Quantity of open bugs
 

Kevin F. Quinn posted on Thu, 10 Mar 2011 20:25:10 +0000 as excerpted:

> Hi all,
>
> I was nosing through bugzilla, and noticed:
>
> * Number of open bugs is greater than 14,000
> * open bugs untouched > 2 years - well over 2000.
> * open bugs untouched 1 - 2 years - well over 2000.
> * open bugs untouched 6 mo to 1 year - well over 2000.
> * open bugs untouched 3 - 6 months - over 2000
>
> The winner is bug #78406, which hasn't been touched for over 2240 days -
> over 6 years - at the time of writing.
>
> I would guess these old untouched bugs aren't actually going to be
> touched, ever - a lot simply won't be relevant any more for one reason
> or another. All they're doing is cluttering up bugzilla.
>
>
> So I'd like to suggest a drastic, perhaps controversial action. Mark
> all bugs that haven't been touched for over (say) 3 months as
> "Resolved:Wontfix", with a polite comment saying that it is closed due
> to lack of resource amongst the volunteer developer community.

You have a point, but for the way Gentoo works, 3 months is rather too
short. Gentoo tracks all sorts of not-ordinarily-considered-bugs as bugs,
and some of the stuff tracked is long-term projects. I've had (gentoo
initscript feature) bugs complete with patches sit for > six months,
before the Gentoo package maintainer had time to look at it and say yeah,
the idea looks good. Another user ended up updating the patch before it
was applied, as stuff /had/ changed, but it /was/ eventually applied.
Meanwhile, both the other user and I (and who knows if anyone else) had
been using the feature in our own initscripts, keeping it working, etc,
but the package didn't turn over /that/ frequently, and over a year to
resolve with a six-month and a three-month idle period wasn't /that/ bad
-- certainly considering that the patch and feature ultimately got in, and
it wouldn't have with your proposal unless someone simply bumped the bug
for no other reason than to keep it open.

Arguably, a year might be better, or possibly six months, but certainly,
the auto-close message should urge the user to re-open if it's still
appropriate. But I'm not sure even that will go over well. Maybe two
years or five years... because arguably at five years, no matter what the
bug is, if it's still valid, it really /needs/ updated, since so much
around it will have changed.

--
Duncan - List replies preferred. No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master." Richard Stallman

Quantity of open bugs
 

On 03/10/2011 02:25 PM, Kevin F. Quinn wrote:

Hi all,

I was nosing through bugzilla, and noticed:

* Number of open bugs is greater than 14,000
* Number of open bugs untouched for more than 2 years - well over 2000.
* Number of open bugs untouched between 1 and 2 years - well over 2000.
* Number of open bugs untouched between 6 months and 1 year - well over
2000.
* Number of open bugs untouched between 3 months and 6 months - over
2000

The winner is bug #78406, which hasn't been touched for over 2240 days
- over 6 years - at the time of writing.

I would guess these old untouched bugs aren't actually going to be
touched, ever - a lot simply won't be relevant any more for one reason
or another. All they're doing is cluttering up bugzilla.
I think Duncan has already covered the major points I was going to
mention: particularly with respect to the fact that we are all
volunteers and thus subject to resource constraints that other projects
might not have. I realize that it is frustrating to a user to have a
bug sit for a year (or more) without ever being resolved (or even looked
at), but there is really only one way to resolve that: get someone who
has the time and expertise to step in and fill the gap. Given that we
can't exactly hold a gun to people's heads and MAKE them work on Gentoo
stuff (nor would I personally be inclined to trust code produced using
such methods), I really don't see another alternative.


We closed a number of bugs related to SELinux recently; many of those
bugs had been open for quite some time and things had changed
sufficiently that we believed that the bug itself was no longer
relevant, or we needed feedback from the user and didn't get any. Some
of those bugs had been open for a couple of years. But we reviewed EACH
of those bugs and made a decision on a case-by-case basis.


I understand and appreciate the desire to close open bugs that are
cluttering up the bugzilla. Not only do they create extra cruft for
everyone to wade through, they also make Gentoo look bad (my GOD,
they've got open bugs dating back to the founding of the Roman
Empire!). However, I'm not convinced that blanket closing bugs that are
over x days (weeks, months, years) is the best (or even desirable) approach.


If a bug is related to a package that no longer exists, then it seems
pretty obvious that there is no need to keep the bug around.


If the bug is waiting on feedback from a user, and that user hasn't
provided the requested feedback in, say, 60 days (after a bump to the
bug) then I'd say that the bug is arguably no longer of importance to
the user, or at least the email address we have on file for that user
doesn't work any more.


Beyond those two conditions, I'd really be loathe to close anything
without good evidence to indicate that it either is no longer relevant,
or it can't be fixed.


