Closed Bug 35839 Opened 21 years ago Closed 18 years ago

REMIND and LATER considered harmful to bugzilla.mozilla.org

Categories

(bugzilla.mozilla.org :: General, defect, P3)

defect

Tracking

()

VERIFIED FIXED

People

(Reporter: ian, Assigned: endico)

References

Details

(The following is an edited copy of comments on bug 13534, which applies to 
Bugzilla itself. This bug applies to the Bugzilla installation at Mozilla.org. 
Jan, I believe you are the person who is in charge of customising Bugzilla, so
I've assigned this straight to you... apologies if I am off-base here!)

I think that we should endeavour to remove REMIND and LATER from Bugzilla. I
believe that they achieve more harm than good.

There are many disadvantages:

(a) They can easily not be found by queries. They are, after all, _unresolved_,
    even though they are marked as RESOLVED/LATER or RESOLVED/REMIND.
    This is a serious problem, as there are frequently new bugs being reported
    that are feature requests that have in fact been marked LATER or REMIND
    but which a quick scan of bugzilla did not find, since the default query
    does not include the LATER and REMIND resolutions.
(b) Reopening them results in comments/activity which clutter the bug report.
(c) They are often set by Netscape employees and hence imply Netscape's 
    agenda for what will go in.  Mozilla, at least, is a open project, where 
    anyone can contribute.
(d) It is difficult to tell exactly when the REMIND and LATER bugs will in fact 
    be reopened.

So, what's the alternative?  The obvious answer is: the milestone.

Remind can already be handled by moving the bugs to a later milestone where
they need to get reevaluated. (There could also be a "remind" keyword, to 
indicate that the bug has just been moved out but not closely looked at.)

For LATER, there needs to be an equivalent milestone. This means that there is
currently no plan for implementation, but it has been considered, and hence
won't appear on groups' new bugs radar. So the milestones would go M1, M2, ... 
M29, M30, LATER.

This, combined  with the introduction of assigning to "nobody@mozilla.org", as 
well as the "helpwanted" keyword and the "Severity: Enhanced" setting, covers 
all the issues of which I am aware.

This would require a little modification of queries by Netscape employees to
exclude untargetted bug reports, but they generally know how to use Bugzilla,
so it would take them little time to adapt. I think it is more important to 
make Bugzilla newbies more productive, and thus reduce the load on everyone.

So in summary: I suggest we remove the LATER and REMIND resolutions, and replace
them with a "LATER" milestone and a "remind" keyword.

[cc'ing the people involved in bug 13534]
CC:ing Dawn, who has actually implemented most of the recent 
mozilla.org- installation specific Bugzilla changes in recent weeks.
I should point out here that I believe there are plenty of Netscapers who agree,
I figure this because I have seen M20 used as a placeholder milestone, and
recently quite a few bugs were moved onto M30 from M20.  A LATER milestone would
never have to be moved out en masse.
I am going to run this by Seamonkey Leads mtg to.  Yeah.  This would be great as 
 milesone...
People should craft their queries to use the advanced restriction "Target
Milestone not equal to LATER", rather than saying that the milestone can be
M1-M30 and later losing records when M31 comes into existence.
Update...got the discussion going on this...hoping for more info to discuss in 
this bug later today.
One additional item to think about on this...REMIND and LATER statuses cause 
dependencies to get crossed off.  This is deceptive, since the dependent bug is 
still (presumably) blocked, and will remain blocked for the foreseeable future.
Overall many engineers do use REMIND and LATER for many reasons.  There are 
currently 353 bugs with these indications.  This is a tool that they use for 
planning their work.  It would be bad to remove this.  

Plus, you'all now have MFuture to move bugs out to that milestone if you like. 
Would like to Resolve this issue as "Won't Fix"...ok?
Ok, so now you've got us curious. *What* do engineers use this for, that 
milestone Future doesn't do? What would happen (besides a lot of spam) if, 
tomorrow, all those 353 bugs which were resolved REMIND/LATER were mass-REOPENED 
and moved to milestone Future?

Would there be anything lost by the removal of REMIND and LATER which could not 
be restored by a couple of controls on the bottom of show_bug.cgi?
|  _                                     _________________
| [_] Remind me about this bug once per [week___________:^]

Not only would this stop the bug from falling into the RESOLVED vacuum, it would 
also have the advantages of (a) being able to be applied to scheduled bugs and 
(b) being able to be used by people other than those who the bugs were assigned 
to (e.g. QA contacts). (Even if (b) would cause too much strain on the database, 
you'd still have (a).)
I agree with Matthew's last comment.
Sounds good to me MPT.  Could you file a bugzilla bug on it?

The absence of this doesn't prevent the removal of LATER however.

I think if possible we should immediately prevent resolved LATER, same for
REMIND if/when this RFE is implemented, and eventually remove the statuses once
the spam would be small enough.
The thing that would happen is that it would mess up a whole lot of people's 
bug lists and make it look like they suddenly have a lot more bugs to work on.
Certain people (and their managers) would need to redo their bug queries. It
would be a lot of busy-work when people should be busy working on shipping
mozilla.

