One thing I notice about most other open source projects is that they have many more flamewars than we do. So, I thought I'd start one:
It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum.
So are we misleading users by having a bugzilla into thinking that if they file a bug there, it'll be fixed when it probably won't?
If so, does it matter?
If we were to simply drop bugzilla, how would it impact the project?
Thoughts, anybody?
thanks -mike
From the standpoint of an near-ordinary* user (watching Boog 1882),
yes. From the standpoint of developers I'd have thought so too.
After all, if you don't have a bug database, how do you know what bugs have been identified?
Bugzilla is (or could be) a valuable place. It acts as a central database and repository of information relevant to issues with WINE. (That can include development as well as bug-fixes). That includes relevant links and trace files. A well-used Bugzilla would be better for searching (although spelling application names correctly would help (see 1882) ).
The newsgroups and IRC channels are more convenient fora for discussions but then the wheat has to be stored somewhere and Bugzilla is one of the places it could be stored.
And I wasn't misled, although I am slightly disappointed. :P
Mike Hearn wrote:
One thing I notice about most other open source projects is that they have many more flamewars than we do. So, I thought I'd start one:
It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum.
So are we misleading users by having a bugzilla into thinking that if they file a bug there, it'll be fixed when it probably won't?
If so, does it matter?
If we were to simply drop bugzilla, how would it impact the project?
Thoughts, anybody?
thanks -mike
Mike Hearn wrote:
It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum.
IMHO, Bugzilla is usually more complex to use, also if it's a lot more clean than using a mailing list, because here we can add dependencies and resolve bugs. Actually, I don't use it either, probably because I usually forget that there's the Bugzilla :) I started using it to submit ebuilds to gentoo, and then for KDE, before that I never used it. It's also true that also posting on wine-users or wine-devel doesn'tmake sure that the bug will be seen: I posted sometime requests for more info about an error or a behaviour of a software under wine, but it's rare that something is found out. Also, there's still my patch for SHLWAPI floating around, because I hadn't had the time to clean it up, but noone answered in the topic AFAICS ( http://thread.gmane.org/gmane.comp.emulators.wine.devel/22665 ).
Maybe a 'promotion' to the use of bugzilla can help address the bigger issues.
Hey,
I'm one of those end-users following development closely. But I rarely post here. I've submitted two bugs in the bugzilla system. One was directly taken up by Lionel and fixed quickly. On the second one I did not get a respond. Here it is (ALSA record problem): http://bugs.winehq.com/show_bug.cgi?id=2046
Mike Hearn wrote:
It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum.
Which leads to question I think. Shouldn't it be used more by developers? Should developers not at least try to mark bugs to some state.
So are we misleading users by having a bugzilla into thinking that if they file a bug there, it'll be fixed when it probably won't?
If so, does it matter?
Yes it does. People have a program that's not working (correctly) and want to submit that somewhere. Or ask for help. If there is nowhere to do that (or there is no reply) you'll simply stop users from using Wine. They'll not be encouraged to look for other problems.
A problem is that most Wine users are coming directly from Windows and are not experienced in writing detailed bug reports. And you'll get the "program x is not working, HEEEELPP!!!!1" 'bug'. That might be the reason developers try to avoid bugzilla. They're tired of this kind of reports.
You could also call Wine not ready for Bugzilla. There are so many bugs that's not needed for users to report bugs. Developers can find them on their own very easy.
If we were to simply drop bugzilla, how would it impact the project?
Well drop it for a period of time, until Wine is more ready for bug reporting. That could be years tough. And I wouldn't like it. It has a bad impact on the project.
I would say, try to have some canned responses for bugs to at least give users the idea someone looked at their bug. And try to fix/assign all bugs that contain usefull information. Some users are able to debug a problem to some extend and then submit a bug to hand over the work they really don't understand.
Kind regards, Paul van Schayck
Well,
I personally think a bugtracking tool (like bugzilla) is needed for any software development project. It helps you collect bugs on a central place and you can keep track on the things that matter.
However, it will only work 'correctly' if both the developers and people that submit bugs work together.
If there is no developer (activily) looking at the submitted bugs, people will complain about it and probably stop submitting and go somewhere else.
If there are no (non developers) submitting bugs, you could still use it to 'implement' features, using the bugtracking tool to keep track of what's happening.
Anyway, I don't know how bugzilla has been used in the past, but if I have a simple look at the 'available' buglist, I see that bug 11 is still marked as NEW (which sounds odd as it was submitted somewhere in 2000)
Maybe a "janitor" project to cleanup the winehq bugzilla is really needed? But you would need some kind of plan to work on the 'backlog'.
You could:
* start on bug 1 going up and checking if it isn't solved yet. Eventually you'll end up at the latest bugs, but that means you'll probably hit a dead end often before you'll reap the benefits out of it.
* start on the last submitted bug going down, because it's most recent found and probably fresh in memory & easily reproduceable. But that might 'annoy' people who have longer outstanding bugs.
* etc..
Doing this might be a good job for developers wanting to start on wine (me?) : trying to reproduce/pinpoint a problem, maybe solve it or hand it over to a developer with enough information for solving the bug?
On the other hand, for now I'm probably just only interested in the bugs I hit into and less interested in other peoples bugs (<sarcastic mode on>after all, I want my program to run correctly, who cares if someone elses program is running</sarcastic mode off>)
Anyway, just some of my thoughts...
Jeroen
Mike Hearn wrote:
One thing I notice about most other open source projects is that they have many more flamewars than we do. So, I thought I'd start one:
It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum.
So are we misleading users by having a bugzilla into thinking that if they file a bug there, it'll be fixed when it probably won't?