Just my $0.02 (not adjusted for currency devaluation)

Later,
Gizmo

Quantity of open bugs
 

On Thu, 10 Mar 2011 20:25:10 +0000
"Kevin F. Quinn" <ml@kevquinn.com> wrote:

> I would guess these old untouched bugs aren't actually going to be
> touched, ever - a lot simply won't be relevant any more for one reason
> or another. All they're doing is cluttering up bugzilla.

I never understand this argument. How does an open bug "clutter" up bugzilla
any more than a closed bug?

It's pretty simple. If a bug isn't fixed, it should be open. Automatically
closing them doesn't make things "better", it just generates shitloads of
useless mail.


--
fonts, gcc-porting, it makes no sense how it makes no sense
toolchain, wxwidgets but i'll take it free anytime
@ gentoo.org EFFD 380E 047A 4B51 D2BD C64F 8AA8 8346 F9A4 0662

Quantity of open bugs
 

On Thu, Mar 10, 2011 at 12:25 PM, Kevin F. Quinn <ml@kevquinn.com> wrote:
> Hi all,
>
> I was nosing through bugzilla, and noticed:
>
> * Number of open bugs is greater than 14,000
> * Number of open bugs untouched for more than 2 years - well over 2000.
> * Number of open bugs untouched between 1 and 2 years - well over 2000.
> * Number of open bugs untouched between 6 months and 1 year - well over
> *2000.
> * Number of open bugs untouched between 3 months and 6 months - over
> *2000
>
> The winner is bug #78406, which hasn't been touched for over 2240 days
> - over 6 years - at the time of writing.
>
> I would guess these old untouched bugs aren't actually going to be
> touched, ever - a lot simply won't be relevant any more for one reason
> or another. *All they're doing is cluttering up bugzilla.
>
>
> So I'd like to suggest a drastic, perhaps controversial action. *Mark
> all bugs that haven't been touched for over (say) 3 months as
> "Resolved:Wontfix", with a polite comment saying that it is closed due
> to lack of resource amongst the volunteer developer community. *I'm
> sure a suitable bugzilla script wiz could do that relatively
> easily. *Users who care about such bugs can still comment on them, or
> talk directly to the assigned dev to highlight it's still a relevant
> issue to them, or even to supply a solution against the current tree.

I'm curious what the root problem is. In general I do not believe
'having lots of bugs open' is an actual problem for Gentoo. Is it
hard to search for bugs? (new bugzilla search non-withstanding.) Are
users upset that their new bug is a dupe of a bug that is already
years old?

-A

>
> It could be an ongoing policy, in which case, users who care about
> them can keep bugs alive simply by posting useful updates to the bug,
> describing how the issue still applies to a new revision for example.
>
> Just a thought from an old ex-dev...
>
> Kev.
>
>
>
>

Quantity of open bugs
 

On 10/03/2011 20:25, Kevin F. Quinn wrote:

Hi all,

I was nosing through bugzilla, and noticed:

* Number of open bugs is greater than 14,000
* Number of open bugs untouched for more than 2 years - well over 2000.
* Number of open bugs untouched between 1 and 2 years - well over 2000.
* Number of open bugs untouched between 6 months and 1 year - well over
2000.
* Number of open bugs untouched between 3 months and 6 months - over
2000

The winner is bug #78406, which hasn't been touched for over 2240 days
- over 6 years - at the time of writing.

I would guess these old untouched bugs aren't actually going to be
touched, ever - a lot simply won't be relevant any more for one reason
or another. All they're doing is cluttering up bugzilla.


So I'd like to suggest a drastic, perhaps controversial action. Mark
all bugs that haven't been touched for over (say) 3 months as
"Resolved:Wontfix", with a polite comment saying that it is closed due
to lack of resource amongst the volunteer developer community. I'm
sure a suitable bugzilla script wiz could do that relatively
easily. Users who care about such bugs can still comment on them, or
talk directly to the assigned dev to highlight it's still a relevant
issue to them, or even to supply a solution against the current tree.

It could be an ongoing policy, in which case, users who care about
them can keep bugs alive simply by posting useful updates to the bug,
describing how the issue still applies to a new revision for example.

Just a thought from an old ex-dev...

Kev.

Why not give userrel a list of 2000 bugs from 6 years old to 6 months
old to elcit help from the community?


A dev could look why a user marked it cantfix e.g. "refers to outdated
version of baselayout" and tick it off the list so it no longer appears.


G

Quantity of open bugs
 

On 20:25 Thu 10 Mar , Kevin F. Quinn wrote:
> Hi all,
>
> I was nosing through bugzilla, and noticed:
>
> * Number of open bugs is greater than 14,000
> * Number of open bugs untouched for more than 2 years - well over 2000.
> * Number of open bugs untouched between 1 and 2 years - well over 2000.
> * Number of open bugs untouched between 6 months and 1 year - well over
> 2000.
> * Number of open bugs untouched between 3 months and 6 months - over
> 2000
>
> The winner is bug #78406, which hasn't been touched for over 2240 days
> - over 6 years - at the time of writing.
>
> I would guess these old untouched bugs aren't actually going to be
> touched, ever - a lot simply won't be relevant any more for one reason
> or another. All they're doing is cluttering up bugzilla.
>
>
> So I'd like to suggest a drastic, perhaps controversial action. Mark
> all bugs that haven't been touched for over (say) 3 months as
> "Resolved:Wontfix", with a polite comment saying that it is closed due
> to lack of resource amongst the volunteer developer community.