Politely evangelizing the Future milestone and how it will make the world
better and bring about world peace and having people change their own bugs
would be fine. Since this has the backing of PDT I suspect that people will
slowly move over to that anyway. Give it time. However at this time I think
it would do more harm than good to try and force people to change their bugs.

I am not going to add anything to bugzilla to prevent people from setting
bugs to later/remind. If there are statuses that we dont' want to use then
we should they should just be removed.
I fully agree.  Marking Resolve/Won't Fix.
Status: NEW → RESOLVED
Closed: 21 years ago
Resolution: --- → WONTFIX
Can someone tell me whether or not REMIND actually does send out email, or was
that just an assumption on MPT's part?
reopening.  REMIND and LATER are silly, always have been, always will be.  Bugs 
in these states are typically lost forever (see some of the older, four-digit 
bugs still in this state) because they are considered closed, rather 
than "open, but not ready to do this at the current time." We now have the 
Future target milestone, so LATER/REMIND are even more redundant.  Dawn, you 
said if no one wants these resolutions anymore, we could remove them.  Well, I 
hardly see anyone use them anymore so...I think it's time.
Status: RESOLVED → REOPENED
Resolution: WONTFIX → ---
Assignee: leger → endico
Status: REOPENED → NEW
reassigning to dawn since Jan is gone, cc lchiang and sairuh

btw, there are currently 158 RESOLVED/LATER bugs, and nearly all of them were 
filed during the period in which MFuture wasn't available...suggesting that, 
had it been available, they would have been filed under it.
Blake, as is said above, mozilla.org doesn't want to do this to lots of bugs in
a bulk change, but rather let the change be gradual.

In particular, 2.5 months ago there were 153 REMIND & 197 LATER bugs.  Now there
are 116 REMIND & 138 LATER bugs.  So they are going away slowly.

Dawn, I suggest that we do keep this bug open, and Future it.  After N6 or
alternatively Moz1, we can expect to see many of these bugs be reopened.  At
this point, I expect the bug count to quite low.

Once it is low enough, it will become practical to do a bulk change on the
remaining bugs and pull any straggler developers into line.  This will also be
after schedule pressure is lessened so fixing their queries is not putting on
undue pressure.  Then we can finally get rid of these resolutions out of b.m.o.

I notice there is no "Future" milestone for mozilla.org.  You might want to add
one.
Err sorry, I think there was a misunderstanding.  Was it established whether
removing the statuses from being able to be used would cause any problems with
existing bugs with that status?

It is certainly true that REMIND and LATER are used very little now, although I
think I saw a REMIND resolution the other day.  Even if we can safely remove the
resolutions, this can be left until after N6 ships I think ...
No, they are not going away slowly.  They are going away because a couple other 
contributors and I took the initiative to email engineers and ask them if they 
minded if we reopened their REMIND/LATER bugs and Future'd them.  All of them 
said we could, and the overwhelming majority of them were surprised that LATER 
and REMIND had been deprecated in favor of Future.  Furthermore, after we 
reopened them, many of the bugs were found to be old or no longer valid, and 
thus were closed up -- because they were RESOLVED, they had long since dropped 
off the engineers' radars.

The resolutions are more trouble than they're worth, and it's silly to think 
that we should just wait around until they magically disappear.  The number 
would still be just as high as it was if we weren't actively searching for 
these resolutions, because they drop off radars and are never seen again.  
Let's educate engineers about the usefulness of Future (since apparently no 
one's taken the time to do this) and get rid of the resolutions.
Two more things:

- How do you expect VERIFIED/REMIND|LATER to just "gradually disappear"?  These 
bugs are as good as gone as far as radars go.

- The notion that Netscape 6's shipping date has anything to do with improving 
Mozilla's bug tracking system is lost on me.  Netscape can feel free to keep 
REMIND and LATER in their instance of Bugzilla, known as Bugscape.
> No, they are not going away slowly.  They are going away because a
> couple other contributors and I took the initiative to email
> ...

Yes I noticed this but wasn't aware it was on so large a scale.  I guess they've
all changed their queries then.

> The resolutions are more trouble than they're worth

Hey you don't have to tell me that, to my knowledge I was the first to actively
complain ...

> these resolutions, because they drop off radars and are never seen again.  

Not quite.  You could expect REMIND and LATER to be mass-reopened by the
engineers once the product has shipped.  Their whole purpose was to drop off the
engineers' radars (and that remains a purpose of MFuture).  Unfortunately there
was the unfortunate side-effect that they dropped off everyone's radars ...

> Let's educate engineers about the usefulness of Future (since
> apparently no one's taken the time to do this)

Agreed.  I was going to post a "State of bugzilla.mozilla.org" message a while
back including this issue but never got around to it.

> How do you expect VERIFIED/REMIND|LATER to just "gradually disappear"?
> These bugs are as good as gone as far as radars go.

Like I said, only until the product ships.

> The notion that Netscape 6's shipping date has anything
> to do with improving Mozilla's bug tracking system is lost
> on me.  Netscape can feel free to keep REMIND and LATER in
> their instance of Bugzilla, known as Bugscape.

