| |||||||||||||
GNOME 3.4 releasedGNOME 3.4 releasedPosted Mar 30, 2012 8:26 UTC (Fri) by josh (subscriber, #17465)In reply to: GNOME 3.4 released by ovitters Parent article: GNOME 3.4 released
Some of the stock answers get a bit insulting as well. For instance, the message that effectively says "we don't know if this bug is fixed or not, but it's old and we've released a new version, so we're closing it anyway". Followed immediately by a bug reporter going "still a bug, reopening it".
(Log in to post comments)
GNOME 3.4 released Posted Mar 30, 2012 9:44 UTC (Fri) by ovitters (subscriber, #27950) [Link]
Eh.. I don't see that as insulting (I mean: at all)?
Sometimes you just don't know and have to rely on the reporter. Maybe some triagers are overactive in using it. E.g. closing just because of a new GNOME version isn't right. But if the bug is old and/or lots of new development happened in between, then it can be ok to ask if it still applies. But it is up to the triager/developer to make a judgement call.
Bugsquad decides entirely on how it is done in general for triagers. They used to ask the reporter and wait a while (e.g. 6 weeks, 3 months). But they noticed that often there was no feedback at all. So instead of leaving it open ("NEEDINFO"), they're closing it more quickly.
The reporter has the ability to reopen, so don't see anything wrong with this? It should say to please reopen, no?
Anyway, if you have suggestions on the stock answer, could you please email gnome-bugsquad@gnome.org (I tend to ignore LWN articles after a few days; hard to track if not a subscriber)? Or followup now and I'll mention it to them. Entire process change is better if you discuss directly (I'd just be a postman).
GNOME 3.4 released Posted Mar 30, 2012 10:11 UTC (Fri) by josh (subscriber, #17465) [Link]
I was talking specifically about the practice of mass-closing bugs due to the release of a new GNOME version, without any particular reason to believe that the specific bug has gone away.
I don't see anything wrong with requesting more information on a specific bug and closing it if the reporter doesn't provide that information. I just object to having to babysit a pile of bugs and mechanically reopen them each time someone mechanically closes them.
GNOME 3.4 released Posted Mar 30, 2012 11:29 UTC (Fri) by alankila (subscriber, #47141) [Link]
It is a tiring tactic. It is done to ensure that user base approaches 0 as time tends to infinity.
GNOME 3.4 released Posted Mar 31, 2012 13:34 UTC (Sat) by ovitters (subscriber, #27950) [Link]
That's happening? It shouldn't be done like that. Care to give pointers?
GNOME 3.4 released Posted Mar 31, 2012 19:30 UTC (Sat) by josh (subscriber, #17465) [Link]
Sure. I did a quick search for bugs closed with the OBSOLETE resolution, sorted from newest to oldest, and found several recent examples.
This one demonstrates particular rudeness: https://bugzilla.gnome.org/show_bug.cgi?id=671744
These just use the stock template I mentioned:
https://bugzilla.gnome.org/show_bug.cgi?id=672014
Valid reasons to close a bug: "can't reproduce that in the current version", or "the entire subsystem you reported the bug against doesn't exist anymore in the current version, and its replacement doesn't have that problem". Invalid reasons to close a bug: "there's a new version and we don't care about the old one anymore", without actually testing that the new version doesn't have the problem, especially when it turns out that it *does*.
GNOME 3.4 released Posted Mar 30, 2012 14:10 UTC (Fri) by ekj (guest, #1524) [Link]
It's fine to close it if it's been triaged, if there's some actual reason to believe it may no longer apply, such as big changes having happened in the subcomponent with the bug after the filing of the bug.
That's not the behaviour I've been seeing though. Instead what I've seen are mechanical mass-closings of the "This is 9 months old, and we're released a new version, so it's being closed", even in cases where the bug was narrowed down to specific functions that hasn't had a single change in those 9 months.
If they instead put it at "NEED INFO" and didn't get any, in a reasonable amount of time, then okay, fine, close the sucker. But I've never once seen that happen. Instead the bug goes straight from "CONFIRMED" to "CLOSED" with no questions posed.
The routine should be to *either* investigate slightly to see if there's any actual reason to think the bug is probably gone, or else, to atleast ask the submitter "Has anyone tested this with a current release ? Is it still relevant?" *then* close it if 3 weeks later there's been no response.
GNOME 3.4 released Posted Mar 31, 2012 13:35 UTC (Sat) by ovitters (subscriber, #27950) [Link]
That shouldn't happen. Please give me the bug numbers or any pointers.
GNOME 3.4 released Posted Mar 31, 2012 15:34 UTC (Sat) by ekj (guest, #1524) [Link]
I'll see if I can dig up the bug-ids, but at the moment I'm in vacation until april 10th, will look at it when I come home.
|
|||||||||||||