugs/where you can read

INVALID
The problem described is not a bug, or not a bug in Mozilla.
versus

INVALID
The problem described is not a bug

in http://bugzilla.mozilla.org/bug_status.html

As a third alternative we might add a new resolution like OBSOLETE, but INVALID
should really cover that. So I suggest changing the description to 
INVALID
The problem described is not a bug, not a bug in Mozilla or outdated.
Discuss.

Comment 1

[Diego Biurrun]

And while we are clarifying this page, we might as well update the description
of  severity critical to include hangs:

Critical
crashes, loss of data, severe memory leak

--->

Critical
crashes, hangs, loss of data, severe memory leak

Comment 2

[Matthew Tuck [:CodeMachine]]

Sounds like a mozilla.org specific issue.  However, resolution descriptions will
be updateable once bug #94534 lands.

Comment 3

[mental]

*** Bug 125637 has been marked as a duplicate of this bug. ***

Comment 4

[Diego Biurrun]

Attachment: update description

I went the mile and changed INVALID to

The problem described is not a bug, not a bug in Mozilla or a bug in an
outdated version of Mozilla.

Comment 5

[Ben Bucksch (:BenB)]

Please see the dup bug, which has some discussion.

Comment 6

[Ben Bucksch (:BenB)]

I don't think that this bug depends on bug 94534.

Comment 7

[Diego Biurrun]

I absolutely agree with you, Ben.  I am removing that dependency.  This is about
the description, not the resolution itself.

Comment 8

[mental]

*** Bug 136596 has been marked as a duplicate of this bug. ***

Comment 9

[Andreas Krüger]

In my opinion, this would be even clearer with an additional "," added:

The problem described is not a bug, not a bug in Mozilla, or a bug in an
outdated version of Mozilla.

(English grammer allows it either way, with or without that ",").

Comment 10

[Andreas Krüger]

But as I am considering the text, I wonder wether the wording is wise.

If I am reporting a bug against Bugzilla, as opposed to Modzilla, - sure, it's
not a bug in Modzilla, I may think.  That by itself does not make it invalid.

I suggest "Modzilla" should be replaced by a more general term:

The problem described is not a bug, not a bug in our product, or a bug in an
outdated version of our product.

This would also be friendly for people like myself who use Bugzilla for their 
own purposes, outside the Modzilla universe.

Comment 11

[Diego Biurrun]

Changing the wording from "Mozilla" to something more generic does not sound
like a bad idea, but is probably outside the scope of this bug.  Bugzilla is
supposed to be templatised in the upcoming version 2.16.  Maybe this can help you.

Comment 12

[Andreas Krüger]

I agree that when the template thing comes through, this whole issue will 
probably be dealt with.

In the meantime, refering to "Modzilla" as the only product name breaks current 
use, both here at the original Bugzilla installation, as well as at many other 
sites.  In my opinion, the proposed fix would introduce a new bug.

This very bug 108514 is not a bug in "Modzilla" (it's a bug in Bugzilla), but 
it is not invalid, either.

Comment 13

[Ben Bucksch (:BenB)]

> as well as at many other sites.

This bug is in the "product" mozilla.org. It applies to bugzilla.mozilla.org
only, not all BugZillas.

BTW: It's "Mozilla", not "Modzilla".

Comment 14

[Diego Biurrun]

The templatisation tracking bug 86168 is targetted for Bugzilla 2.18, the
templatisation of show_bug.cgi is finished, see bug 110012, so it will be in
bugzilla 2.16 to be released real soon now for the last month or two.

This bug is not about Bugzilla.  It is about unclear documentation on b.m.o
leading to unnecessary bug reports about Mozilla.  It is a problem in the
Mozilla community process that I want to have fixed.  Other products have their
own descriptions of INVALID and the other resolutions, just have a look at Red
Hat's Bugzilla.

Comment 15

[Andreas Krüger]

Please excuse the noise!

I came over from bug 136596, which is a bug I filed against the generic product 
BugZilla, and which was marked as a duplicate of this one.  I don't think that 
bug actually is a duplicate of this one, as they correspond to different 
products.

On the other hand, I do no longer think my comments about a generic name are 
valid in the context of this bug.  So I intend leave this bug alone, with only 
a "," to consider.

Comment 16

[WADA:World Anti-bad-Duping Agency]

Adding two bugs in "Depends on:", Bug 295935(NOTABUG) and Bug 295936(NOTMOZ).

Comment 17

[Gervase Markham [:gerv]]

This needs a new patch, to fields.html.tmpl instead of to bug_status.html. And it's pretty generic - it should be in the Bugzilla product.

Gerv

Comment 18

[Frédéric Buclin]

We are not going to mention Mozilla in the documentation by default. What you want is the doc on b.m.o to be updated. -> mozilla.org.

Comment 19

[Dave Miller [:justdave]]

Rearranging this component to not have me as the default assignee, so that it doesn't appear like I'm intending to work on bugs that other people could be taking care of if they didn't think I was already doing them.  If this bug is a software issue on b.m.o and you'd like to fix it, the modified source is now available for the patching (see bug 373688).  Reassigning to the new default owner.

Comment 20

[Wayne Mery (:wsmwk)]

How about
 The problem described is not a bug in this product or it's most current code.

other examples cited in bug 125673 comment 2 could be either WFM or dupe to a fixing or otherwise relevant bug.

Comment 21

[Ben Bucksch (:BenB)]

I strongly disagree. INVALID should be reserved for bugs which were *never* valid, i.e. the reporter was confused.
Bugs which were valid when filed, but are not longer existing (either a) because the problem has been resolved or b) the code/circumstance has gone away, need one or two new flags.
Some installations make bug filer statistics (some people did that for bugzilla.mozilla.org did that), some even base employee ratings on that. Also, it's says a lot about a project whether a bug was filed wrong, thus should not be counted, or the project simply sits out the bugs so many years until the code disappears. Even more so, it's important for the single bug: Entirely different thing whether it never was a bug or whether it was a bug that has disappeared.

So, I propose:
INVALID: The problem described is not a bug, and never was.
NOLONGEREXISTS: The problem existed once, but no longer does, because the code went away or the situation is no longer relevant.
FIXEDBYUNKNOWN: The problem has been fixed in the meantime, but it's unknown by whom or what change.

I deeply feel that the current use of INVALID and WONTFIX states for the latter two is severe abuse and in fact an insult on the reporter, and misstating the reality.

Comment 22

[Gervase Markham [:gerv]]

Ben: the FIXEDBYUNKNOWN and NOLONGEREXISTS situations should be WORKSFORME. 

I think with the addition of INCOMPLETE, which is a polite resolution, the current set covers the range adequately. If someone disagrees, please open a new bug (after checking for dupes) with their reasoning.

Gerv