This was just a temporary issue as a convenience to Netscape's engineers, who,
after all, still make a large proportion of b.m.o users, until they ship.

If people are still using REMIND and LATER in bugzilla.mozilla.org, then that's
definitely something we should do something about ...

I can't speak about Bugscape but I would hope the lessons of REMIND and LATER
have been learnt and they are not present.
unfortunately, in bugscape, the Remind and Later resolutions states are still
there...
I suggest a change in Bugzilla to send a mail every month with the list of REMIND 
and LATER bugs to all the programmers who have some assigned to them.  The list 
should display the summary line in addition to the bug number.  Then let people 
deal with them.  We're grown-ups.  Please do not remove these 2 states from 
Bugzilla, do not force us to put these bugs back under our daily radar.  Too much 
information can be more harmful than not enough.
Pierre, that's a nice thought, and might work in an ideal world.  But it's 
clear that that will have little effect on anything, as evidenced by the fact 
that Asa sent two notices to everyone with bugs targetted to M19 through M30 
(to retarget them), and there are still a little under 700 bugs with those 
milestones.
These bugs can be moved to Target Milestone Future.  This makes it pretty easy
for people to query around them.

What does Verified Remind mean? How is it different from Resolved Remind?  These
just don't make sense as Resolutions.  How is a bug Resolved if it is still
unresolved.  
I found a workaround that may be of some help to my fellow programmers: I'm going 
to open a second bugzilla account where I'll move all the bugs that I don't want 
to see in my daily list.  Nice, eh?
There are a number of tracking bugs for documentation, like bug 51601 or
bug 24689, mine is set on REMIND, as documentation bugs are never really
resolved :-).

What would be the right status for those? Having a doc bugs on daily radar isn't
nice :-(.

And I don't like the idea of another bugzilla account

Thanx to Asa for bringing this to attention by mail.

Axel
Pierre, if you don't want to see a bug on your daily list, you have no business 
owning the bug and you should reassign it to nobody@mozilla.org.

Owning bugs that you do not intend fixing in the near future is almost as bad as 
hoarding them in REMIND or LATER -- it slows down the whole project, by making it 
less likely that someone else will put in the effort to come up with patches to 
the bugs before you get to them yourself.
Well, in my opinion, you should consider replacing a tracking bug with a
keyword, unless you need one of the couple of things you can do with trackers
and not keywords.

If you keep the tracker, the standard way seems to be marking it with the "meta"
keyword and filter out based on that.
Pierre, we have this wonderful little feature near the bottom of the query page
called "Run the remembered query:" where you can save a query for "My Bugs" (or
smesuch name) which excludes Future.  You can even bookmark the query and add it
to your personal toolbar or give it a really cool keyword like "mybugs" so you
just type "mybugs" in the address bar and it will load a list of all of your
bugs that are not set to Target Milestone Future.   If you need help setting up
any of these queries, I'd be glad to assist.  I've lost considerably more time
searching for bugs which I know exist only to find them much later Verified
Remind or Resolved Later than I could possibly lose going around to all 15
people using these and personally setting up a bookmark or saved query.
mpt: that's your opinion, you should qualify it as such. I and many other people
disagree with your assertion, and in fact believe that nobody@mozilla.org should
be retired altogether. Not that I want to enter this argument here again.
*** Bug 66459 has been marked as a duplicate of this bug. ***
QA Contact: asa
Depends on: 66459
Component: Miscellaneous → Bugzilla: Other moz.org Issues
Depends on: 83129
Summary: REMIND and LATER considered harmful → REMIND and LATER considered harmful to bugzilla.mozilla.org
Note for those interested: REMIND and LATER are definitely considered harmful;
when bugzilla.mozilla.org upgrades to the forthcoming 2.16 release of Bugzilla,
a custom template which prevents use of these resolutions will be created.

Gerv
We just went through an upgrade, but remind and later are still in the "resolve
bug, changing resolution to ..." drop down box.
AFAICS this is fixed on the main page, but not the mass change page.
re: comment 34, I just removed it from the mass-change page.

As of 10 minutes ago there were 8 LATER and 6 REMIND bugs remaining.  Asa OKed
me to reopen those, so now there are none.

And this completes our long journey to removing REMIND and LATER. :)
Status: NEW → RESOLVED
Closed: 21 years ago18 years ago
Resolution: --- → FIXED
QA Contact: asa → choess
Can they be removed from the query page too, then?  (That would allow all the
resolutions to fit in the listbox without scrolling.)

Or do you want a separate bug on that?
should have been done as part of this, forgot about that one.  Done.
It's still in Charts, but that makes sense (since those are historical info).
VRFY FIXED.
Status: RESOLVED → VERIFIED
justdave, they're still in http://bugzilla.mozilla.org/queryhelp.cgi#resolution

and i think stephend is going to zap them from http://www.mozilla.org/bugs/ 
Component: Bugzilla: Other b.m.o Issues → General
Product: mozilla.org → bugzilla.mozilla.org
You need to log in before you can comment on or make changes to this bug.