I do come back to bugs after years. They should not be closed if they
are not fixed. "WONTFIX" for me means that there was a decision made
that this will not be fixed, but that is not the case.

+1 for the argument that 14000 open bugs is not a problem. Bugzilla is
not something that needs to be clean and tidy. Closing them would
generate a lot of work because of false positive while there is zero
benefit.

Cheers,
Thomas

> sure a suitable bugzilla script wiz could do that relatively easily.
> Users who care about such bugs can still comment on them, or talk
> directly to the assigned dev to highlight it's still a relevant issue
> to them, or even to supply a solution against the current tree.
>
> It could be an ongoing policy, in which case, users who care about
> them can keep bugs alive simply by posting useful updates to the bug,
> describing how the issue still applies to a new revision for example.
>
> Just a thought from an old ex-dev...
>
> Kev.
>
>
>

--
Thomas Kahle
http://dev.gentoo.org/~tomka/

Quantity of open bugs
 

Il giorno gio, 10/03/2011 alle 20.25 +0000, Kevin F. Quinn ha scritto:
>
> * Number of open bugs is greater than 14,000

For the record, about 10% of those are reported by me through tinderbox
(1506 open bugs reported by me as of today). A number of those date back
in 2008 and earlier, and could be fixed reasonably quickly if the
packages were maintained. And no, me or any other number of people going
through to fix just those is not feasible, see my old post at [1] for
some reasons.

I actually use the fact that these are still open to judge whether a
package has to be removed from the tree, closing them would definitely
be a bad idea for two reasons:

- makes it harder to see whether the package has any maintainer at all;
- would waste my time as I'd be re-opening the same exact bug at the
following tinderbox iteration, as the original bug was closed (and no, I
wouldn't go _reopening_ the bug, since I wouldn't remember there was one
already most of the time, so the load on bugzilla would increase).

So if somebody would still have doubts about this, I think the point is
vetoed to close bugs without action as WONTFIX after any time at all.
Rather get rid of the package in that case. And you can challenge that
with the council if you wish as I'm weighting that in as part of QA,
thank you very much.

OTOH if a bug is waiting for user to report build logs or other kind of
test results, closing as TEST-REQUEST or NEEDINFO is likely a good idea.

[1] http://blog.flameeyes.eu/2009/12/28/the-five-minutes-fix-myth

--
Diego Elio Pettenò — Flameeyes
http://blog.flameeyes.eu/

Quantity of open bugs
 

On 15:45 Sat 12 Mar , Diego Elio Pettenò wrote:
> I actually use the fact that these are still open to judge whether a
> package has to be removed from the tree, closing them would definitely
> be a bad idea for two reasons:

I'm assuming you're talking only about broken builds here and not
"QA-only" bugs. My opinion is that if a tinderbox QA script is the only
thing finding a nonfatal bug, and it's never reported or CC'd by a user,
then it's about as low priority as you can get.

So this might serve as a pointer to potentially unmaintained packages,
but clearly more investigation is required before removal.

--
Thanks,
Donnie

Donnie Berkholz
Sr. Developer, Gentoo Linux
Blog: http://dberkholz.com

Quantity of open bugs
 

Il giorno sab, 12/03/2011 alle 11.09 -0600, Donnie Berkholz ha scritto:
>
>
> I'm assuming you're talking only about broken builds here and not
> "QA-only" bugs. My opinion is that if a tinderbox QA script is the
> only
> thing finding a nonfatal bug, and it's never reported or CC'd by a
> user,
> then it's about as low priority as you can get.

Not really. An user would never report that the package is bundling
libraries, but that is actually pretty high in priority as it can lead
to hidden security issues already resolved in the original library to
sneak in the system.

At the same time, very few users report ignored variables (CC, CFLAGS,
LDFLAGS, ...) but they are just the same a problem. Especially when
hardening flags are not used at all.

> So this might serve as a pointer to potentially unmaintained
> packages,
> but clearly more investigation is required before removal.

There is always the need to do manual investigation. But in general when
you see a package that

- ignores LDFLAGS;
- shows fortify source warnings;
- ignores CC;
- misuses autotools;
- bundle libraries.

you can pretty safely assume neither somebody is looking after it, nor
using it.

--
Diego Elio Pettenò — Flameeyes
http://blog.flameeyes.eu/

Quantity of open bugs
 

On Sat, 12 Mar 2011 15:45:34 +0100
Diego Elio Pettenò <flameeyes@gmail.com> wrote:

> And you can challenge that with the council if you wish as I'm
> weighting that in as part of QA, thank you very much.

Again that language. Why do you talk like that?


jer