If so, does it matter?
If we were to simply drop bugzilla, how would it impact the project?
Thoughts, anybody?
thanks -mike
Mike Hearn wrote:
One thing I notice about most other open source projects is that they have many more flamewars than we do. So, I thought I'd start one:
:)
It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum.
I have been thinking the same thing. The trouble Wine has versus other projects is that useful bug reports are very rare. This isn't really the fault of the people entering bugs into the database, but can be broken down into four factors: 1. Unlike other projects, we rarely have access to the source code to find out what the program is doing. 2. The people entering bugs aren't always developers and don't always have the skills to provide a good bug report. 3. A lot of Wine's code doesn't have a "maintainer". Without someone caring about a particular area or knowledgable about the code there is often no incentive to fix bugs. 4. There are a huge number of possible combinations of Wine's configuration, ranging from Windows version to builtin versus native DLLs to DLLs from different versions of Windows.
So are we misleading users by having a bugzilla into thinking that if they file a bug there, it'll be fixed when it probably won't?
Yes, we are most of the time.
If so, does it matter?
Bugzilla is a useful tool. When it becomes clogged up with bugs that no one is going to fix, this lessens it usefulness.
If we were to simply drop bugzilla, how would it impact the project?
There are many useful bugs/task lists currently in bugzilla and I believe it would harm the project to just drop Bugzilla. However, by changing the usage of bugzilla we could make it more useful again. Perhaps someone needs to go through bugzilla requesting more information and assigning bugs to the correct components, or perhaps we need to put a big message on the front page of it saying something like "only for specific problems - we can't help you if you just say that program X doesn't work"
Rob
On Sat, Jul 24, 2004 at 03:14:04PM +0100, Mike Hearn wrote:
One thing I notice about most other open source projects is that they have many more flamewars than we do. So, I thought I'd start one:
It strikes me, looking at the wine-bugs list, that there is a huge disparity between the number of people maintaining it and the number of people filing bugs in it. It seems to be quite rare for communication on bugfixes to take place there, wine-devel is the more usual forum.
So are we misleading users by having a bugzilla into thinking that if they file a bug there, it'll be fixed when it probably won't?
If so, does it matter?
If we were to simply drop bugzilla, how would it impact the project?
I think it is worth keeping.
Even if there is much cruft, at one time it will get more acceptance.
I took a look in the last hours and there are again some low hanging fruit to clean out.
It will get better over time. Hopefully ;)
Ciao, Marcus
I talked to Dimi, Lionel and Alexandre on IRC about this.
This is what they had to say:
Lionel didn't think it should be scrapped however he proposed a system whereby bugs that didn't have any activity for two releases or more would be automatically closed. If you wanted to keep a bug open users would have to post reports of behaviour with current releases.
Dimi thought it should be kept as well, but thought the UI could be improved a lot: for instance a mail based interface would be good. He also thought a dedicated triage guy would be great. Realistically though that's not likely to happen anytime soon. Alexandre basically agreed with this view, I think.
I don't have a strong opinion either which way ... I used to read wine-bugs back when I worked only as a volunteer but usually people come into IRC with apps they'd like to get working and I tended to do that instead. I still do IRC "app on demand" type bugfixing occasionally :)
thanks -mike
Mike Hearn wrote:
I talked to Dimi, Lionel and Alexandre on IRC about this.
This is what they had to say:
Lionel didn't think it should be scrapped however he proposed a system whereby bugs that didn't have any activity for two releases or more would be automatically closed. If you wanted to keep a bug open users would have to post reports of behaviour with current releases.
http://bugs.winehq.org/show_bug.cgi?id=1236
Summary: Marcus was apparently of the same opinion. I, obviously, am not.
I use bugzilla as a tool to keep track of my "todo" list. If there is a place where I suspect a certain behaviour is broken, I open a bug report (knowing it will only be me who will have to "confirm" it... :-), and leave it "unconfirmed".
I think asking us (wine developers) to move at the same pace Alexandre is releasing is unrealistic. The idea is to make bugzilla more friendly to us, not less.
I see why user-opened bugs could benefit from this policy, but developers opened bugs should probably not. Of all the BiDi related bugs, I think 1235 is the only place where a bug MAY have disappeared "by itself".
Shachar
I see why user-opened bugs could benefit from this policy, but developers opened bugs should probably not. Of all the BiDi related bugs, I think 1235 is the only place where a bug MAY have disappeared "by itself".
Well, when I proposed this solution, I considered this case and had a special Bugzilla state (which could only be used by some 'registered' users) which would let the bug bypass these checks.
Or, even more easy, as soon as a bug is assigned to a known list of users, it would not need this prodding (as, well, we would suppose that Wine developpers would actually care about their assigned bugs :-) ).
Lionel
On Sun, Jul 25, 2004 at 03:02:13PM +0100, Mike Hearn wrote:
[finally back on email, account was busted for a while]
Dimi thought it should be kept as well, but thought the UI could be improved a lot: for instance a mail based interface would be good. He also thought a dedicated triage guy would be great. Realistically though that's not likely to happen anytime soon. Alexandre basically agreed with this view, I think.
There's no reason this shouldn't happen anytime soon. We had a few great guys triaging bugs in Bugzilla. People are looking to help, maybe someone steps up to the task.
-
Colin Wright -
Diego 'Flameeyes' Pettenò -
Dimitrie O. Paun -
Jeroen Janssen -
Lionel Ulmer -
Marcus Meissner -
Mike Hearn -
Paul van Schayck -
Robert Shearman -
Shachar Shemesh